Thesis Van Huyssteen N

A CONSUMER PURCHASE DECISION MODEL
FOR THE SOUTH AFRICAN PERSONAL MOTOR VEHICLE

INSURANCE INDUSTRY
by
NADIA VAN HUYSSTEEN
submitted in accordance with the requirements
for the degree of
DOCTOR OF PHILOSOPHY IN MANAGEMENT STUDIES
at the
UNIVERSITY OF SOUTH AFRICA
Supervisor: Prof. S. Rudansky-Kloppers
January 2022
Open Rubric
DECLARATION
Name: Nadia van Huyssteen
Student number: 5084-289-7
Degree: PhD Management Studies
A consumer purchase decision model for the South African personal motor
vehicle insurance industry.
I declare that the above thesis is my own work and that all the sources that I have
used or quoted have been indicated and acknowledged by means of complete
references.
I further declare that I submitted the thesis to originality checking software and that it
falls within the accepted requirements for originality.
I further declare that I have not previously submitted this work, or part of it, for
examination at Unisa for another qualification or at any other higher education
institution.
_____________________ 24/01/2022
SIGNATURE DATE
(Mrs Nadia van Huyssteen)
ii
ACKNOWLEDGEMENTS
All the honour to God for He gave me the courage and strength to complete my studies
to the best of my ability.
My heartfelt thanks goes to the following people:
My supervisor, Prof. Sharon Rudansky-Kloppers, thank you for your commitment to

my research. Your academic excellence, guidance and continuous encouragement
are highly appreciated.
Andley Wu from the Road Traffic Management Cooperation who tried assisting with
access to their database and referred me to Philip from the Automobile Association.
Thank you for your assistance and advice.
Philip Purnell and his team from the Automobile Association gladly assisted with
access to their database and distributing the survey to their database. Thank you for
your kindness and assistance.
Dr Dion van Zyl for his patience and assistance with the statistical aspects of this
thesis. I have really learned a lot.
Retha Burger for editing my dissertation.
To the love of my life, Arno, thank you for all your patience and encouragement
throughout my studies.
My family and close friends, thank you very much for all the support, patience and love
you have unconditionally provided during these past few years. I am truly blessed to
have such great parents, in-laws, brother, sisters-in-law, brothers-in-law, and friends
who love and believe in me as much as they do. Thank you for the motivating words
and for always showing an interest in my studies.
My colleagues and dearest friends, Adele, Alicia, Amorie, Themari and Petri, thank
you for always listening and giving advice.
iii
ABSTRACT
The low penetration rate of motor vehicle insurance in South Africa places enormous
pressure on the short-term insurance industry. An analysis of individuals in the market
and their purchasing habits will allow short-term insurers to understand their
consumers and purchasing behaviour. This study aimed to develop a model that would
determine the purchase decision process of consumers (insured) and potential
consumers (non-insureds) regarding personal motor vehicle insurance in South Africa.
It is important for short-term insurance companies need to gain an understanding of
purchase behaviour in terms of personal motor vehicle insurance in South Africa, as it
will allow them to satisfy consumers’ needs and influence individuals’ purchase
decisions. Several decision-making models were evaluated to aid in the development
of the conceptual consumer purchase decision for personal motor vehicle insurance
model that illustrates the influences and steps in the purchase decision process.
Quantitative data was collected through an online questionnaire. The study followed a
three-phase model-building strategy of: (1) Exploratory Factor Analysis, (2)
Confirmatory Factor Analysis and (3) logistic regression model building to test and
refine the model until an acceptable model could be established. The study found that
the purchase decision for personal motor vehicle insurance consumers and potential
consumers consists of the following influences and steps: company’s marketing efforts
(include people, process and price of the policy), sociocultural environment factors
(included items such as positive non-commercial sources and negative reference
person or group), communication sources (paid-for social media efforts, buzz agent
and customised messages); psychological attributes (items such as perception about
a certain insurance company and personality), demographic variables (household
income, highest level of education and age), and decision-making process steps (pre-
purchase search requesting quotations and evaluating alternative prices,
quality/standards and brands). The demographic profile of respondents was used to
develop a description of the average personal motor vehicle driver and personal motor
vehicle insurance consumer. This study provides an acceptable model explaining the
purchase decision process of individuals in the personal motor vehicle insurance
market in South Africa.
iv
KEY WORDS:
Purchase behaviour; purchase decision process; external decision-making influences;

short-term insurance company’s marketing efforts; sociocultural environment;
communication sources; psychological attributes; consumer decision-making process;
personal motor vehicle insurance market, South Africa.
v
OPSOMMING
Die lae penetrasiekoers van motorvoertuigversekering in Suid-Afrika plaas enorme

druk op die korttermynversekeringsbedryf. ’n Ontleding van individue in die mark en
hul koopgewoontes sal korttermynversekeraars in staat stel om hul verbruikers se
koopgedrag te verstaan. Die doel van hierdie studie is om ’n model te ontwikkel wat
die koopbesluitnemingsproses van verbruikers (versekerdes) en potensiële
verbruikers (nieversekerdes) rakende persoonlike motorvoertuigversekering in Suid-
Afrika bepaal. Dit is belangrik dat korttermynversekeringsmaatskappye koopgedrag
rakende persoonlike motorvoertuigversekering in Suid-Afrika verstaan, aangesien dit
hulle in staat sal stel om in verbruikers se behoeftes te voorsien en individue se
koopbesluite te beïnvloed. Verskillende besluitnemingsmodelle is geëvalueer om te
help met die ontwikkeling van ’n konsepmodel vir persoonlike
motorvoertuigversekering wat die invloede en stappe in die koopbesluitproses
illustreer. Kwantitatiewe data is deur middel van ’n aanlyn vraelys ingesamel. Die
studie het ’n driefasige modelboustrategie gevolg, bestaande uit: (1) verkennende
faktoranalise, (2) bevestigende faktoranalise en (3) logistiekeregressie-modelbou om
die model te toets en te verfyn totdat ’n aanvaarbare model daargestel kan word. Die
studie het bevind dat die koopbesluit vir verbruikers en potensiële verbruikers van
persoonlike motorvoertuigversekering uit die volgende invloede en stappe bestaan:
maatskappy se bemarkingspogings (insluitend mense, proses en prys van die polis),
sosiaal-kulturele omgewingsfaktore (insluitend items soos positiewe niekommersiële
bronne en negatiewe verwysingspersoon of -groep), kommunikasiebronne (betaalde
pogings op sosiale media, gons-agent- en doelgemaakte boodskappe); sielkundige
kenmerke (items soos persepsie oor ’n bepaalde versekeringsmaatskappy en
persoonlikheid), demografiese veranderlikes (huishoudelike inkomste, hoogste
opvoedingsvlak en ouderdom), en stappe in die besluitnemingsproses
(vooraankoopsoektog wat kwotasies aanvra en alternatiewe pryse evalueer,
gehalte/standaarde en handelsmerke). Die demografiese profiel van respondente is
gebruik om ’n beskrywing van die gemiddelde persoonlike motorvoertuigbestuurder
en verbruiker van persoonlike motorvoertuigversekering te ontwikkel. Hierdie studie
verskaf ’n aanvaarbare model wat die koopbesluitproses van individue in die
persoonlike motorvoertuigversekeringsmark in Suid-Afrika verduidelik.
vi
SLEUTELWOORDE:
Koopgedrag; koopbesluitproses; eksterne besluitnemingsinvloede; korttermyn-

versekeringsmaatskappy se bemarkingspogings; sosiaal-kulturele omgewing;
kommunikasiebronne; sielkundige kenmerke; verbruikersbesluitnemingsproses;
persoonlike motorvoertuigversekeringsmark; Suid-Afrika.
vii
ISIFINQO
Izinga lokungena eliphansi lomshwalense wezimoto eNingizimu Afrika libeka

ingcindezi enkulu embonini yomshwalense wesikhashana. Ukuhlaziywa kwabantu
emakethe kanye nemikhuba yabo yokuthenga kuzovumela abadayisi bomshwalensi
besikhathi esifushane ukuthi baqonde abathengi babo kanye nokuziphatha
kokuthenga. Lolu cwaningo luhlose ukuthuthukisa imodeli ezonquma inqubo yesimo
sokuthenga sabathengi (abanomshwalense) kanye nabathengi abangahle babe nawo
(abangewona) mayelana nomshwalense wemoto yomuntu siqu eNingizimu Afrika.
Kubalulekile ukuthi izinkampani zomshwalense wesikhathi esifushane zithole

ukuqonda mayelana nezindlela zokuziphatha uma umuntu ethengwa umshwalense
wemoto eNingizimu Afrika, njengoba izohambisana nokuthi yanelise izidingo
zabathengi futhi ibe nomthelela ezinqumweni zokuthenga komuntu. Amamodeli
wokuthatha izinqumo amaningana ahlolwa ukuze asize ekuthuthukisweni
kwesinqumo somqondo sokuthenga umthengi semodeli yomshwalense wemoto
yomuntu siqu ebonisa imithelela nezinyathelo zenqubo yesinqumo sokuthenga.
Idatha yobuningi iqoqwe ngohlu lwemibuzo oluku-inthanethi. Ucwaningo lulandele isu

lokwakha imodeli lezigaba ezintathu: (1) Ukuhlaziywa Kwezinto Ezihlolayo, (2)
Ukuhlaziywa Kwezinto Eziqinisekisayo kanye (3) nokwakhiwa kwemodeli yokuhlehla
kwempahla ukuze kuhlolwe futhi ihlolisiswe yini imodeli kuze kusungulwe imodeli
eyamukelekayo. Ucwaningo luthole ukuthi isinqumo somuntu sokuthengwa
komshwalense kanye nabathengi abangase babe ngabathengi siqukethe imithelela

nezinyathelo ezilandelayo: imizamo yokumaketha yenkampani (ihlanganisa abantu,
inqubo kanye nentengo yenqubomgomo), izici zemvelo yezenhlalo kwamasiko
(okufakwe izinto ezifana nezinto ezinhle ezingezona eziqondile) imithombo
yezentengiselwano kanye nomuntu noma iqembu eliyisithenjwa), imithombo
yokuxhumana (ekhokhelwe imizamo yenkundla yezokuxhumana, ibhazi ejenti kanye
nemilayezo engokwezifiso); izici zengqondo (izinto ezinjengombono ngenkampani
ethile yomshwalensi kanye nobuntu), ukuhlukahluka kwabantu (imali engenayo
yasekhaya, izinga eliphezulu lemfundo nobudala), kanye nezinyathelo zenqubo
yokuthatha izinqumo (ukusesha ngaphambi kokuthenga ucela amakhotheshini
nokuhlola ezinye izintengo, ikhwalithi/ amazinga kanye nemikhiqizo). Iphrofayili
yezibalo zabantu abaphendulayo yasetshenziswa ukuthuthukisa incazelo yomshayeli
viii
wemoto yomuntu ophakathi nendawo kanye nomuntu ongumthengi womshwalense
wemoto. Lolu cwaningo luhlinzeka ngemodeli eyamukelekayo echaza inqubo
eyisinqumo sokuthengwa kwabantu ngabanye abasemakethe yomshwalense
wezimoto eNingizimu Afrika.
AMAGAMA ABALULEKILE:
Indlela yokuziphatha kokuthengwayo, inqubo eyisinqumo sokuthenga, amathonya

angaphandle okwenza izinqumo; imizamo yokuthengisa yenkampani yomshwalense
yesikhashana; imvelo yezenhlalo namasiko; imithombo yokuxhumana; izimfanelo
ezingokwengqondo; inqubo yokuthatha izinqumo zabathengi; imakethe
yomshwalense wemoto yomuntu siqu, eNingizimu Afrika.
ix
TABLE OF CONTENTS
Page
DECLARATION ...................................................................................................................... II
ACKNOWLEDGEMENTS ..................................................................................................... III
ABSTRACT ........................................................................................................................... IV
TABLE OF CONTENTS ......................................................................................................... X
LIST OF FIGURES ..............................................................................................................XVI
LIST OF TABLES ................................................................................................................XIX
ABBREVIATIONS AND ACRONYMS ................................................................................XXI
BACKGROUND AND PROBLEM STATEMENT ............................................ 1
INTRODUCTION ..................................................................................................... 1
BACKGROUND TO THE STUDY ........................................................................... 1
LITERATURE REVIEW .......................................................................................... 7
INSURANCE INDUSTRY RESEARCH AND CONTRIBUTION OF THIS
RESEARCH ............................................................................................................ 9
THE RESEARCH PROBLEM ............................................................................... 12
OBJECTIVES OF THE STUDY ............................................................................ 15
Primary research objective .............................................................................15
Secondary research objectives ......................................................................16
RESEARCH DESIGN AND METHODOLOGY ..................................................... 16
Primary research ............................................................................................16
Target population ...........................................................................................17
Means of accessing potential respondents ....................................................17
Sampling methods ..........................................................................................18
Sample size ....................................................................................................19
Data collection ................................................................................................19
Ethical considerations ....................................................................................20
Data processing and analysis.........................................................................20
Research validity and reliability ......................................................................21
LAYOUT OF THE CHAPTERS ............................................................................. 22
SOUTH AFRICAN MOTOR VEHICLE INSURANCE INDUSTRY ................. 25
INTRODUCTION ................................................................................................... 25
GLOBAL INSURANCE HISTORY ........................................................................ 26
HISTORY OF THE SOUTH AFRICAN INSURANCE INDUSTRY ........................ 29
COMPONENTS OF THE SOUTH AFRICAN INSURANCE INDUSTRY .............. 32
Long-term insurance ......................................................................................33
Short-term insurance ......................................................................................35
x
HISTORY AND DEVELOPMENT OF THE GLOBAL MOTOR VEHICLE
INSURANCE INDUSTRY...................................................................................... 42
THE SOUTH AFRICAN MOTOR VEHICLE INSURANCE INDUSTRY ................ 44
The importance of the South African motor vehicle insurance industry .........46
The nature of the motor vehicle insurance product/service ............................46
Different services offered to the motor vehicle insurance market ...................50
THE BUSINESS ENVIRONMENT OF THE SHORT-TERM INSURANCE
COMPANIES......................................................................................................... 52
Micro-environment ..........................................................................................52
Market environment ........................................................................................56
Macro-environment ........................................................................................63
CONCLUSION ...................................................................................................... 77
CONSUMER PURCHASE DECISION-MAKING PROCESS......................... 79
INTRODUCTION ................................................................................................... 79
THE MARKETING FUNCTION OF A SHORT-TERM INSURANCE
COMPANY ............................................................................................................ 80
Components of the marketing function ...........................................................80
CONSUMER BEHAVIOUR ................................................................................... 89
Types of consumers .......................................................................................90
Mental and physical activities of consumers ..................................................91
CONSUMER DECISION-MAKING ....................................................................... 91
Roles in the decision-making process ............................................................93
Decision-making rules ....................................................................................93
Types of decision-making ...............................................................................95
Decision-making styles ...................................................................................96
PURCHASE DECISION-MAKING THEORIES ..................................................... 98
The risk theory ................................................................................................99
The decision theory ........................................................................................99
REVIEW OF CONSUMER DECISION-MAKING MODELS ................................ 104
Andreasen model (1965) ..............................................................................104
Nicosia model (1966) ...................................................................................106
Engel, Kollat and Blackwell model (1968) ....................................................107
Howard and Sheth model (1969) .................................................................109
Perreault and McCarthy model (2002) .........................................................111
Schiffmann, Kanuk and Hansen model (2012) .............................................113
Milner and Rosenstreich model (2013).........................................................114
Schiffman and Wisenblit model (2015 and 2019) .........................................115
CONCLUSION .................................................................................................... 117
A CONSUMER PURCHASE DECISION FOR PERSONAL MOTOR VEHICLE
INSURANCE CONCEPTUAL MODEL ............................................................................... 120
xi
INTRODUCTION ................................................................................................. 120
CONSUMER PURCHASE DECISION MODEL ADAPTED FOR PERSONAL
MOTOR VEHICLE INSURANCE MARKET ........................................................ 120
EXTERNAL INFLUENCES ................................................................................. 122
Marketing efforts ...........................................................................................122
Sociocultural environment ............................................................................124
Communication sources ...............................................................................126
INTERNAL INFLUENCES .................................................................................. 129
Psychological attributes ................................................................................129
Demographics of individuals ........................................................................132
DECISION-MAKING PROCESS ......................................................................... 132
PURCHASE DECISION ...................................................................................... 134
CONCLUSION .................................................................................................... 134
RESEARCH METHODOLOGY .................................................................... 135
INTRODUCTION ................................................................................................. 135
RESEARCH OBJECTIVES REVISITED ............................................................ 135
Primary research objective ...........................................................................136
Secondary research objectives ....................................................................136
THE RESEARCH PROCESS ............................................................................. 137
Steps in the research process ......................................................................139
RESEARCH DESIGN ......................................................................................... 140
RESEARCH TERMINOLOGY ............................................................................ 142
Research paradigm: Ontology, epistemology and axiology .........................142
Deduction and induction ...............................................................................143
Empirical research, primary data and secondary data .................................144
Variables ......................................................................................................145
DESCRIPTORS OF THE OVERALL RESEARCH DESIGN OF THE STUDY ... 146
Research paradigm ......................................................................................147
The degree to which the research questions have been formed..................153
The method of data collection ......................................................................154
The researcher’s control over variables in this study ...................................154
The purpose of the study ..............................................................................155
The time dimension ......................................................................................155
The topical scope .........................................................................................156
The research environment ...........................................................................156
RESEARCH APPROACH ................................................................................... 156
Qualitative research .....................................................................................158
Quantitative research ...................................................................................159
SAMPLING DESIGN ........................................................................................... 160
xii
Population, population element, sample and census defined ......................161
Sampling method .........................................................................................162
Sample size ..................................................................................................172
DATA COLLECTION .......................................................................................... 173
Data-collection instrument ............................................................................174
Data-collection process ................................................................................195
Data-collection errors ...................................................................................195
DATA PROCESSING ......................................................................................... 196
Data capturing ..............................................................................................196
Data coding ..................................................................................................196
Editing ..........................................................................................................197
DATA ANALYSIS ............................................................................................... 197
Descriptive statistics .....................................................................................197
Model fit analysis ..........................................................................................199
RELIABILITY AND VALIDITY ............................................................................ 201
Reliability ......................................................................................................201
Validity ..........................................................................................................202
PRESENTATION OF DATA FINDINGS ............................................................. 204
ETHICAL CONSIDERATIONS ........................................................................... 204
Voluntary participation ..................................................................................205
No harm to respondents ...............................................................................205
Anonymity and confidentiality .......................................................................206
Informed consent ..........................................................................................206
CONCLUSION .................................................................................................... 209
DESCRIPTIVE FINDINGS OF RESEARCH ................................................ 210
INTRODUCTION ................................................................................................. 210
DESCRIPTIVE ANALYSIS OF RESEARCH FINDINGS .................................... 211
DEMOGRAPHIC COMPOSITION OF THE RESPONDENTS ............................ 212
Age ...............................................................................................................212
Population group ..........................................................................................213
Gender .........................................................................................................215
Marital status ................................................................................................216
Highest level of education ............................................................................217
Employment status .......................................................................................219
Province mostly travelled in with personal motor vehicle .............................220
Household income ........................................................................................222
Personal income ...........................................................................................223
Consumer demographic profile ....................................................................224
Respondent demographic profile .................................................................226
xiii
GENERAL PERSONAL MOTOR VEHICLE INSURANCE QUESTIONS .......... 227
Currently have/do not have personal motor vehicle insurance.....................228
Type of personal motor vehicle insurance ....................................................229
Payment party ..............................................................................................230
Reasons for having personal motor vehicle insurance .................................231
Reasons for not having personal motor vehicle insurance ...........................232
INPUT STAGE: EXTERNAL INFLUENCES ....................................................... 234
The company’s marketing efforts .................................................................235
The sociocultural environment .....................................................................244
Communication sources ...............................................................................255
PROCESS STAGE: INTERNAL INFLUENCES ................................................. 263
Psychological attributes: Motivation to purchase personal motor vehicle
insurance ......................................................................................................263
Psychological attributes: Perception about motor vehicle insurance ............265
Psychological attributes: Perception about a certain insurance company ....266
Psychological attributes: Learning ................................................................267
Psychological attributes: Personality ............................................................268
Psychological attributes: Attitude towards motor vehicle insurance .............270
Psychological attributes: Attitude towards a certain short-term insurance
company .......................................................................................................271
PROCESS STAGE: DECISION-MAKING PROCESS ....................................... 272
Decision-making process (Pre-purchase search): Previous experiences ....272
Decision-making process (Pre-purchase search): Request quotations from
more than one short-term insurance company .............................................273
Decision-making process (Evaluation of alternatives): Compare the price of the
different quotations .......................................................................................275
Decision-making process (Evaluation of alternatives): Compare the
quality/standard of the different quotations...................................................276
Decision-making process (Evaluation of alternatives): Compare the brand list
.....................................................................................................................277
CONCLUSION .................................................................................................... 279
MODEL FIT ANALYSIS ............................................................................... 280
INTRODUCTION ................................................................................................. 280
PHASE 1: EXPLORATORY FACTOR ANALYSIS (EFA) WITH
EXPLORATION SAMPLE (N=200) .................................................................... 284
PHASE 2: CONFIRMATORY FACTOR ANALYSIS (CFA) WITH
VALIDATION SAMPLE (N=478) ........................................................................ 287
Model specification .......................................................................................287
Model identification .......................................................................................290
Model estimation and testing ........................................................................291
Model modification .......................................................................................301
xiv
Interpretation of values in CFA revised CPDFPMVI model ..........................307
PHASE 3: LOGISTIC REGRESSION MODEL-BUILDING STRATEGY ............ 310
Variable selection with n=200 exploration sample .......................................310
Logistic regression estimation with n=478 validation sample .......................315
Check the assumption of linearity.................................................................321
Assessing the overall fit of final logistic regression model ............................322
Classification and validation of the final CPDFPMVI model .........................325
CONCLUSION .................................................................................................... 333
CONCLUSIONS AND RECOMMENDATIONS............................................ 334
INTRODUCTION ................................................................................................. 334
OVERVIEW OF RESEARCH AIMS AND OBJECTIVES ................................... 334
REFLECTIONS AND CONCLUSIONS: ADDRESSING THE RESEARCH
OBJECTIVES...................................................................................................... 336
The company’s marketing efforts .................................................................336
The sociocultural environment .....................................................................338
The communication sources.........................................................................340
The psychological attributes .........................................................................341
The decision-making process .......................................................................343
The demographic variables ..........................................................................345
Demographic profile of personal motor vehicle insurance consumers .........346
Demographic profile of personal motor vehicle drivers ................................347
The consumer purchase decision for personal motor vehicle insurance
(CPDFPMVI) model ......................................................................................349
ADDRESSING THE RESEARCH QUESTION ................................................... 352
CONTRIBUTION TO THE BODY OF KNOWLEDGE ........................................ 353
RECOMMENDATIONS FOR COMPANIES IN THE SHORT-TERM
INSURANCE INDUSTRY.................................................................................... 355
RECOMMENDATIONS FOR FUTURE RESEARCH ......................................... 356
LIMITATIONS OF THE STUDY .......................................................................... 357
RESEARCH CONCLUSION ............................................................................... 359
LIST OF REFERENCES ..................................................................................................... 360
APPENDIX A - DATA COLLECTION INSTRUMENT - ...................................................... 414
APPENDIX B - ETHICAL CLEARANCE CERTIFICATE - ................................................. 428
APPENDIX C - FREQUENCY TABLES AND DESCRIPTIVE STATISTICS - ................... 431
APPENDIX D - TABLES AND INFERENTIAL STATISTICS - ........................................... 456
APPENDIX E - PROFESSIONAL EDITING CERTIFICATE - ............................................ 465
xv
LIST OF FIGURES
Page
Figure 1.1: Layout of the thesis .............................................................................................22

Figure 3.1: Andreasen model ..............................................................................................105
Figure 3.2: Nicosia model ....................................................................................................107
Figure 3.3: Engel, Kollat and Blackwell model.....................................................................108
Figure 3.4: Howard and Sheth model ..................................................................................110
Figure 3.5: Perreault and McCarthy model ..........................................................................112
Figure 3.6: Schiffman, Kanuk and Hansen model ...............................................................113
Figure 3.7: Milner and Rosenstreich model .........................................................................114
Figure 3.8: Schiffman and Wisenblit model .........................................................................116
Figure 4.1: A conceptual CPDFPMVI model .......................................................................121
Figure 5.1: The research process ........................................................................................138
Figure 5.2: Components of the research design .................................................................142
Figure 5.3: Link between research paradigms, research design and research methods ....148
Figure 5.4: Population, sample and population element or case .........................................161
Figure 5.5: Selecting a non-probability sampling technique ................................................165
Figure 5.6: Email distributed to the AA database ................................................................167
Figure 5.7: Message distributed on Facebook advertisement marketing tool .....................168
Figure 5.8: Facebook post using snowball sampling ...........................................................170
Figure 5.9: Facebook groups post using snowball sampling ...............................................171
Figure 5.10: Scale development steps ................................................................................178
Figure 5.11: Pilot test process .............................................................................................180
Figure 5.12: Model-building strategy to achieve CPDFPMVI model ....................................199
Figure 5.13: Ethical matters during different stages of the research process ......................208
Figure 6.1: Age (n=678).......................................................................................................213
Figure 6.2: Population group (n=662) ..................................................................................214
Figure 6.3: Gender (n=678) .................................................................................................215
Figure 6.4: Marital status (n=678)........................................................................................216
Figure 6.5: Highest level of education (n=676) ....................................................................217
Figure 6.6: Employment status (n=678) ..............................................................................219
Figure 6.7: Province mostly travelled in with the personal motor vehicle (n=678) ...............220
Figure 6.8: Household income (n=678) ...............................................................................222
Figure 6.9: Personal income (n=678) ..................................................................................223
Figure 6.10: Currently have/do not have personal motor vehicle insurance (n=678) ..........228
Figure 6.11: Type of personal motor vehicle insurance cover (n=379) ................................229
xvi
Figure 6.12: Payment party (n=379) ....................................................................................230
Figure 6.13: Reasons for having personal motor vehicle insurance (n=379) ......................232
Figure 6.14: Reasons for not having personal motor vehicle insurance (n=299) ................233
Figure 6.15: Type of personal motor vehicle insurance offered (n=678) .............................235
Figure 6.16: Advertisements or promotional activities performed (n=678) ..........................236
Figure 6.17: Price of policy (n=678) .....................................................................................237
Figure 6.18: How personal motor vehicle insurance can be purchased (n=678) .................238
Figure 6.19: The company’s physical evidence efforts (n=678) ..........................................240
Figure 6.20: The duration of the process to get a quotation (n=678)...................................241
Figure 6.21: How the employees in the company interact with consumers (n=678) ...........242
Figure 6.22: Partnerships or associations with other companies that you trust (n=678) .....243
Figure 6.23: A positive comment or opinion of a family member (n=678) ...........................245
Figure 6.24: A negative comment or opinion of a family member (n=678) ..........................246
Figure 6.25: The positive opinion of a reference person or group used as a basis of
comparison (n=678).........................................................................................247
Figure 6.26: The negative opinion of a reference person or group used as a basis of
comparison (n=678).........................................................................................248
Figure 6.27: Positive comments of a friend, a newspaper article or the views of an expert on
a blog or other internet platforms (n=678) .......................................................250
Figure 6.28: Negative comments of a friend, a newspaper article or the views of an expert on
a blog or other internet platforms (n=678) .......................................................251
Figure 6.29: The positive opinions of individuals in the same social class (n=678) .............252
Figure 6.30: The negative opinions of individuals in the same social class (n=678) ...........253
Figure 6.31: Cultural and subcultural factors (n=678) ..........................................................255
Figure 6.32: Traditional advertisements about motor vehicle insurance (n=678) ................256
Figure 6.33: Online advertisements about motor vehicle insurance (n=678) ......................257
Figure 6.34: Buzz agent (n=678) .........................................................................................258
Figure 6.35: Customised messages that insurance companies communicate (n=678) ......260
Figure 6.36: Paid-for social media efforts (n=678)...............................................................261
Figure 6.37: Word-of-mouth, advice, recommendations on social media posts about motor
vehicle insurance (n=678) ...............................................................................262
Figure 6.38: Motivation to purchase personal motor vehicle insurance (n=678) .................264
Figure 6.39: Perception about motor vehicle insurance (n=678) .........................................265
Figure 6.40: Perception about a certain insurance company (n=678) .................................266
Figure 6.41: Learning (n=678) .............................................................................................267
Figure 6.42: Personality (n=678) .........................................................................................269
Figure 6.43: Attitude towards motor vehicle insurance (n=678) ..........................................270
Figure 6.44: Attitude towards a certain insurance company (n=678) ..................................271
Figure 6.45: Pre-purchase search - Previous experiences (n=678) ....................................273
xvii
Figure 6.46: Pre-purchase search - Request quotations from more than one short-term
insurance company (n=678) ............................................................................274
Figure 6.47: Evaluation of alternatives - Compare the price of the different quotations (n=678)
.........................................................................................................................275
Figure 6.48: Evaluation of alternatives - Compare the quality of the different quotations
(n=678) ............................................................................................................276
Figure 6.49: Evaluation of alternatives - Compare brand list (n=678) .................................278
Figure 7.1: Model-building strategy to achieve CPDFPMVI model ......................................281
Figure 7.2: Conceptual CPDFPMVI model ..........................................................................283
Figure 7.3: CFA baseline model for CPDFPMVI .................................................................295
Figure 7.4: CFA revised model for CPDFPMVI ...................................................................302
Figure 7.5: Revised CPDFPMVI model after CFA model fit analysis ..................................306
Figure 7.6: Final CPDFPMVI model ....................................................................................328
Figure 8.1: Final CPDFPMVI model ....................................................................................351
xviii
LIST OF TABLES
Page
Table 3.1: Constructs and items in conceptual CPDFPMVI model obtained from reviewed
models .............................................................................................................118
Table 5.1: Research design descriptors .............................................................................147
Table 5.2: Comparison between research paradigms ........................................................152
Table 5.3: Key associations of qualitative, quantitative and mixed method approaches ....157
Table 5.4: Main attributes of qualitative research ...............................................................158
Table 5.5: Main attributes of quantitative research ............................................................160
Table 5.6: Attributes and methods of main sampling methods...........................................163
Table 5.7: Research objectives and self-administered questionnaire questions matrix .....183
Table 6.1: Primary and secondary research objectives......................................................210
Table 6.2: Consumer demographic profile according to this study.....................................225
Table 6.3: A description of the average personal motor vehicle driver...............................226
Table 7.1: Primary and secondary research objectives......................................................280
Table 7.2: Factor findings from EFA ...................................................................................285
Table 7.3: Model specification gained from EFA ................................................................288
Table 7.4: Measurements and thresholds to determine model fit.......................................292
Table 7.5: Model fit statistics for CFA baseline CPDFPMVI model ....................................298
Table 7.6: Reliability and validity of CFA baseline CPDFPMVI model ...............................299
Table 7.7: Model fit for CFA revised CPDFPMVI model.....................................................304
Table 7.8: Reliability and validity of the CFA revised CPDFPMVI model ...........................305
Table 7.9: Index values of the observed variables that represent each construct/latent
variable ............................................................................................................308
Table 7.10: Univariate analysis summarised to identify the important covariances .............313
Table 7.11: Omnibus tests of Model Coefficients for baseline logistic regression model .....316
Table 7.12: Cox & Snell R-Square and the Nagelkerke R-Square values for baseline logistic
regression model .............................................................................................316
Table 7.13: Hosmer and Lemeshow Test for Baseline logistic regression model ................317
Table 7.14: Variables in the Equation for baseline logistic regression model ......................318
Table 7.15: Collinearity Statistics for final logistic regression variables ...............................320
Table 7.16: Variables in the equation to test linearity assumption .......................................321
Table 7.17: Omnibus Tests of Model Coefficients for final logistic regression model ..........322
Table 7.18: Cox & Snell R-Square and the Nagelkerke R-Square values for final logistic
regression model .............................................................................................323
Table 7.19: Hosmer and Lemeshow Test for Baseline logistic regression model ................323
Table 7.20: Variables in the Equation for final logistic regression model .............................324
xix
Table 7.21: Classification table of final CPDFPMVI model...................................................326
Table 7.22: Outcome of research objectives following logistic regression ...........................329
Table 8.1: Reflections and conclusions for Objective 1 ......................................................337
Table 8.9: The CPDFPMVI model ......................................................................................349
Table 8.10: Items that were discarded from the final CPDFPMVI model .............................352
xx
ABBREVIATIONS AND ACRONYMS
AA Automobile Association
AI Artificial Intelligence
AVE Average variance extracted
CEO Chief executive officer
CFA Confirmatory Factor Analysis
CFI Comparative Fit Index
CMIN Normed Chi-Squared/df
CPDFPMVI Consumer purchase decision for personal motor vehicle

insurance
CR Composite Reliability
ECU Electric control unit
EFA Exploratory Factor Analysis
FIA Financial Intermediaries Association of Southern Africa
FPI Financial Planning Institute of South Africa
FSB Financial Services Board
FSCA Financial Sector Conduct Authority
GDP Gross domestic product
GFI Goodness-of-fit
IFRS International Financial Reporting Standards
ILASA Institute of Loss Adjusters of Southern Africa
INSETA Insurance Sector Education and Training Authority
IoT Internet of Things
IRMSA Institute of Risk Management South Africa
IT Integrated technology
KMO Kaiser-Meyer-Olkin
LDV Light delivery vehicle
xxi
MLE Maximum Likelihood Estimation
MSV Maximum Shared Variance
NT National Treasury
NPV Negative predictive value
OEM Original equipment manufacturer
PPV Positive predictive value
RMSEA Root Mean Square Error of Approximation
RTMC Road Traffic Management Corporation
SAIA South African Insurance Association
SAICB South African Insurance Crime Bureau
Sasria South African Special Risks Insurance Association
SAUMA South African Underwriting Managers Association
SEM Structural equational modelling
SEU Subjective expected utility
SQRT Square root
TLI Tucker Lewis Index
US United States of America
xxii
BACKGROUND AND PROBLEM STATEMENT
INTRODUCTION
The purpose of this chapter is to provide an overview of and a background to the study.
The short background discussion and the discussion of the contribution of this
research will explain why this study was conducted. This chapter also includes a brief
review of the relevant literature, presents the research problem and the objectives of
the study, and discusses the research design and methodology that were used to
attain the research objectives. In conclusion, the layout of the chapters is presented
as a framework for the study.
BACKGROUND TO THE STUDY
During 2018 and 2019, the global insurance industry witnessed moderate growth of
4.9% (Binder, Klais & Mußhoff, 2021). However, due to the COVID-19 pandemic, the
gross domestic product (GDP) weakened in many countries (OECD, 2020:26), and
the low interest rates globally had a large impact on many industries, especially the
life insurance industry. Due to the low interest rates, the significant burden on the
capital, liquidity and investment strategies have forced the life insurance industry to
take difficult decisions to balance operational excellence and profitability (Santenac et
al., 2021:26). Furthermore, insurance companies will have to increase premium rates
due to the negative impact of the COVID-19 pandemic on the premiums. In the
personal insurance line specifically, it has been predicted that consumers suffering
economic difficulties caused by the rising unemployment and decreasing disposable
income will shift to purchasing only mandatory insurance products or reducing the
coverage of their existing policies (Binder et al., 2021).
In terms of motor vehicle insurance, the restrictions implemented in countries globally

due to the COVID-19 pandemic caused a drop in mileage on motorways and freeways,
which decreased the risk of consumers becoming involved in accidents (Goldby,
2020:9). Therefore, the strict lockdown measures implemented globally helped
maintain a satisfactory performance for the global motor vehicle insurance industry
due to a decline in claims (S&P Global, 2021).
1
The global insurance industry has a big role to play in the growth of African and South
African insurance companies (KPMG, 2015a:69). The two key contributing factors to
the successful growth of the African insurance market is understanding the changing
consumer demographics and utilising technology. The use of technology such as
telematics devices, online insurance platforms that allow consumers to manage
insurance profiles automatically, online self-inspection offerings for vehicles, and new
features on mobile applications can create value to the market (KPMG, 2015:84;
2017:80).
The change in consumer demographics reveals a growing affluent middle class, an

increase in the working population, a rise in education and financial literacy levels,
which are all beneficial to the insurance market (PwC, 2018:44). Due to this change in
consumer demographics, insurance companies have had to adapt their customer
engagement models (KPMG, 2017:46).
According to Danckwerts, Naran and Swana (2020:3-4), the COVID-19 pandemic has
impacted the South African insurance industry in various ways, such as:
• Employees had to work from home, which increased the associated IT costs, and
impacted the distribution channels.
• Insurance companies reduced premiums and flexibility in premium payment dates

was offered to consumers.
• The technological shortcomings were again highlighted as insurance companies

had to explore the necessary digital capabilities, enhancements and
transformations to operate with minimal disruption.
• The country’s downgrade to BB+ status and low interest rate forced insurance
companies to relook their financial and strategic planning.
Life insurance is particularly dominant within the South African insurance industry as
a large portion of all premiums written are for life insurance products (KPMG, 2015:71;
KPMG, 2019:79). The two different insurance sectors in the South African insurance
industry are long-term insurance and short-term insurance (PwC, 2015). Motor vehicle
insurance makes up the biggest part of the short-term insurance industry, as written
motor net premiums typically account for 43% of all the premiums written by short-
term insurance companies in the industry (Danckwerts et al., 2020:44). The short-term
2
insurance market environment in South Africa is dominated by the four largest
underwriters, namely, Santam, Hollard, Mutual & Federal (re-branded to Old Mutual)
and Guardrisk. These four companies underwrite 66% of the total gross written
premiums in the South African insurance market (Danckwerts et al., 2020:98).
The three types of motor vehicle insurance products offered by short-term insurers
are: comprehensive insurance, third-party, fire and theft insurance, and third party only
cover (Zingwevu & Sibindi, 2014:658). Motor vehicle insurance applies to cars, motor
vehicles, station wagons, 4x4 vehicles. 4x2 vehicles, commercial vehicles, light
delivery vehicles (LDVs), caravans, motorised caravans, trailers, and motorcycles
(Santam, 2014; Van der Walt, 2016).
From 1942 to 1997, it was compulsory for motorists to have motor vehicle insurance,
which mainly covered damage to the motor and bodily injury or death (Zingwevu &
Sibindi, 2014:658). The Road Accident Fund (RAF) that was established in 1997,
started operations on 1 May 1997 (Road Accident Fund, 2021). The RAF replaced
compulsory motor vehicle insurance. However, the RAF covers the bodily injury or
death of third parties only (Zingwevu & Sibindi, 2014:658). The financial challenges
that the RAF has had for decades are continuing to put a strain on their ability to
compensate and offer rehabilitation to road accident victims and their families (Road
Accident Fund, 2016:15).
Insurance companies in South Africa have called on the government to reintroduce

compulsory third-party motor vehicle insurance, as it can reduce costs for motorists
and short-term insurance companies (Bonorchis, 2014; Geldenhuys, 2020a).
Zingwevu and Sibindi (2014:668) demonstrated the urgent need to reintroduce third-
party motor vehicle insurance in South Africa by highlighting the excessive cost of
accidents, the relatively low percentage of insured vehicles on South African roads,
and how this has increased the price of motor vehicle insurance.
One of the reasons South African motorists do not have insurance is the poor
economic conditions in South Africa. Various authors have supported this assertion by
mentioning that affordability is one of the main reasons motorists in South Africa do
not purchase motor vehicle insurance (Auto & General Insurance, 2015; Mokhothu,
2020). Consumers are under severe financial pressure due to the high unemployment
rate and rising costs of living (such as the increase in petrol prices), and they thus
3
cannot afford motor vehicle insurance (KPMG, 2016:95; Van Rensburg, 2018).
However, in South Africa, another reason can be that motorists feel they do not have
to pay for motor vehicle insurance, as they feel that the RAF will assist with personal
injuries and the rehabilitation of road accident victims and families. Furthermore, South
African motorists do not have motor vehicle insurance because they feel it is not worth
it as they are driving motor vehicles that are too old (North, 2021).
According to Chitiga-Mabugu et al. (2021:92), the COVID-19 pandemic and the rising
unemployment rate in the country have severely impacted consumers’ disposable
income. As a result, consumers buy what they need and not what they want (Phillip,
2020). This also applies to the United States of America (US), where expensive
insurance premiums are the main reason why motorists do not purchase motor vehicle
insurance (Vohra, 2019).
As mentioned above, it is also not compulsory to have motor vehicle insurance/cover,

as it is no longer a regulatory requirement. However, in other countries, liability
insurance is a requirement for all motor vehicles. For example, Dutch law specifies
that all motor vehicles should have liability insurance that covers the damage caused
to other drivers and their cars (Zavadil, 2014:868; Van der Kolk, 2021).
Since many insurance companies offer similar motor vehicle insurance products
(Segovia-Gonzalez, Guerrero & Herranz, 2009:278), consumers and potential
consumers have to carefully decide which specific products to buy from which
companies. Motor vehicle insurance can be seen as a grudge purchase (this refers to
products or services that consumers buy because they have to, although they would
rather spend their money on something else) (McDonnell & Bartlett, 2009:69; Pearson,
2016a:2), since the motorist that has insurance pays the premiums each month to
have peace of mind. Should the motor vehicle be involved in an accident, the
insurance company will pay for damage to the motor vehicle (Loader, Goold &
Thumala, 2015:862).
Not only is this buying decision difficult to make due to all the available insurance
products offered by different companies, there are also certain factors that can
influence the individuals’ decisions. For example, the insurance companies’ marketing
efforts and the individual’s sociocultural environment are external factors that can
4
influence the individual’s purchase decision (Schiffman & Wisenblit, 2015:340,
2019:375; Acevedo, 2018:8).
Individuals’ purchase decisions are driven by their needs, and they assess and select
actions that will fulfil their needs. Consumer decision-making, also known as the
purchase decision-making process, is a cognitive process involving the mental
activities that determine what actions the individuals will undertake to eliminate a
tension state caused by a need (Parumasur & Roberts-Lombard, 2014:263; De Mooij,
2019:26). Furthermore, the consumer decision-making process can be explained as
an outline of how individuals make decisions. Not all the individuals who follow these
decision-making process steps become consumers, as they might abandon the whole
idea at any step (Lappeman et al., 2021:154). However, all the individuals involved in
the purchasing process form part of the market. Understanding the purchase
behaviour within a market is thus a difficult but vitally important task for any short-term
insurance business.
Although everyone’s needs, demands, and preferences are unique, many similar
behaviour patterns are evident, indicating that some individuals follow more or less the
same process when it comes to making buying decisions. This process is known as
the consumer decision-making process (Ragunathan, 2016:182). However, the
decision-making periods differ, as some individuals make their decisions quickly and
easily, while others take longer and consider certain factors in more detail before
making the final decision. The speed and ease of a decision to purchase or not to
purchase depend mainly on the type of service, the buying situation and the purchase
involvement (Jooste, 2014:66; Techasurin, Nuangjamnong & Dowpiset, 2020:2).
In terms of the purchasing of motor vehicle insurance, there are broadly two ways in
which individuals can purchase insurance: the traditional third-party method via an
agent or broker, and the direct transformative methods where the insurance
companies are contacted directly via telephone, email, or the companies’ websites
(Stokes, 2012a; Kaesler et al., 2020).
Consumer decision-making also differs according to the type of purchase decision.

The type of purchase decision can be classified according to the degree of buyer
involvement and the degree of differences among services (Lamb, Hair & McDaniel,
2018:403). Individuals that review or consider purchasing motor vehicle insurance will
5
engage in dissonance-reducing behaviour, as the primary cause of dissonance is the
fact that individuals must make a decision between numerous similar alternatives
(Belch & Belch, 2015:125). During this type of behaviour, individuals are highly
involved in the purchase, but they are of the opinion that there are only minor
differences between the insurance products offered by the service brands or short-
term insurance companies (Jooste, 2014:66; Techasurin et al., 2020:2). In other
words, individuals that want to purchase motor vehicle insurance will typically shop
around to see what is currently available in the market, but they will purchase without
thinking about the decision for too long.
Within the short-term insurance market, the purchase decision by the individuals in the
industry can either be to purchase motor vehicle insurance (and be insured) or not to
purchase motor vehicle insurance (and be non-insured). In South Africa, only 30% to
35% of consumers purchase motor vehicle insurance (Buys, 2015:23; Bubear, 2016;
Geldenhuys, 2020a; South African Insurance Association, 2021). This low market
penetration makes it challenging for insurance companies to keep the premiums low
and affordable for current motor vehicle insurers (Fichardt, 2015). The increase in
consumer living costs forces consumers to shop around for the most affordable
premiums (Innovation Group, 2021). Especially at the present moment, during the
COVID-19 pandemic, consumers are moving to value-based purchasing, where
consumers look at the premium cost versus the benefit in much more detail than before
(Moodley, 2021).
The current study aimed to determine the purchase decision process of the insureds
and non-insureds as related to personal motor vehicle insurance by testing a
conceptual model of the consumer purchase decision for personal motor vehicle
insurance (abbreviated as CPDFPMVI). Personal motor vehicle insurance was chosen
for the study because consumers have a choice in the decision whether to purchase
or not to purchase motor vehicle insurance, and whether the motor vehicle is used for
personal or business purposes. It is also important to point out that this study only
determined the purchase decision process of personal motor vehicle insurance and
did not investigate other types of insurance for vehicles such as buses, trucks, or
motorcycles. Despite the researcher in the current study’s best efforts, no similar
studies could be found where unique consumer purchase decision models with regard
6
to individuals in the personal motor vehicle insurance industry of South Africa and
internationally had been developed.
LITERATURE REVIEW
Consumer behaviour can be defined as the process and activities that individuals
participate in when searching for, selecting, purchasing, using, evaluating and
disposing of products or services to satisfy their needs and desires (Belch & Belch,
2015:110; De Mooij, 2019:23-24). A market/audience can be described as a large
number of individuals who have a need for a specific product/service, have the money
to purchase the product/service, are willing to spend the money on it, and are legally
able to buy the product (Erasmus, Strydom & Rudansky-Kloppers, 2019:459).
Considering this description, the personal motor vehicle insurance market will include
individuals who have personal motor vehicle insurance (insureds) and individuals who
do not have personal motor vehicle insurance (non-insureds). However, an individual
can form part of the market but not be a consumer of a product or service (Lappeman
et al., 2021:5).
In order to adapt their marketing strategies to meet the needs of consumers, a

business has to know and understand why individuals in the market behave the way
they do when deciding to purchase or not to purchase products or services (Erasmus
et al., 2019:453). Short-term insurance companies are businesses that offer an
insurance service with an obligation to exchange consumer’s risk (loss through
damage or theft of vehicle) for the financial responsibility provided by the short-term
insurers (Ragunathan, 2016:182). Therefore, it is important for short-term insurance
companies to understand how and why individuals act when they make decisions to
buy or not to buy motor vehicle insurance, as this can assist them in adapting their
current marketing strategies to meet the needs and desires of individuals in their
market.
Various authors have discussed the consumer decision-making processes (Kotler &
Keller, 2012:166; Belch & Belch, 2015:111; Lamb et al., 2017:90; Acevedo, 2018;
Erasmus et al., 2019:454). All the above-mentioned scholars have presented similar
consumer decision-making processes that consist of five steps, namely, need
recognition, information search, evaluation of alternatives, purchase decision, and
post-purchase behaviour. This process is called the consumer decision-making
7
process, and most individuals in a market that have a need for a certain product or
service will typically follow these steps.
The purchase decision step of the process, where the individual makes the purchase
or not, will determine if the individual becomes a consumer or not (Lamb et al.,
2018:112). Potential consumers refer to the individuals who follow the steps in the
purchase decision process but do not make the purchase, whereas consumers are
the individuals who decide to purchase. The current study focused on the thought
processes that lead individuals from identifying a need, generating options, and
choosing to purchase or not to purchase personal motor vehicle insurance until the
purchase decision step. This study thus focused only on the first four steps, until the
purchase decision step, as this study aimed to develop a model that explains the
purchase decision process of consumers and potential consumers in the personal
motor vehicle insurance industry.
A wide range of external and internal factors influence individuals' purchase decisions
(Acevedo, 2018:19). External factors relate to the external stimuli such as the
company’s marketing efforts, sociocultural environment factors and communication
sources (Stankevich, 2017:10). The company’s marketing efforts comprise of the
product/service itself (the policies offered at short-term insurance companies), its
price, its promotion, and where it is sold (Lantos, 2011:9; Stankevich, 2017:10). The
sociocultural environment factors that influence the need to purchase are family,
reference groups (friends and neighbours), other informal and non-commercial
sources, social class, and culture and subcultural memberships (Adams & Lawrence,
2019). The cumulative impact of every company’s marketing efforts, the influence of
family, friends and neighbours and the existing code of behaviour in society are all
external inputs that are likely to affect what the consumers buy and how they use what
is bought (Erasmus et al., 2019:454). The communication sources indicate the
mechanisms that deliver the marketing mix and sociocultural influences to the
individuals (Schiffman & Wisenblit, 2015:340).
Internal factors that influence the individuals’ purchase decision refer to the personal
and interpersonal aspects that are processed in the individual's mind, and which
influence the purchase behaviour (Acevedo, 2018:19; Cunningham, 2018). The
psychological factors that are inherent to each individual are motivation, perception,
learning, personality and attitudes (Schiffman et al., 2012:14; Schiffman & Wisenblit,
8
2015:340), and these factors affect how the external inputs influence the individual’s
recognition of a need (Lamb et al., 2018:95). As such, need recognition is the first step
in the decision-making process.
The decision-making steps demonstrate how individuals make decisions while

purchasing a product/service (Schiffman & Wisenblit, 2015:341; Solomon et al., 2019).
As stated, an individual’s need recognition is triggered when a person is exposed to
either internal or external stimuli (Lamb et al., 2018:95). The above-mentioned internal
and external factors act as a stimulus in triggering the need recognition. The pre-
purchase search is where the individual searches for information that supports a
specific purchase decision (Hoyer et al., 2016:189). During the ‘evaluation of
alternatives’ step, the individual evaluates the information gathered (McIntee,
2014:67). This relates to how individuals process the information obtained in the pre-
purchase search to choose between the alternative brands (Kotler & Armstrong,
2018:177). The individual's purchase decision (to purchase or not to purchase) is the
result, and therefore, the final step of the actual consumer decision-making process
(Milner & Rosenstreich, 2013:115; Schiffman & Wisenblit, 2015:341; Solomon et al.,
2019).
A unique consumer purchase decision model for personal motor vehicle insurance can
assist short-term insurance companies in South Africa in gaining an understanding of
the purchase behaviour of their market (insureds and non-insureds). The literature
review for the current study was used as a framework, together with other consumer
decision-making models, to develop a detailed conceptual model, which explains the
individuals’ purchase decision process within the personal motor vehicle insurance
market. However, as stated before, and discussed in the next section, no evidence
could be found that such a model had previously been developed for the consumer
purchase decision process in the personal motor vehicle insurance industry.
INSURANCE INDUSTRY RESEARCH AND CONTRIBUTION OF

THIS RESEARCH
Research has been conducted on various topics in the non-life insurance industry, as
discussed in this section.
9
The risk aspects of the non-life insurance sectors in the international insurance market
have been extensively researched (Bae, Kim & Kulperger, 2009; Segovia-Gonzalez
et al., 2009; Onafalujo, Abass & Dansu, 2011; Diers, Eling & Marek, 2012; Kanno,
2016; Baione, Biancalana & De Angelis, 2020; Eling & Jung, 2020; Upreti, Adams &
Jia, 2021).
Research has been conducted on consumer behaviour in the international insurance

market (Kaže, 2010; Souiden & Jabeur, 2015; Salleh et al., 2018).
An international study investigated the impact of financial factors on purchase

decision-making within the insurance market, and this study revealed that the
Romanian motor vehicle insurance market is the most price sensitive one (Naghi,
2015). An international study on how price presentation influences consumer choices
with regard to life insurance products, found that neither price bundling nor price optics
had a statistically significant effect on consumers’ evaluation, or on the consumers’
purchase intention of the product (Huber, Gatzert & Schmeiser, 2015).
As part of an international research study, the consumer buying behaviour and

satisfaction regarding motor vehicle insurance policies within the Indian general
insurance market were studied (Das & Rao, 2017). A more recent study conducted
research on how advertising affects consumers’ decision-making in the US auto/motor
vehicle insurance market (Tsai & Honka, 2019). In addition, a study conducted by
Weedige et al. (2019) aimed to investigate the direct and indirect effect of consumers’
insurance literacy on the purchasing decisions related to life insurance.
Within the broader African insurance market, the health insurance sector has been
extensively researched (De Allegri et al., 2006; Lagomarsino et al., 2012; Spaan et al.,
2012; Alatinga & Williams, 2016; Fenny, Yates & Thompson, 2018; Mathew & Mash,
2019; Nicol et al., 2021). A study conducted on the entire insurance market in Nigeria
revealed that the top three barriers to having insurance are a lack of understanding,
affordability and ignorance regarding where to buy insurance (EFInA, 2012:52). Other
research within the Nigerian insurance market focused on the impact of inflation on
life insurance (Epetimehin & Fatoki, 2011) and the socio-economic indicators that
affect life insurance businesses (Mojekwu & Ibekwe, 2015). A study in Ethiopia
(Demissie, 2014), assessed customer satisfaction with motor vehicle insurance
10
services, while a study in Ghana (Agyei et al., 2020), investigated the influence of trust
on customer engagement in Ghana's insurance market.
Locally, research has been done on customer loyalty in the long-term South African
insurance market. The influence of two-way communication and conflict handling on
intentional customer loyalty via Customer Relationship Management (CRM) was
investigated (Du Plessis & Roberts-Lombard, 2013). A study by Mackay, Mostert and
Petzer (2015) that investigated the interrelationship between service quality, relational
benefits, customer satisfaction and behavioural intentions in the South African short-
term insurance market found that consumers' level of satisfaction has a positive effect
on their behaviour intentions, including loyalty, word-of-mouth, and future loyalty.
Another South African study investigated the variables that influence unethical
decision-making and behaviour in organisations in the life insurance sector of South
Africa (Viviers, Van Greunen & Venter, 2012). The purchase influences within the
South African motor vehicle insurance market were described (Mankum, 2019), and
the antecedents of consumer behaviour within the South African life insurance market
was explored (Kruger, 2018).
However, no evidence could be found of studies conducted solely on the purchase

decision process, specifically, the purchase decision process within the international
or South African personal motor vehicle insurance market. Furthermore, no existing
consumer decision-making models in other insurance sectors could be applied in the
South African personal motor vehicle insurance industry, as this industry is a unique
field with different consumer risks than other type of insurance products. Therefore,
this study aimed to determine the purchase decision process of insureds and non-
insureds with regard to personal motor vehicle insurance.
Organisations should have a clear understanding of individuals in their market and

their purchasing habits (Strydom, 2014:275; Erasmus et al., 2019:132). Therefore,
short-term insurers should understand the individuals in their market and their
purchasing behaviour. Parumasur and Roberts-Lombard (2014:262) and Lamb et al.
(2018:92) state that this understanding is vital for organisations to serve and influence
their market. It thus seems that short-term insurance companies can benefit from
having a clear understanding of their consumers and potential consumers, and their
11
purchase behaviour, as they can satisfy these individuals’ needs and may be able to
influence their consumers.
Consumers who decide to purchase products or services make decisions regarding

the acquisition, consumption, and disposition of the various services offered by the
different organisations (Mpinganjira et al., 2013:246; Cunningham, 2018). This can
thus also be applied to the personal motor vehicle insurance market, since the
consumer who decides to buy personal motor vehicle insurance will make decisions
regarding the acquisition, consumption and disposition of the various services offered
by the different short-term insurance companies. Therefore, it will be beneficial for
short-term insurance companies in South Africa to understand the purchase decision
process of both the insureds and non-insureds. They also need to comprehend why
some motorists do not buy personal motor vehicle insurance. Therefore, to meet this
need, the current study developed a model to illustrate the consumer purchase
decision process of insured and non-insured within the personal motor vehicle
insurance industry of South Africa.
THE RESEARCH PROBLEM
Motor vehicle accidents occur every day on South African roads. According to Hogg,
(2015), the estimated economic cost of road accidents in South Africa at the time was
309 billion rand per annum. During 2019, road accidents cost the South African
government over 176 billion rand (Mlambo, 2020). Even though the costs have
decreased over the years, this is still a very large sum of money that is being paid due
accidents per year.
Crime statistics revealed that car/motor vehicle hijacking increased over the 2016-
2017 year by 14.5% in comparison to the previous year. The total number of car/motor
vehicle hijackings reported during this period was 16 717 compared to 14 602 for the
previous period (Africa Check, 2017). Recent crime statistics indicated that during the
2019-2020 year, a total of 18 162 motor vehicles were hijacked, which was an increase
from the 16 026 the year before (Africa Check, 2021). Furthermore, statistics have
revealed that 53 809 motor vehicles were stolen in South Africa during 2016, taking
the total number of vehicles stolen since 2004 to 831 540 vehicles (Crime Stats SA,
2015). The number of motor vehicle theft incidents in South Africa from April 2016 to
March 2017 was 47 586 (Statistics South Africa, 2017a:9). During the 2019-2020 year,
12
46 921 motor vehicles/motorcycles were stolen, which shows a slight decrease (Africa
Check, 2021).
A November 2013 hailstorm in Johannesburg resulted in claims of more than 2 billion

rand. This event had a negative influence on insurance companies’ earnings for that
year (IOL, 2014). During 2017, claims for weather-related damage affected short-term
insurance companies, as claim volumes increased in mostly Johannesburg (Gauteng)
and KwaZulu-Natal, due to storm and flooding events that occurred in these two
provinces (Knoesen, 2017).
The civil unrest that is on the rise in South Africa has an effect on consumers’ motor
vehicle insurance policies, but this does not directly impact the short-term insurers
(Coetzee, 2021). Insurance companies are not obligated to pay out claims for damage
or loss incurred during a strike, civil unrest or protest, as these types of claims are
covered by an additional South African Special Risks Insurance Association (Sasria)
cover. Sasria is a state-owned entity that offers cover for loss and damage to insured
property (such as motor vehicles, houses and business property and inventory) due
to civil unrest, including rioting, strike action, and public disorder (Saria SOC Ltd, n.d.).
Sasria is the only insurer that provides cover for damage caused during these types
of incidents, and it is usually added to the short-term insurance policy for motor vehicle
insurance, however, it is still optional (Coetzee, 2021). Therefore, there is a cost
increase for the consumer (taken that additional cover should be added to their policy),
but at least the consumers will be covered in events like looting and striking.
The repair and replacements costs of motor vehicle accidents, hijacking and stolen
motor vehicles remain exceptionally high due to the radically more expensive original
equipment manufacturer parts which owners are obligated to use while the vehicles
are under warranty (Santam, 2015; Stokes, 2021a). Due to the weakening in the rand
during 2015, the import costs for vehicle parts are higher, and therefore the repair and
replacement costs are higher for short-term insurance companies (Vazeer, 2016b). It
will also be difficult to replace a motor vehicle with the short-term insurer's pay-out
value, taking into consideration that new motor vehicle prices also hiked during the
2020 year, despite the stronger rand (Droppa, 2021).
The excessively high attrition rate (also known as churn rate) in the South African
short-term insurance market means that short-term insurers have to consistently
13
compete on price by decreasing their premiums (Coutts et al., 2019:11; Innovation
Group, 2021). The churn rate is known as a metric for measuring customer retention.
It applies to companies that offer services to their customers through an ongoing
contractual relationship. More specifically, customer churn measures how many
customers a company have lost over a certain time period (Kraus, 2016; Frankenfield,
2021).
Previous research points out that the policy price is the primary reason for switching
to another motor vehicle insurance provider (Sanders & Tyson, 2014; Livingston,
2020). According to Bolhaar, Lindeboom and Van der Klaauw (2015:57) individuals'
willingness to switch from one insurance company to another generates enough
competition to withhold insurers from increasing premiums. Then again, the costs are
increasing for short-term insurers and it can be challenging for them not to increase
premiums.
According to eNaTIS (2021), the live vehicle population (registered vehicles) driving
on South African roads for August 2021 was 12 882 765. The live vehicle population
refers to the number of vehicles that are registered and vehicles that have a renewed
vehicle license (Letshwiti, Stanway & Mokonyama, 2003:2). Only 30 to 35% of these
vehicles (including motor cars, minibuses, buses, midibuses, motorcycles,
quadrucycles, tricycles, light delivery vehicles (LDVs), trucks, and other self-propelled
vehicles) on South African roads are insured (Buys, 2015; Peyper, 2021; South African
Insurance Association, 2021). This indicates an extremely low level of vehicle
insurance penetration, as only a small portion of the vehicle population on South
Africa’s roads is in fact insured.
This low market penetration level creates a general problem for insurers to recover
costs from third parties responsible for accidents where clients are involved. The
chances are high that these third parties do not have vehicle insurance or cannot afford
to pay for the damage (Santam, 2015). The current financial distress of consumers
during the COVID-19 pandemic can also impact the penetration rate when consumers
decide to cancel their current motor vehicle insurance policies (Peyper, 2021).
Given the background above, there is a definite need for this research because of the
dearth of research in the motor vehicle insurance industry, and the low percentage of
consumers who possess motor vehicle insurance in South Africa. There is a dire need
14
for short-term insurance companies to better understand the factors that play a role in
the individual purchase decisions of their insurance products to better persuade
consumers to purchase their products.
Therefore, the current study aimed to answer the following main research question:
Which factors influence the individuals’ purchase decisions in the South African
personal motor vehicle insurance industry?
It is therefore imperative to conduct research to determine the purchase decision

processes of both the motor vehicle insureds and the non-insureds to understand the
behaviour behind motorists’ reasoning for not purchasing personal motor vehicle
insurance. The research objectives are presented in the next section.
OBJECTIVES OF THE STUDY
The purpose of this study was to determine the purchase decision process of
consumers (insured) and potential consumers (non-insureds) regarding personal
motor vehicle insurance, by developing a unique model that will assign a probability to
each of the possible choices based on the context of the decision.
The primary objective of this study was to develop a model which would illustrate the
consumer purchasing decision process in the South African personal motor vehicle
insurance industry. The purchasing behaviour of insureds and non-insureds would be
determined by testing a conceptual model of the consumer purchase decision for
personal motor vehicle insurance (CPDFPMVI). A tested purchase decision model
specifically in terms of personal motor vehicle insurance can assist short-term
insurance companies in understanding the purchase behaviour of their consumers
(insureds) and potential consumers (non-insureds). In order to achieve the primary
objective of this study, several secondary objectives were developed. The primary and
secondary objectives of this study are listed below.
Primary research objective
To develop a model that explains the consumer purchase decision process in the
South African personal motor vehicle insurance industry.
15
Secondary research objectives
The following secondary research objectives were formulated for the current study:
• To determine the company’s marketing efforts that influence the individual’s need
to purchase personal motor vehicle insurance in South Africa.
• To establish the sociocultural environment factors that influence the individual’s

need to purchase personal motor vehicle insurance in South Africa.
• To identify the communication sources that influence the individual’s need to

purchase personal motor vehicle insurance in South Africa.
• To determine the individual’s psychological attributes that influence the need

recognition to purchase personal motor vehicle insurance in South Africa.
• To investigate the steps of the decision-making process for the personal motor
vehicle insurance industry in South Africa.
• To assess the demographic variables that influence the individual’s need to

• To assess the consumer demographic profile within the South African personal
• To compile a general demographic profile that describes the average personal

motor vehicle driver in South Africa.
RESEARCH DESIGN AND METHODOLOGY
The research design refers to the blueprint of how the research objectives will be
fulfilled (Cooper & Schindler, 2014:82; Zikmund et al., 2017:21). This section provides
a discussion of how the primary research for the current study was conducted, and an
explanation of how data was processed and analysed.
Primary research
Primary research is typically conducted to obtain specific data that will allow the study
to achieve the set research objectives. The research paradigm for the current study is
the post-positivism view, as the researcher identified and assessed the causes that
influence outcomes by testing a conceptual model. This paradigm is mostly found
16
when quantitative research is conducted (Creswell, 2014:10; Collis & Hussey,
2021:40).
Furthermore, the research was descriptive in nature, as the research aimed to

determine the characteristics and understanding of the purchasing decision process,
which results in some individuals deciding to purchase and others not to purchase
personal motor vehicle insurance (Babin et al., 2015:127; Babbie, 2016:91; Saunders,
Lewis & Thornhill, 2019:801). Therefore, the research design followed during the
primary research step in the current study was quantitative in nature.
With a quantitative design the research questions/objectives are generally deductive

and certain variables need to be investigated (Leavy, 2017:72). According to Cooper
and Schindler (2014:146), quantitative research attempts the precise measurement of
something, and in business research specifically, quantitative methodologies usually
measure consumer behaviour, knowledge, opinions or attitudes. Quantitative
methodologies have become the research approach commonly used for studying
consumer behaviour (Hackett, 2019).
Thus, for the purpose of the current study, the insureds and non-insureds’ purchase
decision behaviour was assessed by means of a quantitative research approach. The
rest of this section discusses the target population, sampling method, data collection
and data processing and analysis of this research design.
Target population
The target population, also known as the units of analysis, refers to the ‘what’ or ‘who’
being studied (Babbie, 2016:97). According to Babin et al. (2015:170), the important
question to be asked when considering the target population is: Who is the relevant
population?
Therefore, for the current study, the target population were individuals who are vehicle
owners and/or drivers (motorists) in South Africa. This was the target population, as
this population consisted of the individuals that would have the choice whether to
purchase or not to purchase personal motor vehicle insurance.
Means of accessing potential respondents
After identifying the target population, the researcher searched for an available sample
frame by contacting the Road Traffic Management Corporation (RTMC). This
17
government entity is responsible for coordinating and facilitating motor vehicle
registrations in South Africa. The RTMC, therefore, has a database of all the registered
motor vehicle owners in South Africa. However, unfortunately, they could not provide
the researcher with access to a complete database of current contact details and email
addresses of registered motor vehicle owners. Therefore, the researcher had to find
other means of obtaining the contact details of these potential respondents.
The Automobile Association (AA) of South Africa has a database with up-to-date email
addresses of motorists in South Africa, and this database was a means of accessing
potential respondents for this study. In addition, the Facebook Market Insight tool was
used to distribute the survey to individuals who were likely to drive motor vehicles in
South Africa. Thus, Facebook was also used as a means of accessing potential
respondents for this study.
Sampling methods
Availability sampling, also recognised as convenience or haphazard sampling, is used

when the researcher wants to study the characteristics of individuals passing the
sampling point at a specific time, or if other sampling methods are not feasible (Babbie,
2021:192). One of the main reasons researchers consider convenience sampling is
that this method is always feasible when resources are low and when there is no
sampling frame available (Silvia, 2020:60). This type of sampling involves selecting
haphazard individuals that are easily available to invite them to participate in a study.
Convenience sampling is a non-probability sampling method widely used when online

questionnaires are distributed via email or social media platforms like Facebook
(Saunders et al., 2019:324). Convenience sampling was not the researcher’s first
choice for the current study, but it was deemed feasible as no sampling frame was
available, and the target sample included motorists in South Africa. This study targeted
individuals who drive motor vehicles and are personally responsible for purchasing
personal motor vehicle insurance cover in South Africa.
Snowball sampling is another non-probability sampling method that can be utilised to

identify potential respondents. This type of sampling technique is used when a
researcher asks current respondents to identify one or more individuals who meet the
specific criteria and who are willing to participate in the study (Johnson & Christensen,
2017). Snowball sampling was thus used as a complementary method for the current
18
study by posting the online survey on Facebook and convincing Facebook users to
recruit their friends to complete the survey.
Sample size
The sample size refers to the number of respondents included in a study. It is important
to point out that no set rule exists on the ideal size of a sample, as it is dependent on
the design and nature of the study. Numerous aspects were considered when
determining the sample size of the present quantitative study (see Section 5.8.3). A
response rate of less than 1% was achieved (678 responses), which is low.
Nonetheless, it is not uncommon to have a low response rate when using the web-
based/online survey method, and previous research has indicated that online surveys
have a lower response rate than paper-based questionnaires (Nulty, 2008:303; Ebert
et al., 2018).
The survey completeness of online surveys is higher than that for paper-based
questionnaires (Ebert et al., 2018). This was, in fact, a finding in this study as well,
where both online and paper-based questionnaires were used. Respondents fully
completed the online questionnaires, while there were many incomplete paper-based
questionnaires that could not be used. Guinalíu and Díaz De Rada (2020) found that
using mixed modes to collect data can increase the survey's response rate. By using
an online questionnaire and paper-based questionnaires, the response rate can be
improved. Therefore, the researcher decided to use the paper-based questionnaire
method as well, due to the low response rate of the online survey. The data collection
method used in the current study is discussed below.
Data collection
Data was collected by means of a survey, and, more specifically, an online survey
using a self-administered questionnaire. According to Wagner, Kawulich and Garner
(2012:100), survey research is typically used to obtain data from large groups of
respondents. Online surveys specifically allow a researcher to send questionnaires to
a large group of individuals via the Internet.
In the case of the current study, the online survey was distributed via email to the AA
database and accessible on Facebook, and paper-based questionnaire were also
distributed. The self-administered survey (data-collection instrument) was pre-tested
19
to determine the weaknesses in the design and instrument. A pilot test was conducted
where the researcher distributed the questionnaire on Facebook for seven days. The
weaknesses that became apparent were corrected, and after that, the primary data
collection phase started. Ethical considerations were taken into account before the
data was collected.
Ethical considerations
Ethics in business research involves the moral principles and values that influence
how a researcher conducts his/her research activities (Ghauri, Grønhaug & Strange,
2020:23). Ethical principles and standards should be adhered to during each step in
the research process (Babin et al., 2015:41). The research process was directed by
the ethical standards set by the University of South Africa. An ethical clearance
certificate was obtained before the data collection process commenced (see Appendix
B for the Ethical Clearance Certificate).
A participant information sheet was provided to each respondent, and this letter
explained the purpose of the study and all the information that the respondents needed
to be aware of before taking part in the study. Informed consent forms were provided
to all the respondents as a matter of disclosing the procedures of the research design.
Respondents signed the informed consent form on the online questionnaire by ticking
an electronic checkbox (Saunders et al., 2019:260). Furthermore, the researcher
collected data without harming the respondents, and respondents' anonymity and
confidentiality were maintained (Babbie, 2016:63-65; Zikmund et al., 2017:165).
Data processing and analysis
The data processing and analysis of the questionnaires included the following: All the
questions in the questionnaire were coded. Data were exported in an Excel
spreadsheet format from the LimeSurvey program. After that, the data were edited to
ensure any possible errors made during the exporting process were corrected. The
statistical package IBM SPSS (Version 27) was used for the descriptive data analysis.
Thereafter, the descriptive statistical analysis was performed to enable data cleaning,
and the data were summarised by means of figures and tables.
The next step was to conduct the model fit analysis (inferential statistical analysis).
The process underlying the model fit strategy consisted of three phases: the
20
Exploratory Factor Analysis (EFA) phase, Confirmatory Factor Analysis (CFA) phase,
and the logistic regression phase. The following statistical packages were used for the
different phases: EFA and logistic regression model-building strategy used IBM SPSS
(Version 27), and Amos was employed in the CFA. Firstly, EFA was used to explore
and estimate the factor structure of the conceptual model. Secondly, the EFA results
were used to perform a CFA, which tested how well a pre-specified measurement
theory composed of measured variables and factors fit the reality as captured data.
Thirdly, the CFA model fit findings were used in the logistic regression to describe the
relationship between an outcome variable and a set of independent variables. These
outcomes concluded that the logistic regression model demonstrated sufficient
external validity for complete acceptance of the result, as was found with the overall
model fit analysis of the final CPDFPMVI model.
Research validity and reliability
The researcher ensured validity and reliability during the research process of this
study. Validity refers to the extent to which a test measures exactly what the
researcher wishes to measure (Bell, Bryman & Harley, 2019:174). Reliability
comprises of the accuracy and precision of a measurement procedure. Thus, if the
study should be replicated, the same findings would be obtained (Cooper & Schindler,
2014:257; Bryman, 2016:157).
Firstly, pilot tests were conducted to ensure the validity of the self-administered
questionnaire (Dikko, 2016:521). There are different types of validity, namely, content
validity, criterion validity and construct validity (Cooper & Schindler, 2014:257-259;
Zikmund et al., 2017:276-277). The content validity was ensured, as experts in the
field validated the questionnaire (Bell et al., 2019:174). Regarding criterion validity, the
self-administered questionnaire was judged in terms of relevance, freedom from bias,
reliability and availability (Cooper & Schindler, 2014:258; Bell et al., 2019:174).
Construct validity was tested through confirmatory and exploratory factor analysis
(Zikmund et al., 2017:277).
Furthermore, internal consistency was assessed to ensure reliability in this study. The
Cronbach’s alpha coefficient was used to ensure internal consistency, as this
assessed the degree to which the instrument items, where applicable, are
homogenous and whether it reflects the same underlying construct (Cooper &
21
Schindler, 2014:260; Zikmund et al., 2017:274). However, reliability in terms of the
model-building strategy is slightly more complex, and this is discussed in Chapter 7.
LAYOUT OF THE CHAPTERS
The layout of the chapters is illustrated in Figure 1.1, and a short description of the
content in each one of these chapters is provided.
Figure 1.1: Layout of the thesis
22
Chapter 1: Background and problem statement
Chapter 1 introduced the fundamental components of this study, with all the required
background information. A short literature review was given, as well as a motivation
for this study. The research problem was presented, and this led to the research
question. Thereafter, the research objectives were provided. An overview of the
research design and methodology and the layout of the chapters were provided.
Chapter 2: South African motor vehicle insurance industry
Chapter 2 consists of a literature review of the South African insurance industry and
all the components of the market. The motor vehicle insurance industry and the
importance thereof in South Africa are included. The nature of the motor vehicle
insurance product/service and the different products on offer are discussed. The
distinct differences between direct short-term insurance companies and traditional
short-term insurance companies are examined. Finally, the business environment of
the short-term insurance companies is described.
Chapter 3: Consumer purchase decision-making process
Chapter 3 presents a literature review of the types of customers, the three roles of
customers, and their mental and physical activities. Consumer satisfaction and
consumer retention, and the nature of consumer behaviour are discussed. Levels of
consumer decision-making and the four ways of consumer decision-making are
included in this chapter. The different decision-making models are reviewed and
discussed.
Chapter 4: A consumer purchase decision for personal motor vehicle insurance

conceptual model
In Chapter 4, the CPDFPMVI conceptual model is presented and discussed. This

conceptual model was used in the research methodology to develop a questionnaire
to test the model.
Chapter 5: Research methodology
Chapter 5 presents a detailed explanation of the methodology used in the study. This
chapter includes the research design that was followed while conducting the research
study. More topics included in this chapter are the research population, sampling
23
methods used, sampling size, the data collection process and the methods utilised to
process and analyse the data.
Chapter 6: Descriptive findings of research
Chapter 6 presents the descriptive data analysis in detail and the online survey results
are provided. This chapter contains the descriptive analysis where data are described
and interpreted.
Chapter 7: Model fit analysis
This chapter systematically explains the process followed to develop the final
CPDFPMVI model. A unique model-building strategy was utilised to achieve the final
CPDFPMVI model. The process underlying the model fit strategy consists of three
phases, and these are discussed: the EFA phase, CFA phase, and the logistic
regression phase. The decision behind each statistical method in this model-building
strategy is explained in this chapter. Finally, the final CPDFPMVI model is presented.
Chapter 8: Conclusion and recommendations
Chapter 8 outlines the conclusions and recommendations of this study. This chapter
presents a discussion of the research objectives, the research outcomes and also
reports on the conclusions and recommendations in relation to the objectives of this
study. Future research suggestions are made, and recommendations for short-term
insurance companies are presented. The research limitations of the study are
addressed and the final research conclusion is presented.
24
SOUTH AFRICAN MOTOR VEHICLE INSURANCE
INDUSTRY
INTRODUCTION
The global insurance industry has a rich and ancient history, and the countries in the
global insurance industry have some important similarities, which include
deregulation, the rise of competition, generally rapid growth in insurance sales, and
the emergence of new distribution channels. Nonetheless, there are a number of
important differences among insurance markets across the world, such as inflation
rates and regulatory changes, in each country (Niranjan & Rohita Kumar, 2014;
OECD, 2020:7).
The first South African insurance company, the Zuid Afrikaansche Brand en
Lewensversekering Maatschappij, was established in 1835 in Cape Town. The main
components in the South African insurance industry are long-term insurance and
short-term insurance. The short-term insurance industry in South Africa is governed
by the Financial Sector Conduct Authority (FSCA) and the Short-Term Insurance Act
is used as the regulatory framework (Dinnie, Bracher & Patricks, 2013; Krige & Laskov,
2018).
This chapter presents a discussion of the associations that support the short-term
insurance industry, and the different types of short-term insurance services available
in the country are explained. This chapter also presents a summary of developments
in the country’s short-term insurance industry. Short-term insurance companies are
organisations that operate in a rapidly changing business environment (Liakopoulos,
2016:2). Therefore, this chapter makes a distinction between direct short-term
insurance companies and traditional short-term insurance companies.
South Africa is the only country in Africa that has an established, well-developed,
mature and comprehensive short-term insurance market (Maharaj, 2016; Bagus et al.,
2020). Motor vehicle insurance is a type of short-term insurance, and this is the type
of insurance that is sold the most in the short-term insurance market of South Africa
(KPMG, 2016:97; Danckwerts et al., 2020:44). The importance of motor vehicle
25
insurance in South Africa, the nature of this type of insurance and the different types
of motor vehicle insurance are discussed in this chapter. The components of the
business environment, namely, the micro, market and macro-environments within the
short-term insurance industry, as well as the variables or factors that influence the
short-term insurance companies are comprehensively explained. Practical examples
of how these factors influenced the short-term insurance companies in the past and
how future risks can be mitigated are provided.
GLOBAL INSURANCE HISTORY
Insurance has an ancient history. The first form of short-term insurance, called marine
insurance was recorded by the Babylonian and Chinese traders, 3000 years BC
(Thompson, 2021). Merchants limited their losses by dividing their items between
different ships that travelled across treacherous bodies of land and water (Jenkins,
2015). Approximately 1750 BC, the Babylonians developed a system, where
merchants sailing the Mediterranean received loans to fund their shipment. When a
loan was received to fund a shipment, a payment of an additional sum was made to
the leader in exchange for the leader’s guarantee that the loan would be cancelled
should the shipment be lost at sea (Thompson, 2021).
Around 750 BC the inhabitants of Rhodes (Greece) invented the “general average”
concept. Merchants whose goods were shipped together, paid a premium into a fund
that was proportionally divided. That fund was used to reimburse any merchant whose
goods were intentionally sacrificed for the safety of the vessel and remaining property
(Hinkelman, 2008:294).
Thereafter, the Greeks and Romans formed the first types of health and life insurance
with their caring societies that originated around 600 BC. These societies offered care
for families of deceased citizens as well as payment for funeral rituals. These societies
existed in numerous areas in the world for centuries (Thompson, 2021).
In the Achaemenid Empire/First Persian Empire region, from 550 to 330 BC, the
monarchs of Ancient Persia were the first to insure their subjects and they registered
the insuring process in governmental notary offices (Harrington, 2011). During the 12th
century (1101-1200), a type of state insurance was initiated in Anatolia (Turkey). This
state insurance reimbursed traders from the state treasury for their losses (Thompson,
26
2021). During the mid-1300s, separate insurance contracts that were not tied to loans
and other contracts were introduced in Genoa (Italy). In 1343, the earliest known
marine insurance contract was drawn up in Genoa (Hinkelman, 2008:294). This was
one of the first short-term insurance contracts.
The first printed work on insurance was written by Pedro de Santarém, and this book
was published in 1552 (Zeiler, 2012). Up to the late 17th century, many areas still had
friendly societies that gathered money to pay for medical expenses and funerals
(Thompson, 2021). Some forms of insurance developed in London early in the 17th
century. The will of the English colonist, Robert Hayman mentioned two "policies of
insurance" taken out with the diocesan Chancellor of London, Arthur Duck. Worth £100
each, one was related to the safe arrival of Hayman's ship in Guyana, and the other
was with regard to his life. The will of Hayman was signed and sealed on 17 November
1628, but not proved until 1633 (Rahman, 2011).
Modern short-term insurance can be traced back to the Great Fire of London that took
place on 2 September 1666. This fire was the reason why the first insurance company
emerged (Klein, 2001a). The fire was so severe that it destroyed 373 acres of the city
(only a fifth of the city was not destroyed), which contained over 13 200 houses, 84
churches and 44 company buildings (Robinson, 2011). In 1667, Dr Nicholas Barbon
formed the first actual insurance company that was known as the “The Insurance
Office”. The houses and buildings that were insured by The Insurance Office were
protected by fire-fighting teams that were formed to put out fires in the houses and
buildings that were insured (Klein, 2001b).
At the end of the 17th century, London’s importance in the world of trade expanded,
and this also amplified the need for cargo/marine insurance (Zeiler, 2012). In 1688,
Lloyd’s of London was founded at Edward Lloyd’s coffee house near the Royal
Exchange at the London docks on the River Thames. Lloyd’s became the popular
meeting place for insurance companies, merchants, ship owners, ship captains and
other individuals who wanted to insure cargoes, ships and those who were willing to
underwrite maritime ventures (Hinkelman, 2008:294). In support of these ventures, in
1693 astronomer Edmond Halley published a paper that provided a link between life
insurance premiums and average life spans (Krieger, 2011:63).
27
The first insurance company in the US was based in Charles Town, South Carolina. It
opened in 1732 to underwrite fire insurance (Zeiler, 2012). In the 1750s Benjamin
Franklin started a company that collected contributions to stop disastrous fires from
destroying buildings (Thompson, 2021). In 1752, Franklin established his company,
Philadelphia Contributionship, for the Insurance of Houses from Loss by Fire. This
company warned against fire hazards, and they did not insure wooden houses as this
was seen as a big risk (Rahman, 2011). In 1787, the first fire insurance company called
Mutual Insurance Company was founded in New York. This company was renamed in
1846 as the Knickerbocker Fire (Geisst, 2006:214).
Slave insurance was one of the original forms of industrial/business risk management.
The male and female slaves became such valuable assets to planters/farmers that
owners took out life insurance policies for these slaves. During the 1840s, the number
of slaves insured in the South of America was the same as the whites with life
insurance in the North (Ralph & Rankin, 2021).
Throughout the 1880s in Germany, Chancellor Otto von Bismarck introduced old age
pensions, accident insurance, medical care and unemployment insurance. This
formed the basis of the modern European Welfare State (Langley, 2014). In 1897, the
Workers Compensation Act in Britain required employers to insure their employees
against industrial/work-related accidents (Williamson, 1990:95). Furthermore, in 1905,
the Liberal Party and Sir Henry Campbell-Bannerman, leader of the party at that time,
introduced social insurance in Britain (Wooster, 2014). In 1944, social insurance
became the foundation of the modern British Welfare State (Field, 2011).
The domestic insurance industry in China was nationalised during 1949 under the
People’s Insurance Company of China (Fortis, 2001:2). Medical cover as we know it
today was introduced in the US during the 1950s (Health24, 2011). In 1992, the third
non-life directive and third life directive were passed in the European Union, which
created a single market for insurance services in Europe. With the establishment of
this so-called single passport, insurance companies were allowed to offer insurance
anywhere in the European Union if the required permission was granted from the
regulating authority (European Commission, 1990; Campbell, Goldberg & Rai,
2003:126).
28
This short history of the global insurance industry shows the progress and
developments within the industry that started as long ago as 3000 years BC. The next
section provides a brief history of the South African insurance industry.
HISTORY OF THE SOUTH AFRICAN INSURANCE INDUSTRY
The South African insurance industry emerged in 1806 when a number of British
insurance companies opened businesses in the Cape Colony (today known as Cape
Town), where representatives were temporarily dispatched to Cape Town to offer
insurance services to the local residents (Borscheid & Haueter, 2012:326).
On 6 August 1806, the first two insurance agents in South Africa, Alexander
Macdonald and John Houghton, were given power of attorney by the Phoenix
Assurance Company of London. They sold fire insurance, which during that time, was
the most widely held insurance contract in the Colony (Pearson & Yoneyama,
2015:149). Marine insurance also played an important role as there were 170 ships
occupying the Cape port annually and there were risks involved. The integration of
Phoenix Assurance Company of London was confirmed as they translated the
application forms for their insurance policies into Dutch, which showed that they tried
to satisfy the needs of their community.
In 1826, the United Empire and Continental Life Assurance Association, and the
Alliance British and Foreign Fire and Life Insurance Company established two branch
offices in South Africa (Oxford Business Group, 2021). The Royal Insurance Company
was the first British insurance organisation to dispatch a full-time British Home Office
representative to the Cape Colony in 1845 (Borscheid & Haueter, 2012:326). These
early insurance companies primarily offered fire, marine, life, and to a limited extent,
accident insurance (Pearson & Yoneyama, 2015:149).
On 14 March 1831, the South African Fire and Life Assurance Company opened its
doors (Vivian & Todd, 2011; Oxford Business Group, 2021). In December 1835, the
first South African insurance company was established in the Cape Colony and it was
called the Zuid Afrikaansche Brand en Lewensversekering Maatschappij. A decade
later (1845), the Mutual Life Assurance Society of the Cape of Good Hope was
founded, and it modelled its business structure on the Scottish Equitable Mutual Life
Assurance Society, which was later called Old Mutual.
29
General growth in the sector continued until 1861, as there were more than 20
insurance companies operating in the Cape (Borscheid & Haueter, 2012:326). In 1894
the Cape Town Fire Tariff Committee was formed, and similarly, the Johannesburg
Fire Tariff Committee in 1898. Different insurance tariffs were quoted and the purpose
of these committees was to administer the insurance companies across set standards
(Borscheid & Haueter, 2012:327-328). The South African insurance industry further
developed as the international/global insurance industry expanded (Oxford Business
Group, 2021).
The fire insurance stock companies (later called the Council of Fire Insurance
Companies that was established in 1907) managed to form associations in Cape
Town, Durban, Port Elizabeth and Bloemfontein. Until 1930, the English companies
dominated the fire insurance market, as they benefited largely from their British
experience in offering residential, industrial and mining fire insurance (Pearson &
Yoneyama, 2015:150). By the end of the 19th century, the Cape Colony had more than
50 foreign insurance companies providing coverage to the surrounding communities
(Oxford Business Group, 2021).
Marine insurance’s introduction in South Africa was a British initiative. Equitable

Marine, established in 1843, was the first local marine insurance company. Thereafter,
in 1874, the second marine insurance company, Colonial Assurance Company,
opened its doors. By 1898, there were eight marine insurance companies, however,
the life and fire insurance markets had a bigger economic influence on the insurance
industry (Borscheid & Haueter, 2012:328).
At the turn of the 19th century, the three local life insurance companies, the South
African Mutual Life Assurance Company, the Southern Life Association (established
1891) and the Industrial Life Assurance Company of South Africa (established 1894)
dominated the South African life insurance industry. The African Life Assurance
Society (known as the first Transvaal office) was the only insurance company founded
outside the Cape, and this company was founded in Johannesburg, in 1904. African
Life had branches in the Union of South Africa, Khartoum, Aswan, Syria, Cairo and
Alexandria (Borscheid & Haueter, 2012:328).
Soon after the introduction of automobiles/motor vehicles in South Africa, accident

insurance gradually developed. By 1917, accident insurance had a relatively low
30
market penetration, as life and fire insurance were still more dominant. The total
accident insurance premium income during that year was £79 833, in relation to the
£2 million for the life insurance, and the £700 000 for fire insurance (Borscheid &
Haueter, 2012:328).
Motor vehicle accident insurance was introduced in the early 1930s and the Motor
Vehicle Insurance Act of 1942 came into effect in 1946. This act stipulated that third-
party motor vehicle insurance for all drivers and owners are compulsory. The 1976
Soweto youth uprising emphasised the risks associated with riots, as insurance
companies were unwilling to cover the damages caused by political riots. To this end,
Sasria (South African Special Risk Insurance Association) was formed in 1979, as the
South African Government agreed to be the reinsurer, and this enabled Sasria to
provide riot cover (Still & Stokes, 2016:48-49).
In the mid-1980s, health insurance was introduced in South Africa, and by 1989 there
were almost 50 000 policies in place. This industry developed rapidly, and in 1991
there were at least 13 South African insurance companies that offered medical
insurance (Health24, 2011). The dread and disease insurance, also known as critical
illness insurance, was conceptualised by South African heart surgeon, Dr Marius
Barnard in the early 1980s, as he had seen at first-hand the financial stresses patients
experienced living with heart disease and other serious illnesses (Reynolds & Faye,
2015:1).
In 1990, the FSB (Financial Services Board) was established with the enactment of
the Financial Services Board Act (International Monetary Fund, 2010:5). The FSB is
an independent institution that supervises and oversees the South African non-
banking financial services industry. This institution is fully funded by fees and levies
gained from the industry (Financial Services Board, 2017:4).
After the year 2000, there was an increase in the use of the Internet and there were
significant regulatory changes to the insurance industry (Still & Stokes, 2016:49). The
global insurance crises of 2008 to 2009 had an effect on both long-term and short-
term insurance companies in South Africa. Insurance companies faced challenges,
which included, lower investment yield, higher expenses and increases in failed
policies, as consumers struggled to cope with the increased food prices and the
unemployment rate (Jeffreys, 2012:88).
31
In 2010 the industry began to recover and the claims profile of insurance companies
started to appear normal again. The net premiums of short-term insurance companies
increased from 3.78 billion dollars in 2010 to 4.27 billion dollars in 2011. The long-term
insurance industry showed a higher rate of growth in net premiums from 53.2 million
dollars in 2010 to 117.9 million dollars in 2011 (Jeffreys, 2012:89).
Until 2016, short-term insurance companies had to operate in difficult circumstances,

where dissatisfying economic growth, sideways trending markets, high unemployment
and downward trending commodity prices had an effect on insurance companies’ way
of doing business. Insurance companies in South Africa had to adapt to the ‘disruptors’
by making use of new technology in order to gain market share (Still & Stokes,
2016:49). From 1 April 2018, the FSB was replaced by the FSCA, which acted as the
new sheriff for the South African financial services sector (Krige & Laskov, 2018).
As indicated above, the history of insurance in South Africa was mainly derived from
the British colony, as the first insurance agents in South Africa were British. The South
African insurance industry consists of two components, which are discussed in the
next section.
COMPONENTS OF THE SOUTH AFRICAN INSURANCE

INDUSTRY
The insurance industry of South Africa is composed of different insurance businesses

in the long-term and short-term sectors (PwC, 2015).
The long-term insurance sector in South Africa is regulated by the Long-Term

Insurance Act and is governed by the FSCA (Britton & Van der Vyver, 2021). The long-
term insurance companies’ International Financial Reporting Standards (IFRS) in
terms of earnings and key ratios showed substantial improvements during 2015
(Johnson et al., 2016:4). In 2019, the high number of complaints that were recorded
regarding the high premium rates, low claim ratios and high expense ratios, indicated
a potential disruption in the long-term insurance industry (Danckwerts et al., 2020:63).
This was the presumption before the COVID-19 pandemic began.
Long and short-term insurance are discussed in detail below.
32
Long-term insurance
According to the FSB, long-term insurance covers various events, such as death,
retirement and disability (Financial Services Board n.d.:1). Long-term insurance
services include life insurance, funeral insurance, dread and disease/critical illness
insurance, disability insurance, retirement annuities and health insurance (Financial
Services Board, n.d.:1; Duwalcoe Financial Consultants, 2021).
The different kinds of life cover available in South Africa are life insurance, term/fixed
insurance, and endowment policies (Financial Services Board, n.d.), as discussed
below:
• Life insurance is also known as life assurance. This is a way for the consumer to
protect his/her loved ones/dependents financially if the individual should die during
the duration of the policy. Life insurance cannot be seen as an investment or
savings service, as the individual identifies a total amount, and premiums are paid
monthly or annually accordingly (Legal & General, 2021).
• Term/fixed insurance is the same as life cover but for a set period of time. This
service works well if an individual has such a policy in place until the bond of his/her
house is paid off (Financial Services Board, n.d.).
• An endowment policy is practically a life insurance, as it still allows cover for the
life insured, although if this policy matures and the consumer is still alive, the total
amount insured will be paid out to the consumer (Sinha, 2019; Haldane, 2020).
• Funerals can be costly, nevertheless by purchasing funeral insurance the policy

will pay out a cash amount to cover the expenses of the consumer’s funeral. There
are two types of funeral insurance. The first type is the stand-alone policy that is
sold by life insurance companies, their intermediaries and funeral businesses. The
other type of funeral insurance can be included in a life insurance policy (Kearney,
2016; Ardé, 2018).
• Dread and disease insurance, also referred to as critical illness insurance, is a

more expensive cover than disability insurance. However, a dread and disease
policy will pay out a lump sum on the diagnosis of a severe illness (Lamprecht,
2016).
33
• Disability insurance is a type of insurance that protects an individual from loss of
income in the event that he/she will not be able to work anymore due to illness,
injury or accident. This insurance policy will allow individuals to receive a
percentage of their income even though they cannot work anymore (Heathfield,
2020). There are two types of disability cover provided in South Africa, namely,
capital disability cover and income protector disability cover.
o Capital disability cover is when the capital/lump sum for disability cover is
included in a life insurance policy.
o The income protector disability cover is an effective risk cover option and one
form of this cover provides a monthly income with annual increases. Should
the consumer be permanently or temporarily disabled, it can replace the
consumer’s full salary until he/she recovers or dies, or if the policy matures,
whichever comes first (Financial Services Board, n.d.).
• A retirement annuity is a term used for the accumulation of retirement funds. The
individual saves into a retirement annuity while working (Fisher-French, 2010;
Kagan, 2021a). Individuals pay a specific amount of money to a financial institution
and the company then invests this money. After the individual retires, the company
pays out a regular income (Epperson, 2014). There are three types of retirement
annuities, namely, conventional or fixed-interest annuities, the living annuity and
composite annuities.
o A conventional or fixed-interest annuity can be explained as the annuity where

an individual pays in a sum of money and when the policy matures, the
consumer will be paid out a fixed income every month, for the rest of his/her
life.
o A living annuity is when the consumer carries the mortality risk and not the
insurance company. In other words, the income of the consumer depends on
the growth of the investment.
o Lastly, a composite annuity is a combination of conventional and living

annuities. This type of annuity provides the individual with a flexible income
from the living annuity part and a guaranteed income from the conventional
annuity part (Financial Services Board, n.d.).
34
• Health insurance differs from medical aids as health insurance is regulated by the
Long-Term Insurance Act, while medical aids are regulated by the Medical
Schemes Act. The health insurance service is designed to ensure that individuals
receive a fixed amount/lump sum when they need funds for medical purposes. The
amount received will not differ, irrespective of the type of treatment required or
which healthcare providers the individual chooses to use (Independent Financial
Consultants, 2021).
Short-term insurance
The short-term insurance sector in South Africa is regulated by the Short-Term

Insurance Act, and is governed by the FSCA (Dinnie et al., 2013; Krige & Laskov,
2018). The associations and professional standard bodies that serve the short-term
insurance industry of South Africa, include: the Financial Intermediaries Association
of Southern Africa (FIA), The Financial Planning Institute of South Africa (FPI), the
Insurance Sector Education and Training Authority (INSETA), The Insurance Institute
of South Africa (IISA), The Institute of Loss Adjusters of Southern Africa (ILASA), The
Institute of Risk Management South Africa (IRMSA), The South Africa Insurance
Association (SAIA), The South African Insurance Crime Bureau (SAICB), and the
South African Underwriting Managers Association (SAUMA). Short-term insurance
companies’ membership of these associations and professional standard bodies is
voluntary and they operate on a non-profit basis. These representative bodies also
commonly have an executive team funded by membership fees, fundraising,
conference fees and sponsorships (Still & Stokes, 2016:331-339).
This industry comprises the following lines of business: personal, commercial and
corporate business lines (Still & Stokes, 2016:166). These business lines are
characterised according to who the short-term insurance policyholder is. Firstly, the
personal business line involves a short-term insurance policyholder who is a natural
person (Government Gazette, 1998:12). In other words, the personal business line
includes all short-term insurance services that are sold to individuals. Secondly, in
Section 1 of the Short-term Insurance Act (Act No. 53 of 1998) the commercial
business line is defined as “…short-term insurance business in respect of which the
policyholder is a legal person.” A legal person is a term used for a business that has
legal rights and responsibilities according to the law (Cambridge Dictionary, 2021).
35
Commercial insurance is thus when a business takes out insurance and the
policyholder is the business. However, Still & Stokes (2016:166) are of the opinion that
most insurance companies view commercial insurance as insurance that is offered to
small, medium and large businesses from any business sector. Thus, any business
that purchases cover from a short-term insurance company will take out commercial
insurance. Lastly, the corporate line is where large businesses are the policyholders
and these businesses are insured to cover them against certain operational and
liability risks (Still & Stokes, 2016:166).
The personal business line is the largest sector within the short-term insurance
industry. In 2014, 48.2% (R 49 944 billion) of all short-term insurance premiums
resorted under the personal business line (Financial Services Board, 2014:3).
Buthelezi (2020) reports that the personal business line grew in line with inflation
during 2019. As the personal business line within the short-term insurance industry is
the largest, this study will focus only on the personal business line. The developments
within the South African short-term insurance industry are pointed out next.
Development of the short-term insurance industry in South Africa

According to Laing (1988:33), the South African short-term insurance industry
emerged in the years from 1950 to 1958. The two distinct areas that should be
considered are the direct insurance market and the appearance of a professional
reinsurance market. During the early 1950s and mid-1970s, the direct insurance
industry experienced a radical transformation as the scale of operations in this market
increased. The change occurred in the nature of how direct insurance companies
operated. Both of these developments within the direct insurance market influenced
the emergence of the reinsurance market.
On the one hand, direct insurance companies involve insurance companies that
offer/sell insurance services directly to their consumers. These companies do not
make use of traditional intermediaries, such as brokers or agents, to sell their
insurance services to consumers (Shinn, 2021). Reinsurance companies, on the other
hand, take all or a portion of the risks covered under a policy that was issued by an
insurance company, and the insurance company pays the reinsurance company a
premium. Therefore, reinsurance can be seen as a form of an insurance for insurance
companies (The Economic Times, 2021).
36
South Africa is the only country in Africa that has an established, well-developed,
mature and comprehensive short-term insurance market (Maharaj, 2016; Bagus et al.,
2020). The high level of consumer confidence in local insurance providers and the
high market competition are two reasons why the South African short-term insurance
sector has such a high penetration rate (Insight Survey, 2016). Furthermore, the South
African short-term insurance industry can contribute significantly towards the
economic transformation of this country, as this industry provides business
opportunities within different industries. The building, motor vehicle, plumbing, and
panel beating industries are examples of industries impacted by the short-term
insurance industry (Pearson, 2016b).
The global insurance crises of 2008 to 2009 had a ruinous effect on the short-term
insurance companies in South Africa due to lower investment yield, higher expenses
and increases in failed policies. Short-term insurance companies had to face many
challenges, including the increase in food prices and the high unemployment rate
(Jeffreys, 2012:88).
In 2010, the industry began to recover as normal claim profiles of insurance companies
resurfaced. The net premiums of short-term insurance companies increased from 3.78
billion dollars in 2010 to 4.27 billion dollars in 2011 (Jeffreys, 2012:89). According to
Faurie (2014), the insurance industry changed rapidly due to rapid changes in
technology and regulation. Consequently, insurance companies had to be flexible and
had to modify their business models to continue being profitable. The short-term
insurance companies who implemented new technology, such as telematics and
alternative distribution channels, quickly adapted to the change. These short-term
insurance companies were referred to as the ‘disruptors’ in the short-term insurance
industry.
Until 2016, short-term insurance companies had to operate in difficult circumstances

where unsatisfactory economic growth, sideways trending markets, high
unemployment and downward trending commodity prices had an effect on insurance
companies’ way of doing business (Still & Stokes, 2016:49). Short-term insurance
companies continue to adapt to the ‘disruptors’ by upgrading the use of technology in
order to gain market share (Still & Stokes, 2016:49; Van Wyk, 2016). Before the
COVID-19 pandemic hit the country in 2020, the short-term insurance industry was on
37
a strong growth path. However, the COVID-19 pandemic negatively impacted the
short-term insurance industry (Geldenhuys, 2020b).
As previously stated, the direct insurance companies are the ‘disruptors’ within the
short-term insurance industry (Chapman, 2016:10). The differences between direct
short-term insurance companies and traditional short-term insurance companies are
explained below.
Differences between direct short-term insurance companies and

traditional short-term insurance companies
Over the past few years, drastic changes have taken place in the short-term insurance
industry (Moodley, 2019:2). The changes include the changing weather conditions, the
escalation in the use of social media, and other technological advances. Short-term
insurance companies and traditional intermediaries, such as brokers, have to keep up
with this continually evolving environment to ensure that the consumers’ needs are
fulfilled adequately (Roberts, 2016a; Wassink, 2021).
Insurance companies’ business models focus on selling insurance services directly to

the consumers or through traditional intermediaries (Trussell & Reader, 2015:2;
Kaesler et al., 2020). These two components are the distinguishing factors between
direct short-term insurance companies and traditional short-term insurance
companies.
The current study included only direct short-term insurance companies, as these type
of short-term insurance companies have more information about their consumers
because they control the access to the consumers (Schaerer, Wanner & Grinyer,
2011:4; Ernest-Jones, 2019). The differences between these two types of companies
are discussed below.
Direct short-term insurance companies
Direct short-term insurance companies can be explained as insurance companies that

do not use brokers or agents to sell services, and services are sold directly to the
consumers (Shinn, 2021). Technological advances, such as applications, telematics
and other online platforms can be used as intermediaries to sell the insurance services
directly to the consumers (Van Wyk, 2016; Moodley, Forbes & Olivier, 2020).
38
More and more consumers within the short-term insurance market are becoming
comfortable with technology and they tend to be active on social media forums. In
addition, young individuals in the market are more likely to make use of these direct
channels (INSETA, 2014:22-25). There is also an increase in the number of new
entrants in the short-term insurance market with new services in terms of motor vehicle
insurance (PwC, 2014:30; Moodley, 2019:9). The more technology develops, the more
short-term insurance companies will be forced to shift to the direct type of insurance
offerings. Direct insurance companies have more or less the same costs as the
traditional insurance companies, because instead of paying commission to brokers,
direct insurance companies have marketing costs, as well as costs for ongoing
services, such as a call centre and website improvements (Fisher-French, 2016).
Traditional short-term insurance companies
A traditional short-term insurance company sells policies via the traditional distribution
channels, such as brokers and agents (Insurance Europe, 2014:39; Deloitte, 2020:14).
These traditional insurance companies make use of a large broker network to sell the
insurance policies for them (Hesse, 2015a). The brokers or agents have direct control
over the consumer access, therefore, traditional short-term insurance companies
struggle to obtain all the personal information about their consumers from the
traditional intermediaries (Schaerer et al., 2011; Ernest-Jones, 2019).
There are still a lot of consumers who wish to build a relationship with a broker who
understands their needs, with whom they have regular meetings and from whom they
receive sound advice (Roberts, 2016a; KPMG, 2017:40). For example, the non-life
insurance consumers in Germany prefer to get insurance advice from agents (Suter
et al., 2017:45).
There are still a large number of consumers who underestimate the benefits of using
an insurance broker (Roberts, 2016a). Some of these benefits include the detailed
advice and guidance brokers offer to consumers, the brokers’ understanding of the
industry, as well as the risk factors pointed out by the brokers that may affect the
consumer’s insurance requirements in the future (Naik, 2020). As indicated above, the
costs associated with these two business models are approximately the same.
A movement that is becoming more perceptible within the short-term insurance

industry of South Africa, is that the traditional short-term insurance companies are
39
starting to explore the technological advances and are starting to implement some of
the direct channels (Van Eeden, 2017). The traditional short-term insurance
companies that have included some of the new technological direct channels can
therefore not be classified as direct insurance companies, as these companies receive
the highest sales from the traditional distribution channels. Direct short-term insurance
companies have detailed information about their consumers as they sell services
directly and all the information about the consumers is on their systems (Schaerer et
al., 2011; Ernest-Jones, 2019). This is an immense advantage that a direct short-term
insurance company has over a traditional short-term insurance company.
It is important to distinguish between the types of short-term insurance companies that

offer personal motor vehicle insurance. Previous research points out that non-life
insurance consumers in Luxembourg, Spain, Italy and France mostly purchase non-
life insurance products through brokers/agents (Suter et al., 2017:153).
In terms of Africa, a study conducted on Nigerian motor vehicle drivers revealed that
consumers equally prefer methods to purchase directly from the insurance company
and to purchase from brokers (Sodiq, 2020:89). Furthermore, a study conducted within
the Ethiopian motor insurance industry revealed that the use of brokers to sell motor
vehicle insurance is the primary sales method (Regassa, 2018:48). No information
could be found regarding used sales/purchasing methods within the South African
motor vehicle insurance industry. The different types of personal short-term insurance
offerings in the South African short-term insurance industry are discussed below.
Types of short-term insurance

The various types of personal short-term insurance available in South Africa include
motor vehicle insurance, household contents insurance, homeowners insurance,
personal liability insurance, personal all risk insurance, personal accident insurance
and travel insurance (Financial Services Board, n.d.:2; Still & Stokes, 2016:184-192).
Motor vehicle insurance can cover a consumer’s motor vehicle against damage, theft
and fire (Financial Services Board, n.d.:2), depending on the type of motor vehicle
insurance purchased. There are three types of motor vehicle insurance:
comprehensive motor vehicle insurance, third-party, fire and theft insurance, and third-
party only insurance (Doyle, 2021; iWYZE, 2021a). This study will focus on motor
40
vehicle insurance and later in this chapter (Sections 2.6.2 and 2.6.3), this type of
insurance will be explained in detail.
There is a difference between household contents insurance and homeowners

insurance. Household contents insurance covers all the content/possessions in the
consumer’s home against theft, loss, damage and other risks, such as fire or flooding.
Examples of house contents that can be insured are furniture and appliances. Fixed
items like carpets and taps are not covered by home contents insurance (Gray-Parker,
2015; Isaac, 2017). In contrast, homeowners insurance is also known as building
insurance and it protects the consumer’s home against damage to the building, roof
and fittings of the home. Cover is normally provided for the building and outbuildings,
fixtures and fittings, water damages, geysers, sewerage, gas, electricity and telephone
connections, walls, gates, fences and garages or carports. In general, cover is
provided for damage or loss caused by events, such as fire, lightning, explosion,
malicious damage, storm, wind, hail, snow or flood, earthquake and earth tremors,
bursting, leaking or overflowing and impact (South African Insurance Association,
n.d.). This type of policy is usually mandatory if the homeowner has a home loan
(Gray-Parker, 2015; Isaac, 2017).
Personal liability insurance can be explained as the type of insurance that provides
cover when another party is taking legal action against the consumer, in his/her
personal capacity for financial loss or physical injury or death (Kagan, 2021b). In other
words, personal liability insurance covers any form of liability to the policyholder. This
type of insurance is normally included in the household contents insurance policy or
homeowners’ insurance policy. A settlement regarding this type of insurance will
include compensation for medical costs (present and future), restoring or replacing
damaged property, pain and suffering of the injured party, loss of income and legal
costs, as well as expenses (Hesse, 2014; Moon, 2021).
Personal all risk insurance covers all the personal items that policyholders own;
nonetheless, the location and utilisation of personal items are important aspects of this
type of insurance (Araujo & Cheng, 2020). This type of insurance is normally included
in a household contents policy. While the household contents policy only covers items
that are stored and kept in the house, the all risk section will specify specific items that
will be covered when they are moved or transported outside of the house. Examples
41
of personal items that are insured under the personal all risk section are mobile
phones, bicycles, golf clubs, and so forth (Still & Stokes, 2016:192).
Personal accident insurance is another type of personal short-term insurance that can
be a standalone policy, or this cover can be included in personal business line policies
and travel policies. This insurance provides cover for accidental death or permanent
disability caused by bodily injury within a specific time of the accident. Cover can be
provided for the consumer and for the consumer’s family (Still & Stokes, 2016:195).
Travel insurance typically covers medical expenses and costs associated with lost
luggage when travelling to countries other than South Africa (Financial Services
Board, n.d.; Squaremouth, 2021). According to Hesse (2016), travel insurance policies
differ, but the three main categories of risk which most travellers expect to be
adequately covered for are medical treatment and associated costs, loss or theft of
baggage and personal effects, and travel cancellation costs.
As previously mentioned, motor vehicle insurance has the highest penetration rate in
the South African short-term insurance industry. The focus of the current study is on
the personal motor vehicle insurance sector, specifically. Therefore, the history and
development of the global motor vehicle insurance industry is discussed next, before
a discussion of the South African motor vehicle insurance industry will be presented.
HISTORY AND DEVELOPMENT OF THE GLOBAL MOTOR

VEHICLE INSURANCE INDUSTRY
The world’s first modern automobile/vehicle was designed and built in Germany by
Karl Benz in 1885 (Bellis, 2018) and on 3 July 1886, he drove this first automobile in
Mannheim, Germany (Deffree, 2019). The first gasoline-powered vehicle crash in the
world was recorded in Ohio, America in 1891. This accident was minor and the driver
of the vehicle, James W. Lambert, endured minor injuries (Mastinu & Ploechl,
2014:1584). Five years later, on 17 August 1896, a more serious accident occurred
near London’s Crystal Pace in the United Kingdom (UK), when Bridget Driscoll was
knocked down and killed in a vehicle accident (McFarlane, 2010; Brown, 2017).
As vehicles became more popular in the different countries, vehicle accidents

increased globally. The US legal system decided to place the financial burden of an
accident on the individual at fault. This is how motor vehicle insurance was introduced.
42
The first motor vehicle insurance policy was sold by Travelers Insurance Company to
Gilbert L. Loomis in 1897. This motor vehicle insurance policy was written in the same
way as a horse and carriage policy. This policy covered Gilbert L. Loomis’ risks
involved in killing or injuring someone in an accident and damages of the other party’s
property (Ilumba, 2015; Smith, 2021).
In 1907, the first motor vehicle insurance company was established in Providence,
US, by A.T. Vigneron. This small insurance company was called Automobile Mutual
Insurance Company of America, and it offered automotive, fire and theft insurance
(Amica, 2021). In 1922, George J. Mecherle founded State Farm Mutual Automobile
Insurance Company and this company offered auto insurance for farmers in the state
of Illinois, US (Simms, 2004; Encyclopedia.com, 2018). Until this point, motor vehicle
insurance was not mandatory in any country.
Massachusetts in the US was the first state that passed a law to make motor vehicle
insurance mandatory in 1927. Afterwards, New York and North Carolina also
mandated laws, and during the 1950s-1970s many states followed suit (Kane, 2015;
Healey, 2021). In the mid-1990s the internet exploded, and in 1995, Progressive, the
first motor vehicle insurance group, launched the website, auto-insurance.com, and
two years later the name changed to progressive.com (Progressive, 2021). From the
year 2000, the technology changed rapidly and insurance companies all over the world
that offered motor vehicle insurance had to adapt to the changes to provide a variety
of innovative services to gain market share (Hillebrand, 2014).
In 2010, Google announced that they were developing and testing a system for a self-
driving motor vehicle in the hope of reducing the number of motor vehicle accidents
that occur each year by half. The system can detect people and several potential
hazards from a distance. By 2015, Google motors showed that they had travelled more
than one million miles without causing an accident, but these motor vehicles had been
involved in 13 collisions that were not caused by these Google motor vehicles. The
first accident caused by one of the motor vehicles Google was testing occurred in
2016. At that time it was predicted that the self-driving motor vehicle will be launched
commercially in 2020 (Nguyen, 2019). However, experts believe that the full adoption
of these autonomous motor vehicles will not happen before 2030 (Gibbs, 2016).
43
Even though this new self-driving motor vehicle will only affect the motor vehicle
insurance industry in the future, insurers need to include it in their service offering and
redefine the insurance needed for such motor vehicles (Ferenzy et al., 2016:16). A
company called Adrian Flux in the UK introduced the first personal driverless motor
vehicle insurance policy in 2016. This policy was specially designed for consumers
who drive motor vehicles that have driverless features, such as self-parking, or for
consumers who are thinking of buying a motor vehicle with autopilot features (Kollewe,
2016). Telematics and new digital platforms are also examples of technological
improvements that impacted the global motor vehicle insurance industry during 2016.
During that year, digital disruptors within the industry took a narrow focus by offering
on-demand services with a personal touch and rapidly innovating services, such as
telematics and applications, to meet the consumers’ specific need (Ferenzy et al.,
2016:3).
The developments within the global motor vehicle insurance industry have an
influence on the South African motor vehicle insurance industry. The early days of the
South African motor vehicle insurance industry, the legislation and importance of this
industry are discussed in the next section. The motor vehicle insurance services and
the various motor vehicle insurance services available in the South African personal
motor vehicle insurance industry are also reviewed below.
THE SOUTH AFRICAN MOTOR VEHICLE INSURANCE

INDUSTRY
Motor vehicles became more available in South Africa after 1896, when the first motor
vehicle, a Benz Velo, arrived in the country (Wheels24, 2016a). The first recorded road
accident in South Africa occurred on 1 October 1903 in Maitland, Cape Town. Between
1903 and 2003, an estimated number of 393 977 people were killed in accidents in
South Africa (Arrive Alive, 2021a).
During the 1930s, there was a rising concern as the number of motor vehicle accidents
increased and the inability of negligent drivers to pay for the damages became a
problem (Wills, 1986:134). This led to the introduction of the Motor Vehicle Insurance
Act (Act No. 29 of 1942), which made motor vehicle insurance in South Africa
compulsory. This legislation stipulated that all motor vehicles in the country should be
insured and the insurance covered the expenses of the third-party’s motor vehicle
44
damage caused in an accident (Whittaker, 2007:29). In other words, third-party
insurance as we know it today was compulsory.
In the 1960s, it was evident that a number of insurance companies had insufficient
income to operate and were as a result liquidated. The Compulsory Motor Vehicle
Insurance Act (Act No 56 of 1972) was introduced and fell under the legislation of the
Motor Vehicle Accident Fund. This fund was created to act as a reinsurer to carry the
liabilities of insurance companies that sold this compulsory motor vehicle insurance.
The Motor Vehicle Accident Act (Act No. 84 of 1986) replaced the act of 1972, and this
act ensured that the Vehicle Accidents Fund was financed by a levy on motor vehicle
fuel sold in the country (Stokes, 2010).
In 1989, the Multilateral Motor Vehicle Accidents Fund Act (Act No. 93 of 1989) was
introduced, but was replaced with the Road Accident Fund Act (Act No. 56 of 1996),
which commenced on 1 May 1997 (Road Accident Fund, 2021). The Road Accident
Fund (RAF) covers bodily injury or death of third parties only, and does not cover any
expenses due to damages of the motor vehicle. This fund receives a certain
percentage of each one litre of petrol sold (Zingwevu & Sibindi, 2014:658). Thus, motor
vehicle insurance is no longer compulsory in South Africa and only 30 to 35% of all
the motor vehicles driving on South African roads have a type of motor vehicle
insurance (Buys, 2015:23; Bubear, 2016; Geldenhuys, 2020a; South African
Insurance Association, 2021).
There are over 11 million registered vehicles (excluding caravans and trailers), and
the majority (60-65%) of these vehicles are not insured (Automobile Association of
South Africa, 2020). However, it is essential to note that these statistics include the
following vehicle classes: minibuses, buses, bus trains, midibuses, motorcycles,
quadrucycles, tricycles, LDVs, panel vans, other light load vehicles, and trucks. This
study only focused on motor vehicles, which includes the vehicle class ‘Motor cars and
station wagons’.
Since 2014, the motor vehicle insurance industry has been highly competitive and the
increasing numbers of ‘disruptors’ with new service concepts has boosted the
consumers’ expectations, which has also made it a diverse market in which to compete
(PwC, 2014:29). The importance of the motor vehicle insurance industry in South
Africa is discussed below.
45
The importance of the South African motor vehicle insurance industry
The motor vehicle insurance industry plays a significant role in the economic wellbeing
of South Africa (Moig, 2015), as motor vehicle insurance is the most sold insurance in
the South African short-term insurance industry. A total of 43.9% of the total premiums
written in 2015 within the short-term insurance industry, was for motor vehicle
insurance (KPMG, 2016:97). The short-term insurance industry contributed 3%
towards the GDP in 2015, and the financial services sector is the largest contributor
to the country’s GDP (Omarjee, 2016).
The FSCA is a public entity, which is mandated by the South African government to
oversee and enforce compliance with the specific laws that regulate financial
institutions. This entity also promotes financial education, as well as awareness about
financial services, institutions and services (Financial Sector Conduct Authority, 2021).
The short-term insurance companies are regulated by the FSCA, according to the
Short-Term Insurance Act, and in terms of this act, all short-term insurance companies
should be registered with the FSCA. The Ombudsman for short-term insurance plays
a mediating role in dispute resolution between insurance companies and
policyholders/consumers (Stokes, 2021b). Thus, if a motor vehicle insurance
policyholder/consumer should have a complaint, it can be directed to the Ombudsman
for short-term insurance.
The purpose of short-term insurance is to put the consumer in the exact same position
the individual occupied before the loss, dependent on the terms and conditions
stipulated in the policy contract (Vivian & Mushai, 2020:64). Similarly, since motor
vehicle insurance is a type of short-term insurance, the aim of motor vehicle insurance
is thus to place the consumer in the same position, concerning the motor vehicle
occupied before the loss, reliant on the terms and conditions agreed in the policy. The
nature of the motor vehicle insurance product/service is explained below.
The nature of the motor vehicle insurance product/service
The difference between a product and a service should first be described. A product
refers to something that consumers can purchase in a store or shop to satisfy their
needs or wants (Ferrell & Hartline, 2014:151; Layon, 2014:2; Fahy & Jobber,
2019:190). In other words, a product is a physical and/or tangible item that is
purchased. A service, however, is essentially intangible and it does not result in the
46
ownership of an item (Bernstein, 2014:174; Fahy & Jobber, 2019:217). The most
distinct difference between a product and a service is that a product is physically
produced and a service is performed by organisations (Strydom, 2014:201;
Cunningham, 2018). For example, a product is a cleaning item (such as a tile cleaner)
that is available online or in a store to purchase, whereas a service will be a team of
cleaners that is hired to clean the home. Similarly, a motor vehicle is the physical
product, while motor vehicle insurance is a service provided by an insurance company.
Unlike physical items that are produced and placed into inventory, services are
produced and consumed at the same time (Bernstein, 2014:174; Fahy & Jobber,
2019:217). A service is therefore inseparable from the organisation that provides it.
It is important to point out that some organisations can offer both products and services
(Ferrell & Hartline, 2014:151; Kahn, 2015:5; Clatworthy, 2019:130). Organisations that
sell motor vehicles are a good example, as these organisations should offer a good
service before the consumer will decide to purchase the physical motor vehicle. After
sale service also takes place after the purchase and includes the warranty and service
plan of the motor vehicle that was purchased. This is not the case with short-term
insurance organisations, as consumers who purchase insurance will not receive a
physical/tangible product for the money paid. Insurance is a financial service offered
by insurance companies, where consumers purchase financial protection against loss
or harm (Beik, 2018:2).
As explained in Section 2.4.2.3, the short-term insurance industry consists of three

lines of business, namely, personal, commercial and corporate business. Motor
vehicle insurance can thus also be offered in each of these three lines. The personal
business line of short-term insurance is the largest (Financial Services Board, 2014:3).
As explained in Section 2.4.2, the personal business line is when the short-term
insurance policyholder is a natural person and not a business. For the purpose of this
study, the focus was on personal motor vehicle insurance only. Personal motor vehicle
insurance provides cover against the accidental damage or theft of a personal motor
vehicle and not a company-owned motor vehicle (Still & Stokes, 2016:2).
By selling personal motor vehicle insurance policies to consumers, the insurance

company and an individual enter into an insurance contract. This contract specifies
the nature and value of the motor vehicle, the risks or loss events this motor vehicle
47
are insured against, as well as the premium that will be charged for the cover provided
(Still & Stokes, 2016:6).
The premium of each personal motor vehicle insurance policy is different, as insurance
companies calculate the premium based on age, driving experience and the claim
history of the driver (Rankoe, 2014). Most short-term insurance companies use a
computer system to determine the value of the motor vehicle, but the individual must
provide the insurance companies with details, such as the model number, what year
it was manufactured and if any modifications were performed on the motor vehicle
(Van der Walt, 2016). It is, however, also important to indicate whether the insured
value of the motor vehicle is based on the market, retail or trade value. The market
value is the average of the retail and trade value, while the retail value represents the
value that will be paid to purchase the insured motor vehicle, and the trade value is
what a second-hand dealer will purchase the insured motor vehicle for (Arrive Alive,
2009).
According to Auto Accident (2019), statistics indicate that younger drivers are more
likely to be involved in motor vehicle accidents than older drivers. Because of the high
risk associated with younger drivers, the driver’s age has an impact on the premium.
Moodley-Isaacs (2013) concurs that the chances of a young, relatively inexperienced
driver being involved in an accident are high. The main cause of motor vehicle
accidents and motor vehicle fatalities is due to driver behaviour, as 90% of all
accidents are caused by the driver (Ossi in PwC, 2014b:80). Neethling in Hesse (2015)
explains that an inexperienced driver is a higher risk, and even if the driver is older,
with a recently issued driver’s licence, it is also treated as a high risk. The longer an
individual drives, the more experience he/she gains and it is for this reason that
consumers get a slight discount when they are older than 25 (Arrive Alive, 2021b). The
high risk of young and inexperienced drivers is viewed the same by the different short-
term insurance companies, as previous claim experiences revealed that the young
and inexperienced drivers are in fact a high risk (Schoeman, 2016). Seeing that a
driver with little experience in driving might behave/react differently to an experienced
driver, the inexperienced driver can be seen as a high risk.
The driver’s gender also plays a role, as statistics revealed that women are considered
less reckless drivers than men (Masuku, 2018; Dialdirect, 2020). The question can be
asked whether it is fair that women pay a reduced premium due to the fact that they
48
are female? Men are regarded as a higher risk, as statistics revealed that from 2012
to 2015, 60% more men died in road accidents. These numbers are particularly high
in the age groups 20-24, 25-29, 30-34 and 35-39, of which the 25-29 group is the
highest. This report also revealed that the most fatalities in the road user category for
drivers, were male drivers between 20-39 years of age (Arrive Alive, 2021c). Apart
from risks such as the high likelihood that males would drink before driving, and the
high confidence males have when they are driving, the distance they drive is generally
further than that of females (Phillips, 2011).
Where the motor vehicle is parked during the day and night, also has an impact on the
premium. With this factor, the insurance companies aim to determine how safe or risky
the address is where the vehicle is parked. It will, for example, be safer and a lower
risk in a locked garage than when it is parked outside (Van der Walt, 2016). If the
driver chooses to pay a bigger excess when he/she claims, the premium will also be
lower (BUSINESSTECH, 2019). Excess can be explained as the amount a consumer
chooses to pay before the insurance company settles the rest of the claim (Zerbst,
2011; BUSINESSTECH, 2019).
The use of the motor vehicle will also impact the insurance premium, as the motor
vehicle can be used for business or personal purposes. Personal use is when a car is
used to drive from home to work, drive to holiday destinations, go for shopping and so
forth (Masuku, 2018). On the other hand, when a motor vehicle is used for visiting
clients, selling services or even taking orders, the motor vehicle is being used for
business purposes. Sales representatives and estate agents are examples of
individuals that will take out personal motor vehicle insurance and they will have to
stipulate that the motor vehicle is used for business purposes. Motor vehicles used for
business purposes are usually seen as a higher risk, and therefore, this attracts a
higher premium rate (Schoeman, 2016).
The consumer should also indicate who the regular driver of the insured motor vehicle
is as the consumer can own the motor vehicle, but another person might drive it. This
is therefore also a determining factor of the monthly premium paid by a consumer
(Dialdirect, 2020). Furthermore, the inflation and the security measures, such as an
approved car alarm system or tracking device in the motor vehicle, are also factors
that can impact the premium paid per month (MiWay, 2019).
49
There are a number of consumers who believe that red or black-coloured motor
vehicles have a significant influence on the motor vehicle insurance premium (Fourie,
2013a; Hearst Autos Research, 2021). This is just a myth, as most insurance
companies are more interested in the accident history of the driver, the number of
kilometres driven per year, and where in South Africa the consumer lives (Discovery,
2021a). Fourie (2013a) and Cox-Steib (2020) are also of the opinion that the colour of
the insured motor vehicle does not affect the premium. However, other authors
revealed that the colour of the motor vehicle affects the consumers’ premium as it
pertains to the visibility on the road and paint costs, should paintwork be needed for
repairs (Nagel, 2021; WP Motors, 2021).
By taking all the above-mentioned factors into consideration, a personal motor vehicle
insurance premium in South Africa is based on the type of motor vehicle, insured value
of the motor vehicle, modification to the motor vehicle, demographics of the driver,
issue date of the driver’s licence, where the vehicle is parked during the day and night,
excess amount, what the motor vehicle is used for, who the regular driver is, inflation,
and security measures in place. However, the type of motor vehicle insurance that is
selected also has a considerable impact on the premium paid, as prices might differ
for the different types of cover (Masuku, 2018). The different types of motor vehicle
insurance are discussed next.
Different services offered to the motor vehicle insurance market
The majority of short-term insurance companies have three basic types of motor
vehicle insurance offerings: comprehensive motor vehicle insurance, third-party fire
and theft, and third-party only (Van der Walt, 2016; Doyle, 2021).
Comprehensive motor vehicle insurance refers to cover provided for accidental

damage, theft and fire, as well as injury to other parties involved in a motor vehicle
accident (Arrive Alive, 2021b). This type of motor vehicle insurance is the most
expensive, as it is seen as the all-inclusive type of motor vehicle insurance offered in
South Africa (Hesse, 2015c; Expatica, 2021). Depreciation, wear and tear and
mechanical or electrical breakdown are expenses that the comprehensive motor
vehicle insurance does not cover (South African Insurance Association, n.d.:2).
Third-party fire and theft motor vehicle insurance is cover provided for fire, theft and
hijacking incidents, but third-party insurance does not cover the damage to the insured
50
motor vehicle, only the damage to the other party’s motor vehicle involved in the
accident (Holden, 2014; Arrive Alive, 2021b). With this type of insurance, the
consumer will be responsible for covering the costs of the damages to his/her own
motor vehicle (Van der Walt, 2016). Third-party motor vehicle insurance only covers
the costs when the consumer is in an accident and the accident was his/her fault
(Hesse, 2015c; Ardé, 2020). In other words, any damage, theft, fire or hijacking to the
insured motor vehicle will not be covered by the insurance when the accident was the
consumer’s fault and the consumer will have to pay for this out of his/her own pocket.
Naked and King Price Insurance have a unique type of motor vehicle insurance which
is called “Pay As You Drive”, and this type of cover is specifically for individuals who
do not drive a lot as it offers comprehensive motor vehicle insurance but only for the
kilometres that the consumer actually drives. The satellite-tracking device used to
track the kilometres can also act as an anti-theft device, since it can assist the
authorities in tracking the stolen motor vehicle (Knowler, 2020).
Regardless of the type of motor vehicle insurance chosen, most of the short-term
insurance companies offer free roadside assistance programmes, where they will
come and assist when a consumer needs roadside assistance. The insurance
companies might even make arrangements so that the consumer pays reasonable
towing costs (Van der Walt, 2016).
Many of the insurance companies that offer motor vehicle insurance offer a no-claim
bonus/cash-back reward as an incentive for the consumer. Insurance companies offer
this no-claim bonus/cash-back reward if the consumer does not claim in a specific time
period. This prospect of the financial incentive is attractive to consumers. So much so,
that they choose not to claim from the insurance company to avoid losing the cash-
back reward or discount on future insurance premiums (Wheels24, 2016b).
It is not clear which type of motor vehicle insurance most South African drivers
purchase, but a study conducted on US drivers revealed that 78% of American drivers
purchase comprehensive insurance for their motor vehicles (Institute Insurance
Information, 2021). The next section discussed the business environment of the short-
term insurance companies in South Africa.
51
THE BUSINESS ENVIRONMENT OF THE SHORT-TERM
INSURANCE COMPANIES
The insurance industry is changing at a rate that has not been seen for at least a
generation (Liakopoulos, 2016:2). Every organisation operates in a rapidly changing
business environment and an organisation should adapt to the changes in its business
environment (De Beer & Rossouw, 2018:194). The business environment of
organisations, such as short-term insurance companies, consists of the micro, market
and macro-environments. While the micro-environment can be controlled by an
organisation, they cannot control the market and macro-environment (Erasmus et al.,
2019:122-123). Each one of these environments has variables/factors, which can
affect the business environment. These variables/factors can have a negative or
positive impact on the business strategy and survival. Thus, short-term insurance
companies should evaluate and assess these variables/factors regularly to determine
the impact. The components of the short-term insurance companies’ business
environment are discussed below.
Micro-environment
The micro-environment is known as the sum total of all the variables and factors that
occur within an organisation (Ferreira, 2011:14; Botha, 2018:32). Organisations have
control over this, as this refers to their internal environment. The variables affecting
this internal environment are the mission, vision and objectives, areas of management
or business functions, and resources within an organisation, as discussed below
(Erasmus et al., 2019:125-126).
Mission, vision and objectives

The mission and vision statements are vital components in the long-term strategies of
any organisation. These statements include inspiring words which are developed by a
leader of an organisation to clearly stipulate the purpose, direction and driving forces
of an organisation (Epstein, 2021). The mission statement thus stipulates the primary
purpose of an insurance company as an organisation (Ahmed, 2019a). The vision
statement specifies what the insurance company wants to achieve as a company
(Sonkiya, 2020). In other words, the vision explains the future situation of the
insurance company. Objectives are the clear goals that an organisation needs to
achieve so that the organisation can continue to operate (De Beer & Rossouw,
52
2018:3). It is thus clear that the mission, vision and objectives are aspects that a short-
term insurance company can control, as the leaders and managers in the organisation
are the responsible parties who formulate it.
Momentum (2021), a short-term insurance company in South Africa, reveals that their
vision statement is “…to satisfy the wealth creation and preservation, insurance and
income needs of clients through our deep understanding of the upper retail insurance,
savings and investment market in South Africa.” The mission statement of this
company is “We commit to fulfil your lifelong financial needs by providing value-for-
money solutions that are relevant and unique, engaging in a transparent, simplified
and effective manner and building enduring partnerships.” Discovery (2021b) states,
“Discovery is a shared value insurance company whose purpose and ambition are
achieved through a pioneering business model that incentivises people to be healthier,
and enhances and protects their lives.” Another short-term insurance company,
iWYZE (2021b), explains that their mission is to “...ensure that your insurance is a little
easier to understand.” This company’s strategy or vision is: “Responding to our
customers’ needs, we are building a long-term savings, protection, investment and
lending group by leveraging the strength of our people and capabilities in South Africa
and around the world. We will focus, drive and optimise our businesses to enhance
value for customers, shareholders and the communities we serve”.
Areas of management
Koontz, Weihrich and Cannice (2020:2) define management as “… the process of
designing and maintaining an environment in which individuals, working together in
groups, effectively accomplish selected aims.” Murugesan (2012:1) explains
management as the art and science of getting work done through the actions of
employees. Management is a process of giving directions and controlling the
resources and numerous activities of employees, in order to achieve the objectives of
the organisation (Erasmus et al., 2019:202).
The different areas of management or business functions in an organisation are

marketing management, public relations management, financial management,
purchasing management, human resource management, information management,
operations management and general management (Strydom, 2015:136; Botha,
2018:151). These functions should be managed, and the organisation should be
53
structured in such a way that it can control the influences from the micro-environment,
and still function productively within the market environment (Erasmus et al.,
2019:202). For example, the marketing management function is responsible for
recognising the differences in consumers’ preferences and competitors’ movements
in order to develop strategies, to be proactive in countering any influences from the
market environment.
An organisation cannot satisfy the consumers’ needs unless the resources, such as
human, information, physical and financial resources, are not effectively managed.
Every employee in an organisation needs repeated encouragement, and management
is the only party that can provide this (Bose, 2012:5). A manager in an organisation
has four main functions that he/she should perform, namely, planning, organising,
leading and control (De Beer & Rossouw, 2018:13). Thus, in terms of short-term
insurance companies, management refers to the individuals in the organisation that
manages employees who work together, through planning, organising, leading and
controlling resources to achieve the set objectives of the organisation.
In any organisation there are three levels of management, namely, first-line level,
middle/functional level and top/strategic level. A supervisor is an example of a first-
level manager and this type of manager is mainly responsible for managing the daily
activities of four to five subordinates. An example of a middle/functional manager is a
marketing manager, who has the responsibility of managing a department and this
type of manager is involved in the medium-term planning. Top/strategic level
management is a typical chief executive officer (CEO) of an organisation and this type
of manager has to perform the strategic planning within an organisation (Pride,
Hughes & Kapoor, 2017:173). All the managers in an organisation have routines to
gather data and to monitor the business environment. They also understand the
changes and development in the business environment (Aghazadeh, 2016:225).
Every manager in an organisation has the responsibility to take the necessary action
to enable individuals to make the best contribution towards achieving the objectives of
the group. This group can be the team members, departments or the entire
organisation (Koontz et al., 2020:2). Thus, every manager in a short-term insurance
organisation, irrespective of the level of management, has different, tasks,
responsibilities and skills that should be performed in order to achieve the
organisational objectives. Management is an internal environmental variable that can
54
be controlled by the organisation, and the managers in short-term insurance
companies should monitor the business environment on a regular basis.
According to Beer, Finnström and Schrader (2016), previous research revealed that
executives in different industries indicated the lack of access to leadership training and
this can create obstacles for future leadership roles. Painter (2016:3) argues that this
is a big problem in the insurance, also short-term insurance, industry. Insurance
companies provide various services to institutional, commercial and individual
consumers. It is imperative for short-term insurance companies to develop leaders and
this can be done by creating a leadership strategy, which is in line with the insurance
companies’ business strategy.
Resources
Resources can be defined as the objects or assets that can be named, have a
purpose, share a common uniform boundary and can be manipulated through
representations (Johnston & Gamon, 2013:1524; De Beer & Rossouw, 2018:2).
Resources in an organisation can be divided into tangible and intangible resources.
Tangible resources are objects/assets that are observable and easily quantified.
Examples of tangible resources are financial, physical and technological resources.
Intangible resources refer to objects/assets that are harder to identify and more
challenging to quantify. Human, innovation and reputational resources are examples
of intangible resources. Human resources are intangible, as the skills and abilities of
staff members and managers are harder to identify (Peng, 2017:99).
The resources in an organisation are limited, and therefore, should be managed

wisely, so that there are sufficient resources to achieve the greatest possible measure
of prosperity (De Beer & Rossouw, 2018:2). It is the responsibility of managers in
short-term insurance organisations, to manage the tangible and intangible resources
accordingly so that the greatest achievement can be reached with the number of
resources available. This is an internal variable that can be controlled by the short-
term insurance companies.
Managers in short-term insurance companies should be aware of the changes and

trends among the employees (human resources) they are managing. A study
conducted by INSETA (2016:37-38), revealed that employees in the short-term
insurance industry have a greater need for stimulation. It is also a norm that individuals
55
change career paths three to five times in their lifetimes. Furthermore, employees are
not loyal towards their employers anymore and they would gladly change jobs to
improve their current situation, even if it is just for a short period. Headhunting is also
popular, and this results in remuneration which is out of line with that of other
employees on the same level. This can trigger dissatisfaction and low morale among
the other employees (INSETA, 2016:37-38). These findings can assist managers in
short-term insurance companies to incorporate improvements to ensure that
employees are better stimulated. Employees’ loyalty towards the company should be
improved and management should ensure that current employees are not affected
when they headhunt for new employees. In the next section, the market environment
of short-term insurance companies is discussed.
Market environment
The market environment does not exist in isolation, as the micro and macro-
environments impact the market environment. When the managers in organisations
analyse the market environment, they evaluate the possible effects of the
macro/external environment variables or changes on the organisations’ survival and
strategy strength (Aghazadeh, 2016:255). Thus, managers in short-term insurance
companies should evaluate the possible effects that the external environment might
have on the organisational strategy, as this impacts the survival of the insurance
companies. Components of this market environment that should be evaluated include
market/consumer, suppliers, labour markets and unions, intermediaries and
competitors, as discussed below (Erasmus et al., 2019:126).
The market
The market variable refers to the consumers that have a need and the financial power
to purchase a certain product or service (Ferreira, 2011:16; Erasmus et al., 2019:126).
A market/audience can be described as a large number of individuals who have a
need for a specific product/service, have the money to purchase the product or
service, are willing to spend the money on it and are legally able to buy the product
(Erasmus et al., 2019:459). An individual can form part of the market but not be a
consumer of a product or service (Lappeman et al., 2021:5). In other words, the market
variable within the short-term insurance industry will include individuals who have
56
personal motor vehicle insurance (insureds) and individuals who do not have personal
motor vehicle insurance (non-insureds).
The consumer profile should be established by short-term insurance companies and

this is the information that guides them on how to target the audience/market. Most
US licenced drivers fall in the 35-39 age category, which includes 90.9% of the
country's licenced drivers (Buchholz, 2020). In England, there is a large increase in
the number of older individuals (aged 70 and older) that have full driving licences (RAC
Foundation, 2021). No information could be obtained about the age of drivers in South
Africa who have driving licences.
The Black African population makes up the majority of the South African population,
and constitutes approximately 81% of the total population (Maluleke, 2020:23). The
percentage distribution of vehicle ownership by population groups in South Africa (in
other words, percentage of the population group owning motor vehicles) is Black
African 19.8%, Coloured 41.8%, Indian/Asian 81.3%, and White 94.2% (Statistics
South Africa, 2017b:12). Therefore, a very small percentage of the Black African
population are vehicle owners. In contrast, a high percentage of the White population
own motor vehicles, but this population group is also the smallest in the country.
However, seen in total, it is important to realise the percentage of vehicle owners of
each population group in South Africa, which is as follows: Black Africans possess
52.6% of the motor vehicles in South Africa, Coloureds 9.7%, Indian/Asians 6.5% and
Whites 31.2% (Statistics South Africa, 2017b:12). Thus, the Black African population
group have the highest percentage of motor vehicle ownership.
Maluleke (2020:2) specifies that approximately 51.1% of the South African population
are female, and 48.9% are male. Interestingly, 37.6% of the South African male
population own motor vehicles, and only 19.6% of the female population own motor
vehicles (Statistics South Africa, 2017b:12).
A study conducted amongst Ghanaian drivers found that 59.6% are married, 27.1%
are single, and 13.3% are divorced (Atombo et al., 2017:11). No information could be
obtained about the marital status of drivers in South Africa. A study conducted on
Qatari drivers found that the highest education level is secondary education
(equivalent to high school education in South Africa), which made up 33.4% of the
study’s respondents (Soliman et al., 2018:655). Several studies found that drivers'
57
education levels form part of the demographic factors that impact the drivers’ driving
behaviour (Atombo et al., 2017; Mohamed & Bromfield, 2017; Soliman et al., 2018).
No information could be obtained about the highest level of education of drivers in
South Africa.
In a study (Martinussen et al., 2017:2) conducted on drivers’ driving behaviour in

Denmark, the findings related to employment status revealed that 59.3% of the
respondents were employed, 28.5% retired, 5.4% self-employed, 4.8% unemployed,
and 2% students. No information could be obtained about the employment status of
drivers in South Africa.
The province where most drivers drive with their motor vehicles should also be
considered, as this might impact the marketing efforts in each province of the country.
The live vehicle population (registered motor vehicles) in each province of South Africa
are Gauteng 41.4%, Western Cape, 17.2%, KwaZulu-Natal 13.7%, Eastern Cape
6.4%, Mpumalanga 6%, Limpopo 4.8%, North West 4.5%, Free State 4.3% and
Northern Cape 1.7% (eNaTIS, 2021). The number of motor vehicle drivers in South
Africa can also be linked to the population density of each province. The provinces
with the highest population density in South Africa are (1) Gauteng, (2) KwaZulu-Natal
and (3) Western Cape (Maluleke, 2020). Therefore, this explains why most registered
motor vehicles are also in these three provinces.
The disposable income of consumer is also an important factor to consider. The

disposable income of consumers consists of their household income and personal
income. In 2018, a study by Statistics South Africa (2018) revealed that social grants
remain a vital income source for 44.3% of households, particularly in the country's
poorest provinces. Furthermore, the social grants were the second most crucial
income source (45.2%) for households, and salaries (64.8%) remain the primary
income source for South Africa households. This explains the household income
composition in South Africa, but no information could be obtained that clearly specified
the average household income in the country at the time of the current study. The
average personal income in South Africa is R22 500 per month. However, the highest
earnings are the reason for this high average, as most South African workers can
merely desire to reach a salary above R20 000 (Head, 2020).
58
Organisations often segment markets on the basis of demographics as this information
is widely available and it can also relate to the consumers’ purchase decisions (Lamb,
et al., 2013:275, 2018:337). The general demographic variables include age, gender,
income, household size, education, ethnic background and family lifestyle (Boone &
Kurtz, 2014:170; Willan, 2021). The demographic variables of insurance consumers
can impact their purchase decision (Esau, 2015:493; Mazhar, Rehman & Din,
2015:10; Schmidt, 2019:503).
Industry observers predicted in 2014 that a challenge will appear with regard to the
short-term insurance industry’s consumers over the next five years (INSETA,
2014:24). They believed that consumers perceive no or very little difference between
the various short-term insurance brands (Dunn, 2014). Short-term insurance
companies were advised to monitor this carefully, because if this is, in fact, the case
and consumers do not see a difference between the different brands, they have to
improve their brand image to improve the consumers’ perception.
According to Kriel (2017), consumer expenditure is under pressure and this is a result
of the slow economic growth in South Africa, as well as the high unemployment rate.
Johnson et al. (2016:9) also agree with the statement that the consumers are under
pressure due to the economic conditions in the country. This is true, especially now
with the COVID-19 pandemic, impacting consumers’ disposable income and the rising
unemployment rate in the country (Chitiga-Mabugu et al., 2021:92).
Consumers have become more active in their responses to their experiences with
short-term insurance companies (Ernst & Young, 2013:3; Moodley, 2019:5). When
consumers experience bad service from a motor vehicle insurance company, they do
not just accept this and keep quiet. They raise their voices by posting their experiences
online. For example, dissatisfied consumers can post tweets or write a blog about their
bad experience (Parkin, Jooste & Thorpe, 2019) and this will impact negatively on the
short-term insurance company.
The low level of consumer knowledge about financial concepts, such as inflation, can
be seen as a weakness for short-term insurance companies (Sibanda & Sibanda,
2016:5), as this can influence consumers purchase behaviour if they do not have an
understanding of financial concepts. This weakness can, however, be turned into a
strength, if the short-term insurance companies develop strategies to improve the
59
education and knowledge of consumers, as well as executing these strategies. For
example, short-term insurance companies can update and expand their current
marketing efforts to ensure that the education and knowledge of consumers are
improved. An informed consumer can make a better distinction between good and bad
insurance advice (Llewellyn, 2012:4; Woolard, 2015:3; Cherry, 2020a). Many
consumers still prefer the human connection of an insurance professional when buying
insurance (Sodiq, 2020:90). The consumer expectations in the short-term insurance
industry of South Africa are changing, as they seek simplicity, transparency and speed
(Knoesen, 2020). Short-term insurance companies should be aware of all the above-
mentioned consumer demands, needs and preferences as this will have an impact on
their business strategies.
Suppliers
Suppliers refer to the organisations that supply products and services (direct raw
materials used to produce outputs) as well as resources (such as equipment, staff and
consumables) to an organisation (Cox, 2014:10; Botha, 2018:37). Examples of short-
term insurance companies’ suppliers are the companies where they purchase the
computers, computer software/systems, telephone handsets/devices, the Internet,
office equipment and stationary, which the employees use to generate a policy.
Outsourced service providers, such as panel beaters can also be suppliers for short-
term insurance companies.
Short-term insurance companies pay approved panel beaters to repair the damaged
motor vehicles. The repair costs may be high, as the original equipment manufacturer
(OEM) parts of vehicles that are still under warranty, are significantly more expensive.
The weakness of the rand also causes an increase in the costs of imported parts and
most of the OEM parts are imported (Santam, 2015). Coetzee, in Wheels24 (2015),
states that sometimes accident-damaged vehicles are “uneconomical to repair”
because of the excessive costs to use the OEM parts to repair the damage. Santam
introduced a certified aftermarket parts programme which assists in reducing repair
costs, by using parts not directly sourced from vehicle manufactures, to repair
damaged vehicles. This is a more economical alternative, but the downside is that
these aftermarket parts cannot be used on vehicles that are still under warranty
(Wheels24, 2015; Stokes, 2021a). Short-term insurance companies will have to
assess these variables to make the needed changes or to think of innovative solutions.
60
Labour markets and labour unions
The availability of skilled labour and the effect of strikes on organisations are well-
known variables that should be considered by organisations in South Africa (Erasmus
et al., 2019:126). This strict regulations within the short-term insurance industry force
employees to write examinations to obtain the required qualifications (Financial Sector
Conduct Authority, 2021). It was predicted that the increases in regulations and the
increasing cost of compliance within the short-term insurance industry will stress the
need for qualified and skilled compliance officers (INSETA, 2014:3). The rapid pace
of change in these regulations also pose as a key challenge in terms of training
employees in insurance companies (Danckwerts et al., 2020:91).
All the employees working within the various short-term insurance companies have
the right to belong to a trade union. The main purpose of trade unions is to negotiate
on behalf of their members. They will also defend and advance employee’s rights and
improve the wages and working conditions of members (De Beer & Rossouw,
2018:158).
In South Africa, strike action is used as a measure of last resort when there is no
progress in wage negotiations between the employers and the trade unions. During
these strikes, assets such as motor vehicles are often damaged. This can, however,
be covered under an additional Sasria service, which can be included in a policy
(Stokes, 2012b; Coetzee, 2021). Sasria covers damages due to strike action, but not
damages caused by war, insurrection, rebellion, revolution and military or usurped
power (Saria SOC Ltd, n.d.; Faurie, 2014b). As can be seen in the discussion above,
the labour markets and labour unions play a significant role in the short-term insurance
industry.
Intermediaries
Intermediaries are the link between the organisation and the final consumer. Every
organisation is assisted by intermediaries as they can assist in finding consumers.
Brokers and agents are examples of intermediaries (Senapati, 2015:19; Jensen,
2019). The short-term insurance companies make use of two types of intermediaries
at this stage, namely, intermediated insurance and direct insurance. Intermediated
insurance refers to when insurance companies make use of a traditional intermediary,
such as a broker to sell their policies. Approximately 84% of all the total gross written
61
premiums originate from the intermediated distribution channel (Still & Stokes,
2016:173). Direct insurance companies sell the policies directly to the consumers and
web-based or technological applications can be used as distribution channels (Still &
Stokes, 2016:174; Natarajan, 2018).
The online direct sales of personal motor vehicle insurance have been a viable
distribution channel for a number of years. Brokers and agents had to demonstrate
their value so that they can retain and expand their market (INSETA, 2014:7). As the
industry changes to meet the consumer needs and keeps up to date with other
significant changes, the role of intermediaries cannot be underestimated. With all the
developments and uncertainty in South Africa, it becomes vital to assist consumers to
be insured correctly. This is where the expert advice of brokers can ensure that short-
term insurance companies move closer to their consumers (Epic MSLGROUP, 2017).
A threat to the short-term insurance industry is, unfortunately, that many experienced
brokers are leaving the industry, as they feel that the extra legal responsibilities are
overwhelming and it takes them away from their core business (INSETA, 2014:16;
PwC, 2016:52). Potential candidates also do not enter the industry due to the
perception that there are prohibitive barriers to entering (INSETA, 2014:16). Each
short-term insurance company has to evaluate the involvement, strengths and
weaknesses of intermediaries, including brokers and other intermediates, to establish
if it will benefit the company in achieving its end goal.
Competitors
Competitors can be explained as the other organisations that operate and trade in the
same market and have the same consumer groups as the organisation (Cox, 2014:12;
Botha, 2018:37). For instance, all the short-term insurance companies that offer
personal motor vehicle insurance are competitors in the short-term insurance market.
During 2016, the short-term insurance companies struggled to attain their desired
premium rates due to the highly competitive industry (KPMG, 2016:99). Currently,
short-term insurance companies are still under immense pressure due to low market
growth in a highly competitive market (Faurie, 2019a).
The comparison websites/aggregators have assisted short-term insurance companies

to move the power back to consumers by forcing motor vehicle insurance companies
to price the policies more competitively. This can be disruptive as the impact of this
62
price competition can be seen in the local short-term insurance market. To worsen the
matter, short-term insurance companies are already cutting supply chain costs to
remain competitive. The excessively high churn rate/switching cost impacts short-term
insurance companies as they have to constantly compete by lowering premiums and
they still face the pressure of growing costs (Coutts et al., 2019:11). Previous research
points out that the policy price is the primary reason for switching to another motor
vehicle insurance provider (Sanders & Tyson, 2014; Livingston, 2020).
Direct insurance companies that offer personal short-term insurance, have a slightly
higher market share, as it is more convenient and easier to sell policies online with
telesales as a form of support for advice (Still & Stokes, 2016:180). As discussed
earlier, the direct insurance companies can be viewed as disrupters within the short-
term insurance industry. This is mainly due to the high utilisation of technological
advancements to sell their policies directly. This is another example of how the macro-
environment influences the market environment. Thus, short-term insurance
companies have to evaluate their market share in relation to their competitors,
thereafter, decisions and implementations should be made to increase the market
share. The macro-environment and its sub-environments are reviewed below.
Macro-environment
The macro-environment surrounds the organisation and its market environment. This
environment is made up of sub-environments and there are a number of variables in
each environment that will affect the organisation in either a positive or a negative way
(Ferreira, 2011:21; Cherunilam, 2016:11). The organisation has no control over these
factors.
Short-term insurance companies have to manage the risks in the macro-environment,

as it is vital for the sustainability of their companies. These perceived risks are viewed
as threats, but short-term insurance companies can make an effort to convert these
threats into opportunities (Snyders, 2016). The macro-environment has six distinct
sub-environments, namely, the technological, economic, social, physical, institutional-
governmental and international environments. Each one of these sub-environments
has factors/variables that will impact the organisations (Erasmus et al., 2019:128) and
all these sub-environments, their variables and impacts on the short-term insurance
companies are discussed below.
63
Technological environment
Technology can be seen as the most dramatic force that disrupts existing businesses,
markets, and the macro-environment of an organisation (Fleisher & Bensoussan,
2015:161; Botha, 2018:42). The technological environment includes all the aspects
which have to do with the development of new products, and services that are newly
available to the market (Ferreira, 2011:25; Cherunilam, 2016:68). This environment
can also present opportunities for organisations to present innovative products and
services that improve their competitiveness (Aghazadeh, 2016:261).
The technological environment within the short-term insurance industry has changed
the way short-term insurance companies do business (Hoosen, 2016). The new
technological developments that currently play a role in the short-term insurance
companies’ technological environment, specifically with regard to personal motor
vehicle insurance, are discussed below.
Telematics
Telematics can play an integral part in developing sustainable solutions for motor
vehicle insurance companies (Stokes, 2012b; Dharani et al., 2018). The telematics
devices provide insurance companies with comprehensive data about the consumers.
This device does not only have the function of tracking the vehicle, but it also monitors
how the driver drives, whether the driver is prone to speed, sudden braking or quick
accelerations. It can also monitor whether the driver is driving a lot during the evenings
and critical for insurance companies, it can also indicate whether the driver is on
his/her cell phone while driving. All of the information obtained will allow the insurance
companies to view the driver as a high risk or not (Ardé, 2016). There are a number
of South African short-term insurance companies that utilise telematics to track and
collect data about the consumers’ driving behaviour. Discovery Insure has seen the
benefit of using telematics, and they have already gathered more than 14 billion
kilometres of driving data since it was implemented (Stokes, 2021c).
Mobile applications
The utilisation of smartphones and tablet acceptance amongst consumers are

continuing to grow in South Africa. The benefit of mobile applications that can be
downloaded onto a smartphone or tablet is that it enables consumers and other
stakeholders to communicate easily with the short-term insurance companies. Mobile
64
applications are viewed as the perfect solution because it can facilitate a direct
communication link. For example, insurance companies can include an option on the
mobile application to contact an emergency service provider, which will allow the
consumers to ask for immediate assistance with the press of a button on the
application. Furthermore, by incorporating this development in a business strategy, it
can simplify and speed up internal operations radically (Fourie, 2013b; Ferenzy et al.,
2016).
Short-term insurance companies can also use these applications to send SMSs to the
consumers’ mobile phones to warn them of potential danger. Smartphones and the
application on it can be a connection device to the consumers (Balasubramanian,
Libarikian & McElhaney, 2021). For example, insurance companies can warn the
consumers via SMS about a weather prediction for hailstorms and advise that insured
motor vehicle should be parked undercover (Lamprecht, 2014). Many short-term
insurance companies in South Africa still make use of this function.
There are certain challenges that short-term insurance companies will have to
overcome when they decide to make use of these applications. There is a shortage of
application developers in South Africa and this makes the initial developing costs high
(Chutel, 2016). Furthermore, the Internet connectivity in certain parts of the country is
not up to standard, and therefore, may negatively affect application users who might
be in an area with no or poor Internet connectivity (Fourie, 2013b; Hawthorne &
Grzybowski, 2019). However, the opportunity does exist to develop an application
which is wireless (KPMG, 2016:80).
Interestingly, a short-term insurance company developed an application that uses a

smartphone’s accelerometer and GPS to provide detailed driving information such as
harsh braking, distracted driving, and speeding. It can also provide personalised
driving tips, and has a function that ranks a driver’s score against his/her friends, as
well as other drivers in South Africa (Discovery Insure, 2014; Discovery, 2021c).
Social media
In the last few years, social media has become a very important communication
medium and it should thus form part of the marketing strategy in the financial industry
(George, 2017). Social media has become a tool that short-term insurance companies
use to validate consumer claims before they pay out (Knoesen, 2018; Faurie, 2019b).
65
A carefully organised social media presence is a critical aspect of the marketing
strategy.
There are, however, risks or threats of including a social media presence (Conway,
2017). Social media risks involve possible harm to the brand reputation when a
number of consumers post negative content about bad experiences on the company’s
social media platform, employees that disclose confidential information about the
company, corporate identity theft, and legal, regulatory and compliance violations
(Seymour, 2019:7). It is a good idea for short-term insurance companies to ensure
that the compliance team (the employees that ensure that the company is in
compliance with all the applicable laws, rules and regulations) works with the
marketing team, as this can improve the understanding and respect of any social
media regulation (Daniele, 2016; George, 2017).
There are certain requirements that organisations should pay attention to when
deciding to include social media in their marketing strategies (George, 2017). One of
these is the risk of disclosing confidential company information, such as trade secrets,
know-how and other copyright information on a social media platform. The employees’
rights to use social media if their association with an organisation can be seen on
social media should be specified (Daniele, 2016). An organisation should be aware of
any social media regulations and compliance challenges that pertain to its industry
when something is posted. Organisations should develop proper internal procedures
and controls to ensure that associated risks are managed effectively. Organisations
should, furthermore, decide which devices and applications employees may use and
under what circumstances an employee may communicate with consumers on social
media (George, 2017).
Crouth (2016) and Snook (2019) are of the opinion that social media can also be used
by short-term insurance companies to determine what type of risk a consumer is, and
if the insurance claim should be paid out. If, for example, the consumer posts photos
or a status on a public domain/social media site (such as Facebook) the day that a
motor vehicle accident claim was submitted to an insurance company, this content can
be used as evidence. Insurance companies can investigate on these platforms to
make sure that the consumer’s story is true, and this could influence whether the claim
is successful or not.
66
Bracher (2017) and Geospatial World (2020) argue that social media can assist
insurance companies in the future to determine the risk category of the consumers.
The content a consumer posts about him/herself on social media platforms can
determine the premium he/she will pay or if the potential consumer will get insurance
from the insurance company. Insurance companies will, however, only have access
to the consumers’ platforms if the security settings are not activated.
Social media can, therefore, be one of the Big Data sources that insurance companies
can use to obtain data from (Dillard, 2017:5). The term Big Data and how it can be
used by short-term insurance companies is explained next.
Big Data
Big Data is a term for substantial sets of data which are sizable, more varied in a
complex structure that is difficult to store, analyse and visualise for further processes
or results (Sagiroglu & Sinanc, 2013:42; Taylor, 2021). This data technology offers
insurance companies a chance to create, capture and analyse valuable consumer
data, which will assist insurance companies to calculate and manage risks more
effectively (Ferenzy et al., 2016:6). The advances in data have changed the way in
which insurance companies use information. Insurance companies have to make use
of more effective ways to gather new data sources so that they can improve in the
pricing of risks and at the same time, expand consumer relationships (Preston, 2017).
The sources for Big Data, from which short-term insurance companies can obtain data
are: telematics devices installed in vehicles, location-based sensors fitted in offices
and homes, wearable devices, such as watches and step counters, as well as social
media platforms (Dillard, 2017:4).
Big Data has the opportunity to create a more proactive approach within the insurance
industry (Preston, 2017). Big Data can ensure that short-term insurance companies
understand their consumers and it provides the opportunity to obtain data about a
consumer from multiple sources to generate better consumer outcomes. The analysis
of data about consumers will allow short-term insurance companies to communicate
and interact with consumers on all platforms available (Schliesser, 2014:25;
Kaigorodova et al., 2021:38).
By utilising Big Data, the short-term insurance companies can offer consumers a
smooth claiming process, for example, if this is in place, a short-term insurance
67
company will be notified with an alert that an insured motor vehicle was in an accident.
Some short-term insurance companies in South Africa already make use of Big Data
to monitor the driving behaviour of drivers and reward consumers for good driving (De
Vries, 2017). Another advantage for insurance companies using Big Data is that data
analysis can allow short-term insurance companies to offer solutions tailored to
specific consumer needs. This real-time analysis of Big Data can also bring benefits,
such as refined policies, better estimations gained from consumer analysis, assisting
in meeting the regulatory and compliance demands of the industry faster and cost
effectiveness (Preston, 2017). A further advantage of Big Data is fraud reduction, by
obtaining data from new and external sources and using sophisticated algorithms to
combine data sets, it detects patterns that can assist short-term insurance companies
to reject fraudulent claims (Lewarne, 2017).
The implementation of Big Data amongst insurance companies has been slower than
what was expected, but it definitely captured the attention of many insurance
companies as this can create a competitive advantage (Preston, 2017). Customisation
will become effortless as data capture and analysis expand and become more
affordable to insurance companies (Ferenzy et al., 2016:6). The future looks promising
for those short-term insurance companies that see the opportunity and invest in Big
Data (Havlik, 2016). By considering the benefits provided above, short-term insurance
companies can definitely view Big Data as an opportunity.
Cloud computing
Cloud computing is an integrated technology (IT) which can change the way
organisations store large amounts of data and it makes provision for designing and
deploying applications. Some insurance companies have started to write applications
to the cloud, and make use of the cloud-based infrastructure to support the core
applications, such as policy administration, claims and billing systems (Lambert, 2014;
Betz & Payne, 2021). According to Smith in De Vries (2017), the back-end
administration systems of insurance companies are still outdated as these systems
have been built in phases over a period of time.
The rise of cloud computing enabled greater access to IT systems, but particularly, the
back-end systems which provide access to intermediaries, and it can expedite the risk-
rating process, as well as speeding up the sales process extensively (Kaigorodova et
68
al., 2021:38). This type of technology has become very popular in the life insurance
industry, but in the other industries, other innovations at a system level that require
less capital expenditure have been implemented (Field, 2017). Short-term insurance
companies can investigate this advancement and determine if it can be used to
streamline processes.
Internet of Things (IoT)
IoT relates to the various physical devices available that can be connected to the
Internet to collect and share data (Ranger, 2020). A large number of insurance
consumers are willing to share more information in exchange for savings on their
policies, and these IoT devices automate the sharing of data (Canorea, 2021). In terms
of motor vehicle insurance, telematic tracking devices are used to collect significant
volume of data and this helps short-term insurance companies to determine the motor
vehicle risk factors (Moodley, 2019:9). This data allows insurance companies to
determine a policy rate/price based on the risk factors (Kaigorodova et al., 2021:38).
The IoT can definitely be seen as an opportunity for short-term insurance companies
in South Africa, as it can enhance the customer experience, add important safety
benefits, assist insurance companies to be more proactive, improve efficiency, save
costs by reducing risks (Deetjen et al., 2019).
Artificial Intelligence (AI)
Advani (2021) defines AI as the intelligence demonstrated by machines/robots. These

machines can learn with experience and perform human-like tasks. Insurance
companies currently use AI-driven tools to gather deeper data, more detailed data and
connect to consumers in more significant ways (Choudhary, 2020). By adopting AI,
short-term insurance companies can assess the damage to a motor vehicle at the
beginning of the claim process and generate estimates, which allows effectivity for the
insurers and greater clarity for the policy holders (Manning, 2020). AI can therefore be
regarded as an opportunity for short-term insurance companies.
Self-driving vehicles
As discussed earlier in this chapter, self-driving/autonomous motor vehicles, which are

still in the developing phase, will definitely have an impact on the motor vehicle
insurance industry, as motor vehicle insurance policies have to be adjusted
accordingly. Through automation, vehicles can operate more cheaply and
69
consistently, reliably and safely (Bornstein, Rakow & Clemente, 2017; Beachey,
2020).
In a self-driving motor vehicle, the technology in the vehicle takes over for an entire
trip. Limited self-driving motor vehicles are vehicles that have built-in technology which
takes over all the driving functions in certain traffic/environmental conditions (Venter
2017; Lutkevich, 2019). The self-driving vehicles will most likely reduce the risk of
vehicle accidents, but it will not eliminate the risk totally (Aon Risk Solutions, 2014:3;
Filiz, 2020). The industry perspective is that self-driving vehicles will likely reduce the
claim frequency and the monthly premium, and changes will have to be made to the
underwriting policy (KPMG, 2015b:21; O’Very, 2018).
Although self-drive motor vehicles are still in the testing phase and not available to
buy, there are a number of newly manufactured vehicles that individuals can purchase
that are fitted with assistive technology, such as autonomous braking. In the UK, the
first self-driving policy was introduced to accommodate these consumer needs
(Kollewe, 2016). In South Africa, insurance companies still have to formulate a
legislative framework for self-driving vehicles (Lovells, 2016; Jacobs, 2020b).
Cybercrime
While technological developments can be opportunities for short-term insurance

companies, there are also risks involved with these developments. The threat of
cybercrime, where these technological advances can be hacked, should be
considered (Epic MSLGROUP, 2017). Cybercrime is a rising concern in the short-term
insurance industry, especially because many of the functions in this industry are
computer-based (Sutherland, 2015; Faurie, 2019c; Johansmeyer, 2021).
There has been an increase in cyber-attacks in South Africa where cyber criminals
attack organisations, however, many organisations have made provision for these
types of attacks and have amended their strategies (Alfreds, 2016; Singh, 2021).
Short-term insurance companies should thus develop strategies to mitigate this threat.
With the introduction of self-driving motor vehicles, future cybercrimes should also be
considered by short-term insurance companies. Bornstein et al. (2017) state that the
future of self-driving vehicles is an overwhelming challenge and the stakes are high
when it comes to cybercrime. Fortunately, some of the cyber risks involved in self-
driving motor vehicles have been confronted before. A potential danger was
70
highlighted in 2015, when two computer security specialists (white hat hackers) broke
into the protected system and network of a Jeep Cherokee and cut its transmission on
the highway. This was part of a research initiative and Chrysler had to recall 1.4 million
motor vehicles to resolve the deficiency in the electric control units (ECUs), to ensure
that hackers cannot gain access to the ECUs of these motor vehicles in the future
(Toews, 2016).
Hackers can also obtain personal information from the ECU of a vehicle, and this is
also a risk that motor vehicle manufacturers are aware of (Ranft et al., 2016:17). The
risks that have been confronted before can be used as learning opportunity and motor
vehicle manufacturers can improve the self-driving motor vehicles’ ECUs to prevent
hackers from gaining access to these systems (Nash et al., 2017:14). The automotive
industry can still make improvements on these vehicles as they can use the advice
related to the cyber security systems from the banking and financial services
(Bornstein et al., 2017; Noonan, 2019).
Short-term insurance companies can utilise the technological developments

discussed above to design tailor-made insurance services (Schliesser, 2014:25;
Balasubramanian et al., 2021). Insurance companies in the short-term insurance
industry have to evaluate the technological environment constantly as this is a fast-
changing environment that offers a number of opportunities. However, if short-term
insurance companies do not adapt to these changes competitors will implement these
changes first.
Economic environment
An organisation has no control over the economic environment, as factors, such as
inflation, exchange rates, recessions, and monetary and fiscal policy impact the
organisation (Erasmus et al., 2019:128). This environment also includes factors, such
as the personal disposable income and the purchasing behaviour of consumers. A
consumer’s disposable income is affected by the countries’ inflation, exchange rates,
recessions, and economic growth rates (Ferreira, 2011:21; Cherunilam, 2016:75).
During 2015, the weakening rand had a negative effect on the GDP rate. In addition,
the possible downgrade of South Africa's sovereign rating was another challenge for
short-term insurance companies (KPMG, 2016:137). The economic growth is of
enormous concern for short-term insurance companies and the consumers. Short-
71
term insurance companies are already struggling to cope with the high inflation rate,
as the value of claims is higher due to the imported parts that are needed to repair the
damage to motor vehicles. The weakening in the rand is another reason why insurance
premiums continue to increase (Roberts, 2016b). Consumers are becoming poorer,
and the rise in the unemployment rate is one of the reasons why consumers’
disposable income has decreased. Thus, short-term insurance services are seen as
luxury services, as consumers cannot afford these services anymore (KPMG,
2016:95).
At the beginning of 2017, two of the foreign-currency rating companies (Standard &
Poor’s Global Ratings and Fitch Group) downgraded South Africa to junk status (BB+),
which is the first level of the sub-investment grade. This downgrade heightened the
economic concerns amongst short-term insurance companies (Pearson, 2017). This
event impacted the short-term insurance companies and consumers, as the motor
vehicle repair costs will increase due to the motor vehicle parts being imported and it
will cause an escalation in premiums. In addition, the country’s downgrade to BB+
status and the low interest rate forced insurance companies to relook their financial
and strategic planning (Danckwerts et al., 2020:3-4).
It was predicted that in 2017, the living costs of consumers will rise and short-term
insurance will become unaffordable for consumers (Govender, 2017). The current
financial distress of consumers during the COVID-19 pandemic can also impact the
penetration rate when consumers decide to cancel their current motor vehicle
insurance policies (Peyper, 2021). Short-term insurance companies will have to
identify innovative ideas to manage these threats in the economic environment.
Social environment
The social environment consists of all the consumer characteristics, such as
demographic variables, age, culture, geographic location and so forth. These
characteristics are the aspects that affect the consumers’ buying behaviour (Del
Marmol & Feys, 2015:10; Perera, 2017:12). The consumers’ lifestyles, habits and
values can influence the organisation (Erasmus et al., 2019:128). For instance, the life
stage of a consumer can affect the consumer’s behaviour and purchase decisions,
which in turn, affects the organisation. The personality of consumers can also
72
influence the consumers’ purchase behaviour, as well as how and when they use the
products/services of an organisation (Middlebrook, 2020).
A report compiled by PwC (2014:54) revealed an important demographic factor for

short-term insurance companies. They found that the black middle class is emerging
in the industry. They also found that changes in literacy and an increase in education
can encourage short-term insurance companies to improve the awareness and
consumption of insurance services. Another report by Delaporte and Bastid (2016:5)
indicated that a new type of consumer has started to make a mark in the global short-
term insurance industry. This is the Generation Y, also known as the Millennial
generation. The millennial generation was born in the 1980s and 1990s and they are
comfortable with digital technology. These individuals are current and immediate
future potential consumers, and these are the individuals that are starting to make a
mark in the global short-term insurance industry (Delaporte & Bastid 2016:5). Some
short-term insurance companies can view this as a threat, and it can disrupt the way
traditional short-term insurance companies do business, as many make use of brokers
and minimum technological channels.
The social unrest in South Africa was amplified due to the COVID-19 lockdown, and
service delivery protests and riots are likely to increase, as a result (Moodley, 2021).
As explained earlier, this does not directly impact insurance companies in terms of
personal motor vehicle insurance (as it is covered by an add-on to the policy by
Sasria), but consumers will ask about this and it should be explained to them. More
consumers are working from home, another consequence of the COVID-19, and this
has a direct impact on the consumers’ risk profile for short-term insurance companies
(Moodley, 2021). Short-term insurance companies should take into account all the
social environmental changes and should ensure that the business strategy is
amended to make provision for these changes.
Physical environment
The physical environment refers to the natural resources in a country, which the
organisation can use as inputs to produce end-products and services (Aghazadeh,
2016:262). This explanation of the physical environment does not explain how this
environment can affect short-term insurance companies. However, this environment
can account for risks in the short-term insurance industry. For example, a hailstorm in
73
a certain area can cause damage to motor vehicles, which will result in a high claim
rate during a certain period, and therefore, the physical environment is the reason why
some risks emerge in the short-term insurance industry. The worrying climate change,
natural disasters and weather conditions are also aspects that can impact on this
environment (Del Marmol & Feys, 2015:11; Lappeman et al., 2021). All the aspects of
the physical environment that impact the short-term insurance companies are
discussed below.
The significant hailstorm that occurred in Gauteng in 2013 resulted in great losses for
insurance companies. The losses experienced from this storm exceeded the total
losses sustained from the six hailstorms of 2012. In order to reduce future potential
insurance losses, insurance companies started to inform consumers about expected
hailstorms to mitigate the risk (Vazeer, 2016a). The drought of 2015 and 2016 was the
worst drought in South Africa in 111 years. Water shortages were a concern and
harvesting losses increased. Furthermore, these conditions increased the chances of
fire losses for the consumers of short-term insurance companies (KPMG, 2016:95).
Weather conditions thus have an impact on the short-term insurance companies as it
can affect their claim rates and income (Du Plessis, 2017). There is a continuing
concern that weather catastrophes will impact the sustainability of insurance
companies (Madzingira, 2021). Short-term insurance companies have to be able to
act proactively in these conditions to minimise their risks.
Institutional-governmental environment
The institutional-governmental environment involves the political and legislative
components, and organisations should establish how the legislation and political
aspects can influence the organisation (Erasmus et al., 2019:128). The short-term
insurance industry is governed by the National Treasurer (NT) and the FSCA (Still &
Stokes, 2016:253; Krige & Laskov, 2018).
The South African short-term insurance industry is under increased legislative

pressure, but at least, the industry is well legislated (INSETA 2014:16). There are
several pieces of legislation that have an impact on how short-term insurance
companies should operate, specifically, as related to motor vehicle insurance policies.
These legislations are the Consumer Protection Act (Act No. 68 of 2008), Financial
Advisory and Intermediary Services Act (Act No. 37 of 2002), Companies Act (Act No.
74
61 of 1973), Financial Institutions Protection of Fund Act (Act No. 28 of 2001),
Financial Services Scheme Act (Act No. 37 of 2004), Inspection of Financial
Institutions Act (Act No. 80 of 1998), Short-Term Insurance Act (Act No. 53 of 1998),
Protection of Personal Information Act (Act No. 4 of 2013), Financial Intelligence
Centre Act (Act No. 38 of 2001) (Still & Stokes, 2016:260-271), and the Financial
Sector Regulation Act (Act No. 9 of 2017) (Gibson, 2021). Policies that short-term
insurance companies offer to consumers should be in line with these acts (Tshidi &
Sithole, 2016:9).
The Consumer Protection Act (Act No. 68 of 2008), stipulates that any consumer
agreement (policy), which intends to limit the risk or liability of a supplier (a short-term
insurance company), must be highlighted to the consumer in writing and in simple
language to ensure that the consumer understands the facts, nature and effects of the
agreement (Joffe, 2011; Barendse, 2015; Dinnie, 2015). The Protection of Personal
Information Act (Act No. 4 of, 2013) ensures that short-term insurance companies
conduct themselves in a responsible manner when collecting, processing, storing and
sharing personal information of consumers (Tadman, 2017; Sithole, 2021). Short-term
insurance companies can be held liable if any personal information about a consumer
is disclosed (Financial Services Board, 2016:6). These are two examples of how acts
direct the conduct of short-term insurance companies in relation to the policies offered
to consumers. Short-term insurance companies had until 30 June 2021 to ensure that
their daily operations comply with the POPI Act. Failure to comply can face penalties
by the Information Regulator (Sithole, 2021).
Political risks have consequences that are difficult to anticipate (Cape Business News,
2017; Buthelezi, 2021). With regard to a short-term insurance company in South
Africa, political action will not necessarily be focused on short-term insurance
companies, but the political actions, such as strikes that occur, can present risks for
these organisations, as it can result in policy claims. Civil unrest is on the rise in South
Africa (Coetzee, 2021), which is a matter of concern for insurance companies
(Buthelezi, 2021). However, as discussed earlier, short-term insurance companies do
not cover these events, and an additional Sasria service should be included in a policy.
Sasria provides special risk cover, such as civil commotion, public disorder, strikes,
riots and terrorism, for all individuals that own assets in South Africa. Sasria SOC Ltd
is also part of the short-term insurance industry as this is a special risk short-term
75
insurance company (Saria SOC Ltd, n.d.). All these political aspects should be
assessed and the legislation impacting the short-term insurance companies should be
taken into account to direct their conduct when offering policies to consumers.
International environment
The international environment consists of the local and foreign political trends or
events that may have an impact on an organisation (Erasmus et al., 2019:128).
Changes in the international environment of the short-term insurance industry will
impact the South African short-term insurance industry. Preparations should be made
to update strategies for the changes to come.
Regulatory changes in the international market were identified as one of the risks
faced by the international insurance industry, as there are broadening costs involved
and it can sidetrack management (Strydom, Naidoo & Botha, 2014:23; Chakar,
2020:5). This global regulatory/legislative pressure is already impacting the South
African short-term insurance industry and innovative ways to overcome this challenge
should be considered.
At a global level, several digital movements are rapidly emerging to form completely
new types of insurance organisations. The consistently connected consumer who
expects fulfilment when seeking information about the insurance services that are
available, acts as the motivation behind this new type of organisation. The delivery of
tailored digital products, IoT that increases the need for streaming analytics, and an
increased focus on algorithmic risk assessments are examples of current global
developments (Daskal, 2021). The global shift has already impacted the South African
short-term insurance companies, as was discussed earlier under the technological
environment. In general, the short-term insurance industry of South Africa has been
slow to adapt to these digital movements (Moyo, 2016).
The two foreign-currency rating companies that downgraded the country can also be
seen as an international factor that impacts the short-term insurance companies in
South Africa. This was discussed in the economic environment section above. The
self-driving motor vehicle is another international movement that will affect the South
African insurance companies, but no legislative framework has been introduced thus
far. Short-term insurance companies thus have to consider these changes when they
develop their policies.
76
Going forward, competition will escalate globally and partnerships between tech-savvy
(technological advanced) officials and well-funded new market entrants are expected
to increase. This will promote further innovation and change in the industry. At the
same time, more comprehensive data regulations will develop and insurance
regulators, as well as data privacy rules, will become more important. Insurance
companies will be able to use data to influence the level of service customisation for
consumers (Ferenzy et al., 2016:20). These are aspects short-term insurance
companies have to investigate to ensure that they compete as sustainable short-term
insurance companies in the market.
As can be gathered from all the examples and events that have occurred in the macro-
environment of short-term insurance thus far, the impacts are immense. Short-term
insurance companies should manage and evaluate the macro-environment regularly
to mitigate future risks.
CONCLUSION
To conclude, the global insurance industry has a rich and ancient history. Insurance
has evolved in South Africa from when the first insurance company was established
in Cape Town in 1835. The main components in the South African insurance industry,
long-term insurance and short-term insurance, were discussed. The role of the FSB
and FSCA in the short-term insurance industry was highlighted. The associations that
support the short-term insurance industry were specified, and the different types of
short-term insurance services available in the country were explained. Developments
in the short-term insurance industry of the country were summarised. The distinction
between direct short-term insurance companies and traditional short-term insurance
companies was made. This study included only direct short-term insurance
companies, as these companies have more detailed information about their
consumers.
This chapter presented that South Africa is the only country in Africa that has an
established, well-developed, mature and comprehensive short-term insurance market.
Motor vehicle insurance is part of the short-term insurance industry, and is the type of
insurance that is the most sold insurance in the short-term insurance industry of South
Africa. The importance of motor vehicle insurance in South Africa, the nature of this
type of insurance, and the different types of motor vehicle insurance were discussed.
77
As the personal business line within the short-term insurance industry is the largest,
this study focused only on personal motor vehicle insurance. The components of the
business environment, namely, the micro, market and macro-environments within the
short-term insurance industry were reviewed. Lastly, the variables or factors of the
business environments that influence the short-term insurance companies were
comprehensively explained. As can be gathered from the discussion of the business
environment in which short-term insurance companies operate, it is clear that the
variables in the micro, market and macro-environments have a huge impact. Short-
term insurance companies should manage and evaluate these three sub-
environments regularly to alleviate the future risks.
78
CONSUMER PURCHASE DECISION-MAKING PROCESS
INTRODUCTION
Marketing is the central bridge between the organisation and its business environment,
as this function connects a company and its market, which consists of the consumers
and potential consumers (Erasmus et al., 2019:436). The basic underlying principle of
the marketing function in an organisation is to understand the consumers (Sethna &
Blythe, 2019:6). This includes understanding the individuals’ buying behaviour, habits,
needs, and how consumers make their purchasing decisions (Parumasur & Roberts-
Lombard, 2014:2; Cant, 2020:3).
Consumers refer to the individuals or groups, for example, families or businesses, who
purchase, use, maintain and dispose of products and services to increase their life
satisfaction and fulfil their needs (Goldsmith, 2021:9). Personal motor vehicle
insurance consumers will therefore be the individuals or groups who purchase,
maintain or cancel their personal motor vehicle insurance policies. Potential
consumers are also part of the market/audience, but these individuals do not
necessarily purchase and consume the product or service (Lappeman et al., 2021:5).
Therefore, motor vehicle drivers that do not purchase/have personal motor vehicle
insurance will be potential consumers. Since consumers and their decision-making
when purchasing an organisations products/services are central to the marketing
strategies of a company, it is important to give a brief background of what marketing
entails before a discussion of the consumer decision-making models is given.
The marketing function of a business, such as a short-term insurance company, will

have to be aware of the individuals’ needs, wants, and demands to satisfy these
needs, wants and demands (Schiffman et al., 2014:6; Kotler & Armstrong, 2018:17).
These aspects are discussed in Section 3.2.1, which deals with the components of the
marketing concept. The marketing mix and service mix (motor vehicle insurance is a
service) elements are also explained in this chapter. Consumer research is discussed
and the consumer behaviour concept is reviewed. Roles in the decision-making
process, decision-making rules, different types of decision-making, and decision-
making styles are also considered. These aspects are important to the marketing
79
function of short-term insurance companies to determine individuals’ behaviour during
purchasing decisions. In addition, the theories that are applicable to analysing the
purchase decision process in efforts to increase consumer demand in the personal
motor vehicle insurance market are explained.
Most of the consumer decision-making theories or models assume that the consumer
decision-making process consists of several steps. It is, however, important to point
out that these steps may vary from product to product or from service to service, but
individuals all pass through similar types of steps (Prasad & Jha, 2014; Mahmood &
Baharun, 2019). Several consumer decision-making models that have been
developed by scholars over the years, will be evaluated. All the limitations and
expansions of the different decision-making models are considered to develop a
conceptual purchase decision model for personal motor vehicle insurance. Lastly, the
conceptual consumer purchase decision for personal motor vehicle insurance
(CPDFPMVI) model is presented.
THE MARKETING FUNCTION OF A SHORT-TERM INSURANCE

COMPANY
As explained in Chapter 2, marketing is a business function of a short-term insurance

company. The marketing function of an organisation is responsible for identifying and
anticipating consumers’ needs and wants, as well as introducing and delivering
products and/or services to satisfy the consumers’ needs (Lantos, 2011:3-4; Erasmus
et al., 2019:436). Consumer preferences are changing and the business environments
are rapidly evolving, therefore the marketing function of a business needs to have a
good understanding of their consumers’ behaviour (Schiffman et al., 2014:6; Coggins,
Robinson & Wilkie, 2020). In order to understand the role of the marketing function in
a business, the components of the marketing function will be discussed next.
Components of the marketing function
An uncomplicated definition of marketing is the managing of profitable customer

relationships (Porral & Stanton, 2017:19). The American Marketing Association (2021)
defines marketing as “…the activity, set of institutions, and processes for creating,
communicating, delivering, and exchanging offerings that have value for customers,
clients, partners, and society at large.” Consistent with this definition, is the one from
80
Pride et al. (2015:5) that defines marketing as “the process of maximising returns to
stakeholders by developing exchanges with valued customers and creating an
advantage from them.” According to Armstrong et al. (2015:4) and Kotler and
Armstrong (2018:82), an organisation's marketing function is responsible for delivering
value to consumers, managing consumer relationships, ensuring that organisational
objectives are achieved, and at the same time benefits should be provided to
shareholders. The basic activities of a marketing function are thus to understand
consumer needs, determine a target market, develop products/services that offer
value, price the products/services correctly, and distribute and promote the
products/services correctly (Nieuwenhuizen, 2015:97; Botha, 2018:33). These
activities form the components of the marketing function, which will be discussed
below. However, the marketing concept, which forms the basis of the marketing
function, is first explained.
The marketing concept

Babin and Zikmund (2016:8) define the marketing concept as a predominant idea in
modern marketing thinking that focuses on how an organisation provides value to the
consumers beyond the physical product and production processes. Schiffman and
Wisenblit (2019:34) concurred with the description above, by stating that the marketing
concept has to do with identifying a target market and fulfilling the needs of this target
market. Kotler and Armstrong (2018:37) explain the marketing concept as an
organisation’s achievement of goals, depending on determining the needs and wants
of their target market and delivering the desired satisfaction more effectively than their
competitors. When an organisation adopts the marketing concept, it develops a
marketing orientation (Babin & Zikmund, 2016a:8; Cant, 2020:8).
Strydom (2018:304) agrees with this view by stating that the marketing concept affirms
that an organisation should attempt to satisfy consumers’ needs through coordinating
activities to allow the organisation to achieve its objectives. The essence of the
marketing concept lies in four principles, namely, profitability, consumer orientation,
social responsibility and organisational integration (Nieuwenhuizen, 2015:96;
Erasmus et al., 2019:439).
The survival of an organisation depends on its profitability, as an organisation has to

experience profitability to be sustainable (Babin & Zikmund, 2016a:10). Organisations
81
should rather attempt to achieve a specific rate of return in the long term, rather than
achieving high returns in the short-term (Strydom, 2018:300). Therefore, long-term
profitability is the primary objective of any organisation in the free-market system, as
it can result in customer satisfaction and retention (Erasmus et al., 2019:439).
Consumer orientation, also known as market orientation, relates to the organisation’s

task to research and determine consumers’ needs and wants, and then designing a
system to satisfy the needs and wants of consumers (Boone & Kurtz, 2016:10;
Tomczak, Reinecke & Kuss, 2018:3). This means that all the actions of an organisation
should be aimed at satisfying the consumers’ needs, demands and wants (Babin &
Zikmund, 2016a:8).
The social responsibility principle places the focus on how marketing affects the
society as a whole in operating with scarce resources, environmental destruction and
worldwide competition (Strydom, 2018:303). Organisations are responsible for the
community in which the marketing task is performed (Cant, 2020:9), and can engage
in projects where money is spent on improving the community. For example,
organisations can sponsor sports events and educational institutions, such as at
schools and universities. However, organisations need to be mindful and act
responsibly by abiding by the laws of the country in which they operate (Erasmus et
al., 2019:441).
Organisational integration is the principle that emphasises the need for marketing
information to be provided and used by all the organisation's functions (Babin &
Zikmund, 2016a:10). All the decisions within an organisation should be coordinated in
such a way that it will lead to the successful marketing of products and services
(Strydom, 2018:304). Therefore, all the functions of an organisation should work
together to achieve the same objective in marketing products and services to the
consumers. The consumer needs, wants and demands were mentioned in the
consumer orientation principle above, but are explained in detail below.
Consumers’ needs, wants and demands

Organisations have to understand their consumers’ needs, wants and demands
(Schiffman et al., 2014:73; Saurav, 2020) in order to provide customer satisfaction.
According to Kotler and Armstrong (2018:30), human needs “…are the states of felt
deprivation.” The human needs include the basic physical needs for food, clothing,
82
water, and so forth; social needs for belonging and affection; and self-esteem or self-
actualisation needs for knowledge and self-expression (Kotler & Armstrong, 2018:30).
A need occurs when an individual’s existing level of satisfaction does not meet the
desired level of satisfaction (Ferrell, Hartline & Hochstein, 2021:118).
Wants are defined by Kotler and Armstrong (2018:30) as “…the form human needs
take as shaped by culture and individual personality.” A want can also be explained
as the individual’s desire for a certain product or service which will satisfy the need
(Ferrell et al., 2021:118).
When individuals have the buying power to purchase a product or service that will
satisfy their needs, it is referred to as a consumer demand (Kotler & Armstrong,
2018:30). The distinguishing difference between an individual’s wants and demands
is that a demand only occurs when the individual is able and willing to pay for the want
(Ferrell et al., 2021:142). Need recognition is an important step in most of the
consumer decision-making modules designed by scholars (see Section 3.6).
Consumer satisfaction can be described as the consumer’s fulfilment response after

purchasing a product or service (Smith, 2019:22). Schiffman et al. (2014:25) and
Acevedo (2018) state that consumer satisfaction relates to the consumer’s perception
of the performance of a specific product or service in respect to the consumer’s
expectations. Solomon, Russell-Bennett and Previte (2013:302), and Staffaroni (2019)
describe consumer satisfaction as a reflection of how a person feels about a product
or service after it has been purchased from an organisation.
As can be understood from the definitions, a consumer is satisfied when the

performance of a product or service is in line with the consumer’s expectations.
Satisfied consumers can lead to personal recommendations, and in such a way, the
organisation can gain new consumers (Sethna & Blythe, 2016:104; Strydom,
2018:308). In order to better identify and satisfy the needs, wants and demands of
individuals in the market, the market can be segmented into different groups. This
concept of market segmentation is discussed below.
Market segmentation
Individuals in a target market can be classified into groups according to their common
needs and characteristics. This is called market segmentation (Schiffman et al.,
2012:6; Dolnicar, Grün & Leisch, 2018:6). Market segmentation is defined as the
83
process of splitting individuals in the market into different groups and segments
(McDonald & Dunbar, 2012; Cant, 2020:164). A more comprehensive definition of
market segmentation is the process where a group of individuals within a broad mass
market is divided into smaller groups which will consist of people with similar
characteristics who have similar expectations and who are seeking similar benefits
from a product or service (Burrow & Fowler, 2016:187; Strydom, 2018:312). For
example, short-term insurance companies segment their markets into smaller
homogeneous groups of individuals to ensure that the correct marketing efforts are
directed at the correct group of individuals.
The geographical location, demographic, physiographic and behavioural variables can

be used as the basis or criteria to segment a market (Burrow & Fowler, 2016:187-188;
Strydom, 2018:314). The use of geographic location (by region of a country or the
world) as basis to segment a market is based on the core assumption that individuals
have diverse needs and wants because of where they live (Cahill, 2006; McDonald &
Dunbar, 2012; Lamb et al., 2018:399). However, geographic location as a
segmentation approach does not ensure that all the individuals in a location will make
the same purchase decisions, but it will assist in finding and establishing general
patterns (Boone & Kurtz, 2016:277; Rajagopal, 2019:89).
Organisations often segment markets on the basis of demographics, as this

information is widely available and it can also relate to the individuals’ purchase
decisions (Lamb et al., 2013:275, 2018:337). The general demographic variables
include age, gender, income, household size, education, ethnic background and family
lifestyle (Boone & Kurtz, 2014:170; Willan, 2021).
Physiographic segmentation is segmentation on the basis of variables such as

personality, motives, lifestyles and geodemographics, which cluster potential
consumers into neighbourhood lifestyles (Lamb et al., 2017:143). Behavioural
segmentation is based on what individuals do, for example, what hobbies they have,
what food they eat, how they travel, where they work and how they spend their spare
time (Sethna & Blythe, 2019:11).
A general consumer profile can be compiled only after an organisation has segmented
its market and categorised all the homogenous groups (DeVault, 2021). A general
consumer profile will allow the organisation to select a promising segment that can
84
potentially be targeted. The selected segment will form the target market, and the
organisation views this market as potential consumers (Erasmus et al., 2019:447). A
target market refers to a group of potential homogeneous individuals who have similar
needs that the organisation decides to serve (Armstrong et al., 2015:185; Berkowitz
2017:81; Ward, 2020).
After a target market has been determined, the organisation can position itself by
creating an image and identity about the products/services offered in the potential
consumers’ minds (Schiffman & Wisenblit, 2019:35). Short-term insurance companies
should perform market segmentation to establish their target market, after which they
position the company, with the correct marketing efforts to form the image and identity
about personal motor vehicle insurance policies in the potential or existing consumers’
minds. Another essential marketing component that plays an important role in
consumer behaviour is the marketing mix, which is discussed below.
The marketing mix

The concept ‘marketing mix’, also known as the ‘four Ps’, refers to the four marketing
activities, namely, product, pricing, distribution and promotion. Organisations have to
plan, organise, implement and control these four mix elements to ensure that
consumers’ needs, wants and demands are met (Percy & Rosenbaum-Elliott,
2016:143; Pride & Ferrell, 2021:6). The original marketing mix, which consisted of 12
elements (product planning, pricing, branding, channels of distribution, personal
selling, advertising, promotions, packaging, display, servicing, physical handling, as
well as fact finding and analysis), was developed by Neil Borden in the 1950s.
However, Eugene McCarthy simplified it to the current basic marketing mix in 1960
(Baines, Fill & Rosengren, 2017:18). This marketing mix is viewed as the basic
marketing toolkit that the marketing function in an organisation can use to implement
their marketing plans (Ramaswamy & Namakumari, 2013:173; Baker & Saren,
2016:205; Masterson, Phillips & Pickton, 2021).
As the understanding of the marketing functions within organisations developed,

marketing professionals and scholars suggested that the basic marketing mix should
be expanded (Pride et al., 2018:15). Two scholars, Booms and Bitner (1981), included
three additional Ps (physical evidence, process and people) into the basic marketing
mix. This was done specifically for services, as the services context differs from the
85
product context (Baines et al., 2017:20, 2021:12). The manner in which organisations
market products differs from how services are marketed. Thus, the expanded
marketing mix was proposed. The expanded marketing mix is also called the service
mix (Chaffey & Smith, 2017:54). Physical evidence, process and people are additional
Ps in the service mix, but another P (partnership) has been suggested by Pride et al.
(2018:17) to form part of the service mix. Each one of the Ps is explained below.
Firstly, the original ‘four Ps’ consist of the product, price, place and promotion. Product
as an element of the marketing mix, refers to decisions and actions associated with
the products/services an organisation manufactures or delivers to consumers and how
it meets the individuals’ needs (Percy & Rosenbaum-Elliott, 2016:144; Masterson et
al., 2021:25). The packaging and labelling of products are also included in the product
element of the marketing mix (Baines et al., 2021:11). With regard to short-term
insurance companies, the product refers to a service, and includes the motor vehicle
insurance policies and the insurance packages (as explained in Chapter 2, Section
2.6.2) offered to the market to satisfy the individuals’ needs.
The price element relates to the decisions and actions linked with determining the
prices of products and services to be offered. This includes every decision and activity
regarding the pricing objectives, pricing policies and ensuring that consumers receive
value for money (Pride & Ferrell, 2021:7). For example, short-term insurance
companies have to make decisions and take action regarding the prices of motor
vehicle insurance policies to ensure that consumers receive value for their money.
The place element refers to the distribution of products/services and includes the
distribution channels and physical distribution activities (Baker, 2014:296; Masterson
et al., 2021:25). This element is important, as the distribution strategies will ensure
that consumers find the products/services in the correct quantities, at the right time
and the right place (Boone & Kurtz, 2016:47). For consumers to be satisfied,
organisations such as short-term insurance companies have to ensure that products
and services are available at the right time and in/at convenient locations (Pride &
Ferrell, 2021:7) to satisfy the needs of their consumers. The distribution strategy can
include a direct approach, traditional approach, or may be a combination of the two
approaches, as discussed in Chapter 2, Section 2.4.2.2.
86
Promotion is the last element of the original marketing mix concept, and this element
refers to the use of communications to persuade individuals and create awareness
related to purchasing a product or service (Baines et al., 2017:40). This element is
also concerned with all the activities involved in informing and educating individuals
about product and service offerings (Pride & Ferrell, 2021:8). Promotion can be viewed
as the link between the sellers and buyers. Organisations can make use of
salespeople/brokers to communicate messages about a product or service, but they
can also convey the message directly through advertisements (Boone & Kurtz,
2016:48) or other promotion mix elements such as sales promotion, direct marketing,
interactive marketing, public relations and personal selling (Fennis & Stroebe,
2021:27). Short-term insurance companies have to consider the different ways of
communicating to inform, educate, persuade and remind individuals about their motor
vehicle insurance policies.
These ‘four Ps’ remain central to the marketing mix of an organisation as they
represent the core strategic marketing focus (Pride & Ferrell, 2021:6). However, short-
term insurance companies also have to incorporate the service mix, as these types of
companies offer a financial service to the market. The four additional Ps to form the
service mix include physical evidence, process, people and partnerships.
The physical evidence element relates to the tangible components regarding service
delivery, such as the building where the organisation operates from or the uniforms
the employees wear while they deliver a service (Baines et al., 2021:12). The
equipment used and the physical environment or atmospherics whilst delivering a
service also form part of the physical evidence element (Ramaswamy & Namakumari,
2013:713; Pride et al., 2018:16). The online physical presence also has to be in place,
and this refers to the online digital presence through using the brand image/logo on
the organisation’s website and updating the content regularly (Chaffey & Smith,
2017:94). For example, the physical evidence components within a short-term
insurance company will include the employees' uniforms (including the contracted-out
employees), the emails from consumers, the logo on the email, and so forth.
The process element involves the internal and external processes and transactions
and the communication within the organisation, which are requirements to be fully
operational (Chaffey & Smith, 2017:94). The timing of the service delivery will be
impacted by the processes used to deliver the service. The degree of personalisation
87
and automation whilst delivering a service is also important (Pride et al., 2018:16).
Short-term insurance companies should consider these activities to provide a quick
service. During the quotation stage, for example, the short-term insurance companies
can do this process quicker by using Big Data and automated computer systems
(discussed in Section 2.7.3.1).
People, as an element in the service mix, refers to the employees who work at an
organisation, in other words, the individuals who ensure that a service is delivered
(Sethna & Blythe, 2019:13). This also relates to how the employees treat consumers
while a service is being delivered (Pride et al., 2018:15). The human element in
offering services is the employees as they get paid to prepare, produce and deliver
the services (Masterson et al., 2021:268). The employees of short-term insurance
companies who interact with the consumer when a policy is sold or when a claim is
logged, thus constitute the people element.
The last service mix element is partnerships, and this entails the associations and
agreements with other organisations, where two organisations form business
partnerships, alliances and joint ventures (Pride et al., 2018:17). The fact that a major
deal of marketing today is based on strategic partnerships and alliances has provided
the need to include this element in the marketing mix (Chaffey & Smith, 2017:51).
More insurance companies focus on partnerships or associations to drive growth,
create new channels and drive innovations (Reader, 2016:4). Insurance companies
are starting to recognise that the greatest opportunities will likely come from outside
of the organisation. An example of how short-term insurance companies can include
this element is to form a partnership with brokers and with specific panel beaters to
ensure that a quality service is delivered to the consumer. Another example is where
short-term insurance companies form partnerships with academics to generate new
ideas (Perkmann & Salter, 2012; OECD, 2015:42).
A short-term insurance company will benefit from including these service mix elements
in their marketing plan, since the implementation of these elements can assist in
successfully directing the marketing efforts to the target market. The way individuals
think, feel and act during and after a purchase relate to the consumers’ behaviour. The
consumer behaviour aspect is discussed below.
88
CONSUMER BEHAVIOUR
Jacoby (1978:87) was one of the first authors who stated that, in essence, consumer
behaviour “…encompasses the acquisition, composition, and disposition of goods,
services, time and ideas by decision-making units (for example, individuals, families,
organisations, etc.).” Lamb et al. (2011:191) provide a more general definition of
consumer behaviour, which is: “Consumer behaviour describes how consumers make
purchase decisions and how they use and dispose of purchased goods or services.”
Schiffman et al. (2012:2), however, define consumer behaviour, as “… the behaviour
that consumers display in searching for, purchasing, using, evaluating and disposing
of products and services that they expect will satisfy their needs.” Schiffman et al.
(2014:4) as well as West, Ford and Ibrahim (2015:80) agree with this definition of
consumer behaviour.
Consumer behaviour also refers to the study of how individuals purchase products
and services (Kardes, Cronley & Cline, 2015:7; Hoyer et al., 2018:5). It is, however,
clear from all the above definitions that consumer behaviour suggests more than just
how products and services are acquired. The three activities highlighted in all the
definitions are how consumers acquire, use, and dispose of products and services.
When consumers acquire products and services it also involves their decisions about
time and money (Hoyer et al., 2018:6). Personal motor vehicle insurance consumers
can obtain motor vehicle policies from brokers, online directly from short-term
insurance companies, or contact the companies directly to purchase the policy. The
time and effort spent to obtain the desired personal motor vehicle insurance policy will
also be part of the acquisition activity. How consumers use products and services
refers to consuming (Sethna & Blythe, 2016:7; Rajagopal, 2019:4).
Regarding personal motor vehicle insurance, it will be difficult to determine how

consumers use (or consume) a policy, as it is a service offered until a claim is logged.
To dispose of a product or service refers to how the consumers get rid of the product
or services when it is worn out or used up (Hoyer et al., 2018:6; Sethna & Blythe,
2019:9). How consumers cancel or terminate their personal motor vehicle insurance
policies with short-term insurance companies when they do not want to have insurance
or cannot afford it anymore, will reveal the disposal activity involved in consumer
behaviour.
89
According to Schiffman et al. (2014:4), the study of consumer behaviour involves how
individuals in a market think (the purchase decision-making process), feel (the
emotions of consumers) and behave (the physical actions of consumers which is the
result of decisions and feelings). In other words, consumer behaviour involves how
individuals spend their available resources, such as money, time and effort, on
products and services. However, Sethna and Blythe (2019:10) illustrated that the basic
attitudes (made up of thought, emotion and intended behaviour) and the personal and
environmental factors are the general aspects of consumer behaviour. The purchase
decision-making process is analysed and discussed in more detail in Section 3.6 of
this chapter.
Consumer behaviour research is important for organisations. The information obtained

can assist strategic marketing in terms of segmentation, targeting, and positioning by
identifying the consumer differences and similarities to ensure correct consumer
profiling is performed (West et al., 2015:82; Rajagopal, 2019:3). Other benefits of
conducting consumer behaviour research, include improving business performance
by creating consumer-focused strategies, the ability to influence public policies (for
example, if the study is about the nutrition of foods), and it can assist in educating
consumers to make better decisions (Kardes et al., 2015:11-12; Ahmed, 2019b). The
sections below review the type of consumers and consumers' mental and physical
activities.
Types of consumers
Consumers refer to the individuals or groups, for example, families or businesses, who
make the decisions to purchase, use, maintain, and dispose of products and services
to increase their life satisfaction and fulfil their needs (Goldsmith, 2021:9). A consumer
also pays for the product or service (Parumasur & Roberts-Lombard, 2014:2; Khartit,
2021). The individuals or groups who decide to purchase motor vehicle insurance and
pay for the policy will be viewed as motor vehicle insurance consumers.
Burrow and Fowler (2016:316) indicated that service organisations, such as short-term
insurance companies, have two types of consumers, namely, individual consumers
and business consumers. Parumasur and Roberts-Lombard (2014:2) and Long (2020)
agreed that individual consumers purchase for their households, whereas business
consumers are part of households but purchase in the capacity of a business. The
90
individual consumers are thus the individuals that purchase personal motor vehicle
insurance and the business consumers are the businesses (big or small) that purchase
motor vehicle insurance for their company-owned motor vehicles. This study focused
be on the individual consumers who decided to purchase personal motor vehicle
insurance, but this study also investigated why individuals decide not to purchase
personal motor vehicle insurance. Individuals who follow the purchase decision-
making steps but decide not to purchase, are referred to as potential consumers.
Mental and physical activities of consumers
The definition of consumer behaviour refers to both mental and physical activities. The
mental, also known as cognitive activities, relate to the individual’s thought processes,
their opinions, beliefs, attitudes and intentions about a product or service (Kardes et
al., 2015:10; Pride & Ferrell, 2021:145). These activities occur when the individuals go
through the decision-making process and evaluate how well the product/service
satisfies their need (Parumasur & Roberts-Lombard, 2014:5; Erasmus et al.,
2019:459).
The physical activities involved with consumer behaviour relate to actions that are
taken by the consumers regarding how they purchase and how they compare the
information about products/services (Einstein, 2017:55). The different mental and
physical activities of consumers are discussed in the sections below.
CONSUMER DECISION-MAKING
A decision is an individual's response to a situation, and it entails judgment,

expectations and evaluations (Oliveira, 2007:16; Ismagilova et al., 2019:11). Another
definition of a decision is that a decision occurs when an individual selects an option
between two or more available options (Schiffman et al., 2012:63; Smith, 2019:90).
Individuals can therefore only select/choose an option if there are alternative options
available.
In the personal motor vehicle insurance market, consumers and potential consumers
have numerous options to select from, as quite a number of short-term insurance
companies offer personal motor vehicle insurance. Consumers have to decide to
purchase or not to purchase, how to purchase the service (which distribution channel),
and how to pay for the service (Schiffman et al., 2012:63; Acevedo, 2018:17). This
91
study focused on the consumer and potential consumer’s decision to purchase or not
to purchase personal motor vehicle insurance.
The steps that consumers generally follow when they purchase products or services
is known as the consumer decision-making process (Lamb et al., 2011:191; Cant,
2020:125). In other words, when an individual decides to purchase a personal motor
vehicle insurance policy, he/she passes through various steps, which form part of the
consumer decision-making process. Parumasur and Roberts-Lombard (2014:264)
define the consumer decision-making process as “…a sequential and repetitive set of
psychological and physical activities ranging from problem recognition to post-buying
behaviour.” There are, however, some individual (motivation, social, perception,
learning, personality and attitudes) and external (marketing efforts of company,
economic environment and the consumer’s reference group) influencing factors that
can impact an individual’s decision-making (Kaser, 2013:48; Stankevich, 2017:10).
Furthermore, the consumer decision-making process, also known as the purchase

decision-making process, can be explained as an outline of how individuals generally
make decisions. Not all the individuals who follow these purchase decision-making
process steps become consumers, as they might abandon the whole idea at any point
in the process (Lappeman et al., 2021:154). Even though most scholars refer to the
consumer decision-making process, all the individuals in a market that have a need
for a certain product or service will follow these steps. Not only the consumers, but
also potential consumers, follow the consumer decision-making process, whether they
actually purchase or not.
Consumer decision-making is similar to problem-solving because a problem occurs

when individuals perceive a difference between their current situation and what they
want (Parumasur & Roberts-Lombard, 2014:263-264; Khanna & Hanspal, 2019). Due
to the individual’s problem, a need exists and this need should be satisfied. All the
steps included in the consumer/purchase decision-making process, as well as the
influences on the individuals’ decisions, were introduced in Chapter 1. These steps
will be discussed comprehensively in Section 3.6, where they will be illustrated in the
different decision-making models. The roles in the decision-making process, decision-
making rules, types of decision-making and the different decision-making styles are
explained below. These can impact the short-term insurance company’s
understanding of the individuals’ purchasing decision-making behaviour.
92
Roles in the decision-making process
Organisations can attempt to understand their consumers’ behaviours during

purchase decision-making by identifying the different roles an individual can play in
decision-making (Gbadamosi, Bathgate & Nwankwo, 2013:73; Aghazadeh,
2017:332). The five different roles that can be identified in most decision-making
processes are initiator, influencer, decider, purchaser and user, as explained below
(Parumasur & Roberts-Lombard, 2014:4; Kotler & Armstrong, 2018:192):
• The initiator is the individual that first suggests making the purchase (Lamb et al.,
2018:112).
• The influencer is the individual whose views or advice is taken into consideration
when the purchase decision is made, and this individual thus influences the
purchase decision (Armstrong et al., 2015:150; Kotler & Armstrong, 2018:192).
• The decider is the individual who makes the decision to purchase or not to
purchase, what to purchase, how to purchase or where to purchase (Kotler et al.,
2013:228; Gbadamosi, 2019:296).
• The individual who physically makes the purchase is the buyer or purchaser (Lamb
et al., 2018:112).
• Finally, the user is the individual who uses the product or service (Gbadamosi,
2019:296).
By identifying the main participants and the roles played by participants in a decision-
making process, organisations can direct their marketing efforts accordingly (Kotler &
Armstrong, 2018:192). Short-term insurance companies can therefore benefit by
understanding the participants (the market) and the roles these participants play
during the decision-making process. This will allow them to design the correct
marketing efforts aimed at the correct participants.
Decision-making rules
The decision rules, also referred to as information processing strategies, are the
procedures used by individuals to facilitate brand choices/decisions (Sethna & Blythe,
2019:57). The two main categories are compensatory and non-compensatory decision
rules (Schiffman et al., 2012:77; Arunangshu, Pradip & Satakshi, 2021:94-95).
93
Compensatory decision rules refer to how the deficiencies in one criterion of the brand
can be compensation for the strengths or advantages of another criterion. Individuals
often use this rule as it emphasises the balance and trade-offs between options as
individuals search for their choice (Emilien, Weitkunat & Lüdicke, 2017:244). For
example, when an individual decides to purchase motor vehicle insurance, he/she
might be able to compensate on the fact that the car rental option is not included but
the price will be more affordable.
The non-compensatory decision rules are strategies used by individuals when a

product/service does not meet the decision rule for one attribute/feature and this
cannot be compensated for by its strengths or advantages in other areas (Sethna &
Blythe, 2019:56). In other words, this rule does not allow individuals to balance the
positive evaluation of a brand on one feature against the negative evaluation on
another feature. Three non-compensatory decision rules can be considered, namely,
the conjunctive decision rule, the disjunctive decision rule and the lexicographic
decision rule (Schiffman et al., 2012:78; Arunangshu et al., 2021:95-96), which can be
explained as follows:
• The conjunctive decision rule implies that the individual will establish a minimum
benchmark level/cut-off point to evaluate the brand feature. Therefore, brands that
do not meet the benchmark level will not be further considered (Sethna & Blythe,
2019:57). When applying this rule, an individual can use a certain price as a
benchmark/cut-off point to make a decision between the different personal motor
vehicle insurance policies.
• The disjunctive decision rule relates to when an individual sets a minimum cut-off
point separately for each feature. This is similar to the conjunctive decision rule,
but the cut-off level might be higher than that normally established in the
conjunctive decision rule (Schiffman et al., 2012:78; Arunangshu et al., 2021:95).
With this rule, an individual who is busy making the decision to purchase personal
motor vehicle insurance can use a certain price as the cut-off point, but the price
might be higher than that in the conjunctive decision.
• When individuals make a lexicographic decision, the rule is that the individual first
ranks the features according to perceived relevance or importance, and then the
various alternatives are compared in terms of the feature that is considered most
94
important. If one option is ranked the highest then this option will be selected
(Sethna & Blythe, 2019:56). For instance, when an individual considers technology
as the most important feature when selecting a short-term insurance policy, he/she
will have to work through all the quotations and rank them in order of importance.
The option with the most technological features will be the highest ranked and this
option will be selected.
Types of decision-making
The main types of decisions that consumers and potential consumers make during the
decision-making process are complex, dissonance-reducing, variety-seeking,
habitual/routine, and impulsive decision-making (Du Plessis, Strydom & Jooste,
2012:83-84; Masterson et al., 2021:99). In addition, the type of decision an individual
makes depends on the type of product/service purchased, the purchase situation, and
the level of purchase involvement (Lamb et al., 2011:197; Pride & Ferrell, 2021:146).
During complex decisions, individuals are highly involved in a purchase and they are
aware of significant differences between the brands. An individual usually makes this
type of decision when purchasing an expensive product/service that is not bought
frequently and involves risks, for example, when purchasing a motor vehicle,
computer, furniture, and so forth (Du Plessis et al., 2012:83; Berger-Grabner, 2021).
Dissonance-reducing decisions occur when individuals think there is a small difference

between brands, and although they are highly involved in the purchase, they put little
effort into searching for information about a brand. Purchasing electrical tools or
insurance are examples of a purchase where dissonance-reducing decisions occur
(Masterson et al., 2021:99).
Habitual/routine decisions are taken by individuals when there is a low involvement in

a purchase, and when they perceive few significant differences between the different
brands. If consumers are satisfied with a product/service, it will be purchased again,
therefore, forming a habit. Toothpaste and soap are examples of items purchased
during a habitual decision (Parumasur & Roberts-Lombard, 2014:271; Berger-
Grabner, 2021). Individuals make habitual/routine decisions when purchasing low-cost
items that require little search and decision effort, such as buying a bottle of water or
batteries (Pride & Ferrell, 2021:179).
95
A variety-seeking decision is made when individuals understand the criteria for
evaluating the product/service, but they have not yet established a clear brand
preference. Individuals have a low involvement in the purchasing process as they
purchase different brands (takeaway foods, restaurants, entertainment) for the sake
of having a variety (Kardes et al., 2015:231; Masterson et al., 2021:99).
An individual takes an impulsive decision if it is an unplanned action or a purchase is

made on the ‘spur of the moment’. Therefore, the decision is made at the point of
purchase. Items such as bubble gum and magazines are examples of items that are
purchased when an impulsive decision is made. (Du Plessis et al., 2012:84; Cant,
2020:125).
Individuals that review or consider purchasing financial service policies/products, such

as personal motor vehicle insurance, will engage in dissonance-reducing behaviour,
as they must make a decision between numerous similar alternatives (Belch & Belch,
2015:125; Katrodia, 2021:61). During this type of behaviour, individuals are highly
involved in the purchase, but they are of the opinion that there are only minor
differences between the insurance products offered by the short-term insurance
companies (Jooste, 2014:66; Techasurin et al., 2020:2). In other words, individuals
wanting to purchase personal motor vehicle insurance will typically shop around to see
what is currently available in the market, but they will purchase without thinking about
the decision for too long.
Decision-making styles
The concept of a decision-making style was first defined by Sproles and Kendall
(1986:267) as the mental orientation that characterises an individual’s approach to
making choices. As explained earlier, various consumer and potential consumer
characteristics, such as personality and attitude, contribute to the individual’s process
of making decisions between alternatives. The decision-making styles can be
categorised into three main approaches, namely, the psychographic/lifestyle
approach, consumer typology approach, and the consumer characteristic approach
(Bandara, 2014:5; Rašković et al., 2020:497).
The common theme among these three approaches is the perception that the
individual has to consider a number of dimensions (for example, time spent on
searching, cost sensitivity, importance of brand) while making the decision and
96
following the decision-making process (Merviö, 2015:290; Sandell, 2019:1383). The
consumer characteristics approach was recognised as the approach with the highest
explanatory power, as it provides consumers’ affective and cognitive orientation within
the decision-making process (Piacentini & Cui, 2012:32; Bandara, 2014:5; Prakash,
Singh & Yadav, 2018:90).
Sproles and Kendall (1986:269) identified the following eight categories of decision-
making styles: perfectionism or high-quality conscious, brand conscious, novelty-
fashion conscious, recreational and hedonistic conscious, price and ’value for money’
conscious, impulsive, confusion by overchoice, and habitual or brand-loyal. Each one
of these styles is discussed briefly below as it can assist short-term insurance
companies in understanding how certain characteristics can influence the individuals’
decision-making process.
• For individuals who are perfectionistic or high-quality conscious, quality becomes

the determining factor when choosing between alternatives (Merviö, 2015:288;
Acevedo, 2018:26).
• The brand conscious individual is orientated to purchase the more expensive and
well-known brands (Mokhlis & Salleh, 2009:575; Berger-Grabner, 2021).
• The novelty-fashion conscious individuals are individuals who seek for new things
and these consumers are innovative and try to keep up with new trends
(Information Recources Management Association, 2017:25; Reilly, 2021:123).
• Recreational and hedonistic conscious individuals purchase because they find it

enjoyable (Grable, Archuleta & Nazarinia, 2011:294; Acevedo, 2018:26).
• The price and ‘value for money’ conscious individuals are low-price conscious and
typically make comparisons to look for the best value for money option (Chebat &
Oumlil, 2015:170; Reilly, 2021:123).
• An impulsive individual is someone who will purchase on the spur of the moment
and who is not bothered by how much is spent, and he/she wants to get the ‘best
buys’ (Mokhlis & Salleh, 2009:575; Acevedo, 2018:27).
• Confusion by overchoice refers to the individuals that experience difficulty and

confusion due to the large amount of information about the product/service and the
97
number of alternatives they have to choose from during the decision-making
process (Merviö, 2015:290; Reilly, 2021:123).
• Lastly, the habitual or brand-loyal style of consumer decision-making relates to the

individual who has favourite brands and/or favourite stores. The individual tends to
purchase the same brand and/or at the same store over and over again (Chebat &
Oumlil, 2015:170; Acevedo, 2018:27).
It is important for an organisation's marketing function to have an understanding of the

decision-making styles, as these styles influence the consumer and potential
consumer’s behaviour during decision-making (Walsh et al., 2001:117; Atadil et al.,
2018:618; Sandell, 2019:1383). According to an individual’s characteristics, these
styles can assist short-term insurance companies in directing their marketing efforts
at the right type of consumers. The purchase decision-making theories applicable to
the motor vehicle insurance market are discussed in the next section.
PURCHASE DECISION-MAKING THEORIES
Insurance companies should continuously strive to respond and satisfy the changing
needs and wants (market demand) of insurance consumers and potential consumers
in their efforts to ensure their sustainability (Desrosiers, 2012:1; Muia, 2017:36).
Currently, the market penetration of motor vehicle insurance is extremely low, as only
30 to 35% of all the motor vehicles on South Africa’s roads are insured (Buys, 2015:23;
Bubear, 2016; Geldenhuys, 2020a; South African Insurance Association, 2021). Motor
vehicle insurance is also not mandatory in South Africa, as discussed in Chapter 2.
Knowing why some individuals in the market decide to purchase motor vehicle
insurance, while some decide against it, will give short-term insurance companies an
understanding of why consumers and potential consumers behave the way they do.
With this information they can implement strategies to encourage market demand. The
purchase decision process of insureds and non-insureds can assist in obtaining
information on the reasons why individuals decide to purchase or not to purchase
motor vehicle insurance, which can be used to stimulate the consumer demand. There
are various theories on the purchase decision or consumer decision-making. The two
theories discussed below are applicable to analysing consumers’ purchase decisions.
98
The risk theory
The risk theory refers to the likelihood that a catastrophic loss, which can lead to
physical and emotional harm of any kind, can be experienced (Onafalujo et al.,
2011:285; Xu & Fan, 2019:550). This is in fact why most people purchase insurance,
since there is a risk that they can lose/damage the asset. The risk is that which is being
ensured (Thoyts, 2010:4; Jankensgard & Kapstad, 2021:102). The term risk is defined
by the Oxford English and Spanish Dictionary (2021) as “a situation involving exposure
or danger” and “the possibility that something unpleasant or unwelcome will happen.”
These definitions point out that which constitutes risk is the element of uncertainty and
the potential of loss. If it is unknown whether a particular event will occur, this
possibility is known as uncertainty, whereas, if it is unknown whether a particular event
will occur but the probability that it will occur can be estimated, this possible occurrence
is then called a risk (Thoyts, 2010:4; Jankensgard & Kapstad, 2021:27).
Risk perception is also an important term that should be explained as each individual
can view a risk differently. Physiological evidence of a study conducted by Slovic
(1987) revealed that individuals perceive risks differently. The way in which individuals
view a risk can be described as their risk perception, however, the risk perception can
be influenced by objective facts and also by the individuals’ values and beliefs
(Drennan, McConnell & Stark, 2015:70). Some other factors that can influence the
individual’s risk perception include the characteristic of the risk source itself, value
attitude, ethnic-cultural and the socioeconomic background of the individual, and the
level of information about the risk (Onafalujo et al., 2011:286; Hanger et al., 2018:681).
These above-mentioned factors can influence the risk perception of motor vehicle
insurance consumers and potential consumers in the South African personal motor
vehicle insurance market. An individual who perceives the risk of being in an accident
as low, might view this as the underlying reason for not purchasing motor vehicle
insurance. This theory was, however, not tested in the current study as the consumers’
perceptions of risks were not analysed or measured.
The decision theory
The decision theory distinguishes between normative/rational decision theories and

descriptive theories (Hansson, 2005:6; Oliveira, 2007:12; Coxon, 2012:30; Takemura,
2014:10, 2021:6). The normative decision theory, also known as the rational decision
99
theory, is a theory of how individuals should make decisions (Coxon, 2012:30; Briggs,
2014; Takemura, 2021:6). A descriptive decision theory is a theory that explains how
people make decisions (Hansson, 2005:6; Briggs, 2014; Takemura, 2021:6) and it also
explains how the psychological aspects influence the individuals’ decisions, as well as
the reasoning behind decisions (Oliveira, 2007:12; Takemura, 2021:6). Both the
normative and descriptive theories reveal the nature of actual human decision-making
to a degree (Takemura, 2014:10, 2021:6).
As the purpose of the current study was to determine why consumers decide to, or
decide not to purchase motor vehicle insurance, the psychological elements were
investigated, as well as the reasoning behind why certain decisions are made or not
made. Therefore, all the rational and descriptive theories, and their development over
the years, are explained below to consider if one or more of these theories might fall
within the scope of this study.
Rational decision theories

Many seminary scholars have contributed towards these theories. The utility theory
was developed by Von Neumann and Morgenstern (1944). The term utility is defined
as the state or quality of being useful (Dictionary.com, 2021). In economics the utility
theory is used to measure the pleasure or happiness with a product or service, and
how it relates to the purchasing decisions made by individuals (Simon, 1959:258;
Moffatt, 2019; Pirgmaier, 2021).
The expected utility theory
The expected utility theory, which is based on individual preferences, was first
proposed by Bernoulli (1954). According to this theory, individuals do not evaluate
prospects by the expectation of their monetary outcomes, but rather by the expectation
of the subjective value of these outcomes. According to Oliveira (2007:13) and
Casquilho (2015:186), the expected utility theory can be interpreted in two ways,
namely, analytically and synthetically. The analytic view is where the choices
represent the revelation of preferences, which are defined as utilities, and the synthetic
view is about the decision-makers’ evaluation of utilities and probabilities, and the
integration of these judgements lead to a decision.
However, the expected utility theory makes predictions that are not the same as
consumers’ actual behaviour, and does not take the social, situational and cognitive
100
factors that are likely to influence the consumers’ purchase behaviour, into account
(Kunreuther & Pauly, 1982:26; Kahneman & Tversky, 1984:349). The individuals might
also respond differently towards risk than what is predicted in the expected utility
theory, since individuals perceive risks differently (Richter, Schiller & Schlesinger,
2014:93; Lewandowski, 2017:288).
The subjective expected utility (SEU) theory
The expected utility theory was extended as long ago as 1954 by Savage (1954),
where he combined the utility theory and the expected utility theory to form the
subjective expected utility (SEU) theory. The SEU theory is a theory of decision-
making that predicts that a decision-maker chooses the alternative that maximises
subjective expected utility. SEU can be defined as “the expected utility of an event or
outcome calculated from subjective probabilities rather than from objective
probabilities based on relative frequencies of observable events” (Colman, 2015:738;
Chen, 2021).
This theory has to do with the utility function and the probability function that are
combined to measure the decisions of individuals (Fishburn, 1981:190). Furthermore,
this theory can be interpreted as a product of the performance of two tasks, namely,
judgement and decision-making. For the judgement task, the information regarding
each available alternative needs to be reduced to a set of distinct outcomes and their
subjective probabilities. The decision-making task includes considering the utilities of
these outcomes by their subjective probabilities and selecting the alternative with the
highest expected utility (Erev, 1992:117). This was viewed as the dominant rational
theory, and it assumes that people behave as though they make the most of the sum
of products of utility and probability (Slovic, Fischhoff & Lichtenstein, 1977:9).
This theory is not applicable to this study, as the aim of this study was to determine
the purchase decision process in the personal motor vehicle insurance market and not
to make predictions about the consumers’ choices.
The theory of reasoned action
Fishbein and Ajzen (1975) proposed the theory of reasoned action, which predicts
behaviour intentions during the decision to purchase a product or service. The beliefs
and antecedents to behavioural intentions are divided into two distinct sets, namely,
the normative and behavioural. The behavioural beliefs are suggested as being the
101
underlying influence on the attitude of an individual towards performing the behaviour,
and the normative beliefs have an impact on the individual’s subjective norm about
performing the behaviour (Madden, Ellen & Ajzen, 1992:3).
Fishbein and Ajzen (1975) indicated that there were three boundary conditions that
could affect the extent of the relationship between intentions and behaviours. The
three boundary conditions are: the degree to which intentions and behavioural
measures correspond with regard to their levels of specificity, the stability of intentions
between time as a measurement and performance of behaviour, and the degree to
which carrying out the intention is under the volitional/decisional control of the
individual (Madden et al., 1992:4).
The theory of planned behaviour
The theory of planned behaviour was suggested by Ajzen (1985) to extend the
boundary condition of pure volitional/decisional control. This theory allows researchers
to predict intentions and behaviours with regard to the purchase or use of a single
brand/product, as well as the purchase choice and use between more than one
brand/product (Ajzen, 2015).
Although both of these last two theories imply that individuals make behavioural
decisions based on careful consideration of available information (Conner & Armitage,
1998:1430), they were not relevant to the current study as the purchase intentions and
behaviours of individuals were not predicted.
Descriptive decision theories

Festinger (1957) first introduced the theory of cognitive dissonance, which suggests
that individuals have an inner need to hold all their attitudes and beliefs in harmony
and avoid disharmony, which is known as dissonance. In other words, individuals
experience tension or discomfort when their beliefs or attitudes are not equal to their
behaviours (Cherry, 2020b). Cognitive dissonance is the term used to describe
feelings of discomfort from having conflicting/inconsistent attitudes, beliefs and
thoughts, especially with regards to behavioural decisions (McLeod, 2018; Cherry,
2020b). For example, when this discomfort exists between an individual’s
beliefs/attitudes/thoughts and the information contradicts the individual’s
beliefs/attitudes/thoughts, the individual experiences a feeling of regret after making a
purchase (Bolia, Jha & Jha, 2016:148).
102
This theory only applies to the post-purchase stage of the decision-making process,
as feelings of regret indicate that cognitive dissonance occurred after purchasing or
not purchasing personal motor vehicle insurance. As explained in the section
regarding the type of decisions (Section 3.4.3), when personal motor vehicle insurance
is purchased, it involves dissonance-reducing behaviour, but it might be that
consumers regret their decisions to purchase. This theory was not tested in this study,
as the post-purchase step of the consumer decision-making process was not part of
the scope of the current study.
The prospect theory
The prospect theory developed by Kahneman and Tversky (1979) is one of the first
theories that began to provide a more descriptive analysis of decision-making (Coxon,
2012:32; Desrosiers, 2012:7; Thomas & Loughran, 2018). The prospect theory
suggests that individuals begin by mentally discarding information that is of little use
to making decisions. Individuals then use a neutral reference point from which to make
judgements of potential losses or gains. Thereafter the prospects of the decision are
evaluated according to the individual's attitude towards risk (Coxon, 2012:32).
The prospect theory considers the outcomes of decisions as obtaining gains or losses
rather than utilities (Coxon, 2012:32; Stoner & Wankel, 2012:30). This theory also
predicts that individuals are more sensitive to minor losses than to minor gains, and
the perception of loss or gain is dependent on the reference point (Tversky &
Kahneman, 1991:1039). The reference point refers to the current asset position
(Hansson, 2005:47; Karni, Maccheroni & Marinacci, 2015:919). This theory indicates
that when individuals dislike losses more than an equivalent gain, they are more willing
to take risks to avoid a loss (Samson, 2016:116).
This theory was not tested in the current study, but the results might reveal, that if
individuals are more sensitive to minor losses (for example, the monthly policy fee)
than major gains (for example, when the motor vehicle is damaged and the insurance
company pays to repair it), they are willing to take the risk of not having personal motor
vehicle insurance.
The theory of buyer behaviour
The theory of buyer behaviour, developed by Howard and Sheth (1970), attempts to
explain the brand choice behaviour of a buyer or consumer. This theory also assumes
103
that the brand choice is not random, but is rather systematic, and therefore presents
a model. It should be pointed out that this theory is based on products, and not
services, such as personal motor vehicle insurance. This model and its shortcomings
will be discussed in Section 3.6.4.
The last few years of experimental and empirical research in behavioural research
have suggested that researchers should consider the various underlying behavioural
processes. However, no theory could explain observed behaviour (Harrison &
Rutström, 2009; Burghardt & Bodansky, 2021:2-3). Theoretical models form the basis
of most empirical and experimental research (Richter et al., 2014:86; Flache et al.,
2017).
As this study assessed the individuals’ behaviour during their purchase decisions, all
the consumer decision-making models are reviewed in the section below.
REVIEW OF CONSUMER DECISION-MAKING MODELS
Milner and Rosenstreich (2013), as well as Mahmood and Baharun (2019),

acknowledged that the existing theories of purchase/consumer decision-making are
not appropriate for the financial services market and there is a need to develop a new
conceptual model. Most of the consumer decision-making models assume that the
decision-making process consists of several steps. Not all the individuals who follow
these steps become consumers, as they might abandon the whole idea at any step
(Lappeman et al., 2021:154). While it is important to point out that these steps may
vary from products to services, consumers and potential consumers all pass through
similar steps (Prasad & Jha, 2014:335; Panwar et al., 2019:36).
As discussed earlier, although most scholars refer to the consumer decision-making

process, not only consumers follow these steps, as all the individuals in a market that
have a need for a certain product or service will follow these steps. Scholars have
developed several consumer decision-making models over the years, and the
expansion of these models is discussed below.
Andreasen model (1965)
Andreasen (1965) presented one of the earliest consumer behaviour models, as

illustrated in Figure 3.1 on the next page.
104
Figure 3.1: Andreasen model
Source: Adapted from Andreasen in Horner & Swarbrooke (2016:47)
105
As can be seen in Figure 3.1, this model emphasises the importance of information in
the decision-making process, as the direct flows are indicated by solid arrows and the
feedback by the dotted line arrows. The information processing during the decision-
making process includes the attainment of information, interpreting the information and
evaluating the information to make the final decision (Viksne et al., 2016:233). The
model also highlights the importance of attitudes, however, it fails to highlight individual
attitudes in relation to the repeat purchase behaviour of individuals (Horner &
Swarbrooke, 2016:46; Panwar et al., 2019:36).
Repeat purchase falls within the post-purchase step that does not form part of the
scope of the current study. The model indicates that beliefs, norms, values, feelings
and personalities might influence the filtration of information, but no psychological
factors and sociocultural factors are included.
Nicosia model (1966)
A decision-making model developed by Nicosia (1966) was one of the first models that
shifted the focus from the action of purchase to the more complex decision process
that consumers engage in regarding products and services (Batra & Kazmi, 2008:374;
Prasad & Jha, 2014:339; Mahmood & Baharun, 2019:479).
The Nicosia model concentrates on the purchase decision of a new product or service
(Horner & Swarbrooke, 2016:46). As can be seen in Figure 3.2, this model was
presented in a flow chart format, where four major fields or components were
highlighted. The four fields are firstly, communication of information to affect the
consumer’s attitude (influences from organisation’s attributed and consumer
attributes), secondly, the search and evaluation process, thirdly, the decision which is
the act of purchase, and lastly, the consumption/storage and experience/feedback.
106
Figure 3.2: Nicosia model
Source: Adapted from Nicosia in Horner & Swarbrooke (2016:48)
A major limitation of the Nicosia model is that it is based from a marketer’s perspective
and is not based on the consumer’s perspective. The consumers’ activities are
generally defined and not explained in detail (Milner & Rosenstreich, 2013:108;
Mahmood & Baharun, 2019:480). While little empirical work has been done to support
the model, the descriptions of purchase decision-making have been criticised, as this
model does not have the ability to describe or predict the actual consumer behaviour
with accuracy (Tuck in Foxall, 2004:13).
Engel, Kollat and Blackwell model (1968)
A more detailed consumer decision-making process that was presented by Engel,

Kollat and Blackwell (1968) can be seen in Figure 3.3. This model highlights the input
factors, information processing, decision process, and variables influencing the
decision process.
107
Figure 3.3: Engel, Kollat and Blackwell model
Source: Adapted from Engel, Kollat & Blackwell in Viksne et al. (2016:234)
What differentiates this model from the others discussed thus far is its emphasis on
the external variables/factors that might have an impact on the decision process.
According to this model, the influences include individual characteristics such as
values and motives, social influences such as culture and reference group, and
situational influences. This model, unfortunately, gives little guidance on which specific
situational influences can impact the decision process (Hunt, 1983:12).
In terms of short-term insurance, situational factors, such as the individuals’ physical

factors, social factors, time factors, the purchasing reason and their moods might
influence the individuals’ decisions. As the situational influences are not specified, this
model was not suitable for the current study.
The linear nature of the model has also been criticised, as the elements of the
purchase decision-making process do not necessarily occur in a set sequence
(Brinberg & Lutz, 1986; Bray, 2008:14; Viksne et al., 2016:237; Omotoyinbo, Worlu &
108
Ogunnaike, 2017:112). As it is difficult to predict in which order consumers will act in
the decision-making process, it is suggested that a non-linear/not sequential decision-
making process stage should be included in a model for personal short-term
insurance.
This model has another limitation, which is the implicit assumption that consumers
have the capability to evaluate the alternatives and make a rational decision on the
best suited alternative (Milner & Rosenstreich, 2013:109; Mahmood & Baharun,
2019:483). The insurance industry is unique in the sense that consumers might have
difficulty evaluating the decision to purchase directly after the purchase has been
made. For example, the advice received from the broker leads the consumer to make
a decision to purchase a personal motor vehicle insurance policy, but the purchase
decision can only be evaluated when a claim is made and the short-term insurance
company has to pay for the damaged motor vehicle to be repaired/replaced.
Loacker (2015:95) agreed by stating that the actual utility of insurance often does not
occur before an insured event has taken place. However, the post-purchase
evaluation step falls out of the scope of this study, as only the purchase decision
process of personal motor vehicle consumers and potential consumers South Africa
was determined.
Howard and Sheth model (1969)
Another effort to create a comprehensive model of consumer decision-making was the

model developed by Howard and Sheth (1969). This model was based on the Howard
and Sheth theory of buyer behaviour, and this theory attempted to explain the
purchase behaviour of consumers buying a brand choice, as well as other types of
buying (Hunt, 1983:12). The Howard and Sheth model is illustrated in Figure 3.4.
109
Figure 3.4: Howard and Sheth model
Source: Adapted from Howard & Sheth in Horner & Swarbrooke (2016:49)
As can be seen in the figure above, the flow of information is presented in four main
concepts, namely, inputs (marketing and social stimuli), perceptual constructs
(attention and information search), learning constructs (motives, choice criteria, brand
comprehension, leading to an attitude, confidence, intention and satisfaction) and
outputs (purchase, intention, attitude, brand comprehension and attention). A fifth
concept, the exogenous variables, is also visible. The exogenous variables concept
illustrates the influences on perceptual and learning constructs, such as the
importance of the purchase, the individual’s background, reference groups,
personality traits, time available and financial status (Howard & Sheth in Milner &
Rosenstreich, 2013:110). The output stage is also part of this illustration and the
intention aspect is an important aspect as this indicates the individual’s intention to
repurchase or not.
This model is seen as an important step towards the more recent input-process-output
models (Milner & Rosenstreich, 2013:109; Prasad & Jha, 2014:339). However, the
fact that this model is in linear form is also problematic, as explained in the critique of
the previous model.
110
The Howard and Sheth model improved the complexity of the purchase process with
various variables and relationships to internal processes and external influences, as
well as information sources. It is, however, important to state that the complexity of
this model can also be seen as a shortcoming (Milner & Rosenstreich, 2013:110;
Mahmood & Baharun, 2019:484).
It must also be mentioned that the initial testing of this model was highly unfavourable,
with non-confirmatory results (Farley & Ring, 1970). However, the initial testing by
Farley and Ring (1970) raised uncertainties regarding the data, and a call for
improvements to the data collection and procedures before the constructs and
relationships claimed by the Howard and Sheth model could be confirmed (Milner &
Rosenstreich, 2013:110).
The model testing by Farley and Ring (1970) became a subject of criticism, leading to
a discussion on how consumer decision-making models should be tested in general
(Hunt & Pappas, 1972; Lutz & Resek, 1972; Taylor & Gutman, 1974; Milner &
Rosenstreich, 2013). It is essential to state that Taylor and Gutman (1974) are of the
opinion that a comprehensive model can provide value conceptually, even if it is not
possible to test it. The conceptual CPDFPMVI model, which was tested in this study,
is presented in Chapter 4.
Perreault and McCarthy model (2002)
The Perreault and McCarthy model (Figure 3.5) is a modification of the Engel, Kollat
and Blackwell model, with the problem recognition or need-want awareness step as
the first step in the decision process. The psychological and social influences are also
visible in this model, as well as the additional purchase situation’s impacts on the
decision process.
111
Figure 3.5: Perreault and McCarthy model
Source: Perreault & McCarthy (2002:173)
The marketing mix/four Ps (discussed in Section 3.2.1.4) is also a new influence that
is illustrated in this model. Post-purchase evaluation, which involves feedback, is also
part of this model. However, the weaknesses of this model are that the three distinct
stages (input, process and output) are not clearly indicated in this figure, and it is not
clearly indicated how the psychological variables influence the process (Raiguru &
Misra, 2017:664).
112
Schiffmann, Kanuk and Hansen model (2012)
The Schiffman, Kanuk and Hansen model (Figure 3.6) is one of the most recent
consumer decision-making models. It is also a relatively comprehensive model, as it
clearly illustrates the three main stages (input, process and output) of the decision-
making process. The input stage indicates that the external influences, which consist
of the firm’s marketing efforts and sociocultural environments, influence the
consumers’ decision-making. The consumer decision-making stage directly impacts
the post-decision behaviour. However, the post-decsion behaviour did not form part of
the conceptual model of the current study.
Figure 3.6: Schiffman, Kanuk and Hansen model

Source: Schiffmann et al. (2012:36)
113
This model is a further modification of the Perrault and McCarthy model. The
marketing mix, labelled the firm’s marketing efforts, is also included in this model.
However, the marketing mix elements only refer to the four original Ps and the service
marketing Ps are not included. This is thus a limitation as it cannot be applied to
services, such as insurance. The social influences have been expanded and the
cultural issues are included in the sociocultural influences section of this model. Lastly,
the process stage is clearly indicated and the psychological experiences are additional
variables that can influence the need recognition or pre-purchase search, or the
evaluation of alternatives. This model could therefore be considered for adaption to
the personal motor vehicle insurance market.
Milner and Rosenstreich model (2013)
The Milner and Rosenstreich model, illustrated in Figure 3.7, is a decision-making

model specifically for the financial services market. This model also illustrates three
different stages, namely, inputs, process and outcomes.
Figure 3.7: Milner and Rosenstreich model

Source: Milner & Rosenstreich (2013:111)
114
The same three stages as in the Schiffman, Kanuk and Hansen model are illustrated
in this model, although the last stage is called the outcome stage. The purchase
situation, consumer characteristics and information sources are the input factors,
according to this model. These input factors are very broad and no specific variables
or points are provided to explain each factor. Milner and Rosenstreich (2013:112)
explain in the discussion of the model that the purchase situation refers to the external
drivers, while the consumer characteristics consist of the sociocultural and
psychological influences, and the information sources relate to sources from
companies’ marketing efforts and the sociocultural variables. The information sources,
as input factor, is an expansion of the Schiffmann, Kanuk and Hansen model.
The steps in the consumer decision-making process stage are also illustrated in a
linear form, but more financial terms are used. For example, the information utility is
known to be the pre-purchase search step, and the need evaluation of solutions refers
to the evaluation of the purchase alternatives. The criteria development step is,
however, additional in this model. No option is included to make provision for the
consumer that does not purchase, therefore, this can be seen as a limitation. A
substantial shortcoming of this model is that the customers’ experience is not taken
into consideration.
Although there were shortcomings, this model was suitable to be considered for the
development of a conceptual purchase decision model specifically for the personal
motor vehicle insurance market.
Schiffman and Wisenblit model (2015 and 2019)
Figure 3.8 presents the Schiffman and Wisenblit model. This is the most recent
decision-making model, and it is an expansion of the Schiffman, Kanuk and Hansen
model.
115
Figure 3.8: Schiffman and Wisenblit model
Source: Schiffman & Wisenblit (2015:20, 2019:375)
Similar to the Schiffman, Kanuk and Hansen model (and also the Milner and
Rosenstreich model), the Schiffman and Wisenblit model includes the three stages in
decision-making. With regard to the input stage, the communication sources are an
expansion of the Schiffman, Kanuk and Hansen model. The communication sources
indicate the mechanisms that deliver the marketing efforts and sociocultural influences
to the consumer (Schiffman & Wisenblit, 2015:340, 2019:376). This model also does
116
not include all the service marketing mix elements, which can be seen as a limitation,
given that personal motor vehicle insurance is a financial service.
In this model and in the Schiffman, Kanuk and Hansen model, the psychological
factors are part of the decision process stage and these psychological factors also
have an influence on the consumers’ decision (Milner & Rosenstreich, 2013:112;
Erasmus et al., 2019:454).
Seeing that the current study was developing a conceptual model specifically for the
personal motor vehicle industry, the relevant type of decision would be a dissonance-
reducing type of decision (as explained in Section 3.5.3). It was, therefore, for the
purpose of the current study, necessary to include these psychological factors as
individual/internal factors that would influence the purchase decision.
The actual decision to purchase or not to purchase is illustrated in the output stage,
which is an expansion of the previous models discussed. The rest of the items in this
model will not be discussed as the conceptual model used in this study ended with the
purchase decision step.
This section discussed the expansion and limitations of the different decision-making
models. The next chapter presents the conceptual CPDFPMVI model, considering all
the criticisms and benefits of the models discussed above.
CONCLUSION
To conclude, the marketing function of a short-term insurance company has to be

aware of the consumers’ needs, wants and demands to satisfy the consumers’ needs,
wants and demands. The service mix elements were specified in this chapter, and
taking into consideration that motor vehicle insurance is a service, the service mix
should be included in short-term insurance companies’ marketing plan.
Different roles can be played in the decision-making process and consumers have
decision-making rules/strategies, which they use to make a decision. Short-term
insurance companies should consider several types of decision-making and various
decision-making styles, as these aspects are important in determining the individual’s
behaviour during purchase decisions.
117
It was evident that not only one theory would apply to the current study and an adapted
purchase decision model would be used to conduct the empirical research. Several
consumer decision-making models that have been developed by various scholars over
the years, were evaluated. All the limitations of and expansions to the different
consumer decision-making models were considered to develop a conceptual
CPDFPMVI model. The table below summarises the constructs and items from the
models reviewed in this chapter which were included in the conceptual CPDFPMVI
model.
Table 3.1: Constructs and items in conceptual CPDFPMVI model obtained from
reviewed models
Obtained from which consumer

Constructs Items
decision-making model
Company’s ▪ Type of policies offered Schiffman, Kanuk and Hansen

marketing ▪ Promotion model
efforts
▪ Price of a policy
▪ Distribution channels
▪ Physical evidence Not in one of the reviewed models.

These four items were added
▪ Process based on the expanded/service
▪ People marketing mix elements (Chaffey
▪ Partnerships & Smith, 2017:54; Pride et al.,
2018:17).
Sociocultural ▪ Positive/negative family The initial items were obtained

environment member from the Schiffman, Kanuk and
factors ▪ Positive/negative reference Hansen model. However, the
person or group literature revealed that positive
and negative comments from the
▪ Positive/negative non- sociocultural groups (as obtained
commercial sources from the Schiffman, Kanuk and
▪ Positive/negative social Hansen model) could influence the
class decision-making process.
▪ Cultural and subcultural Therefore, the distinction was
factors made between the positive and
negative items.
Communication ▪ Traditional advertisements The advertising item from the

sources ▪ Online advertisements Schiffman, Kanuk and Hansen
model was renamed to distinguish
between the traditional
advertisements and online
advertisements.
▪ Buzz agent The rest of the items in this

construct were obtained from the
118
Obtained from which consumer
Constructs Items
decision-making model
▪ Customised messages Schiffman, Kanuk and Hansen
▪ Paid-for social media efforts model.
▪ User-generated social media
posts
Psychological ▪ Motivation Schiffman, Kanuk and Hansen

attributes that ▪ Perception about motor model
influence the vehicle insurance
need
▪ Perception about a certain
recognition
short-term insurance
company
▪ Learning
▪ Personality
▪ Attitude towards motor
vehicle insurance
▪ Attitude towards a certain
short-term insurance
company
Decision- ▪ Pre-purchase search More detailed steps were included

making process considering previous as items which were included was
experiences gained from the discussion about
▪ Pre-purchase search the Schiffman, Kanuk and Hansen
requesting quotations model. According to Schiffman
and Wisenblit (2015:340,
▪ Evaluate alternative prices 2019:376)
▪ Evaluate alternative qualities
▪ Evaluate alternative brands
Demographics ▪ Age Not in one of the models reviewed.

▪ Gender The general demographic
variables of consumers (Boone &
▪ Marital status Kurtz, 2014:170; Willan, 2021)
▪ Highest level of education were included.
▪ Employment status
▪ Household income
Decision to Schiffman, Kanuk and Hansen

purchase (Yes) model
Decision not to Schiffman, Kanuk and Hansen

purchase (No) model
The table above illustrates from which reviewed models each construct and item in the
conceptual CPDFPMVI model was obtained. The next chapter presents the
conceptual CPDFPMVI model for the current study.
119
A CONSUMER PURCHASE DECISION FOR PERSONAL
MOTOR VEHICLE INSURANCE CONCEPTUAL MODEL
INTRODUCTION
The existing theories of purchase/consumer decision-making (as discussed in Chapter

3) are not appropriate to the financial services market, and there is a need to develop
a new conceptual model (Milner & Rosenstreich, 2013:106; Mahmood & Baharun,
2019). Personal motor vehicle insurance is a type of financial service offering. A
conceptual model that illustrates the purchase decision process of personal motor
vehicle insurance consumers and potential consumers can be used to describe the
purchase behaviour. In this chapter, the conceptual CPDFPMVI model is presented
and discussed.
CONSUMER PURCHASE DECISION MODEL ADAPTED FOR

PERSONAL MOTOR VEHICLE INSURANCE MARKET
This study aimed to determine the purchase decision process of consumers (insured)
and potential consumers (non-insureds) regarding personal motor vehicle insurance,
by developing a unique model that will assign a probability to each of the possible
choices based on the context of the decision. Despite the researcher’s best efforts, no
similar studies could be found where unique purchasing decision models with regard
to individuals in the personal motor vehicle insurance industry of South Africa and
internationally had been developed.
A tested consumer purchase decision model, specifically in terms of personal motor

vehicle insurance can assist short-term insurance companies to understand the
purchase behaviour of their consumers (insureds) and potential consumers (non-
insureds). After considering the limitations of and expansions to the different decision-
making models, the conceptual CPDFPMVI model in Figure 4.1 was developed to
illustrate the purchase decision process for personal motor vehicle insurance
according to the theory reviewed in Chapter 3.
120
Figure 4.1: A conceptual CPDFPMVI model
121
The recent consumer decision-making models reviewed in Chapter 3 illustrate the
three stages: input, process, and output. These three stages are also illustrated in the
conceptual CPDFPMVI model above. The steps outlined in the conceptual CPDFPMVI
model above are explained in detail below.
EXTERNAL INFLUENCES
As illustrated in Figure 4.1 above, four types of influence can affect the need
recognition step, which impacts the purchase decision of individuals: the company’s
marketing efforts, sociocultural environment, communication sources, and
psychological attributes. External influences relate to the outside stimuli that affect an
individual during the purchase decision process (Stankevich, 2017:10). The external
influences are divided into the company’s marketing efforts, the sociocultural
environment and communication sources. All these external influences are discussed
in the sections below. Psychological attributes and the demographics of individuals
are internal influences, and this type of influence is discussed in Section 4.4
Marketing efforts
The company’s marketing efforts refer to the marketing mix elements (price, product,
place/distribution and promotion) that a short-term insurance company should perform
as part of its marketing plan. As explained in Section 3.2.1.4, personal motor vehicle
insurance is a financial service. Therefore, the model should include the additional four
Ps (physical evidence, process, people and partnerships), which the other models
have not included.
A company’s marketing efforts should be a direct attempt to reach, inform and

persuade individuals that need personal motor vehicle insurance (Schiffman &
Wisenblit, 2015:19, 2019:375), but these efforts can also be the reason why individuals
decide not to purchase, or why consumers choose, or do not choose, a certain short-
term insurance company, as discussed below:
• For example, if the price is too high for a policy, the individual will choose a cheaper
policy. Individuals might also feel that the price for motor vehicle insurance is too
high and they cannot afford it, and therefore decide not to purchase motor vehicle
insurance at all.
122
• The product/service is the type of policies (for example, comprehensive motor
vehicle insurance, third-party fire and theft, and third-party only) offered by a
company. This will also be an influencing factor, as the specific need related to
insurance determines the type of policy the consumer will look for.
• The promotion efforts of a company can positively influence an individual, if the

company has launched a promotional campaign that attracts an individual, and this
influences the purchase decision for motor vehicle insurance.
• The distribution channels used to sell motor vehicle insurance can also impact the
individual’s purchase decision process. For example, if an individual prefers not to
contact a direct short-term insurance company but would instead use a broker, a
short-term insurance company that uses brokers will be selected.
• The physical evidence efforts (the appearance of the company’s building,

employee uniforms and online physical presence) can impact an individual’s
purchase decision.
• The process efforts can also influence the purchase decision. For instance, when
an individual asks for a quotation, and it takes longer than what he/she expected,
consumers might decide not to purchase.
• The way that the people in the company deliver the service can also influence the
individual’s purchase decision. If a potential consumer contacts a short-term
insurance company directly and the employee delivering the service is rude and
not helpful, then this might influence his/her decision to purchase personal motor
vehicle insurance from that company.
• Partnerships or associations with other companies that an individual dislikes or

likes can have an effect on the individual’s purchase decision process. For
example, certain short-term insurance companies have motor partners where
individuals can purchase an extended motor warranty after the manufacturer motor
warranty has expired and this can convince potential consumers to purchase
insurance from a company that has this association.
The next section discusses the sociocultural environment is discussed.
123
Sociocultural environment
The sociocultural environment consists of family, reference groups, non-commercial

sources, social class, cultural and subcultural factors that can affect the purchase
decision process of consumers or potential consumers. The sociocultural inputs relate
to all the non-commercial influences on the need recognition of individuals (Schiffman
& Wisenblit, 2015:19, 2019:376).
Positive/negative opinion from a family member

The family influence can be a comment or an opinion of a family member that might
influence the decision-making stage (Loacker, 2015:111). This comment or opinion by
a family member can influence the decision process negatively or positively. A
consumer’s family forms the environment where the individual learns values, and it
develops and shapes their personality (Kotler & Armstrong, 2018:164). The
individual’s family environment offers the possibility of developing opinions about
certain subjects, particularly personal motor vehicle insurance.
Positive/negative opinion from a reference group

A reference group is any person or group of people that serve as a point of
comparison/reference. This reference group can influence why individuals form their
values and attitudes, or be the reason why they behave the way they do (Strydom,
2018:307). The reference group thus provides some points of comparison to
consumers and potential consumers regarding their behaviour, lifestyle or
preferences. Furthermore, a reference group influences individuals' self-image and
influences their purchase behaviour (Kotler & Armstrong, 2018:162). The influence
from a consumer’s reference group can form a positive or negative opinion, which will
impact their decision to purchase. In this case, the positive or negative opinion of a
reference person or group can be used as a basis of comparison, which will influence
the consumers’ need to purchase personal motor vehicle insurance.
Positive/negative opinion from non-commercial sources

Non-commercial sources can be the comments of a friend, a newspaper article and
the views of an expert on a blog or other internet platforms (Schiffman et al., 2012:68;
Hollensen, 2019). This type of source can influence the purchase decision of an
individual when deciding about short-term motor insurance (Ismagilova et al., 2017:9).
The positive or negative comments (personal words and recommendations) from
124
trusted friends and other customers are viewed as more credible than those coming
from commercial sources, such as advertisements (Kotler & Armstrong, 2018:163).
A study conducted by Facebook IQ (2019) on US insurance consumers found that

80% of the insurance buyers regard recommendations from family and friends as
important when deciding on insurance brands and different policies. This just shows
the influence word-of-mouth from non-commercial sources can have on the
consumers’ decision to purchase personal motor vehicle insurance.
Positive/negative opinion from social class

Social class can be explained as a group of people who have the same prestige and
status in society (Parumasur & Roberts-Lombard, 2014:7; Schiffman & Wisenblit,
2019:268). The social class includes individuals that share similar values, interests
and behaviours (Kotler & Armstrong, 2018:162). The social class reflects three
different components of an individual’s social status, namely income, education and
occupation. Other aspects include home value, area of residence, employment
positions, interests and recreation (Schiffman & Wisenblit, 2019:268). Thus,
individuals in the same social class give their opinions (whether positive or negative).
These opinions gained from members with the same values and interests can
influence an individual's purchase decision.
With regard to personal motor vehicle insurance, the positive or negative opinion of
members in the individual’s social class is a factor that influences the need to purchase
(the first step in the decision-making process).
Cultural and subcultural factors

An individual's cultural and subcultural factors shape personal perceptions,
dispositions and behaviours (East et al., 2017:115). A person’s culture is the basic
contributing factor of what he/she wants and how he/she behaves. Individuals grow
up in a society, and as a child, they learn the basic values, perceptions, wants and
behaviours from their families. Subcultures include population groups, native
language, religion and racial groups (Kotler & Armstrong, 2018:159). For example, an
individual’s language preference might influence his/her purchase decision if the
service is not offered in his/her preferred language. The section below discusses the
communication sources.
125
Communication sources
The communication sources that are gathered from the company’s marketing efforts
and sociocultural environment include the mechanisms that are used to deliver the
message to the consumers (Schiffman & Wisenblit, 2015:340, 2019:376). As depicted
in the Schiffmann and Wisenblit model, communication sources from a company’s
marketing efforts are advertising, buzz agents, customised messages and owned or
paid-for social media. Word-of-mouth, advice, recommendations and social media
posts from sociocultural influences (user-generated social media content) can be
communication sources that deliver a message to consumers.
The communication sources indicate the mechanisms that deliver the marketing mix
and sociocultural influences to the consumer (Schiffman & Wisenblit, 2015:340,
2019:376). The communication sources discussed below can impact an individual's
need to purchase personal motor vehicle insurance.
Traditional and online advertisements

The advertising activities performed by organisations, such as short-term insurance
companies, whether traditional or online, shape the image of the insurance companies
in the consumers’ consciousness (Borda & Jędrzychowska, 2012:28; Juska, 2018:4).
Advertisements have a significant influence on the purchase behaviour of individuals
(Haider & Shakib, 2017). Traditional advertisements refer to advertisements on
television (TV) and radio, whereas examples of online advertisements are video
advertisements and banner ads on websites. Organisations invest more in social
media and mobile advertisements, which also shift towards digital marketing (Haider
& Shakib, 2017). A study conducted by Facebook IQ (2019) on UK and Australian
insurance consumers found that 33% of the UK and 37% of the Australian insurance
consumers show an interest in a motor vehicle insurance brand or product once they
have seen an advertisement on Facebook and other social media platforms.
Buzz agents
Advertising efforts by insurance companies can serve as an effective tool to generate
buzz among consumers, and this can drive traffic towards a certain motor insurance
company and its offerings (Jacobs, 2020a). Buzz agents are individuals enlisted by
companies or even created opinion leaders to serve as ‘brand ambassadors’ that
spread the word about a company’s products (Kotler & Armstrong, 2018:163).
126
Consumers can also act as buzz agents. Consumers that are a source of a product
referral and tell other individuals about a certain product or a service are regarded as
buzz agents (Claro & Bortoluzzo, 2015:210). Therefore, any person who is a product
referral source and tells other individuals about motor vehicle insurance can be
regarded as a buzz agent within the motor vehicle insurance market.
Customised messages
Customised messages that are communicated to consumers form part of the direct
marketing efforts of companies. Since direct marketing is more targeted, a customised
message is usually directed at a specific consumer or group of potential consumers,
and it is personalised (Kotler & Armstrong, 2018:440). There is a district move toward
personalised marketing within the motor vehicle insurance market, as this allows
insurers to build a better relationship with consumers (Adamova et al., 2018).
Personalised/customised messaging strategies can be executed to engage with
consumers, and this is an effective way of reaching insurance consumers and potential
consumers (Moorcraft, 2021). Thus, the customised messages that short-term
insurance companies communicate should be personalised to build better
relationships with consumers.
Paid-for social media content

Paid-for social media content is another communication mechanism that
organisations, such as short-term insurance companies, can utilise to communicate
their service offerings to their market. Social media marketing activities include
actions/posts on social media platforms that encourage individuals to choose products
and brands (Bilgin, 2018:129; Parkin et al., 2019).
An advertising campaign on Facebook is an example of paid-for social media content,

but the new robust way of paying for social media content is to make use of social
media influencers (Appel et al., 2020:82). Organisations can pay money to an
influencer with a large following population to mention a brand or product/service in a
social media post (Carter, 2018). The influencer's opinion of the brand or
product/service shared on social media is seen by a considerably larger number of
people, and these are people who trust the influencer (Barysevich, 2020). One of the
primary reasons individuals trust influencers is that they can relate to the influencers
127
on a personal level (Barker, 2020). Nearly 40% of Twitter users indicated that they
made a purchase as a direct result of an influencer’s tweet (Santora, 2021).
Previous research shows that social media marketing activities significantly affect
consumer brand awareness, brand image, and brand loyalty (Bilgin, 2018:142).
Therefore, paid-for social media content can be regarded as a communication
mechanism that influences the purchase decision of personal motor vehicle
consumers and potential consumers.
User-generated social media content

User-generated social media content can be explained as personal words about a
company or service, advice, and recommendations shared/posted on a social media
platform. These posts are generated by the individual’s family members or someone
from the individual’s reference group. User-generated content on social media is
changing the way consumers purchase products/services (Agarwal, 2020:3076).
Recent research found that word-of-mouth using user-generated social media is

essential in the motor vehicle insurance market to enable consumers to talk about the
company’s motor vehicle insurance products (Ekanayake, 2021). Consumer reviews
on the internet are a powerful source of recommendations, and the consumers’
purchase decisions are strongly influenced by this electronic form of word-of-mouth
(Teixeira, Pereira & Dionísio, 2018:361). This shows the influence that user-generated
social media posts can have on the individuals’ decision to purchase personal motor
vehicle insurance.
Although advertising (online and traditional), customised messages and paid-for social
media are promotional efforts performed by the firm, and buzz agents can be a
distribution effort by the firm, the communication sources refer to the mechanisms that
deliver the message to the individuals. Word-of-mouth, advice and recommendations
as well as self-generated social media by a family member, someone from the
reference group, non-commercial sources or a member of the social class, are all
different mechanisms that can deliver a message to the individual, and that can
influence the individual’s purchase decision. In other words, these communication
sources can be seen as the methods or platforms that are used to communicate a
message to consumers. For example, a post on Facebook that was created and
posted by the short-term insurance company can convey a message to an individual,
128
and Facebook is thus the social media platform used as a communication source to
convey the message. This message can influence the need recognition, or it can be
the reason why potential consumers never reach the decision-making process stage.
All the factors discussed in this section do not only influence the need recognition step
but can also influence the individual to stop with the decision process and force them
to make the decision not to purchase a personal motor vehicle insurance policy. In the
following section, the process stage is explained.
INTERNAL INFLUENCES
The individual factors that influence individuals during the consumer decision process
are known as internal influences (Kaser, 2013:48; Stankevich, 2017:10). The internal
factors that influence the individual’s purchase decision refer to the personal and
interpersonal aspects that are processed in the consumer's mind, which influence the
purchase behaviour, and specifically, the need to purchase (Acevedo, 2018:19;
Cunningham, 2018).
As indicated in Figure 4.1, the psychological attributes and the individuals’

demographics are internal influences that form part of the process phase. The
influence of these individual factors on the need recognition step is explained below.
Psychological attributes
As illustrated in the conceptual model, psychological attributes are an internal

influence, as discussed below. Motivation, perception, learning, personality, and
attitudes form part of the psychological attributes that influence the need recognition
step.
Motivation
Motivation can be seen as the driving force that drives individuals to fulfil their needs
and wants (McIntee, 2014:72; Rajagopal, 2019:17). A motive, also known as the drive,
is the need that adequately pushes an individual to seek satisfaction (Kotler &
Armstrong, 2018:169). If an individual has a specific need, he/she will be motivated to
purchase it (Strydom, 2018:306). A consumer tries to satisfy the most important need
first, and a motivated person is ready to act (Kotler & Armstrong, 2018:172).
129
The motivation of consumers to purchase insurance is determined by two main factors,
namely, risk expectation and risk sensitivity. Risk expectation refers to the financial
value of insurance for the item/object (motor vehicle) being insured. Risk sensitivity
relates to the concerns consumers have, and these concerns directly affect the moral
benefit of having insurance or not (Meral, 2019). The risk sensitivity is also related to
the individual’s peace of mind in knowing that they are ‘protected’ in the case of an
event occurring (Suter et al., 2017:84). For example, the risk sensitivity might be higher
if the individual has recently been in a motor vehicle accident, and the individual will
not think twice about purchasing motor vehicle insurance again in the future. The other
side of the coin is if an individual views himself/herself as a good driver because the
motor vehicle they currently drive was not damaged in an accident in a number of
years. The risk sensitivity of such an individual will be low and he/she might not see
the benefit of purchasing insurance.
Perception
As mentioned above, a motivated individual is ready to act, and how an individual acts
is affected by his/her perception of the situation (Kotler & Armstrong, 2018:172).
Perception is a process of establishing the properties of stimuli by means of using one
or more senses such as vision, hearing, taste, smell and touch (Hoyer et al., 2018:80).
Perceptions determine what individuals pay attention to and what stimulates their
interests. In addition, it can influence them to attach their own interpretations to a
message (Erasmus et al., 2019:455).
Each individual within the personal motor vehicle insurance market may have a
different perception of motor vehicle insurance, and also of a specific short-term
insurance company. It will be difficult to establish what each consumer’s perception is,
but their perceptions about motor vehicle insurance and about a specific short-term
insurance company is a psychological factor that impacts their need to purchase
personal motor vehicle insurance.
Learning
The learning attribute refers to the process of how individuals purchase and consume
information, as well as the experience that they can apply to future decisions
(Parumasur & Roberts-Lombard, 2014:6; Erasmus et al., 2019:455). When individuals
act during the decision-making process, they learn (Kotler & Armstrong, 2018:173).
130
The experience gained from participating or acting in the decision to purchase
personal motor vehicle insurance changes their future purchase decisions. Therefore,
the learning/experiences gained from previous motor vehicle insurance purchases can
influence future purchase decisions.
Personality
Personality is related to the personal differences of individuals and it determines and
reflects how individuals think (Strydom, 2018:306). The inner characteristics (specific
qualities, attributes, traits, factors and mechanisms that distinguish one individual from
another) form an individual’s personality (Schiffman & Wisenblit, 2015:67, 2019:90). It
is, however, essential to point out that personality reflects individual differences,
personality is generally consistent and enduring, and although it is enduring, it can
change (Schiffman & Wisenblit, 2015:68; Sethna & Blythe, 2019). For example, an
individual’s personality can change or be altered by a major life event like the birth of
a child, the death of a loved one and so forth. This could influence his/her decision to
purchase short-term personal motor vehicle insurance. Each individual’s distinct
personality influences his/her purchase decision (Kotler & Armstrong, 2018:168).
Attitude
An individual can have a positive, negative or neutral attitude towards a
product/service or a brand. This depends on their past experience with the
product/service, such as insurance and the insurance company (Strydom, 2018:306).
An individual's attitude describes relatively consistent evaluations, feelings, and
tendencies toward an object or an idea (Kotler & Armstrong, 2018:173). Previous
research suggests that attitudes regarding insurance are mainly formed from past
purchase experiences (Suter et al., 2017:155).
It is clear to see how these psychological attributes form part of the individual’s
experience, and how these attributes can influence the need recognition step towards
making the decision to purchase personal motor vehicle insurance. These factors not
only influence the need recognition step, but can also influence the individual to
abandon the decision process and force them to make the decision not to purchase a
personal motor vehicle insurance policy.
The demographics of individuals and how this influences the need to purchase is
discussed below.
131
Demographics of individuals
The demographics of individuals can also impact the decision to purchase. Therefore,
this can also be viewed as an internal influence. As discussed in Section 3.2.1.3,
organisations often segment markets on the basis of demographics and it can also
relate to the individuals’ purchase decisions (Lamb et al., 2013:275, 2018:337). The
general demographic variables include age, gender, income, household size,
education, ethnic background and family lifestyle (Boone & Kurtz, 2014:170; Willan,
2021).
None of the decision-making models reviewed in Chapter 3 included the

demographics of individuals. This can be viewed as a shortcoming, as these
demographic variables may influence the need to purchase, which affects the
individual’s purchase decisions. The conceptual CPDFPMVI model (in Figure 4.1)
includes all the demographic variables that might have an influence on the consumers
or potential consumers’ need to purchase personal motor vehicle insurance.
Age, gender, marital status, highest level of education, employment status and
household income are all the demographic variables identified for the current study
that could possibly have an impact on the respondents need to purchase personal
motor vehicle insurance (as discussed in Section 7.4.1). The decision-making process
is discussed below.
DECISION-MAKING PROCESS
As mentioned above, the decision-making process explains how consumers and

potential consumers make decisions (Schiffman & Wisenblit, 2015:341, 2019:376),
and it starts with need recognition. This step can also be called problem recognition
or problem awareness, as illustrated in some of the models discussed in Chapter 3.
An unsatisfied need is recognised when an individual notices a difference between the

desired state of affairs and the actual state of affairs (Lamb et al., 2013:186; Sethna &
Blythe, 2019). The desire to recognise the need is dependent on two factors. Firstly,
the magnitude of the difference between the actual and desired state (the bigger the
difference, the higher the desire will be to recognise the need). Secondly, the
importance of the need is considered. The more important the need, the higher the
desire will be to satisfy the need (Du Plessis et al., 2012:86; Reyes, 2020:32). If an
132
individual does have a need for personal motor vehicle insurance, he/she will continue
with the pre-purchase search.
The type of decision step suggested by the Schiffman and Wisenblit model was not
included in the proposed conceptual CPDFPMVI model, as it was already established
that individuals who consider purchasing personal motor vehicle insurance would
engage in dissonance-reducing behaviour (Belch & Belch, 2015:125; Katrodia,
2021:61). Therefore, it was decided not to include this step in the conceptual
CPDFPMVI model.
The pre-purchase search or search for information step is where the individual
searches for information that supports a specific purchase decision (Hoyer et al.,
2018:189). The search for information prior to the purchase may be internal
(remembering facts about a policy or recalling experiences of previous service
deliveries) and/or external, which include reading about possible policies, searching
on the Internet or visiting a broker (Sethna & Blythe, 2019). Regarding personal motor
vehicle insurance, this will typically be the step where individuals think about previous
experiences (internal search) and request a few quotations of policies from different
short-term insurance companies (external search). Motor vehicle insurance
consumers can also search for information online regarding a specific company and
its policy offerings. The findings of this search remain important in the consumers’
purchasing decisions (Mau, Pletikosa & Wagner, 2018:1139). When the individual has
performed this pre-purchase search, he/she will move to the evaluation of alternatives
step. However, if the information obtained does not meet the individual’s requirements
or needs, he/she can decide to stop the process.
Evaluation of alternatives is where the information gathered by the individual is

evaluated (McIntee, 2014:67; Strydom, 2018:308). This is how individuals process the
information obtained in the pre-purchase search to choose between the alternative
brands (Kotler & Armstrong, 2018:177). Individuals use criteria such as price, quality,
performance standards, ethical characteristics, and aesthetic qualities to measure
whether the desired product/service will contribute to their need satisfaction and
lifestyle (Erasmus et al., 2019:458). A list of brands from which individuals plan to
make their selection can also be used to evaluate alternatives (Schiffman & Wisenblit,
2015:342, 2019:379).
133
During this step, individuals who obtain information about personal motor vehicle
insurance will use the criteria mentioned above, such as price (price of the policy on
quotation) and quality (the standard of the quotation) or the brand list (list of different
companies), to compare the different quotations and measure if there is a policy that
will contribute to their need satisfaction. After the information and alternatives have
been evaluated, the individual has to make a choice to purchase or not to purchase.
The purchase decision step is discussed below.
PURCHASE DECISION
The purchase decision or choice made by the consumer is the result, and therefore,
the final step illustrated in the output stage (Milner & Rosenstreich, 2013:115;
Schiffman & Wisenblit, 2019:375). This is the final step in the purchase process, where
individuals take action to buy or not to buy the service/product (Erasmus et al.,
2019:458), in this case, personal motor vehicle insurance. The purchase decision step
is where the individual makes the purchase or not, and this is also the step that
determines if the individual becomes a consumer or not (Lamb et al., 2018:112).
Individuals who decide to make the purchase will become consumers, and those who
decide not to purchase cannot be regarded as consumers. The post-decision
behaviour will not be covered as this falls out of the scope of the current study.
CONCLUSION
None of the existing consumer decision-making models illustrates the purchase

decision behaviour within the financial service market, specifically for the personal
motor vehicle insurance industry. Therefore, the conceptual CPDFPMVI model was
presented and discussed in this chapter. This proposed CPDFPMVI conceptual model
was tested in the current study. The next chapter presents and discusses the research
methodology regarding the testing of this conceptual model.
134
RESEARCH METHODOLOGY
INTRODUCTION
The literature review for this study was discussed in Chapters 2 and 3. The conceptual
model named the consumer purchase decision for personal motor vehicle insurance
(CPDFPMVI) was presented and discussed in the previous chapter (Chapter 4). This
chapter aims to present and explain the research methodology used in this study. A
detailed discussion of the research design, critical methodology, measurement
process, sampling design and techniques, data collection process and data analysis
methods are provided. The specific steps in the empirical research process are
explained and the analytical methods used are justified. The measures taken to ensure
reliability and validity are outlined. All the ethical considerations and how they were
implemented during the research process are comprehensively explained.
Consequently, the primary research process is outlined in this chapter.
RESEARCH OBJECTIVES REVISITED
The purpose of this study was to determine the purchasing decision process of
consumers (insured) and potential consumers (non-insureds) regarding personal
motor vehicle insurance. This study aimed to develop a unique model that will assign
a probability to each of the possible choices based on the purchase decision context.
The primary objective of this study is to develop a model that explains the purchase
decision of the consumers’ purchasing decision process in the South African personal
motor vehicle insurance industry. The purchasing behaviour of insureds and non-
insureds was determined by testing a conceptual CPDFPMVI model. A tested
consumer purchasing decision model, specifically in terms of personal motor vehicle
insurance, can assist short-term insurance companies in understanding the purchase
behaviour of their consumers (insureds) and potential consumers (non-insureds).
In order to achieve the primary objective of this study, several secondary objectives
were developed. The primary and secondary research objectives of this study are
presented next.
135
The following secondary research objectives were formulated for the current study:





The primary and secondary research objectives outline the purpose of this study.
Setting the research objectives is the first step in the research process, as these
objectives are the goals to be achieved when conducting the research (Zikmund et al.,
2013:50). The research process followed to conduct the research to achieve the set
objectives is explained next.
136
THE RESEARCH PROCESS
Research cannot be conducted without a purpose; therefore, the research

question/problem of this study was identified in Chapter 1, and the primary and
secondary objectives were developed to answer this research question or to solve the
research problem. In order to solve the problem and to answer the research question,
decisions need to be made by the researcher to determine which appropriate process
will solve the problem or answer the research question.
The research process is an arrangement of clearly defined steps that should be

followed during a research study (Cooper & Schindler, 2014:12). Although the
research process steps are illustrated in a certain sequence, certain steps might
overlap, and some of the steps will have to be revisited more than once. Each time a
step is revisited, the researcher has to reflect on the issue at hand, and the ideas might
need to be redefined (Saunders et al., 2019:11).
The research process followed in this study is illustrated in Figure 5.1.
137
Figure 5.1: The research process
Source: Adapted from Babbie (2016:114) and Zikmund et al. (2017:17)
The steps depicted in the figure above are explained below.
138
Steps in the research process
As illustrated in Figure 5.1, the following steps were performed during the research
process:
• Step 1 of the research process includes the problem discovery and problem
definition (refer to Chapter 1). During this step, the researcher identified and
explained the research problem and the gap in knowledge that is worthy of
research and that can contribute to the existing body of knowledge. From this
research problem, research objectives and research questions were formulated to
ensure that the research problem can be addressed. The research problem of this
study can be summarised as follows: Only 30 to 35% of the motor vehicles that
drive on South African roads have motor vehicle insurance (Buys, 2015:23;
Bubear, 2016; Geldenhuys, 2020a; South African Insurance Association, 2021).
Therefore, it is imperative to understand the behaviour behind motorists’ decisions
regarding the purchasing of motor vehicle insurance. The development of a model
that illustrates the motorists’ purchasing decision when deciding to purchase
personal motor vehicle insurance can assist short-term insurance companies to
understand the purchase behaviour of their consumers (insureds) and potential
consumers (non-insureds).
• Step 2 is to critically review the relevant literature (refer to Chapter 2 and 3). A
literature review is a written argument that supports a research topic by building a
case from credible evidence gained from previous research (Machi & McEvoy,
2016:5; Efron & Ravid, 2019:156). The literature review of a study also included
two important factors, namely, conceptualisation and operationalisation (Babbie,
2021). During this step in the research process, the researcher reviewed all the
previous and academic research to write an argument that supports the research
topic of this study. Conceptualisation was provided by exploring the key concepts,
such as consumer purchasing decisions and the other variables that are included
in the consumer purchasing decision process. Operationalisation was incorporated
by developing a conceptual CPDFPMVI model, and this model indicates how each
variable will be measured.
• Step 3 of the research process deals with selecting a specific research problem,
question or hypotheses (see Section 5.2). The formulation of a specific research
139
problem is vital as it provides a direction of the study. The research problem, as
was identified in Step 1, was sufficient, but it was refined after the literature review
had been conducted. As the research problem determines the nature (qualitative
or quantitative) of the research that will be conducted, the testing of a conceptual
model requires statistical analysis, and it was decided to make use of a quantitative
survey.
• Step 4 involves planning and selecting the research design (see Sections 5.4,
5.6, 5.7 and 5.8). This step includes the research methods, selection of population
and sampling methods to be used. For the purposes of this study, a quantitative
survey method was selected. The population selected was motorists in South
Africa, and two sampling methods were used, namely, convenience sampling and
snowball sampling. An online self-administered questionnaire was used to collect
the primary data for this study.
• Step 5 relates to the data collection process (see Section 5.9).
• Step 6 deals with the data processing (see Section 5.10).
• Step 7 includes the statistical data analysis and presentation of results (see
Section 5.11)
• Step 8 involves the development and refinement of the model. The model-
building strategy utilised to achieve the final CPDFPMVI model is illustrated in
Figure 5.12.
• Step 9 is the last step which involves presenting the final model and drawing
conclusions. The final step in the research process is to present the data findings
(see Section 5.13).
The research design used in the current study is discussed below.
RESEARCH DESIGN
The research design of a study is the plan of action that indicates how to achieve the
research objectives. This plan will specify the methods and procedures that will be
used to collect and analyse the needed data (Zikmund et al., 2017:21). A research
design consists of the strategy used to access the information to conduct research,
what the units of analysis/individuals are and how to access them, and the variables
and measurements used to test and address the research objectives and questions
140
(Ang, 2021:126). Another short description of a research design is that it is the
blueprint for collecting, measuring, and analysing data (Cooper & Schindler, 2014:125;
Cooper, Schindler & Sharma, 2019).
From the above definitions and descriptions the essential elements of a research
design can be identified as follows:
• A plan of action and time-based plan.
• A plan is always based on the research question or research objectives.
• A plan to access sources and types of information.
• A plan for specifying the relationships of the variables of the study.
• A procedural outline for every primary research activity.
In addition to these elements, the research design selected for a study should also
reveal the decisions made during the research process. Some reflections include:
• How the causal connections between variables are expressed if the results found
can be generalised in a bigger/another sample population than what was chosen;
• How particular behaviours should be explained and interpreted.
• In what manner the interconnections and changes in the social phenomena should
be explained (Bryman, 2016:40).
The research design serves two main functions: to provide detail on the steps and
procedures taken to conduct the study, and it ensures that the procedures are
adequate to acquire valid, objective and accurate answers to research questions and
objectives (Kumar, 2019).
It can be argued that there is no single best research design, as every research study
is different. Therefore, each study will have a unique research design that has the
most appropriate methods and procedures to collect the specific data required
(Zikmund et al., 2017:27). The research design is essential to develop the criteria for
evaluating research (Mårtenssona et al., 2016:594) and it guarantees that a
researcher reaches valid findings, comparisons and conclusions (Kumar, 2019).
A research design consists of certain components, namely, the research methodology,

data-collection methods, data analysis techniques and instrument development
141
(Cooper & Schindler, 2014:89; Cooper et al., 2019). Figure 5.2 illustrates the
components of the research design.
Figure 5.2: Components of the research design

Source: Adapted from Cooper & Schindler (2014:124); Cooper et al. (2019)
Key issues, such as validity, reliability, replication, trustworthiness, generalisability and

authentication, should be included in the research design (Conrad & Serlin, 2011:149;
Adams, Khan & Raeside, 2014:65; Leavy, 2017:155). This, however depends on the
broad approach followed, whether qualitative or quantitative. For example, in
qualitative research, trustworthiness is relevant, while in quantitative research validity
and reliability are important.
All of these components and key issues are included in the research design of this
study and will be discussed in detail throughout this chapter. Firstly, all the important
research terminology involved in the research design and research process is
presented next.
RESEARCH TERMINOLOGY
This section presents a brief discussion of the terminology concerning the research
process used in the current study and the rest of this thesis.
Research paradigm: Ontology, epistemology and axiology
The term paradigm originated from the Greek word “paradeigma” which means pattern
(Killam, 2013:5). A paradigm is a way of thinking or looking at the world, and it is
142
composed of specific philosophical assumptions that guide and direct a researcher’s
thinking and action (Mertens, 2019). Ontology, epistemology and axiology are the
philosophical assumptions that describe a research paradigm (Chilisa, 2019).
Ontology relates to how individuals are connected in societies and how they make
sense of the reality in which they live (Edelheim, 2015:30; Raghvan, 2021:14). Thus,
ontology is the philosophical assumption that explains the social reality. In other words,
of concern for a study will be what the nature of reality is. Epistemology, also known
as the study of knowledge, explains why individuals equally decide that certain things
are true and others are not (Saunders et al., 2019:133). Therefore, the philosophical
assumption that relates to the ways of knowing is epistemology. Axiology involves the
study of value or goodness, like ethics in research, and should be a fundamental part
of the research planning and implementation process (Mertens, 2019). It indicates the
ethics and value procedures in a study.
To summarise, the research paradigm of a study is a way of explaining a worldview

that is informed by three philosophical assumptions regarding the nature of social
reality (ontology), ways of knowing (epistemology) and ethics/value procedures
(axiology). These three terms and philosophical assumptions are essentially part of
the research paradigm, which is discussed in more detail in Section 5.6.1.
Deduction and induction
The two research approaches that can be followed in a research process are
deduction and induction. Deduction relates to the approach used to develop a
hypothesis or hypotheses based on existing theory, and designing a research strategy
to test the hypothesis/hypotheses (Wilson, 2014:13). The deductive approach is
essentially about testing or verifying an existing theory and not to develop a new theory
(Babbie, 2016:51). The following steps are included in the deductive approach: the
researcher draws what is known about a particular theoretical topic in order to develop
hypotheses, data should be collected on concepts/variables of the hypotheses; lastly,
the hypotheses are confirmed or rejected based on the findings (Bryman, 2016:21). A
survey is normally associated with the deductive approach (Saunders et al.,
2019:153), and this type of research is related to quantitative research (Bryman,
2016:21). With the deductive approach, the researcher works from the theory to the
143
data. Whereas, with the inductive approach, the researcher works from the data to the
theory (Wilson, 2014:13).
As indicated above, the inductive approach is the opposite of the deductive approach
as it involves the development of a theory after the data has been collected and
analysed (Saunders et al., 2019:153). This approach can also be explained as the
approach used to generate conclusions from one or more facts or parts of the
evidence, and the conclusions explain the facts (Cooper & Schindler, 2014:658). This
type of research is associated with qualitative research (Wilson, 2014:13; Bryman,
2016:24).
After distinguishing between the two possible approaches, it is evident that this study
needed to follow the deductive approach. The theory has been presented in Chapters
2, 3 and 4, and will be tested during the data collection and data analysis steps of the
research process by using quantitative methods.
Empirical research, primary data and secondary data
The term empirical research relates to making planned observations (Patten, 2017:3).
A more detailed explanation of empirical research is that it refers to a process of
collecting data and analyses of primary data, which was based on observations or
experiments, to test a hypothesis (Cooper & Schindler, 2014:66; Patten, 2017:3).
Empirical research is also conducted to determine the effect or causes and empirical
evidence is expressed in generalisations, laws and propositions (Rosenberg, 2018).
To summarise, empirical research is the research process followed when primary

and/or secondary data are collected and analysed through observations or
experiments to gain effects and causes or testing hypotheses. In this case, empirical
research will be conducted as primary and secondary data were collected.
Primary data are the raw data collected for the first time by a researcher for a specific
purpose to achieve the research objectives of a study (Cooper & Schindler, 2014:663;
Cooper et al., 2019). This type of data also relates to the ‘fresh data’ obtained in a
study (Bradley, 2013:112; Xian & Meng-Lewis, 2018:103). Primary data can be in a
qualitative or quantitative form (Cooper & Schindler, 2014:96; Cooper et al., 2019) and
with regard to this study, quantitative primary data were collected. The results are
presented in Chapters 6 and 7 after it was analysed to draw up conclusions. Most
144
research objectives and questions are achieved and answered by using a combination
of primary and secondary data (Saunders et al., 2019:338).
Secondary data are the data that was gathered and recorded by someone else for
other purposes than the current research study (Zikmund et al., 2017:99). This type of
data includes both quantitative and qualitative data (Saunders et al., 2019:338). Still,
it has at least one level of interpretation inserted between the event (the first time that
the data were collected) and when it was recorded in a study (Cooper & Schindler,
2014:86; Cooper et al., 2019). Secondary data were not collected in this study as no
existing data were used in the analysis phase. Only a detailed summary of previous
studies was used to compile the literature review.
Variables
A variable can be defined as the element or factor that is under investigation

(Williamson & Johanson, 2018:340). A more comprehensive definition is that a
variable is “a symbol of an act, trait, characteristic or attribute, which can be measured
and to which a value can be assigned” (Cooper & Schindler, 2014:55; Cooper et al.,
2019). Stated in another way, to test a hypothesis or a conceptual model, it is
compulsory to identify specific items (variables) that will be tested or measured
(Stokes & Wall, 2014:105).
It is common to distinguish between the different variables that can be used during a
research process and those variables that can be included in questionnaires used for
survey research, namely:
• Dichotomous variables which only have two values reflect the presence or absence
of a property. It can therefore also be nominal or ordinal. Yes and no questions are
typically dichotomous variables, but values such as male and female may also be
included, if these are the only options.
• Continuous variables are measured in a range and it may include decimal

numbers. Income and age are examples of continuous variables.
• Nominal values are impossible to define in a numerical category or value and it

cannot be ranked. Examples of such variables are gender, geographic location and
occupation.
145
• Ordinal variables refer to variables that can be ranked in order of magnitude, but
the values are not different. For example, a five-point Likert-scale is typically used
to measure this type of variable.
• Interval variables include the variables or categories that have identical distances
between them. The Celsius temperature measured is a good example of interval
variables as the zero degrees variable does not represent “no temperature”, but it
represents a value like any other variable. For example, 13 degrees and zero
degrees are 13 degrees less than if the temperature was 13 degrees.
• Ratio variables also have identical distances or amounts between them. The
difference between an interval variable and a ratio variable is that there is an
absolute zero value. Take for example, if a company makes a profit of R50 000 in
one year and R100 000 the following year, the profits have doubled in the next
year, whereas if the profit for the following year was R50 000 again, there would
be no increase and the true value will be zero (Cooper & Schindler, 2014:55;
Bryman, 2016:51; Saunders et al., 2019:511).
According to Zikmund et al. (2017:48), variables can also be categorised as dependent

and independent variables. A dependent variable relates to the variables that can be
explained or predicted and the variable that is of primary interest to the researcher
(Sekaran & Bougie, 2020:85). The independent variables are measurable
characteristics that can theoretically influence the dependent variables positively or
negatively. Empirically it might help explain variation with no direct influence or
causality (Babbie, 2021:17). Dependent variables change in response to changes in
other variables, whereas independent variables in most cases, cause changes in
dependant variables (Saunders et al., 2019:511). The specific variables used in this
study are provided in Table 5.7, which is presented later in this chapter.
DESCRIPTORS OF THE OVERALL RESEARCH DESIGN OF

THE STUDY
Various descriptors can be used to form the specific research design used in a study
by referring to the different design dimensions (Cooper & Schindler, 2014:126; Cooper
et al., 2019). The eight dimensions are illustrated in Table 5.1.
146
Table 5.1: Research design descriptors
Category Options
Research paradigm ▪ Post-positivism

▪ Constructivism
▪ Pragmatism
▪ Realism
▪ Postmodernism
The degree to which the research question ▪ Exploratory research questions

has been formed ▪ Descriptive research questions
▪ Causal research questions
The method of data collection ▪ Monitoring

▪ Communication study
The power of the researcher to produce ▪ Experimental

effects in the variables under study ▪ Ex post facto study
The purpose of the study ▪ Reporting

▪ Descriptive study
▪ Causal
▪ Explanatory
▪ Predictive
The time dimension ▪ Cross-sectional study

▪ Longitudinal
The topical scope of the study ▪ Case study

▪ Statistical study
The research environment ▪ Field setting

▪ Laboratory setting
▪ Simulation
Source: Adapted from Cooper & Schindler (2014:126); Sekaran & Bougie (2016:43); Cooper et al.
(2019)
Each one of these descriptors is explained in the sections below.
Research paradigm
This topic was briefly discussed in Section 5.5.1. The starting point of the research
design is to determine a research paradigm (Collis & Hussey, 2021:10). Research
depends on the link/relationship between the research design, research method and
the research paradigm. This link/relationship is illustrated in Figure 5.3. Only the
research paradigm (also known as the philosophical worldview) will be discussed in
this section. The other elements will be discussed later in the chapter.
147
Figure 5.3: Link between research paradigms, research design and research
methods
Source: Adapted from Creswell (2014:5); Creswell & Creswell (2017); Saunders et al. (2019:144-145)
The term paradigm refers to a set of assumptions or a belief system shared by the
members of a scientific community (Cooper, 2018). The research paradigm can be
explained as the framework that guides how research should be conducted, based on
the individual’s philosophies regarding their assumptions about the world and the
nature of knowledge (Collis & Hussey, 2021:10). A paradigm can also be explained
as one of the fundamental models or frames that researchers use as a reference to
organise observations and reasoning (Babbie, 2016:33). Essentially, a paradigm
becomes the lenses through which research is perceived and executed (Leavy,
2017:12). In other words, the way in which the researcher views the world, therefore
also known as ‘philosophical worldviews’ (Creswell, 2014:5; Creswell & Creswell,
2017). As depicted in Figure 5.3, the types of assumptions or beliefs an individual
researcher holds based on the factors mentioned above will determine the
researcher’s choice of research design and research methods. The main paradigms
will be briefly discussed below.
Post-positivism
The post-positivism/positivism philosophical belief originated in the natural sciences
and promotes an objective, patterned and identifiable reality (Leavy, 2017:12). Post-
148
positivists hold a deterministic philosophy, since causes will lead to the outcome
(Creswell, 2014:7; Creswell & Creswell, 2017). The research paradigm for the current
study is the post-positivist view, as the researcher had to identify and assess the
causes that influence outcomes by testing a conceptual model. This paradigm is
mostly found when quantitative research is conducted (Creswell, 2014:10; Collis &
Hussey, 2021:40).
The positivism approach generally relies on the following principles (Bell et al.,
2019:30):
• The only phenomena which can be observed by senses and can be warranted as
sources of knowledge.
• The purpose of theory is to generate hypotheses that can be tested. Normally

quantitative research starts with testing a theory.
• Knowledge is gained through gathering facts that provide the basis for universal
propositions.
• Science should be conducted in such a way that it is value-free. Research methods

and conclusions should not be biased.
• There should be a distinction between scientific statements and normative

statements. Senses cannot confirm normative statements. Thus scientific
statements should be formed.
As a final point, the sole purpose of research according to a paradigm is to determine

and confirm theories by testing them empirically. The current study tested a conceptual
model (theory), and therefore, the positivism approach was utilised.
Constructivism
The constructivism/interpretivism approach is generally used in qualitative research
(Creswell, 2014:8; Gray, 2020:27). This paradigm emphasises that individuals
produce the idea of several realities which exist in the context of social interactions
and subjective meanings via social interaction (Roller & Lavrakas, 2015:3).
Constructivism highlights the “worldview of experiences as it is lived and assessed by
the social actors” (Klenke, 2016:22).
Most researchers who use the constructivism approach accept that the position cannot
be pushed to the extreme, but they instead tend to question the individuals whether
149
there is a truth to be told in meeting with society (Pasian, 2015:62). Constructivism
implies that social beliefs and categories are not only produced through social
interaction, but are constantly being revised (Bryman, 2016:33).
Studies that are conducted with a constructivist view usually include open-ended
interviews, observations and narrative documents like diaries, unlike post-positivism
research that includes structured interviews and/or questionnaires (Gliner, Morgan &
Leech 2017:9). Thus, this paradigm was not suitable for the current study.
Pragmatism
The pragmatism approach involves mixing methods, techniques and procedures
associated with quantitative or qualitative methods, which is a necessity (Gray,
2020:31), as this approach offers a practical and applied research philosophy that is
orientated towards action (Klenke, 2016:26). Researchers who follow this research
paradigm, focus on the ‘what’ and the ‘how’ of the research problem (Creswell,
2014:11; Creswell & Creswell, 2017). Furthermore, pragmatism recognises that there
are many ways to conduct research and interpret the worldview, and therefore, it is
vital to include multiple research methods (Saunders et al., 2019:151). Since this study
only included quantitative research methods, this approach was not applicable.
Realism
The essence of realism is what researchers’ sense reality is, which implies that the
objects have an existence independent from the human mind (Saunders, Lewis &
Thornhill, 2012:136; Sekaran & Bougie, 2020:24). Objects of research such as culture,
the organisation and corporate planning exist, and act independently from the
observer. This is why researchers who follow this approach are available for
systematic analysis as a natural phenomenon (Gray, 2020:29). In other words, realism
assumes that objects exist independently from human minds, and it takes the natural
sciences approach towards developing knowledge.
There are two major forms of realism, namely empirical/direct realism and critical
realism (Bryman, 2016:26). Empirical/direct realism means that by using the
appropriate methods, reality can be understood (Bell et al., 2019:31). Therefore, direct
realists believe what they see is what they get. Differently stated, what direct realists
experience through their senses represents the world accurately (Saunders et al.,
2019:147). Critical realism, on the other hand, implies that what researchers
150
experience are sensations, the images/representations of things in the real world and
not the things directly (Saunders et al., 2019:147). Realism was not used as a
paradigm in this study.
Postmodernism
Postmodernism is explained by Saunders et al. (2019:160), as a philosophy that
…emphasises the world-making role of language and power relations.

Postmodernists seek to question the accepted ways of thinking and
give voice to alternative worldviews that have been marginalised and
silenced by dominant perspectives. Postmodernists deconstruct data to
expose the instabilities and absences within them.
Postmodernism draws attention to the social, historical and/or political construction of

knowledge, people, and social relations (Mora et al., 2012:44). Postmodernism is
mainly applied in the artistic and social sciences. The loose alliance of intellectual
perspectives which collectively create a challenging critique of fundamental premises
is the basis of the scientific research (INtgrty, 2016).
The typical methods used in postmodernism research are typically deconstructive in

nature, such as in-depth investigations of anomalies, silences, and absences. The
data are normally qualitative in nature (Saunders et al., 2019:145). As the current
study only included quantitative research methods, this approach was not suitable.
To conclude, Table 5.2 summarises the philosophical worldviews as discussed above.
151
Table 5.2: Comparison between research paradigms
Post-positivism Constructivism Pragmatism Realism Postmodernism
Ontology ▪ Reality is objective. ▪ Reality is socially ▪ Multiple realities. ▪ Reality is objective. ▪ Reality is socially
▪ Independent of constructed. ▪ Paradigm that ▪ Exists independently constructed.
social actors. ▪ Might change. answers research from human thoughts. ▪ Some meanings are
▪ Multiple realities. question the best is ▪ Interpreted through dominated and
selected. social conditioning silenced.
Epistemology ▪ Only observable ▪ Subjective ▪ Both observable ▪ Observable ▪ What counts as the
phenomena can meanings. phenomena and phenomena can “truth” and “knowledge”
provide credible ▪ Focus on the subjective meanings provide credible facts. is decided by dominant
data. details of situation. can provide ▪ Insufficient data result ideologies.
▪ Should focus on knowledge. in inaccuracies in ▪ Focus on
▪ Reality behind
causality and law- these details. ▪ Focus on practical sensations. absences/silences/repr
like generalisations. applied research. ▪ Focus is on essed
▪ Subjective meanings/interpretation
▪ Phenomena are meanings motivate ▪ Integrate different explaining within a
reduced to simplest perspectives to assist context or contexts. s/voices.
the actions.
elements. in interpreting the ▪ Exposure of power
data. relation and challenge
of dominate views as
the contribution
Axiology ▪ Research is ▪ Research is value ▪ Values play a big role ▪ Research is value ▪ Research is value
undertaken in a bound. in interpreting results. laden. bound.
value-free manner. ▪ Researcher is part ▪ Researcher adopts ▪ Researcher is biased ▪ Research embedded in
▪ Researcher is of the research. both objective and by worldviews, power relations.
independent of data. subjective points of experiences and ▪ Some research
▪ Maintains an view during data upbringings, which narratives are
objective stance. interpretation. will have an impact repressed and
on the research. silenced.
▪ Researcher is radically
reflective.
Source: Adapted from Saunders et al. (2019:144-145)
152
The table above illustrates the comparison between the different research paradigms
discussed in this section. As presented, the current study followed the post-positivism
approach where a conceptual model (theory) was tested.
The following section explains the degree to which the research questions have been
formed.
The degree to which the research questions have been formed
Research questions are the key questions that the research process will address, and
they are generally the basis of the research objectives (Saunders et al., 2019:815).
The three basic types of research questions that research studies can address are
exploratory, descriptive and causal research questions (Sekaran & Bougie, 2020:58).
Studies with exploratory research questions are usually qualitative in nature (Sekaran
& Bougie, 2020:58); their purpose is exploration and usually to develop hypotheses or
questions for further research. In other words, exploratory research usually is
unstructured to generate future research tasks (Cooper & Schindler, 2014:126;
Cooper et al., 2019). Furthermore, this type of research is conducted to clarify
ambiguous situations or to determine ideas that may be potential business
opportunities (Babin & Zikmund, 2016a:54; Zikmund et al., 2017:20).
The two more formal types of research questions are the descriptive and causal type
of research objectives. The formal type of studies starts with the hypotheses or
research questions and it includes the precise procedures and data sources
specifications (Cooper & Schindler, 2014:126; Cooper et al., 2019). Descriptive
research questions are designed to collect data that describe characteristics and it can
be quantitative or qualitative in nature (Sekaran & Bougie, 2020:58). This type of
research describes objects, people, groups, organisations or environments and it can
be used to help describe market segments (Babin & Zikmund, 2016a:54; Zikmund et
al., 2017:24). Causal research objectives aim to delineate one or more factors that are
causing a problem. A typical example of a causal research question is “How does
perceived value affect consumer purchase intentions?” (Sekaran & Bougie, 2020:57).
The research question for the current study are descriptive in nature, as the research
aimed at determining the characteristics and understanding of the purchasing decision
process which results in some individuals deciding to purchase and others not to
153
purchase personal motor vehicle insurance (Babin et al., 2015:127; Saunders et al.,
2019:801). The section below discusses the data-collection method selected for the
current study.
The method of data collection
Data can be collected through monitoring or communication. Monitoring refers to the

process where data are collected through observations. In contrast, communication
relates to the process where the researcher asks respondents questions and the
responses are captured (Cooper & Schindler, 2014:127; Cooper et al., 2019).
This study's research questions and objectives were answered by following the
communication process, as data were collected by asking respondents questions and
capturing the responses. Data were collected through a web-based survey and
respondents were invited to complete an online self-administered questionnaire (see
Appendix A). The researcher’s control over the variables in this study is discussed in
the following section.
The researcher’s control over variables in this study
A researcher can manipulate the variables of a study and this depends on whether the
study follows an experimental or ex post facto design (Cooper & Schindler, 2014:127;
Cooper et al., 2019). Studies that follow the experimental design use predictions rather
than research questions (Saunders et al., 2019:190). The researcher attempts to
control/manipulate the variables in the study as certain variables create effects in other
variables (Cooper & Schindler, 2014:127; Cooper et al., 2019). With the ex post facto
design, there is no manipulation of the variables as the researcher has no control over
it (Sekaran & Bougie, 2020:181). Thus, the researcher can only report what happened
and what is happening (Cooper & Schindler, 2014:127; Cooper et al., 2019).
For the purposes of the current study, the researcher did not attempt to manipulate or
control the variables in the study and merely reported on the purchasing decision
process of personal motor vehicle insurance consumers and potential consumers in
South Africa. It is for this reason that this study followed the ex post facto design and
statistical tests on findings and strict sampling procedures were used to ensure that
bias was not introduced.
The purpose of the study is discussed next.
154
The purpose of the study
Any study should have a purpose and the purpose of each study can differ with regard
to the following aspects: reporting, descriptive, causal-explanatory or causal-predictive
(Cooper & Schindler, 2014:127; Cooper et al., 2019). A reporting study will provide a
summary of the data. The data will be reported to achieve a deeper understanding or
to determine statistic comparisons. If the aim of the research is to determine who,
what, where, when, or how much, the study is a descriptive study. Studies that are
causal-explanatory of nature aim to establish how one variable leads to changes in
another variable. Lastly, causal-predictive studies strive to predict an effect on one
variable by manipulating another variable, while holding all the other variables
constant (Cooper & Schindler, 2014:127; Cooper et al., 2019).
Taking into account that the aim of this study’s primary and secondary objectives was
to determine who, what, where, when, or how much, it is a descriptive study. This
study also attempted to create comparisons between the data so that the purchasing
decision process of personal motor vehicle insurance consumers and potential
consumers can be better understood, therefore reporting was also an aim of the study.
Therefore, the purpose of this study was both reporting and descriptive.
The time dimension with regard to this study is covered in the next section.
The time dimension
The time dimension of a study will depend on whether the study is a cross-sectional
study or a longitudinal study (Cooper & Schindler, 2014:128; Cooper et al., 2019). If
the data for a study are only gathered once over a period of days or weeks or months,
it is a cross-sectional study (Sekaran & Bougie, 2020:113), but if it is repeated two or
more times over an extended period, it is a longitudinal study (Zikmund et al.,
2017:137).
The current study used the cross-sectional method, since data was only collected once
to investigate the consumer purchase decision process within the personal motor
vehicle industry of South Africa.
The next section discusses the topical scope of the current study.
155
The topical scope
The scope of a study can be statistical in nature, and within a case study design, the
researcher can also employ statistical analysis. A statistical study aims to determine
a population’s characteristics and the aim is to draw conclusions from the
characteristics of a sample. Furthermore, hypotheses are tested and generalisations
are drawn from the findings according to the representativeness of the sample, as well
as the validity of the design (Cooper & Schindler, 2014:128; Cooper et al., 2019). Case
studies focus on gathering data about a certain object, event or activity (Sekaran &
Bougie, 2020:106). Studies that are based on a case study are often more qualitative
in nature, but not definite, as these studies aim to find important themes (Zikmund et
al., 2017:68).
The topical scope of the current study is statistical, as this study aimed to determine
the characteristics of personal motor vehicle insurance consumers and potential
consumers within South Africa. The section below explains the research environment
of this study.
The research environment
A study can be conducted in a contrived (under artificial, staged or manipulated

conditions) or non-contrived setting. A study that is performed in a non-contrived
setting is known as a field study, and a researcher uses the same natural environment
in which the respondents normally function (Sekaran & Bougie, 2020:109). Research
that is conducted in a laboratory setting is normally artificial or staged, or the conditions
are manipulated. Simulations can be used in research to replicate the essence of a
system or a process (Cooper & Schindler, 2014:128; Cooper et al., 2019).
The current study was conducted in a field setting, as the data were collected in the
same environment in which the respondents normally function, without any conditions
being manipulated or simulated. The research approach of this study is discussed in
the following section.
RESEARCH APPROACH
As mentioned above, the paradigm, the research design and the research methods
contribute to the research approach. The research approach of a study refers to the
plans and procedures for research that stipulate the steps and detailed methods of
156
data collection, analysis and interpretation (Creswell, 2014:3; Creswell & Creswell,
2017). The three main research approaches to be followed are qualitative, quantitative
and a mixed methods approach. Table 5.3 illustrates the differences between these
three research approaches.
Table 5.3: Key associations of qualitative, quantitative and mixed method

approaches
Key Qualitative Quantitative Mixed method

associations approaches approaches approaches
Research ▪ Constructivist ▪ Post-positivist ▪ Pragmatic or critical

philosophy knowledge claims knowledge claims realist knowledge
claims
▪ Sequential, concurrent
and transformative
Methods ▪ Inductive ▪ Deductive ▪ Inductive or deductive

▪ Open-ended ▪ Close-ended ▪ Both open- and
questions, questions, closed-ended
emerging predetermined questions, both
approaches, non- approaches, numeric emerging and
standardised, text data predetermined
or image data ▪ Probability and non- approaches, and both
▪ Likely to use non- probability sampling quantitative and
probability sampling methods can be used qualitative data and
methods analysis
▪ Statistical techniques
used to analyse data ▪ Probability and non-
probability sampling
methods can be used
Research ▪ Has a variety of ▪ Examines ▪ Collects both

strategies strategies relationships between quantitative and
▪ Each strategy has a variables qualitative data
specific emphasis ▪ Experimental and ▪ Develops a rationale
or scope survey (questionnaire for mixing methods
▪ Strategies are: or structured ▪ Integrates the data at
action research, interviews or different stages of
case study structured inquiry
research, observation) research
strategies ▪ Use research
ethnography, strategies of both
grounded theory qualitative and
and narrative quantitative research
research.
Source: Saunders et al. (2019:176-183)
The key associations of the main three research approaches are illustrated in the table
above. The qualitative and quantitative research methods will be discussed next. The
mixed method approach is a combination of the qualitative and qualitative methods,
and does not need to be discussed separately.
157
Qualitative research
Qualitative research is an approach that is used to explore and understand the

meaning that individuals or groups ascribe to a social or human problem (Creswell,
2014:4; Creswell & Creswell, 2017). In qualitative research, non-numeric data (words,
images and video clips) are collected through the use of a technique such as an
interview (Saunders et al., 2019:179). This type of research focuses on discovering
new insights and inner meanings, and it is also less structured than most quantitative
studies (Babin & Zikmund, 2016a:109). Qualitative research mainly answers the ‘how’
(process) and ‘why’ (meaning) things happen, therefore developing an in-depth
understanding of the problem/phenomenon (Cooper & Schindler, 2014:144; Cooper
et al., 2019).
Table 5.4 summarises the main attributes of qualitative research.
Table 5.4: Main attributes of qualitative research
Research aspect Attribute
Focus Understand and interpret the problem/phenomenon.
Purpose Discover ideas; used in exploratory research with general

research objectives.
Approach Observe and interpret data.
Data-collection approach Unstructured and free-form.
Researcher’s independence Researcher is intimately involved; therefore results are

subjective.
Sample design Nonprobability and purposive sampling usually utilised.
Sample size Small samples used, often in natural settings.
Research design Exploratory research designs are used most often.
Data type and preparation Verbal or pictorial descriptions that are reduced to verbal
codes.
Data analysis Human analysis following computer or human coding.
Source: Adapted from Cooper & Schindler (2014:147); Zikmund et al. (2017:66)
Taking into consideration all the attributes of qualitative research provided in the table
above, it is clear that this approach would not be suitable for the current study, and it
would rather follow a quantitative approach. In the next section, the quantitative
research approach is discussed.
158
Quantitative research
Quantitative research is the approach that is used to collect data that generates
numeric data of numerical nature (Saunders et al., 2019:176). When researchers want
to test hypotheses or answer specific research questions, they will follow a quantitative
approach, as this type of research answers the questions that are necessary to
determine a course of action (Zikmund et al., 2017:66). Furthermore, quantitative
research strives to verify the theory by focusing on gathering the facts to establish
valid claims (Gray, 2020:210). More specifically, quantitative studies answer questions
related to how much, how often, how many, when, and who (Cooper & Schindler,
2014:146; Cooper et al., 2019).
Quantitative research assesses the relationship between variables measured

numerically, and that are analysed using several statistical techniques (Saunders et
al., 2019:178). Quantitative research follows a deductive research approach (Bell et
al., 2019:35) and attempts to measure something precisely. This type of research
usually measures consumer behaviour, knowledge, opinions, or attitudes (Cooper &
Schindler, 2014:146; Cooper et al., 2019). Quantitative methodologies have become
the research approach commonly used for studying consumer behaviour (Hackett,
2019).
A deductive approach is evident when determining the relationship between theory

and research (Bryman et al., 2014:62). The four main features of deductive reasoning
in quantitative research are:
• Measurement: The data-collection instrument has to be valid and reliable.
• Causality: This feature explains why things are the way they are by establishing
the causal impact between independent variables and dependent variables.
• Generalisation: The results of a study can be generalised to a population outside

of the selected sample.
• Replication: The findings of a study can be replicated or reproduced by any

researcher as it was originally conducted independently and objectively, but the
study should be capable of replication.
Table 5.5 summarises the main attributes of quantitative research.
159
Table 5.5: Main attributes of quantitative research
Research aspect Attribute
Focus Describe, explain or predict.
Purpose To test hypotheses or answer specific research

questions.
Approach Measure and test data.
Data collection approach A structured approach and categories are provided.
Researcher’s independence Researcher acts as an observer; thus results are

objective and free from bias.
Sample design Probability sampling is utilised.
Sample size Large sample sizes are used to produce generalised

results.
Research design Descriptive and/or causal research designs are most

often used.
Data type and preparation Verbal descriptions are reduced to numerical codes for
computerised analysis.
Data analysis Computerised analysis (statistical and mathematical

analysis) that upholds a clear distinction between facts
and judgements.
Source: Adapted from Cooper & Schindler (2014:147); Zikmund et al. (2017:66)
These main attributes provided in the table above explain the quantitative approach
that was followed in the current study. A quantitative design was followed in the
primary research stage of this study, as research questions/objectives are generally
deductive and certain variables needed to be investigated (Leavy, 2017:72).
According to Cooper and Schindler (2014:146), quantitative research attempts the

precise measurement of something, and in business research specifically, quantitative
methodologies usually measure consumer behaviour, knowledge, opinions or
attitudes. Thus, the insureds and non-insureds’ purchase decision behaviour had to
be assessed by means of a quantitative research approach. The sampling design and
approach followed in this study are provided in the following section.
SAMPLING DESIGN
The previous sections presented all the important concepts involved in the nature of
this study. This section discusses how the population that the researcher wanted to
160
study was determined. The concepts regarding the sampling design are defined and
the specific methods selected for this study are explained in detail in this section.
Population, population element, sample and census defined
The term population can be explained as the entire group of people, events, or things
of interest that a researcher wishes to study or investigate (Sekaran & Bougie,
2020:222). The target population is an essential aspect of the sampling plan of a study.
It refers to the total number of elements of a particular population that is relevant to a
research study (Babin & Zikmund, 2016b:69). A population element or case relates to
the individual respondent from which the measurement is taken (Cooper & Schindler,
2014:338; Cooper et al., 2019). A sample is the subset of the population, whereas a
census refers to the number of individual respondents in the population (Saunders et
al., 2019:292). These concepts are illustrated in Figure 5.4.
Figure 5.4: Population, sample and population element or case

Source: Saunders et al. (2019:296)
The target population of this study was individuals who are vehicle owners and/or
drivers (motorists) in South Africa. More specifically, the target population consisted
161
of individuals who have the choice to purchase or not to purchase personal motor
vehicle insurance. The number of registered vehicles in South Africa was 12 882 765
in August 2021 (eNaTIS, 2021). Since this is the live motor vehicle population on the
roads in the country, the drivers or owners of these motor vehicles can be seen as the
theoretical population of this study. It was, however, challenging to include the entire
target population, as the contact details of all these elements were not available; thus
a sample had to be drawn from this target population (as will be discussed in the
following section).
This target population was selected as it holds the following significant academic
reference relevance:
• The current study only aimed to determine the purchasing decision process of
personal motor vehicle insurance consumers and potential consumers, not other
insurance types for vehicles such as buses, trucks, motorcycles, and so forth. It is
for this reason that the motor vehicle owners or drivers that drive on South African
roads were selected as the target population and not drivers or owners of vehicles
such as buses, trucks, motorcycles, and so forth.
• The market perspective from the purchasing decision process was investigated. A
market/audience can be described as a large number of individuals who have a
need for a specific product/service, have the money to purchase the
product/service, are willing to spend the money on it and are legally able to buy the
product (Erasmus et al., 2019:459). An individual can form part of the market but
not be a consumer of a product or service (Lappeman et al., 2021:5). Considering
the explanation above, the personal motor vehicle insurance market includes
individuals who have personal motor vehicle insurance (insureds) and those who
do not have insurance (non-insureds).
Sampling method
As the size of certain populations is too large, researchers choose to select a sample
from within the entire population for the research study. However, if a census approach
is followed, the entire population is researched. Sampling is a valid alternative to a
census, if it is impractical to survey the entire population, and when time and costs are
constraints to surveying the entire population (Saunders et al., 2019:249).
162
The current study used sampling. Sampling includes any process that draws
conclusions based on measurements of a portion of the entire population (Babin &
Zikmund, 2016a:69). When a sample is selected for a study, it should represent the
full set of cases/elements in a manner that is meaningful to the study, and it should be
justified (Cooper & Schindler, 2014:84; Cooper et al., 2019). There are two main
sampling types, namely, probability and non-probability sampling. Probability sampling
relates to techniques that are based on random sampling procedures, whereas non-
probability sampling refers to techniques that rely on the judgement of the researcher
(Zikmund et al., 2017:351-356).
Table 5.6 tabulates the attributes and methods of these two sampling procedures.
Table 5.6: Attributes and methods of main sampling methods
Probability sampling Non-probability sampling
Attributes
▪ Controlled random selection. ▪ Arbitrary sample selection process.

▪ Every individual has an equal chance ▪ Probability of individual being selected in
of being selected. unknown.
▪ Findings can be generalised. ▪ Normally used if generalisation is not the
▪ Sampling bias is eliminated. goal.
▪ Simpler to carry out.
▪ Less time and money are required.
▪ Seen as the most feasible method.
Methods
▪ Simple random sampling. ▪ Convenience sampling.

▪ Systematic sampling. ▪ Purposive (judgement) sampling.
▪ Stratified sampling. ▪ Quota sampling.
▪ Cluster sampling. ▪ Snowball sampling.
▪ Multistage sampling.
Source: (Cooper & Schindler 2014:348-361; Cooper et al., 2019)
This study used non-probability sampling methods to select a sample from the entire
population. Firstly, it is essential to discuss the means of accessing the potential
respondents and then the specific sampling techniques employed.
Means of accessing potential respondents

After identifying the target population, the researcher searched for an available sample
frame by contacting the Road Traffic Management Corporation (RTMC). This
163
government entity is responsible for coordinating and facilitating motor vehicle
registrations in South Africa. The RTMC, therefore, has a database of all the registered
motor vehicle owners in South Africa. However, this organisation could not provide the
researcher with access to a complete database of current contact details and email
addresses of registered motor vehicle owners. Therefore, the researcher had to find
other means of obtaining these potential respondents.
The Automobile Association (AA) of South Africa has a database with up-to-date email
addresses of motorists in South Africa, and this database was a means of accessing
potential respondents for the study. In addition, the Facebook Market Insight tool was
used to distribute the survey to individuals who are likely to drive motor vehicles in
South Africa. Thus, Facebook was also used as a means of accessing potential
respondents of this study.
The current study employed more than one non-probability sampling technique to
sample the respondents. The figure below illustrates the steps taken to select the non-
probability sampling techniques used to sample the respondents.
164
Figure 5.5: Selecting a non-probability sampling technique
Source: Saunders et al. (2019:316)
The steps outlined in Figure 5.5 were followed to select the sampling techniques. Only
the sampling techniques applicable to the study will be discussed.
165
Convenience sampling
Availability sampling, also recognised as convenience or haphazard sampling, is used
when the researcher wants to study the characteristics of individuals passing the
sampling point at a specific time, or if other sampling methods are not feasible (Babbie,
2021:192). One of the main reasons researchers consider convenience sampling is
that this method is always feasible when resources are low and when there is no
sampling frame available (Silvia, 2020:60). This type of sampling involves the
haphazard selection of individuals that are easily available to invite them to a study.
Convenience sampling is a non-probability sampling method widely used when online
questionnaires are distributed via email or social media platforms like Facebook
(Saunders et al., 2019:324). Convenience sampling was not the researcher’s first
choice, but it was deemed feasible as no sampling frame was available, and the target
sample included motorists in South Africa.
An advantage of convenience sampling is the ease of availability, in the sense that

individuals are easy to identify and reach (Silvia, 2020:60). The downside to using
convenience sampling is that there is a potential of bias (Zikmund et al., 2017:358).
This was, however, limited as the first two questions in the questionnaire were
screening questions. These two questions restricted the sample to South Africans who
drive motor vehicles and are personally responsible for their decision to purchase
personal motor vehicle insurance cover.
This sampling method does not facilitate any control over the representativeness of
the sample (Babbie, 2021:192), and caution regarding generalisability was thus
exercised by the researcher (Saunders et al., 2019:324; Babbie, 2021:192). In other
words, the findings of this study cannot be overgeneralised as this sampling method
is restrictive in that sense. However, the findings gained from this study could still
provide valuable new insights for future replication studies.
By distributing the survey via the AA database, which is a database of motorists in

South Africa, the researcher used convenience sampling. The individuals in this
database were invited using an email and each respondent had to click on the survey
link in the email to be able to complete the questionnaire. The figure below is a
screenshot of the message that was distributed to the database of South African
motorists.
166
Figure 5.6: Email distributed to the AA database
167
The other sample frame, namely Facebook, was used in three ways to distribute the
survey to potential respondents. Firstly, the Facebook advertisement marketing tool
was used to implement convenience sampling to recruit respondents. By explicitly
targeting individuals who showed an interest in motor vehicles within the nine different
provinces in the county, the researcher posted a Facebook advertisement with the
questionnaire link, and the respondents had to click on the link to be able to view and
complete the questionnaire. Figure 5.7 illustrates the message distributed on the
Facebook advertisement marketing tool.
Figure 5.7: Message distributed on Facebook advertisement marketing tool

The figure above demonstrates how the message was posted on the Facebook page
of everyone who was targeted. The individuals still had to click on the link to complete
the online questionnaire.
168
Snowball sampling
Snowball sampling is another non-probability sampling method that can be utilised to
identify potential respondents. This type of sampling technique is used when a
researcher asks current respondents to identify one or more individuals who meet the
specific criteria and who are willing to participate in the study (Johnson & Christensen,
2017).
Snowball sampling was used in this study as a complementary method by posting the
online survey on Facebook and convincing Facebook users to recruit their friends to
complete the survey. Advantages of this sampling method include low costs and large
sampling sizes. The downside to using snowball as a sampling technique is that
individuals tend to refer individuals who are like-minded and similar; thus, the sample
cannot be generalised (Kosinski et al., 2016; Allen, 2017:481). It is for this reason that
the researcher decided to use both the convenience and snowball sampling methods
in this study.
Figure 5.8 illustrates the post used to recruit respondents by using snowball sampling.
This post was shared on the researcher’s Facebook page and it requested individuals
to complete and share the questionnaire link, and in such a way, invite more
respondents to the study.
169
Figure 5.8: Facebook post using snowball sampling
The initial date of the post can be seen in the figure above. This post was posted more
than once to ensure that the maximum number of individuals were recruited.
Facebook groups were also used to invite respondents to partake in the study. The
same message (as indicated in Figure 5.9 below) was posted in Facebook groups
from all nine provinces. Groups like Garden Route Classifieds - Selling and Buying -
From Mosselbay to PE, Nelspruit/Barberton/Hazyview/Bushbuckridge Community
Trade Zone, Advertise Port Elizabeth, Free State Buy and Sell, advertise and sell
anything, Potch Koop en Verkoop, Sell It! Buy It !! joburg ,CPT, Durban, were used to
post the invitation. The researcher had to ask permission from some of the groups to
post the invitation. Snowball sampling was used to invite respondents to the study by
posting the message below (Figure 5.9) in the above-mentioned groups.
170
Figure 5.9: Facebook groups post using snowball sampling
The figure above illustrates the message that was distributed on several Facebook
groups that facilitate the buying and selling of products in all nine provinces of South
Africa. This message was also posted more than once in each group.
Even though the data-collection instrument was developed to be an online

questionnaire, the researcher struggled to obtain views/responses from individuals
without personal motor vehicle insurance. Thus, the questionnaire was printed and
distributed to individuals who drive motor vehicles. The researcher asked current
respondents to identify one or more individuals who drive motor vehicles and who
171
might be willing to participate in the study, and to complete the paper-based
questionnaire. With this last attempt to invite respondents to the study, the researcher
used a combination of the snowball and convenience sampling method. Snowball
sampling, since the current respondents were asked to invite individuals to partake in
the study, and convenience sampling, because the invited individuals had to drive a
motor vehicle and be easily accessible.
Sample size
The sample size refers to the number of respondents that are included in a study. The
sample size is dependent on a number of factors, such as cost available, the overall
size of the population, level of precision required, type of questions asked, and the
type of method used to collect data (Clark, Foster & Bryman, 2019:177-179). The
sample size used in each statistical analysis method is guided by the type of statistical
analysis method used (see Section 5.11.2). Taking into consideration that no set rule
exists on the ideal size of a sample, as it is dependent on the design and nature of the
study, the points below were followed in determining the sample size of the current
study.
Hall (2020:107) identified a few points that should be considered when determining
the sample size of a quantitative study:
• With a larger population size, the sample size will have to increase in relation to a
smaller population size. The size of the sample will level off as the population
increases beyond around 100 000.
• The margin of error can be reduced by increasing the sample size.
• The more viability associated with the measures that are implemented, the larger
the sample that is required to ensure a given level of precision.
• The costs connected with sampling have to be balanced against the increase in
precision obtained from large samples.
Concerning this study, a sample size of at least 600 respondents was considered
sufficient, as adequate statistical analyses methods could have been successfully
conducted with this sample size. Individuals had to click on the link to complete the
questionnaire, which is called ‘the click-through rate’ (Laubenstein, 2018). Taking into
account that the click-through rate for online surveys can be very low, a minimum
172
response rate of 600 was required. Given the data analysis planned for this study, the
target sample size would consist of 300 respondents with motor vehicle insurance and
300 who do not have motor vehicle insurance. The researcher hoped that the sample
would be spread between the different age groups, ethnicity groups, gender groups,
marital status groups, education levels, employment groups, income groups and
provinces in which respondents drive the motor vehicles to provide an overall
impression of the population and assist the researcher in developing a consumer
purchasing decision model for personal motor vehicle insurance. In addition, the
researcher aimed for a larger sample size than projected in order to offer more
accurate results.
A response rate of less than 1% was achieved (678 responses), which is low.
Nonetheless, it is not uncommon to have a low response rate in using the web-
based/online survey method, and previous research has indicated that online surveys
have a lower response rate than paper-based questionnaires (Nulty, 2008:303; Ebert
et al., 2018).
The survey completeness of online surveys is usually higher than that of paper-based
questionnaires (Ebert et al., 2018), and this was, in fact, also a finding in this study.
Proportionally more respondents fully completed the online questionnaires in
comparison to the paper-based questionnaires, where some questionnaires were not
fully completed. Guinalíu and Díaz De Rada (2020) found that using mixed modes to
collect data can increase the survey's response rate. Therefore, the researcher
decided to use the paper-based questionnaire method as well, as there was a low
response rate to the online survey. The data collection of this study is discussed in the
section below.
DATA COLLECTION
The data can be collected once the sample has been identified. Various methods can
be used to collect data from the respondents, but the most popular methods in
business research are interviews, observations and questionnaires (Sekaran &
Bougie, 2016:111). Each of these data-collection methods has its own advantages
and disadvantages, and the type of method should complement the type of data
collected. As discussed before, a quantitative research method was employed, and it
is for this reason that a questionnaire was the data-collection tool or instrument. The
173
data-collection instrument used in this study and the data-collection process followed
in this study are discussed in detail below.
Data-collection instrument
Data were collected by means of a survey strategy and, more specifically, online and
paper-based surveys using a self-administered questionnaire (see Appendix A). The
presence of an interviewer is not required when self-administered questionnaires are
used, as the respondents are responsible for reading and answering the questionnaire
themselves, and the questionnaires can be distributed in various ways (Babin &
Zikmund, 2016a:187). A self-administered questionnaire can be in a paper-based or
electronic/online form, and the purpose is that the respondents answer the questions
by themselves and return the completed questionnaire in the method requested by the
researcher (Sekaran & Bougie, 2020:159).
The questionnaire should be developed in such a way that the respondents will be
able to answer it without needing any assistance. Therefore, attention should be given
to instrument development and design. According to Bryman and Bell (2015:245-248)
and Saunders et al. (2019:536-537), there are a number of principles that should be
considered when it comes to the questionnaire design phase, namely:
• A clear title should be provided for the survey and a cover letter explaining the
purpose of the survey, any ethical considerations and measures taken to protect
respondents’ anonymity and confidentiality should be included.
• Questions should be short, but enough detail should be provided to collect the
appropriate data. The questionnaire should be presented in a formal manner and
the formatting should be done in such a way that the questionnaire is pleasant and
easy to read.
• The questionnaire should be clearly presented so that the respondents know what
is expected of them in each question. The wording of the questions should be
explicit to ensure that there is no ambiguity or misunderstandings.
• In order to minimise confusion, a vertical format for answering the questions is

preferred.
• Clear instructions on how questions should be answered must be provided.
• The questions and answers should not be split and should be kept together.
174
• Closing remarks should be provided at the end of the questionnaire.
All these principles were considered during the questionnaire design phase. The cover
letter for the questionnaire (see Appendix A) includes the purpose of the study, the
approximate time it will take to complete the questionnaire, and the assurance that the
respondents’ anonymity and confidentiality will be upheld. Clear instructions were
provided (respondents were requested to place a tick in the box they wish to select)
and it was stated that there were no right or wrong answers.
Respondents were assured that they can withdraw at any stage and that their
participation was voluntary. Respondents were further informed that there are no
benefits for participating in the study; it is envisioned that the findings of this study will
be used in a PhD thesis; and the findings could also be shared in conference
presentations, the publication of academic articles and other communication. In case
the respondents would like to receive feedback regarding the results of this study, they
were informed that they could contact the main researcher. The respondents were
assured that the researcher did not foresee any negative consequences to them from
completing the survey. The respondents gave consent by proceeding with the survey.
Each of the questions and its answers were placed together on the same page and all
the questions were in electronic format online. The questionnaire was pretested (the
pretesting phase is discussed later in this section).
As discussed earlier, the online survey by means of using a self-administered

questionnaire was initially selected for this study as it is a cost-effective method and it
is less time-consuming than interviews and observations (Sekaran & Bougie,
2020:144). Even though the data-collection instrument was developed to be an online
questionnaire, the researcher struggled to obtain views/responses from individuals
without personal motor vehicle insurance. Thus, the decision was made to use the
paper-based survey method as well to collect data from respondents who do not have
access to the Internet or Internet-operated devices.
According to Wagner, Kawulich and Garner (2012:100), survey research is normally

used to obtain data from large groups of respondents. Online surveys specifically allow
a researcher to send questionnaires to a large group of individuals via the Internet,
and data can also be collected through email, websites and mobile phones (Kumar,
2019). As explained earlier, the online survey link to the questionnaire was distributed
175
via email to the AA database and posted on Facebook. Respondents could then
choose if they wanted to complete it on their computers, laptops or mobile phones.
Facebook is becoming a powerful research tool because it provides a large pool of
respondents (Kosinski et al., 2016:70). Relevant Facebook users were invited to this
study, as the Facebook Market Insight tool was used to distribute the survey to
individuals who are likely to drive motor vehicles in South Africa. Regarding the paper-
based questionnaire, the researcher asked current respondents to identify one or more
individuals who drive motor vehicles and who might be willing to participate in the study
and to complete the paper-based questionnaire. Therefore, the paper-based
questionnaires were distributed by hand to the current respondents who said they
could assist and these respondents handed questionnaires to individuals who drive
motor vehicles and who were willing to participate in the study.
Self-administered surveys do have some disadvantages. This method introduces a

larger chance of non-response (due to respondents not understanding the question or
skipping the question accidentally) as well as non-response error (Sekaran & Bougie,
2020:144). These can, however, be limited by presenting the questions in such a way
that the respondents clearly understand the questions. Additionally, in the online
format, the respondents were not able to go on to the next question if everything was
not answered on that page.
There are some other disadvantages of an online self-administered questionnaire that

have to be outlined as well. Firstly, many consumers are not computer literate. For this
study, consumers had to be computer literate and had to have access to a device and
Internet to be able to complete the questionnaire. Secondly, not all respondents are
willing to participate in a survey, and respondents in this study had to be willing to
complete the questionnaire. This is, however, a disadvantage for any type of survey.
The length of the questionnaire should also be taken into consideration by ensuring
that the questionnaire is not too long, and it should not include many open-ended
questions. As theory is being tested in the current study, the focus, was on collecting
quantitative data, and no open-ended questions were included in the questionnaire.
Questions 6 and 7 requested additional specifications if the option for ‘Other’ was
selected. These options were, however, still numerical and cannot be seen as open-
ended questions. Question 13 is a fill-in question that requests the age of the
respondents, which is also numerical.
176
While the researcher has little or no control over who completes the questionnaire,
questions can be included to ensure the correct target population completes it. These
questions are called screening questions. The first two questions in the questionnaire
allowed that only the individuals who drive motor vehicles are included, and also the
individuals that are personally responsible for their decision to purchase personal
motor vehicle insurance cover in South Africa. If respondents selected option 2 in one
of these two questions, these questionnaires were discarded and respondents were
instructed to stop completing.
There is also a risk of receiving incomplete or incorrectly completed questionnaires

back from the respondents. Therefore, the questionnaire is submitted to pilot test
phase where the questionnaire is pretested to reduce the possibility of incomplete
questionnaire submissions and also to make sure that it is not too long. In order to
avoid incorrectly completed questionnaires, the researcher has to ensure that the skip
questions and instructions are clear and also tested in the pretesting phase.
Online self-completion surveys have a lower response rate than other types of surveys
(Bell et al., 2019:237). However, there are ways to increase the response rate, and
this was done by using more than one sampling method (convenience and snowball
sampling) to ensure that the largest sample was invited to partake in the study. The
low response rate has been stated as a limitation and it will be addressed accordingly.
In the next section, the instrument design and development are discussed.
Instrument design and development

All the above-mentioned survey instrument design principles were taken into account
during the first draft development, together with the detailed literature review in
Chapter 3. Barry et al. (2011:97-98) warn researchers not to develop their own scales
in questionnaires without following the correct steps and procedures. These authors
provide a guide that researchers can use to develop successful survey instruments.
By making use of this guide, Figure 5.10 illustrates the steps taken to develop scales
in this study’s questionnaire.
177
Figure 5.10: Scale development steps
Source: Barry et al. (2011:98)
The figure above presents the steps followed to develop the instrument design
(questionnaire) for this study. The researcher could obtain good measures, but this
was because the researcher had previously gathered an evidence-based
understanding of the intended behaviour measure by reviewing the existing theories
and various decision-making models (as discussed in Chapter 3) during the literature
review.
178
Phase 1 of the study was performed during the literature review, where the existing
theories and various decision-making models were extensively reviewed to outline the
constructs. The existing theories and various decision-making models discussed in
Chapter 3 led to the development of a conceptual CPDFPMVI model. This model could
theoretically determine the purchasing decision process of consumers (insured) and
potential consumers (non-insureds) regarding personal motor vehicle insurance. The
constructs that were identified are illustrated in Table 5.7 below.
Phase 2 was performed by developing a suitable scale design and structure according
to the analysis plan and appropriate response scale to facilitate the analysis.
Therefore, a five-point Likert-scale was developed for each construct (see Table 5.7
below).
In Phase 3, broad and comprehensive sample items were developed for each
construct according to the theory.
Phase 4 started by assessing content validity (Step 1). Two individuals (supervisor
and statistician) both knowledgeable in the content area, reviewed the item pool of
each construct to determine the clarity and relevance to each construct. These
suggestions and alterations were taken into consideration, and the needed
improvements were made. Step 2 was performed by asking one respondent to read
through the scale to assess each item's cognitive and motivational qualities. Feedback
from this assessment was recognised, and the needed improvements were made.
Step 3 was performed by conducting a pilot study (refer to the next section). Steps 4
and 5 were also performed during the pilot test process and are explained in the
section below. The pilot testing phase is explained below.
Pilot testing of data-collection instrument

A pilot test refers to a small-scale data collection process among respondents similar
to those in the final data collection of the study (Babin & Zikmund, 2016a:64). It is vital
that the selected respondents of a pilot study form part of the target population and
that the same procedure and steps are followed as in the primary data collection
phase. The pilot test should be performed to determine the accuracy of instructions
and to establish any weaknesses in the design of the instrument (Bryman & Bell,
2015:272). A pilot test was done in this study and the steps are presented in the figure
below.
179
Figure 5.11: Pilot test process
Source: Researcher’s own compilation
The self-administered survey instrument for this study was pilot tested as follows:
1. The first questionnaire draft was reviewed by the supervisor, who is a professor
specialising in marketing management (an expert from a theoretical perspective).
Therefore, content validity was ensured. Appropriate revisions were made.
2. The second questionnaire draft was assessed by a statistician to ensure that the
questions included in the questionnaire would offer statistically sound information.
180
3. Revisions were made and the questionnaire was submitted for ethical clearance
to the Business Management Ethics Review Committee.
4. The needed corrections were made as suggested by the Committee and the
questionnaire, as well as the study was approved. Only after the ethical clearance
certificate has been issued could the respondents be contacted.
5. Respondents with valid driver’s licenses who are drivers of personal motor
vehicles were asked to complete the pilot test. The researcher firstly performed a
paper-based pilot test where a printed questionnaire was delivered/handed to the
respondents. Thirty completed questionnaires were received back and the
researcher had an in-depth interview with three of the respondents who were
randomly selected. The purpose of these interviews was to test their
understanding of the questions, the length and the flow of the questionnaire (Step
4 of the scale development steps was also performed here). After the interviews
and hard copy questionnaires were received back from the respondents, the
needed amendments were made to the questionnaire.
6. Statistical analysis was conducted on the responses obtained just to ensure that
the questions had a statistical significance (Step 5 of the scale development steps
was performed here). The statistician made recommendations and the needed
corrections were made.
7. The revised questionnaire was again reviewed by the supervisor to make sure all
the changes had been implemented and that it was ready for the online capturing
on LimeSurvey.
8. The questionnaire approved in Step 7 was developed online on LimeSurvey.
9. After it was captured on LimeSurvey, the online questionnaire was tested by the
researcher to ensure that all the skip questions were correctly directed, and that
all the information on the online questionnaire was correct.
10. Only when the researcher was satisfied that the questionnaire was correctly
captured on LimeSurvey, was it released for the online pilot test.
11. The online pilot test was performed and it was open for seven days, and 22
respondents completed the questionnaire.
181
12. The researcher received feedback from three of the respondents, which indicated
small errors and these revisions were made and the final changes were captured
on LimeSurvey.
13. The online self-administered questionnaire was ready to be released for the
primary data collection phase.
The pilot test process as explained above, led to the final online self-administered
questionnaire. The section below provides the details of the self-administered
questionnaire used for this study.
The self-administered questionnaire

The self-administered questionnaire (see Appendix A) used in this study consists of
an introductory paragraph, 12 questions regarding the consumer purchasing decision
process and demographic questions at the end. The introductory paragraph states the
purpose of the study, the time it will take to complete the questionnaire and guarantees
the respondents’ anonymity and confidentiality.
The way in which the questions are formulated is extremely important as it impacts the
type and quality of information that one receives from the respondents. The wording
and structure of the questions should be clear and it should be based on the research
objectives to ensure that the correct data will be collected (Kumar, 2019). It is for this
reason that each question should be linked to a research objective/question.
Table 5.7 indicates the link between the questions asked in the questionnaire and
research objectives (as stated in Chapter 1).
182
Table 5.7: Research objectives and self-administered questionnaire questions matrix
Research objective Corresponding section in questionnaire Sources used to

generate
Number Question Classification of Type of question
question/justification of
variable inclusion
To develop a model Q1 – Primary objective which n/a n/a n/a

that explains the Q21 will be answered when
consumer purchase all the results are
decision process in combined.
the South African
personal motor
vehicle insurance
industry.
Screening questions Q1 and Drivers of motor Nominal Dichotomous question Used to determine if the
Q2 vehicles and personally respondent is eligible to
responsible for complete the survey. If
purchasing personal respondents selected
motor vehicle insurance option two in one of these
cover questions, they were asked
to stop completing the
questionnaire.
Filter question Q3 Have or do not have Nominal Dichotomous question This filter question
personal motor vehicle determined with which set
insurance of questions the
respondent would begin
the questionnaire.
183
generate
variable inclusion
To determine the Q8 Construct: Influence of marketing efforts on need recognition

company’s marketing
efforts that influence Q8.1 Type of personal motor Ordinal 5-point Likert response Forms part of the
the individual’s need vehicle insurance format measuring Schiffmann and Wisenblit
to purchase personal policies influence ranging from decision-making model
motor vehicle no influence (1) to (Schiffman & Wisenblit,
insurance in South extremely large 2015:20, 2019:375).
Africa. influence (5)
Q8.2 Advertisements or Ordinal 5-point Likert response Forms part of the

promotional activities format measuring Schiffmann and Wisenblit
influence ranging from decision-making model
no influence (1) to (Schiffman & Wisenblit,
extremely large 2015:20, 2019:375).
influence (5)
Q8.3 Way to purchase Ordinal 5-point Likert response Forms part of the
format measuring Schiffmann and Wisenblit
influence (5)
Q8.4 Price Ordinal 5-point Likert response Forms part of the

format measuring Schiffmann and Wisenblit
influence (5)
184
generate
variable inclusion
Q8.5 Physical evidence Ordinal 5-point Likert response Element of the service
efforts format measuring marketing mix (Chaffey &
influence ranging from Smith, 2017:54).
no influence (1) to
extremely large
influence (5)
Q8.6 Process Ordinal 5-point Likert response Element of the service

format measuring marketing mix (Chaffey &
no influence (1) to
extremely large
influence (5)
Q8.7 Employees Ordinal 5-point Likert response Element of the service

format measuring marketing mix (Chaffey &
no influence (1) to
extremely large
influence (5)
Q8.8 Partnerships or Ordinal 5-point Likert response Element of the service

associations format measuring marketing mix (Chaffey &
no influence (1) to
extremely large
influence (5)
185
Research objective Corresponding section in questionnaire Sources used to generate
Number Question Classification Type of question question/justification of
of variable inclusion
To establish the Q9 Construct: Sociocultural environment influences on need recognition

sociocultural
environment factors Q9.1 Positive opinion from Ordinal 5-point Likert response format Author-generated, opinion from
that influence the family member measuring influence ranging a family member forms part of
individual’s need to from no influence (1) to the Schiffmann and Wisenblit
purchase personal extremely large influence (5) decision-making model
motor vehicle (Schiffman & Wisenblit,
insurance in South 2015:20, 2019:375).
Africa.
Q9.2 Negative opinion Ordinal 5-point Likert response format Author-generated, opinion from
from family member measuring influence ranging a family member forms part of
from no influence (1) to the Schiffmann and Wisenblit
extremely large influence (5) decision-making model
(Schiffman & Wisenblit,
2015:20, 2019:375).
Q9.3 Positive opinion from Ordinal 5-point Likert response format Author-generated, opinion from
reference member measuring influence ranging a reference member forms part
from no influence (1) to of the Schiffmann and Wisenblit
2015:20, 2019:375).
Q9.4 Negative opinion Ordinal 5-point Likert response format Author-generated, opinion from
from reference measuring influence ranging a reference member forms part
member from no influence (1) to of the Schiffmann and Wisenblit
2015:20, 2019:375).
186
Q9.5 Positive comments Ordinal 5-point Likert response format Author-generated, opinion from
of a friend, a measuring influence ranging non-commercial sources forms
newspaper article or from no influence (1) to part of the Schiffmann and
the views of an extremely large influence (5) Wisenblit decision-making
expert model (Schiffman & Wisenblit,
2015:20, 2019:375).
Q9.6 Negative comments Ordinal 5-point Likert response format Author-generated, opinion from
of a friend, a measuring influence ranging non-commercial sources forms
newspaper article or from no influence (1) to part of the Schiffmann and
the views of an extremely large influence (5) Wisenblit decision-making
expert model (Schiffman & Wisenblit,
2015:20, 2019:375).
Q9.7 Positive opinions of Ordinal 5-point Likert response format Author-generated, opinion from
individuals in the measuring influence ranging social class forms part of the
same social class from no influence (1) to Schiffmann and Wisenblit
2015:20, 2019:375).
Q9.8 Negative opinions of Ordinal 5-point Likert response format Author-generated, opinion from
individuals in the measuring influence ranging social class forms part of the
same social class from no influence (1) to Schiffmann and Wisenblit
2015:20, 2019:375).
Q9.9 Cultural and Ordinal 5-point Likert response format Forms part of the Schiffmann
subcultural factors measuring influence ranging and Wisenblit decision-making
187
from no influence (1) to model (Schiffman & Wisenblit,

extremely large influence (5) 2015:20, 2019:375).
To identify the Q10 Construct: Influences of communication sources on need recognition

communication
sources that influence Q10.1 Traditional Ordinal 5-point Likert response format Author-generated,
the individual’s need advertisements measuring influence ranging advertisements forms part of
to purchase personal from no influence (1) to the Schiffmann and Wisenblit
motor vehicle extremely large influence (5) decision-making model
insurance in South (Schiffman & Wisenblit,
Africa. 2015:20, 2019:375). A
distinction should be made
between traditional and online
advertisements.
Q10.2 Online Ordinal 5-point Likert response format Author-generated,

advertisements measuring influence ranging advertisements forms part of
from no influence (1) to the Schiffmann and Wisenblit
2015:20, 2019:375). A
distinction should be made
between traditional and online
advertisements.
Q10.3 Buzz agent Ordinal 5-point Likert response format Forms part of the Schiffmann
measuring influence ranging and Wisenblit decision-making
extremely large influence (5) 2015:20, 2019:375)..
188
Q10.4 Customised Ordinal 5-point Likert response format Forms part of the Schiffmann
messages measuring influence ranging and Wisenblit decision-making
Q10.5 User-generated Ordinal 5-point Likert response format Forms part of the Schiffmann
social media posts measuring influence ranging and Wisenblit decision-making
Q10.6 Paid-for social Ordinal 5-point Likert response format Forms part of the Schiffmann
media efforts measuring influence ranging and Wisenblit decision-making
To determine the Q11 Construct: Psychological attributes that influence the desire to recognise a need
individual’s
psychological Q11.1 Motivation Ordinal 5-point Likert response format Forms part of the Schiffmann
attributes that measuring influence ranging and Wisenblit decision-making
influence the need from no influence (1) to model (Schiffman & Wisenblit,
recognition to extremely large influence (5) 2015:20, 2019:375).
purchase personal
motor vehicle Q11.2 Perception about Ordinal 5-point Likert response format Forms part of the Schiffmann
insurance in South motor vehicle measuring influence ranging and Wisenblit decision-making
Africa. insurance from no influence (1) to model (Schiffman & Wisenblit,
Q11.3 Perception about a Ordinal 5-point Likert response format Forms part of the Schiffmann
certain short-term measuring influence ranging and Wisenblit consumer
insurance company from no influence (1) to decision-making model
extremely large influence (5)
189

2015:20, 2019:375).
Q11.4 Purchase and Ordinal 5-point Likert response format Forms part of the Schiffmann
consume information measuring influence ranging and Wisenblit decision-making
Q11.5 Personality Ordinal 5-point Likert response format Forms part of the Schiffmann
measuring influence ranging and Wisenblit decision-making
extremely large influence (5) 2015:20, 2019:375)..
Q11.6 Attitude towards Ordinal 5-point Likert response format Forms part of the Schiffmann
motor vehicle measuring influence ranging and Wisenblit decision-making
insurance from no influence (1) to model (Schiffman & Wisenblit,
Q11.7 Attitude towards a Ordinal 5-point Likert response format Forms part of the Schiffmann
certain short-term measuring influence ranging and Wisenblit decision-making
insurance company from no influence (1) to model (Schiffman & Wisenblit,
190
To investigate the Q12 Construct: Decision-making process to purchase

steps of the decision-
making process for Q12.1 Search for Ordinal 5-point Likert response Author-generated. Internal
the personal motor information - format measuring to which search for information, where
vehicle insurance previous extent usage ranging from no individuals remember facts about
industry in South experiences. extent (1) to extreme extent a policy or recall experiences of
Africa. (5) previous service deliveries
(Sethna & Blythe, 2019).
Q12.2 Search for Ordinal 5-point Likert response Author-generated. External

information - request format measuring to which search of information, where
quotations extent usage ranging from no individuals read about possible
extent (1) to extreme extent policies, searching on the
(5) Internet or visiting a broker
(Sethna & Blythe, 2019).
Q12.3 Compare the price Ordinal 5-point Likert response Author-generated. Individuals
of the different format measuring to which use price to measure whether
quotations to extent usage ranging from no the desired product/service will
measure extent (1) to extreme extent contribute to their need
(5) satisfaction and lifestyle
(Erasmus et al., 2019:458).
Q12.4 Compare the quality Ordinal 5-point Likert response Author-generated. Individuals
of the different format measuring to which use quality to measure whether
quotations to extent usage ranging from no the desired product/service will
measure extent (1) to extreme extent contribute to their need
(5) satisfaction and lifestyle
(Erasmus et al., 2019:458).
191
Q12.5 Compare the brand Ordinal 5-point Likert response Author-generated. Individuals
list to measure format measuring to which use performance to measure
extent usage ranging from no whether the desired
extent (1) to extreme extent product/service will contribute to
(5) their need satisfaction and
lifestyle (Erasmus et al.,
2019:458).
To assess the Q13 Age Nominal Open-ended question Demographic criterion. A factor
demographic that determines the policy
variables that premium, as discussed in
influence the Chapter 2. Demographics relate
individual’s need to to the individuals’ purchase
purchase personal decisions (Lamb et al., 2013:275,
motor vehicle 2018:337).
insurance in South
Africa. Q14 Population group Ordinal Multiple-choice question Demographic criterion
requiring a single response
and
To assess the Q15 Gender Nominal Dichotomous question Demographic criterion. A factor
consumer that determines the policy
demographic profile premium, as discussed in
within the South Chapter 2. Demographics relate
African personal to the individuals’ purchase
motor vehicle decisions (Lamb et al., 2013:275,
insurance industry. 2018:337).
and Q16 Marital status Ordinal Multiple-choice question Demographic criterion. A factor
To compile a general requiring a single response that determines the policy
demographic profile premium, as discussed in
Chapter 2. Demographics relate
192
that describes the to the individuals’ purchase

average personal decisions (Lamb et al., 2013:275,
motor vehicle driver in 2018:337).
South Africa.
Q17 Highest level of Ordinal Multiple-choice question Demographic criterion. A factor
education requiring a single response that determines the policy
and please specify option premium, as discussed in
Chapter 2. Demographics relate
to the individuals’ purchase
decisions (Lamb et al., 2013:275,
2018:337).
Q18 Employment status Ordinal Multiple-choice question Demographic criterion

requiring a single response
Q19 Which province Ordinal Multiple-choice question This question was included to
travel with motor requiring a single response ensure that the sample is
vehicle representative in South Africa
according to the number of
registered motor vehicles in each
province. Demographics relate to
the individuals’ purchase
2018:337).
Q20 Monthly income of Ordinal Multiple-choice question Essential to include, as

your immediate requiring a single response consumers might not be able to
family afford these services (KPMG,
2016:95). Demographics relate
193

2018:337).
Q21 Monthly personal Ordinal Multiple-choice question Essential to include, as

income requiring a single response consumers might not be able to
afford these services (KPMG,
2016:95). Demographics relate
2018:337).
Additional questions Q4 Type of personal Ordinal Multiple-choice question Author-generated

motor vehicle requesting a single response
insurance cover
Q5 Payment party Ordinal Multiple-choice question Author-generated

requesting a single response
Q6 Reason/s why Ordinal Multiple-choice question Author-generated

respondent has requesting multiple
insurance responses and ‘please
specify’ option
Q7 Reason/s why Ordinal Multiple-choice question Author-generated

respondent does not requesting multiple
have insurance responses and ‘please
specify’ option
194
Data-collection process
After developing the data-collection instrument, the researcher should continue to the
data-collection phase. The raw data gathered from the self-administered questionnaire
have to be edited, coded and captured (data processing) before it can be utilised to
answer the research questions. The data processing phase is discussed in Section
5.10 below.
The data-collection process for this study started in January 2020 and it ended in
November 2020, during the lockdown period of the COVID-19 pandemic in the
country. Data were collected during these 11 months by sending out the online self-
administered survey link via email and posts on Facebook and distributing paper-
based questionnaires (as indicated in Figures 5.6, 5.7, 5.8 and 5.9). The collection
period was extended a number of times due to challenges caused by the lockdown
and the low response rate. Thus, the survey link was emailed and posted several
times, and the paper-based questionnaire was also distributed during these 11 months
in efforts to increase the response rate. No incentives were offered to the respondents
for completing the questionnaire during the data collection process. The next section
explains data collection errors.
Data-collection errors
Some errors can occur during the data collection of a study, and it can never be totally
eliminated (Hair, Page & Brunsveld, 2020:36). It is, however, important for the
researcher to outline these errors and to minimise errors as far as possible. The errors
can be intentional, for example, if the researcher incorporates deliberate falsifications,
or unintentional, in the case of systematic or random sampling errors that arise during
the research process.
Once the data has been collected, the researcher has to check for errors. There might
be data missing in the questionnaire, or responses might suggest that the respondents
did not understand the question or the instructions were not properly followed (Hair et
al., 2020:36).
In the case of the online survey for the current study, there cannot be any missing data
as the online survey does not allow the respondent to continue to the next question if
the previous question is not answered. A pilot test was conducted to ensure that all
195
the respondents understood the questions. The respondents cannot misinterpret the
instructions as the questions will only appear on the screen; all the instructions are
built into the online questionnaire.
The data processing step is outlined in the next section.
DATA PROCESSING
The data processing activities are data capturing (or data entry), coding and editing
(Biemer et al., 2017:191). Data processing can also be done simultaneously while data
is still being collected, as is the case in an online self-administered questionnaire
(Fielding, Lee & Blank, 2017:214). The data processing activities implemented in this
study are discussed in this section below.
Data capturing
Data capturing refers to the step where data are transformed into a quantitative form
while also being converted into a machine-readable format that can be read and
manipulated by computer software (Babbie, 2021:413). The data for this study were
captured via LimeSurvey, as the respondents completed the questionnaire online. The
captured data were retrieved in an Excel spreadsheet (machine-readable form) before
it was uploaded to IBM SPSS (Version 27), the computer software used.
Data coding
According to Saunders et al. (2019:570), all data have to be coded by using numerical
numbers. By doing this, the data are converted into controllable and logical texts that
can be processed by a computer. When data are coded, each quantitative response
is given a code and this should be done before data analysis starts (Kara, 2019:39).
In essence, coding is the process of categorising information, and numerical coding is
the process of categorising and numbering information for quantitative analysis
(Adams & Lawrence, 2019). Numerical coding was conducted in this study, as data
were quantitatively analysed.
Almost all the questions in the self-administered questionnaire used in this study were
pre-coded, except for question 13, which is a fill-in question. Categories and numbers
were assigned for each question during the design of the questionnaire. Therefore,
196
coding was done before the questionnaire was captured online. No coding had to be
done after the data collection process.
Editing
Once the data capturing and coding have been conducted, data editing should be
performed. Data editing relates to the detecting and correcting of inconsistent data on
the spreadsheet before it is transferred into the software package (Babbie, 2016:390).
In other words, coding refers to the cleaning of the data to ensure that there are no
errors in the spreadsheet (Maylor, Blackmon & Huemann, 2017:299).
In this study, the data were carefully examined and edited to ensure that the errors
were identified and minimised, and all the inconsistencies were cleared. The data on
the Excel spreadsheet was reviewed to confirm that all the numerical values given of
each response were valid and correct. Once the data were cleaned and edited, the
data analysis process commenced, as discussed in the next section.
DATA ANALYSIS
After the data collection process where the data have been collected and processed,
the data can now be analysed. Data analysis can be described as applying reasoning
to understand the data collected (Zikmund et al., 2017:30). The quantitative data
analysis phase includes different methods for coding, categorising and assigning
meaning to data. This data is numeric and it is usually comprised of statistical
measures (Gliner et al., 2017:9). Data analysis for quantitative research contains
descriptive and inferential statistical analysis (model fit analysis), which each have
their own set of analysis techniques. These techniques are discussed below.
Descriptive statistics
Descriptive statistics are utilised to provide a broad description of the dataset and
descriptions of variables in the dataset. In addition, descriptive statistics should include
the size of the dataset, characteristics of the sample, and descriptions of the variables
associated with the main questions in the study (Williamson & Johanson, 2018:439).
In other words, the descriptive statistics in a study describe and summarise the data.
The researcher makes use of charts and/or graphs when it comes to the writing up of
descriptive statistics in a study (Gray, 2020:663). The types of descriptive statistics
197
are frequency counts, measures of central tendency and measures of dispersion
(Fallon, 2016:16-18). Each one of these descriptive statistics is explained below:
• Frequency counts are used when the number of occurrences of a category is

counted, and frequencies are usually reported in percentages (Leavy, 2017:11).
The purpose of frequency counts is to indicate how many respondents are in each
category. This is typically depicted in a graph or a chart to interpret it in a logical
manner (Gliner et al., 2017:169).
• Measures of central tendency describe the central point of a measure in terms

of a mean, mode or median (Maylor et al., 2017:302). The mean refers to the
average of the data, and the median is the point on the scale below where half of
the values lie, and the mode is the value with the most frequent occurrence
(Williamson & Johanson, 2018:436).
• Measures of dispersion indicate the degree to which the data is spread or

distributed. The measures used are range, interquartile, variance and standard
deviation. Range describes the difference between the maximum and minimum
values, and interquartile range is when data values are separated in four equal
sections and the two middle sections are used. Variance is the average of the
squared deviations from the mean of the distribution, and standard deviation is the
average deviation from the mean of the distribution (Williamson & Johanson,
2018:437). The variance and standard deviation are seen as the most useful
measures of dispersion (Zikmund et al., 2017:434).
As indicated above, descriptive statistics can be presented in a graphic format by

means of using diagrams and charts. Frequency tables, histograms, bar charts and
pie charts are mostly used as graphical representations of data when it comes to
descriptive statistics (Cooper & Schindler, 2014:534; Gliner et al., 2017:169; Cooper
et al., 2019).
In conclusion, frequency counts, measures of central tendency, and measures of

dispersion form part of the descriptive statistics in quantitative research. The
descriptive statistics of this study are presented in Chapter 6 and once this step was
completed, the researcher continued to the inferential statistics (model fit analysis).
198
Model fit analysis
The primary objective of this research is to develop a model that explains the purchase
decision of the consumers’ purchasing decision-making process in the South African
personal motor vehicle insurance market. It is, therefore, necessary to choose a
statistical method to analyse the structural relationships and determine the
acceptability of the proposed model.
Several statistical methods can be used to develop a consumer purchasing decision

model. It is, however, important to establish if an already tested model will be used or
a new conceptual model, as this will determine the model-building strategies. As
indicated earlier in this chapter, the theory in the literature review chapters (Chapters
2 and 3) was used to compile a conceptual CPDFPMVI model. Taking that the
CPDFPMVI conceptual model has not been tested before, a distinctive model-building
strategy ideal for this study was developed. The model-building strategy utilised to
achieve the final CPDFPMVI model is illustrated in Figure 5.12.
Figure 5.12: Model-building strategy to achieve CPDFPMVI model

199
The process underlying the model fit strategy consists of three phases, namely, the
Exploratory Factor Analysis (EFA) phase, Confirmatory Factor Analysis (CFA) phase
and the logistic regression phase. All these were, in essence, part of the process of
informing the logistic regression model. Model fit was just one aspect that was
considered as part of assessing the measurements. The decision behind each
statistical method in this model-building strategy is explained below.
EFA is a statistical method used to discover the underlying structure of a relatively

large set of variables (Hair et al., 2019:124). Taken that the conceptual model has not
been tested before, the EFA would be ideal for exploring and estimating the factor
structure of the current theoretical model (Matsunaga, 2010:98). EFA with a holdout
sample of 200 was performed to explore the factor structure as Phase 1 of this study.
CFA is a type of structural equational modelling (SEM) method used to test how well
a pre-specified measurement theory, composed of measured variables and factors,
fits the reality as captured data (Hair et al., 2019:660). CFA is also broadly used to test
the factor validity required in this step to ensure the model specification and
identification. Previous studies found it beneficial to use CFA as the follow-up strategy
on EFA (previous stage) to confirm the existence of a specific factor structure (Nayeem
& Casidy, 2015:70). CFA with a validation sample of 478 was performed to determine
a model fit.
The logistic regression model-building strategy aims to find the best fitting and most
parsimonious, interpretable model to describe the relationship between an outcome
(dependent) variable and a set of independent (predictor or explanatory) variables
(Hosmer, Lemeshow & Sturdivant, 2013:1). Furthermore, logistic regression is a
statistical model-building strategy utilised when the dependant variables are
categorical variables (binary or dichotomous), and the independent variables are
metric variables (Hair et al., 2019:551).
In this study, the dependent variable is the answers to Question 3 in the questionnaire
(see Appendix A). This question was a dichotomous question, and it will be the
outcome of this model. The independent variables are the latent variables tested and
accepted in Phase 2 (the CFA). Also, certain demographic variables, as tested in the
questionnaire, are included in the independent variables/covariances. An overall
sample size of over 400 is recommended to achieve the best results while using the
200
logistic regression model-building strategy (Hair et al., 2019:557). The overall sample
in this study was over the recommended 400, therefore, this model-building strategy
was fitting. An exploration sample of 200 was used to conduct the univariate analysis
to identify all the important covariances. Furthermore, a validation sample of 478 was
used for model estimation in this last phase.
These methods and all the other possible statistical tools that can be used to test the
conceptual model for this study are comprehensively discussed in Chapter 7.
Furthermore, a motivation for the use of each statistical method used in the model-
building strategy is provided in Chapter 7. In the next section, the reliability and validity
measures are discussed.
RELIABILITY AND VALIDITY
The survey research instrument (the self-administered questionnaire in this study)

should be reliable and valid in order to ensure that both the results and conclusions
are accurate and precise (Ruel, Wagner & Gillespie, 2016:78). Reliability means
consistency, and validity refers to the accuracy (Bryman, 2016:157; Clark et al.,
2019:52). The validity and reliability measures followed in this study are discussed
next.
Reliability
Reliability can be defined as the degree to which measures are free from random error,
and consequently, produce consistent results (Zikmund et al., 2017:274). Simply put,
reliability measures whether or not consistent results will be obtained if the research
is repeated (Ruel et al., 2016:78).
According to Mitchell in Saunders et al. (2019:518), there are three ways to test the
reliability at the questionnaire design stage, namely, test re-test, internal consistency,
and alternative form. These three ways to test reliability are explained below.
• Test re-test includes the administration of the same scale or measure to the same
respondents at two separate times to test the stability. When the measure is stable
over time, a test that is administered under the same conditions will obtain similar
results each time (Zikmund et al., 2017:174).
201
• Internal consistency can be tested by means of split-half reliability, or the more
common method to test for internal consistency is Cronbach’s alpha (Patten &
Newhart, 2018:144). The split-half method is when the researcher randomly
divides the scale items in half, and takes the results obtained from one half of the
scale items and checks it against the results obtained from the other half (Zikmund
et al., 2017:275). To get the Cronbach’s alpha, the researcher has to calculate the
average of the coefficients from all the possible combinations of split halves. An
alpha coefficient ranges between one (indicating perfect internal consistency) and
zero (indicating no internal consistency) (Bell et al., 2019:173). Values of 0.7 and
above reveal that the questions combined in the scale are measuring the same
thing (Saunders et al., 2019:518).
• Alternative form relates to the method that is used when responses are compared
to alternative forms of the same question or groups of questions (Saunders et al.,
2019:518).
Internal consistency was assessed to ensure reliability in this study. The Cronbach
alpha coefficient was used to ensure internal consistency as this assesses the degree
to which the instrument items, where applicable, are homogenous, and whether it
reflects the same underlying construct (Cooper & Schindler, 2014:260; Zikmund et al.,
2017:274; Cooper et al., 2019).
Reliability in terms of the model-building strategy is a bit more complex. The reliability
of a CFA model is reported by considering the CR (Composite Reliability) value, which
indicates the reliability and internal consistency of a latent variable (Hair et al.,
2019:763). The rule of thumb is that the CR value should be higher than 0.70 to
indicate internal consistency, while in exploratory research, 0.60 to 0.70 is considered
acceptable (Hair et al., 2017:122).
The final CFA model demonstrated overall reliability, and it was regarded as an
acceptable model to use in Phase 3 (the logistic regression model-building strategy).
The reliability aspects also are comprehensively discussed in Chapter 7.
Validity
Validity determines whether or not the self-administered questionnaire measures what

it is supposed to measure (Bell et al., 2019:174). Firstly, the researcher conducted
pilot tests of the self-administered questionnaire to ensure validity (Dikko 2016:521).
202
There are, however, different types of validity, namely, content (or face) validity,
criterion validity and construct validity (Zikmund et al., 2017:276-277). These different
types of validity and how they were implemented in this study are explained below.
• Content or face validity determines if the measure (self-administered

questionnaire) reflects the content of the concept in question. Individuals with
experience in the field can be asked to act as judges in determining whether or not
the measure reflects the concept in question (Bell et al., 2019:174). The content
validity was ensured as experts in the field validated the questionnaire.
• Criterion validity is established when the following question can be answered:

“Does my measure correlate with other measures of the same construct?” Criterion
validity can be categorised as concurrent validly or predictive validity (Zikmund et
al., 2017:277). Concurrent validity indicates how well the measure can estimate
the present performance, and predictive viability is determined when a new
measure predicts a future event (Bell et al., 2019:174). The self-administered
questionnaire used in the current study was judged in terms of relevance, freedom
from bias, reliability and availability.
• Construct validity is determined during the statistical analysis phase of a study

as it indicates the degree to which the measure confirms the concept/construct
gained from the theory (Zikmund et al., 2017:277). Therefore, it is important to
understand the theoretical basis of the measurements obtained. Construct validity
was tested through EFA and CFA.
Validity in each phase of the model-building strategy is unique to each analysis method
used. For the EFA, discriminate validity, which refers to measures of constructs that
should theoretically not be related to each other (Fornell & Larcker, 1981; Hair et al.,
2019:761), was used.
Three types of validity had to be determined during CFA. Firstly, construct validity,
which is the main objective of CFA, deals with the measurement's accuracy. Once a
good fit is established, the four additional components of construct validity should be
evaluated (Hair et al., 2019:675). These components include: conforming to its
conceptual definition, that it is unidimensional, and meets the necessary levels of
reliability (Hair et al., 2019:162). All these components will be evaluated in Section
7.3.4 when a model fit is achieved.
203
Secondly, convergent validity assesses the extent to which a measure correlates
positively with others (Civelek, 2018:32). Therefore, the AVE (average variance
extracted) value should be higher than 0.5 (Hair et al., 2017:122). However, it is
important to note that while the AVE values should preferably be higher than 0.05, the
context of the CR should also be taken into account.
Thirdly, discriminant validity refers to the extent to which a latent variable is distinct
from other latent variables or indicators (Hair et al., 2019:676). The Fornell and Larcker
criterion can be used to assess discriminant validity, and this method compares the
square root of the AVE with the correlation of latent variables. The square root of each
construct’s AVE should have a higher value than the correlations with other latent
variables (Fornell & Larcker, 1981; Hair et al., 2019:761). All three of these types of
validity were ensured with the final CFA model.
With the logistic regression model-building strategy, the validation sample of n=478
used to perform this logistic regression was particularly useful, as it provides evidence
of external validity. External validity refers to the extent to which the results of a specific
analysis can be generalised in other situations/populations (Hair et al., 2019:373). A
more detailed discussion on the validity aspects in each phase of the model-building
strategy is provided in Chapter 7. The next section presents the data findings.
PRESENTATION OF DATA FINDINGS
The final step in the research process is to present the data findings. This final stage
involves the interpretation of information and the drawing of conclusions (Zikmund et
al., 2017:30). Therefore, it is vital that the findings are presented effectively and is not
merely a summary of the findings. The data findings are presented in Chapter 8 of this
thesis. The section below covers the ethical considerations of this study.
ETHICAL CONSIDERATIONS
Ethics in business research involves the moral principles and values that influence
how a researcher conducts his/her research activities (Ghauri et al., 2020:23). Ethical
principles and standards should be implemented during each step in the research
process (Babin et al., 2015:41). Ethical principles are divided into four main categories,
as follows (Bell et al., 2019:114; Gray, 2020:83):
204
• No harm to respondents;
• Obtain informed consent from respondents;
• Respect respondents’ privacy; and
• Avoid the use of deception.
The research process in this study was directed by the ethical standards as set by the
University of South Africa. Ethical clearance (Appendix B) was obtained from the
Business Management Research Ethics Review Committee at the University of South
Africa before the pilot test and data collection started. In addition, the researcher
consulted UNISA’s Ethical Policies and Procedures to ensure that ethical research
was conducted. All the ethical requirements outlined in this policy were adhered to,
specifically, during the data collection and data analysis phases and these ethical
considerations are discussed below.
Voluntary participation
No one should be forced to participate in a survey; therefore it should be voluntary,

and if an individual decides that he/she does not want to partake in the survey the
individual’s decision should be respected (Sekaran & Bougie, 2020). Respondents
should be given a choice to withdraw from the survey at any stage.
The cover letter to the current study’s questionnaire informed respondents that
participation in this study is voluntary and they could withdraw at any point.
No harm to respondents
It is the responsibility of the researcher to determine if there is a possibility of harm to

the respondents, and if there is a possibility, measures should be taken to minimise
the harm (Bell et al., 2019:114). Research can be considered harmful to the
respondent if it triggers embarrassment, ridicule, belittlement or general mental
distress (Sudman in Gray, 2020:83), or if it generates anxiety or stress to the
respondents or even if it generates negative emotional reactions (Gray, 2020:83).
This study’s purpose was to investigate consumers’ purchasing decision process in

the South African personal motor vehicle insurance market, and as such, carried low
risk, as human respondents were asked to complete an anonymous online self-
administered questionnaire. There were no physical, psychological, social or any other
205
risks to any of the respondents. The questionnaire did not include sensitive questions,
and the respondents were reassured that there would be no right or wrong answers in
the questionnaire. The only foreseeable risk is one of potential inconvenience or
discomfort to the respondents in completing the questionnaire.
Anonymity and confidentiality
Anonymity means that the respondents should not be identified without their
permission, and confidentiality refers to the respondents’ information that should be
kept confidential (Zikmund et al., 2017:165). Personal information obtained from the
respondents should not be shared, especially if permission was not requested from
them, and respondents should not be victimised when participating in the study.
Anonymity and confidentiality were ensured in this study by informing respondents in

the respondent information sheet about the following:
• The survey is anonymous and confidential.
• The researcher will have no way of connecting the information provided by

respondents to them personally, and the answers will be used for research
purposes only.
• Findings are anonymously processed and will be used in a PhD dissertation, and
it could also be shared in conference presentations, the publication of academic
articles and other communication.
Informed consent
Respondents have to be informed about the nature of the study (Zikmund et al.,
2017:165), and it is vital that informed consent is obtained from all respondents
included in the study (Sekaran & Bougie, 2020:159). Informed consent means that the
respondents are provided with sufficient information about the research study for them
to be able to make an informed decision as to whether or not they want to partake in
the study (Bell et al., 2019:118). The researcher has to disclose and explain the
procedures of the research design to the respondents before permission to carry out
the study is asked in the informed consent section of the questionnaire. If an
organisation is also involved in the study, consent should be given by the organisation
as well (Gray, 2020:84). Only the respondents’ informed consent was required in this
study as no organisations were included.
206
According to Saunders et al. (2019:260), a signed informed consent in an online
questionnaire can be facilitated using an electronic checkbox. This study used this
method, as the respondents gave informed consent by clicking on the “next” button.
Appendix A includes the information about the study and the instructions provided, as
well as the informed consent.
The above-mentioned points are only a few of the important ethical considerations that
were used to guide the data-collection process of the current study. The different
ethical aspects were considered during the different stages of the research process.
These considerations are outlined in Figure 5.13.
207
Figure 5.13: Ethical matters during different stages of the research process
Source: Adapted from Saunders et al. (2019:264)
The ethical matters in the figure above were considered and assured during each one
of the research stages. The conclusion of this chapter is provided in the following
section.
208
CONCLUSION
The purpose of this chapter was to discuss the research methodology used in the
current study. This chapter revisited the study’s research objectives and provided a
figure that stipulates the steps followed in the research process. The specific research
design was discussed, and the research terminology used in the rest of the chapter
was briefly explained. The descriptors of the overall research design of this study were
covered and the quantitative research approach used was explained and motivated.
A further discussion of the sampling design followed in this study can be found in this
chapter. Two sampling methods were used, namely, convenience sampling and
snowball sampling. An online self-administered questionnaire was developed and
used as the data-collection instrument.
Furthermore, the data-collection process followed in this study was explained in detail.
The primary data was collected by means of an online self-administered questionnaire,
which was thoroughly edited, captured, coded and cleaned before the data could be
analysed. An in-depth discussion of the data processing steps that were followed in
the current study was presented, as well as a framework of the data analysis steps
executed during this study. A number of statistical analyses were utilised to process
and analyse the data obtained. The statistical data analysis can be found in the next
chapter, and it will be discussed extensively. This chapter concluded with a discussion
of the reliability and validity of the research instrument, as well as the ethical
considerations that were implemented throughout the research process.
209
DESCRIPTIVE FINDINGS OF RESEARCH
INTRODUCTION
A quantitative research approach was followed to determine the purchasing decision

process of consumers (insured) and potential consumers (non-insureds) regarding
personal motor vehicle insurance by developing a unique model that will assign a
probability to each of the possible choices based on the context of the decision. The
descriptive findings presented in this chapter and the inferential statistical analysis
(revealed in Chapter 7) were provided to address this study's primary and secondary
research objectives, as specified in Table 6.1.
Table 6.1: Primary and secondary research objectives
• To determine the company’s marketing efforts that influence the individual’s




vehicle insurance market in South Africa.

210

The descriptive statistics presented in this chapter mainly reveal the respondents’
(who are all personal motor vehicle drivers) demographic composition, and a
demographic profile of the average personal motor vehicle insurance consumer and
personal motor vehicle driver were developed. Additionally, the descriptive statistics
offer a response profile of the statements investigating respondents’ purchase
decision process regarding personal motor vehicle insurance. As mentioned in
Chapter 5, the data analysis process was followed in a systemic manner. The
descriptive statistics of this study are presented in the sections that follow.
DESCRIPTIVE ANALYSIS OF RESEARCH FINDINGS
The frequency tables (see Appendix C), which indicate the respondents' frequency or
incidences that selected an option and valid percentages, were composed on IBM
SPSS (Version 27), a statistical software package. Figures and tables were then
created on Microsoft Excel. The measuring instrument (refer to Appendix A) was
developed to measure respondents’ purchase decision process regarding personal
motor vehicle insurance. The 21 questions in the instrument, which include a range of
items or statements, were intended to determine respondents’ purchase decision
process regarding personal motor vehicle insurance. In theory, these items in the
instrument should deliver a consumer purchase decision model for the South African
personal motor vehicle insurance industry.
As mentioned above, this chapter reveals the descriptive analysis of the data and
illustrates the respondents' demographic composition, using non-probability sampling
methods.
211
DEMOGRAPHIC COMPOSITION OF THE RESPONDENTS
The following demographic information was collected from the respondents to

compose a demographic profile of personal motor vehicle drivers and consumers in
South Africa:
• Age
• Population group
• Gender
• Marital status
• Highest level of education
• Employment status
• Province mostly travelled in with personal motor vehicle
• Household income
• Personal income
The demographic profiling results of the questions asked are illustrated in the
frequency tables (see Appendix C). The sections below present all the most critical
findings from the data analysis.
Age
The respondents were asked to specify their age. A nominal open-ended question was
provided, and respondents were requested to type or write their ages on the space or
line provided in the questionnaire. Table C1.1 in Appendix C reveals the statistics for
the age groups 18 – 30, 31 – 40, 41 – 50, 51 – 60, and Over 60.
Figure 6.1 indicates the result from the ‘age’ variable.
212
Age
30,0%
25,7%
25,0%
19.7%
20,0% 18,9% 18,9%
16,8%
15,0%
10,0%
5,0%
0,0%
18-30 31-40 41-50 51-60 over 60
Figure 6.1: Age (n=678)

As shown in the figure above, the ages are grouped, and the largest age group
represents 25.7% of the respondents (174 respondents) in the 31-40 age group. The
41-50 age group consists of 19.7% of the respondents (134 respondents), and both
the age groups 18-30 and 51-60 equally represent 18.9% of the respondents (128
respondents). The over 60 age group is the smallest group of 16.8% of the
respondents (114 respondents).
As mentioned in Chapter 2 (Section 2.7.2.1), most US licenced drivers fall within the
35-39 age category, which is 90.9% of the country's licenced drivers (Buchholz,
2020). According to the findings of this study, the largest group of drivers (25.7%)
also fall within a similar age category of 31-40 years. In contrast to the findings in
this study where the over 60 age group represented the smallest group, in England,
there is a large increase in the number of older individuals (aged 70 and older) that
have full driving licences (RAC Foundation, 2021).
Population group
All the respondents were requested to indicate their population group. This ordinal,
multiple-choice question requiring a single response had six options: Asian, Black
African, Coloured, Indian, White, and Other with a ‘please specify’ line. The frequency
213
statistics (Table C2 in Appendix C) were used to develop Figure 6.2, which illustrates
the results of the ‘population group’ variable.
Population group
70,0%
60,0%
60,0%
50,0%
40,0%
30,0% 26,1%
20,0%
8,7%
10,0% 4,2%
0,5% 0,5%
0,0%
Asian Black African Coloured Indian White Other
Figure 6.2: Population group (n=662)

Figure 6.2 shows that the largest population group in this study is White, comprising
of 60% or 397 respondents. The second-largest population group is Black African,
which consists of 26.1% of the respondents (173 respondents). Thirdly, the Coloured
population group makes up 8.7% of the respondents (58 respondents), and the Indian
population group 4.2% of the respondents (28 respondents). Three respondents
indicated that they form part of the Asian population group, which is 0.5% of the
respondents. The rest of the of respondents (0.5% or 3 respondents) selected the
Other option, but none of them specified another population group.
The findings of this study reveal that the white population group is the largest
population group participating in the study. As indicated in Chapter 2 (Section
2.7.2.1), the percentage distribution of vehicle ownership by population groups in
South Africa are as follows: Black African 19.8%, Coloured 41.8%, Indian/Asian
81.3% and White 94.2% (Statistics South Africa 2017b:12). Taking into
consideration that the Black African population is the largest population group in the
country, and only 19.8% of the entire Black African population own a motor vehicle.
The 19.8% represents the biggest population group that owns motor vehicles in
214
South Africa, while the white population the smallest population group. Therefore, it
would have been ideal to obtain a similar response rate to these statistics.
Gender
The respondents were asked to indicate their gender. This nominal dichotomous
question had two options, namely, male and female. In Table C3 (see Appendix C),
the frequency statistics of this question are provided. Figure 6.3 presents the
descriptive statistics of this question which tested the ‘gender’ variable.
The figure below reveals that 52.5% of the respondents (356 respondents) indicated
that they are male, and the other 47.5% of the respondents (322 respondents)
indicated that they are female. Thus, male motorists represent the biggest gender
group of motorists who completed this survey.
Gender
Female Male
47.5% 52.5%
Male Female
Figure 6.3: Gender (n=678)

As stated in Chapter 2 (Section 2.7.2.1), Maluleke (2020:2) specified that

approximately 51.1% of the South African population is female, and the other 48,9%
is male. Interestingly, 37,6% of the South African male population own motor
vehicles and only 19.6% of the female population own motor vehicles (Statistics
215
South Africa 2017b:12). This matches this study's finding, which reveals that men
are the largest gender group under South African motorists.
Marital status
The respondents were requested to answer an ordinal multiple-choice question

requiring a single response to indicate their marital status. The marital status options
included single, married, living with a partner but not married, divorced and widowed.
Figure 6.4 presents the marital status variable's descriptive findings, derived from the
frequency table (Table C4 in Appendix C).
Marital status
Widowed
Divorced 3.6%
Living with partner, 7,1%
but not married
10,5%
Single
23.2%
Married
55,6%
Single Married Living with partner, but not married Divorced Widowed
Figure 6.4: Marital status (n=678)

As depicted in the figure above, more than half of the respondents indicated that they
are married, representing 55.6% of the respondents (377 respondents). The single
category consists of 23.2% of the respondents (157 respondents), while 10.5% of the
respondents (71 respondents) specified that they are living with a partner but are not
216
married, 7.1% of the respondents (48 respondents) indicated that they are divorced,
and 3.6% of the respondents (25 respondents) are widowed.
As indicated in Chapter 2 (Section 2.7.2.1), a study conducted amongst Ghanaian

drivers found that 59.6% are married and 27.1% are single (Atombo et al., 2017:11).
It is interesting to note that these findings look very similar to this study’s findings
related to marital status.
Highest level of education
The respondents were asked to indicate their highest level of education by answering
the ordinal multiple-choice question requiring a single response. The following options
were available: Less than Grade 12, Grade 12 Matric, post-Matric Certificate, Higher
Certificate, Diploma, Higher Diploma, Post Higher Diploma (Master’s or Doctoral
Diploma), Bachelor Degree, Honours Degree and Doctorate Degree. The respondents
were instructed to select only one option.
The figure illustrates the descriptive findings of the “highest level of education”
variable, and the frequency table (Table C5) is available in Appendix C.
Highest level of education

30,0%
25,7%
25,0%
20,0%
13,6% 13,8%
15,0%
9,2% 8,9%
10,0% 6,7% 7,5%
5,0% 4,3%
5,0% 3,4%
1,9%
0,0%
Figure 6.5: Highest level of education (n=676)

217
Figure 6.5 presents the descriptive findings, and the respondents indicated the
following:
• The highest level of education group is Grade 12, making up 25.7% of the
respondents (174 respondents).
• The Bachelor’s Degree category scored 13.8% of the respondents (93

respondents).
• In the Diploma category, 13.6% of the respondents (92 respondents) revealed that
their highest education level is a diploma.
• Of the respondents, 62 (9.2% of the respondents) indicated that their highest

education level is an Honours Degree.
• Of the respondents, 60 (8.9% of respondents) specified that a Master’s Degree is

their highest level of education.
• The Higher Diploma category makes up 7.5% of the respondents (51 respondents).
• The respondents who indicated that their highest level of education is less than
grade 12 made up 6.7% of the respondents (45 respondents).
• Of the respondents, 34 (5% of the respondents) revealed that their highest level of
education is a Post-Matric Certificate.
• The Higher Certificate category presents 4.3% of the respondents (29

respondents).
• The respondents who indicated that a doctorate degree is the highest level of
education made up 3.5% of the respondents (23 respondents).
• Only 1.9% of the respondents (13 respondents) specified that their highest level of
education is a Post Higher Diploma.
The findings of this study align with a study conducted in Qatari (mentioned in
Chapter 2, Section 2.7.2.1), which found that most drivers with a driver’s license
have a secondary education level (Soliman et al., 2018:655). Secondary education
in Qatari is equivalent to high school education in South Africa. In addition, several
studies found that drivers' education levels form part of the demographic factors that
218
impact the drivers’ driving behaviour (Atombo et al., 2017; Mohamed & Bromfield
2017; Soliman et al., 2018).
Employment status
All the respondents were requested to answer this ordinal multiple-choice question
requiring a single response which included the following options: Student,
Unemployed, Temporarily employed (including Fixed Term Contracts), Permanently
employed on a full-time basis, Permanently employed on a part-time basis and
Retired.
Figure 6.6 depicts the descriptive statistics of the ‘employment status’ variable. Table
C5 in Appendix C provides the frequency counts.
Employment status
70,0%
57,8%
60,0%
50,0%
40,0%
30,0%
17,8%
20,0% 10,5%
2,5% 6,2% 5,2%
10,0%
0,0%
Student Unemployed Temporarily Permanently Permanently Retired
employed employed on a employed on a
(including Fixed full-time basis part-time basis
Term Contracts)
Figure 6.6: Employment status (n=678)

The table above illustrates that more than half of the respondents, 57.8% (392
respondents), are permanently employed on a full-time basis, and 17.8% of the
respondents (121 respondents) are retired. The temporarily employed category
presents 10.5% of the respondents (71 respondents), the unemployed category
consists of 6.2% of the respondents (42 respondents), and the permanently employed
on a part-time basis category makes up 5.2% of the respondents (35 respondents).
Only 2.5% of the respondents (17 respondents) indicated that they are students.
This study found that most respondents (57.8%) are permanently employed on a
full-time basis. Interestingly, as specified in Chapter 2 (Section 2.7.2.1), employment
219
status findings in a study conducted on drivers’ driving behaviour in Denmark
revealed that 59.3% of the respondents were employed, 28.5% retired, 5.4% self-
employed, 4.8% unemployed and 2% students (Martinussen et al., 2017:2). Even
though the current study used different categories, these statistics appear to be in
line with another global study’s findings.
Province mostly travelled in with personal motor vehicle
An ordinal multiple-choice question requiring a single response was included in the

questionnaire to determine in which province the respondents mostly travel with their
personal motor vehicles. Respondents were asked to identify in which of the nine
provinces (Gauteng, Western Cape, KwaZulu-Natal, Free State, North West, Eastern
Cape, Mpumalanga, Limpopo and Northern Cape) they mostly travel with their
personal motor vehicles.
Figure 6.7 illustrates the descriptive findings from the ‘Province mostly travelled in with
the personal motor vehicle’ variable. The frequency table (Table C7) is available in
Appendix C.
Province mostly travelled in with the personal motor vehicle

50,0% 43,4%
45,0%
40,0%
35,0%
30,0%
25,0%
20,0% 12,8%
15,0% 10,9% 10,1%
10,0% 5,8% 5,6% 4,9% 3,7% 2,8%
5,0%
0,0%
Figure 6.7: Province mostly travelled in with the personal motor vehicle (n=678)
The figure above presents the following findings:
• Of the respondents, 43.4% (294 respondents) indicated Gauteng.
• The Western Cape category consists of 12.8% of the respondents (87

respondents).
220
• Of the respondents, 10.9% (74 respondents) revealed that KwaZulu-Natal is the
province they mostly travel in.
• Free State accounts for 10.1% of the respondents (69 respondents).
• The North West province represents 5.8% of the respondents (39 respondents).
• The Eastern Cape category makes up 5.6% of the respondents (38 respondents).
• Of the respondents, 4.9% (33 respondents) indicated Mpumalanga.
• Of the respondents, 3.7% (25 respondents) revealed that Limpopo is the province
they travel in most.
• Only 2.8% of the respondents (19 respondents) specified Northern Cape.
As shown in Figure 6.7 above, the three provinces in which respondents mostly travel
with their personal motor vehicles are Gauteng (43.4%), Western Cape (12.8%) and
KwaZulu-Natal (10.9%).
As indicated in Chapter 2 (Section 2.7.2.1), the live vehicle population (registered

motor vehicles) in each province of South Africa are Gauteng 41.4%, Western Cape,
17.2%, KwaZulu-Natal 13.7%, Eastern Cape 6.4%, Mpumalanga 6%, Limpopo
4.8%, North West 4.5%, Free State 4.3% and Northern Cape 1.7% (eNaTIS, 2021).
Therefore, the three provinces with the largest live vehicle population in the country
are Gauteng, Western Cape and KwaZulu-Natal, which are also the three provinces
where respondents mostly travel with their personal motor vehicles.
The number of motor vehicle drivers in South Africa can also be linked to the
population density of each province. The provinces with the highest population
density in South Africa are (1) Gauteng, (2) KwaZulu-Natal and (3) Western Cape
(see Section 2.7.2.1). Therefore, this explains why most registered motor vehicles
are also in these three provinces.
221
Household income
The respondents were asked to indicate which of the following income categories best
describe the monthly income of their immediate family from all sources (including the
respondents' income), before tax and other deductions: R0 – R800, R801 – R1 200,
R1 201 – R2 500, R2 501 – R6 000, R6 001– R16 000, R16 001 – R26 000, R26 001
– R36 000, R36 001 – R46 000, R46 001 – R56 000, R56 001 or more, and Prefer not
to say/Don’t know/NA. The purpose of this ordinal multiple-choice question requiring
a single response is to determine respondents' household income.
Table C8 (in Appendix C) presents the frequency count, and Figure 6.8 illustrates the
descriptive statistics of the ‘household income’ variable.
Household income
25,0% 21,4%
19,8%
20,0%
15,0% 11,4% 11,2%

10,0%
10,0% 6,5% 7,2% 6,0%
5,0% 3,5%
1,2% 1,8%
0,0%
R0 – R801 – R1 201 – R2 501 – R6 001– R16 001 –R26 001 –R36 001 –R46 001 – R56 001 Prefer
R800 R1 200 R2 500 R6 000 R16 000 R26 000 R36 000 R46 000 R56 000 or more not to
say/Don’t
know/NA
Figure 6.8: Household income (n=678)

Figure 6.8 indicates that 21.4% of the respondents (145 respondents) indicated that
their household income falls in the R6 001-R16 000 category, while 19.8% of the
respondents (134 respondents) specified that they have a household income of R56
001 and more. The R16 001-R26 000 category represents 11.4% of the respondents
(77 respondents), the ‘prefer not to say’ category represents 11.2% of the respondents
(76 respondents), the 36 001-R46 000 category represents 7.2% of the respondents
(49 respondents), the R2 501-R6 000 category represents 6.5% of the respondents
(44 respondents), the R46 001-R56 000 category represents 6% of respondents (41
respondents), the R0-R800 category represents 3.5% of the respondents (24
respondents), the R1 201-R2 500 category represents 1.8% of the respondents (12
222
respondents), and the R801-R1200 category represents 1.2% of the respondents (8
respondents).
The findings indicate that 75.8% of the respondents have a monthly household
income of more than R6 000. As mentioned in Chapter 2 (Section 2.7.2.1), while
information could be obtained about the household income composition in South
Africa, nothing could be obtained about the average household income currently in
South Africa.
Personal income
The respondents were asked to indicate which of the following income categories best
describe their monthly personal income from all sources, before tax and other
deductions: R0 – R800, R801 – R1 200, R1 201 – R2 500, R2 501 – R6 000, R6 001–
R16 000, R16 001 – R26 000, R26 001 – R36 000, R36 001 – R46 000, R46 001 –
R56 000, R56 001 or more, and Prefer not to say/Don’t know/NA. The purpose of this
ordinal multiple-choice question requiring a single response is to determine
respondents' personal income. Table C9 (in Appendix C) presents the frequency
count, and Figure 6.9 illustrates the descriptive statistics of the ‘household income’
variable.
Personal income
15,0%
16,0%
13,1% 12,5%
14,0% 11,4% 11,5%
12,0% 9,3%
10,0% 8,0%
8,0% 5,8% 6,2%
6,0% 3,2% 4,0%
4,0%
2,0%
0,0%
Figure 6.9: Personal income (n=678)

As shown in Figure 6.9, the highest earned personal income category is R6 001-R16
000, representing 15% of the respondents (102 respondents). The R16 001-R26 000
223
category makes up 13.1% of the respondents (89 respondents), the R56 001 or more
category is made up of 12.5% of the respondents (85 respondents), the R26 001-R36
000 category is made up of 11.5% of the respondents (78 respondents), the R0-R800
category is made up of 11.4% of the respondents (77 respondents), the ‘prefer not to
say’ category is made up of 9.3% of the respondents (63 respondents), the R2 501-
R6 000 category is made up of 8% of the respondents (54 respondents), R46 001-
R56 000 category is made up of 6.2% of the respondents (42 respondents), the R36
001-R46 000 category consists of 5.8% of the respondents (39 respondents), the R1
201-R2 500 category consists of 4% of the respondents (27 respondents), and the
R201-R1 200 category is made up of 3.2% of the respondents (22 respondents).
The findings reveal that 49.1% of the respondents earn more than R16 000 per
month. As specified in Section 2.7.2.1 (Chapter 2), the average personal income in
South Africa is R22 500 per month. However, the individuals with the highest
earnings in the country are the reason for this high average, as most South African
workers can only desire to reach a salary above R20 000 (Head, 2020).
Consumer demographic profile
The descriptive demographic statistics presented above can explain the consumer
profiles (consumers who purchase personal motor vehicle insurance) that participated
in this study. Only the respondents who indicated that they have personal motor
vehicle insurance (answered yes in question 3 of the questionnaire, see Appendix A)
were considered for this consumer demographic profile as they have purchased
personal motor vehicle insurance in South Africa. Therefore, this study's consumer
demographic profile can be used to develop a description of the average personal
motor vehicle insurance consumer in South Africa.
Table 6.2 presents the statistics and a description of the average personal motor
vehicle insurance consumer.
224
Table 6.2: Consumer demographic profile according to this study
Construct Average personal motor vehicle insurance

consumer according to the study
Age The average age is 50 years old.

(Standard deviation=15.9)
Population group Likely to be a white consumer (85.2%).

*[Accessed 95%LCL=80.9% and 95%UCL=98.1%]
Gender Equal probability in terms of gender.

(Male=47.6% and Female=52.4%)
*[Accessed Male - 95%LCL= 41.7% and 95%UCL=
53.6%]
*[Accessed Female - 95%LCL= 46.4% and 95%UCL=
58.3%]
Marital status Likely to be married (64.3%).

Highest level of education Likely to have a Post Higher Diploma (Master’s or

Doctoral Diploma) or higher education level (56.3%).
Employment status Likely to be permanently employed on a full-time basis

(63.9%).
[Accessed *95%LCL=58.1% and 95%UCL=69.5%]
Province mostly travelled in Likely to travel most in Gauteng (61.3%).

with personal motor vehicle *[Accessed 95%LCL=55.4% and 95%UCL=67%]
Household income Likely to have a household income of more than R16

001 (89,5%).
Personal income Likely to earn more than R16 001 personally (86.8%).
*95.0% Lower Confidence Level (LCL) and 95.0% Upper Confidence Level (UCL) indicates
a range of values that is likely to encompass the true value
The table above describes the average personal motor vehicle insurance consumer.
This description was developed from the demographic profile of respondents who
have purchased personal motor vehicle insurance and who participated in this study.
This description explains the different characteristics of the average personal motor
225
vehicle insurance consumer. Short-term insurance companies can use this
information about consumers in their marketing efforts to target the correct audience.
Respondent demographic profile
The descriptive demographic statistics presented above can explain the respondent
profiles that participated in this study (Sections 6.3.1 to 6.3.9). Therefore, this study's
demographic profile can be used to develop a description of the average personal
motor vehicle driver in South Africa. Table 6.3 presents a description of the average
personal motor vehicle driver.
Table 6.3: A description of the average personal motor vehicle driver
Construct Average personal motor vehicle driver according to

this study

(Standard deviation=15.1)
Population group Likely to be a white driver (58.6%).

*[Accessed 95%LCL=54.1% and 95%UCL=63%]
Gender Likely to be a male driver (55.2%).

*[Accessed 95%LCL= 50.8% and 95%UCL= 59.6%]
Marital status Likely to be married (54.4%).

Highest level of education Likely to have a Post-Matric Certificate (Grade 12) or

higher education level (66.9%).
Employment status Likely to be permanently employed on a full-time basis

(63.8%).
*[Accessed 95%LCL= 59.4% and 95%UCL= 68%]
Province mostly travelled in Likely to travel most in Gauteng (43.9%).

with personal motor vehicle *[Accessed 95%LCL=39.5% and 95%UCL=48.4%]
Household income Likely to have a household income of more than R6 001

(76.9%).
*[Accessed 95%LCL=61% and 95%UCL=95.6%]
Personal income Likely to earn more than R2 501 personally (79.6%).

*95.0% Lower Confidence Level (LCL) and 95.0% Upper Confidence Level (UCL) indicates
a range of values that’s likely to encompass the true value
226
The table above describes the average personal motor vehicle driver, and this
description was developed from the respondents' profiles in this study. This description
explains the different characteristics of the average personal motor vehicle driver.
Short-term insurance companies can make use of this information in their marketing
efforts to target the correct audience.
The next section presents the findings of all the general personal motor vehicle
insurance questions, which were included in determining the personal motor vehicle
insurance status.
GENERAL PERSONAL MOTOR VEHICLE INSURANCE

QUESTIONS
The first two questions were used as filter questions to include respondents who drive
motor vehicles and are personally responsible for their decision to purchase personal
motor vehicle insurance. These questions were included to ensure that the correct
individuals who decide to purchase or not purchase personal motor vehicle insurance
are included in the study. The first question requested the respondents to indicate
whether they drive a motor vehicle. Only one out of 706 respondents indicated that
they do not drive a personal motor vehicle (refer to Table C10 in Appendix C). Thus,
this respondent was excluded from the study.
In the second question, the respondents were asked to specify if they are personally
responsible for the decision to purchase personal motor vehicle insurance. Only 27
respondents indicated ‘no’, and these respondents were also excluded from the study
(see Table C11 in Appendix C). A total of 678 respondents were thus included in the
study.
Some general personal motor vehicle insurance questions were included to obtain
some background information related to the personal vehicle insurance status, the
current type of personal vehicle insurance cover, the responsible party that pays for
the personal vehicle insurance, and reasons for having/not having personal motor
vehicle insurance.
227
The general personal motor vehicle insurance results of the questions asked are
illustrated in the frequency tables (see Appendix C). In the discussions below, all the
most critical findings from the data analysis are provided.
Currently have/do not have personal motor vehicle insurance
This question aimed to determine how many respondents currently have a personal
motor vehicle and insurance and how many do not. The respondents were asked to
answer this nominal dichotomous question by indicating yes or no. The answer to this
question determined which set of questions the respondents would be directed to next.
In Figure 6.10, the descriptive statistics of the ‘decision to purchase personal motor
vehicle insurance’ variable are provided, and the frequency count is available in Table
C12 (Appendix C).
Currently have/do not have personal motor vehicle

insurance
No
44.1% Yes
55.9%
Yes No
Figure 6.10: Currently have/do not have personal motor vehicle insurance (n=678)
As shown above, more than half of the respondents, 55.9% (379 respondents),
indicated that they currently have personal motor vehicle insurance, and 44.1% (299
respondents) specified that they currently do not have personal motor vehicle
insurance.
According to this finding, 55.9% of the respondents’ motor vehicles are insured, and
44.1% are not insured. As stated in Chapter 2 (Section 2.6), in South Africa, there
are over 11 million registered vehicles (excluding caravans and trailers), and the
228
majority (60-65%) of these vehicles are not insured (Automobile Association of
South Africa, 2020). However, it is essential to note that these statistics include
minibuses, buses, bus trains, midibuses, motorcycles, quadrucycles, tricycles,
LDVs, panel vans, other light load vehicles and trucks. In contrast, this study only
includes respondents who drive personal motor vehicles. Taking this study’s results
into consideration, the assumption can be made that only 44% are willing to take the
risk of not having personal motor vehicle insurance (as discussed in the prospect
theory, Section 3.5.2.2).
Type of personal motor vehicle insurance
The respondents were asked to reveal which type of personal motor vehicle insurance
cover they currently have. This ordinal, multiple-choice question requesting a single
response had the following options: Comprehensive, Third-party fire and theft, Third-
party only, Pay as You Drive, and not certain.
The table below illustrates the descriptive findings for the ‘type of personal motor
vehicle insurance cover’ variable, and the frequency count is shown in Table C13 (see
Appendix C).
Type of personal motor vehicle insurance cover

120,0%
95,5%
100,0%
80,0%
60,0%
40,0%
20,0%
1,3% 1,3% 0,3% 1,6%
0,0%
Comprehensive Third-party fire and Third-party only Pay As You Drive Not certain
theft
Figure 6.11: Type of personal motor vehicle insurance cover (n=379)

Figure 6.11 reveals that almost all the respondents, 95.5% (362 respondents),
indicated that they have Comprehensive cover. The Third-party fire and theft, and
Third-part only category both equally represent 1.3% of the respondents (5
229
respondents), while 1.6% of the respondents (6 respondents) specified that they are
not certain, and only 0.3% of the respondents (1 respondent) revealed that he/she has
the Pay as You Drive cover.
Comprehensive insurance is the most purchased type of personal motor vehicle

insurance under the respondents. As mentioned in Chapter 2 (Section 2.6.3), a total
of 78% of American drivers purchase comprehensive insurance for their motor
vehicles (Institute Insurance Information, 2021), which is considerably less than the
findings of the current study.
Payment party
The respondents were asked to identify the party responsible for paying the personal
motor vehicle insurance. The options were: I pay for it myself, my family member pays
for it, and my employer pays for it as part of my employment package. The ordinal,
multiple-choice question requesting a single response tested the ‘payment party’
variable.
The descriptive findings are presented in Figure 6.12. In Table C14 (Appendix C), the
frequency count of the descriptive findings is provided.
Payment party
100,0% 90,0%
90,0%
80,0%
70,0%
60,0%
50,0%
40,0%
30,0%
20,0% 8,7%
10,0% 1,3%
0,0%
I pay for it myself My family member pays for it My employer pays for it as part of
my employment package
Figure 6.12: Payment party (n=379)

As depicted in the figure above, almost all the respondents, 90% (341 respondents),
indicated that they pay for the insurance themselves. The respondents who specified
230
that a family member pays for their insurance represents 8.7% of the respondents (33
respondents), and only 1.3% of the respondents (5 respondents) revealed that their
employers pay for their insurance.
Most (90%) of the respondents revealed that they pay for the motor vehicle
insurance themselves. As the respondents pay for this financial service themselves,
they indeed have a view on what impacts their decision to purchase motor vehicle
insurance, which can contribute greatly to this study's results.
Reasons for having personal motor vehicle insurance
The respondents were asked to answer this ordinal, multiple-choice question with
various options and a ‘please specify’ option. The following options were provided: Not
willing to take the chance that I cannot cover the costs when in an accident,
Replacement value of my motor vehicle is high, Motor insurance is a necessity, I know
that I can benefit from having motor vehicle insurance, I have motor vehicle insurance
because my father/mother has/had it, Due to the high crime rate in the country, and
other.
The descriptive findings for the ‘reasons for having personal motor vehicle insurance’
variable are illustrated in Figure 6.13. The frequency tables (Table C15.1-C15.6) can
be viewed in Appendix C. The respondents could choose more than one option for this
question.
As depicted in the figure below, most of the respondents, 82.1% (311 respondents),
specified that they are not willing to take the chance that they cannot cover the costs
when in an accident. More than half of the respondents, 57.3% (217 respondents),
revealed that motor insurance is a necessity, 45.6% of the respondents (173
respondents) indicated that the replacement value of their motor vehicles is high,
39.6% of the respondents (150 respondents) specified the reason as the high crime
rate in the country, and the ‘I know that I can benefit from having motor vehicle
insurance’ category represents 27.4% of the respondents (104 respondents).
The respondents who selected the “other” option make up 6.3% of the respondents
(24 respondents). The main other reasons specified include: it is a requirement from
banks/financial houses that finance the motor vehicle, had losses/damages in the past,
231
and too many drivers on roads that do not have insurance. Only 2.6% of the
respondents (10 respondents) indicated that they have motor vehicle insurance
because their father/mother has/had it.
Reasons for having personal motor vehicle insurance

90,0%
82,1%
80,0%
70,0%
60,0% 57,3%
50,0% 45,6%
39,6%
40,0%
30,0% 27,4%
20,0%
10,0% 6,3%
2,6%
0,0%
Not willing to Replacement Motor I know that I I have motor Due to the Other
take the value of my insurance is a can benefit vehicle high crime
chance that I motor vehicle necessity from having insurance rate in the
cannot cover is high motor vehicle because my country
the costs insurance father/mother
when in an has/had it
accident
Figure 6.13: Reasons for having personal motor vehicle insurance (n=379)
The main reason respondents purchase motor vehicle insurance is thus because
they are not willing to take the chance that they cannot cover the costs when in an
accident. The second reason is that they view motor insurance as a necessity.
Thirdly, the respondents indicated that the motor vehicle's replacement value is too
high not to have motor vehicle insurance cover.
Reasons for not having personal motor vehicle insurance
The respondents were requested to answer this ordinal, multiple-choice question

requesting multiple responses and a ‘please specify’ section, which had the following
232
options: Insurance companies refused coverage for personal motor vehicle, I do not
believe that I can benefit from having insurance, I am dissatisfied with previous
insurance policy or insurance company, I do not need insurance, I cannot afford
insurance, The replacement value of my motor vehicle is low, I do not know what motor
insurance is, and ‘other’.
The descriptive findings for the ‘reasons for not having personal motor vehicle’ variable
are illustrated in Figure 6.14. The frequency tables (Table C16.1-C16.8) are presented
in Appendix C. The respondents could choose more than one option.
Reasons for not having personal motor vehicle insurance

70,0%
58,2%
60,0%
50,0%
40,0%
34,4% 32,8%
30,0%
20,0% 15,1%
13,0%
7,7% 8,7%
10,0%
2,0%
0,0%
Insurance I do not I am I do not need I cannot The I do not Other
companies believe that I dissatisfied insurance afford replacement know what
refused can benefit with insurance value of my motor
coverage for from having previous motor insurance is
personal insurance insurance vehicle is low
motor policy or
vehicle insurance
company
Figure 6.14: Reasons for not having personal motor vehicle insurance (n=299)
As illustrated in the figure above, more than half of the respondents, 58.2% (174
respondents), indicated that they cannot afford insurance, 34.4% of the respondents
(103 respondents) specified that they do not believe that they can benefit from having
insurance, 32.8% of the respondents (98 respondents) revealed that the replacement
value of the motor vehicle is low, 15.1% of the respondents (45 respondents)
answered I do not need insurance, 13% of the respondents (39 respondents) selected
the ‘I am dissatisfied with a previous insurance policy or insurance company’ option
233
and 8.7% of the respondents (26 respondents) indicated that they do not know what
motor insurance is. The respondents who selected the ‘Insurance companies refused
coverage for personal motor vehicle’ option represent 7.7% of the respondents (23
respondents), and only 2% of the respondents (6 respondents) selected the ‘other’
option. The other reasons specified are insurance companies are looking for reasons
not to pay out, saving money and will not insure me.
The findings reveal that most respondents (58.2%) indicated affordability as the
main reason why they do not have personal motor vehicle insurance. As discussed
in Chapter 1 (Section 1.2), various authors mentioned that affordability is one of the
main reasons why motorists in South Africa do not purchase motor vehicle insurance
(Auto & General Insurance 2015; Mokhothu, 2020), but no statistics are available to
provide insight on how important each reason is to drivers in South Africa.
The following section presents and discusses the descriptive statistics related to the
external influences (see conceptual model, Figure 4.1), which are seen as input factors
in the purchase decision process of the insureds and non-insureds.
INPUT STAGE: EXTERNAL INFLUENCES
The item ‘External influences’ was measured by looking at three constructs: the
company’s marketing efforts, sociocultural environment and communication sources.
Each of these constructs consists of various elements that influence the need to
purchase, which is the first step of the decision-making process. Questions were
asked on each of these elements relating to the external influences on the purchase
decision process. The scale used for all three construct elements was a 5-point Likert-
type scale where: (1) was no influence, (2) small influence, (3) moderate influence, (4)
large influence and (5) extremely large influence.
The descriptive statistics for each one of the elements in each construct relating to the
input factors: external influences are presented below in terms of Company’s
marketing efforts, Sociocultural environment, and Communication sources.
234
The company’s marketing efforts
Type of personal motor vehicle insurance offered

The respondents were requested to indicate the extent of influence the ‘Type of
personal motor vehicle insurance policies offered at a company’ has on the need to
purchase personal motor vehicle insurance. This element’s descriptive findings are
illustrated in Figure 6.15 and Table 17.1 in Appendix C.
Type of personal motor vehicle insurance offered
Extremely large 10,2%
Large 22,0%
Moderate 23,7%
Small 18,3%
None 25,8%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.15: Type of personal motor vehicle insurance offered (n=678)

Of the respondents, 25.8% (175 respondents) indicated that the type of personal motor
vehicle insurance policy offered does not influence their need to purchase. The
respondents who indicated that it has a moderate influence represent 23.7% of the
responses (161 respondents), 22% of the respondents (149 respondents) revealed
that it has a large influence, 18.3% of the respondents (124 respondents) specified
that it has a small influence and only 10.2% of the respondents (69 respondents)
indicated that it has an extremely large influence on need to purchase.
The results indicate that the majority of the respondents (74.2%) signalled that the
type of personal motor vehicle insurance policy offered does influence their need to
purchase personal motor vehicle insurance to some extent. Not only does this
finding correspond with the literature (see Chapter 4, Section 4.3.1), which explains
that the type of product influences the need to purchase, it also has implications for
235
the type of service offering insurance companies offer to the consumers. The
additional add-on features to a policy, pay out bonuses and rewards for good driving
behaviour can be included in the type of financial service offered to personal motor
vehicle insurance consumers. Since this is an aspect that has an influence on
consumers’ decisions, it has implications for insurance companies.
Advertisements or promotional activities performed

The respondents were asked to indicate the extent of influence the ‘Advertisements
or promotional activities performed by a short-term insurance company’ have on the
need to purchase personal motor vehicle insurance. This element’s descriptive
findings are illustrated in Figure 6.16 and Table 17.2 in Appendix C.
Large 9,7%
Moderate 20,6%
Small 32,3%
None 34,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.16: Advertisements or promotional activities performed (n=678)

As shown in the figure above, 34.1% of the respondents (231 respondents) revealed
that an insurance company's advertisements or promotional activities do not influence
their need to purchase. The respondents who indicated that it has a small influence
make up 32.3% of the respondents (219 respondents), 20.6% of the respondents (140
respondents) specified that it has a moderate influence, 9.7% of the respondents (66
respondents) indicated it has a large influence and only 3.3% of the respondents (22
respondents) revealed that it has an extremely large influence on their need to
purchase.
236
More than half of the respondents (65.9%) indicated that the advertisements or
promotional activities performed by insurance companies influence their need to
purchase personal motor vehicle insurance to some extent. However, more
respondents (66.4%) indicated that advertisements or promotional activities
performed by insurance companies influence their need to purchase personal motor
vehicle insurance to a small or no extent. This is a finding that has implications for
insurance companies' promotion strategies, whether it be to focus on online
advertising or traditional advertising, or both. Furthermore, this finding agrees with
the literature (see Chapter 4, Section 4.3.1), which explains that insurance
companies' advertisements or promotional activities influence the need to purchase.
Price of a policy
The respondents were asked to indicate the extent of influence the ‘Price of policy’
has on the need to purchase personal motor vehicle insurance. This element’s
descriptive findings are illustrated in Figure 6.17 and Table 17.3 in Appendix C.
Price of a policy
Large 35,3%
Moderate 20,1%
Small 11,1%
None 7,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.17: Price of policy (n=678)

As depicted in the figure above, 35.3% of the respondents (240 respondents) indicated
that a policy's price has a large influence on their need to purchase, and 26.4% of the
respondents (179 respondents) specified that it has an extremely large influence. The
respondents who revealed that it has a moderate influence represent 20.1% of the
237
respondents (136 respondents), while 11.1% of the respondents (75 respondents)
indicated it has a small influence, and 7.1% of the respondents (48 respondents)
revealed that it does not influence their need to purchase.
The main finding is thus that more than half of the respondents (61.7%) believe that
the policy price largely influences their need to purchase personal motor vehicle
insurance. This is a meaningful finding which has implications for the pricing
strategies of insurance companies. Considering the financial constraints and
considerations that consumers might have due to the COVID-19 pandemic, it will be
wise for insurance companies to incorporate pricing strategies that will still draw
consumers even during these difficult financial times. Furthermore, this finding
corresponds with the literature (see Chapter 4, Section 4.3.1), explaining that the
policy’s price influences the need to purchase.
How personal motor vehicle insurance can be purchased

The respondents were asked to indicate the extent of influence the ‘Way in which you
can purchase personal motor vehicle insurance’ has on the need to purchase personal
motor vehicle insurance. This element’s descriptive findings are illustrated in Figure
6.18 and Table 17.4 in Appendix C.
How personal motor vehicle insurance can be purchased
Large 25,5%
Moderate 27,7%
Small 19,9%
None 17,6%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.18: How personal motor vehicle insurance can be purchased (n=678)
238
The figure above shows that 27.7% of the respondents (188 respondents) indicated
that the way in which personal motor vehicle insurance can be purchased has a
moderate influence on their need to purchase, 25.5% of the respondents (173
respondents) specified that it has a large influence and only 9.3% (63 respondents)
revealed that it has an extremely large influence. The respondents who indicated it
has a small influence make up 19.3% of the respondents (135 respondents), and
17.6% of the respondents (119 respondents) specified that it does not influence their
need to purchase.
Even though only 34.8% of the respondents indicated that the way personal motor
vehicle insurance can be purchased has a large influence on their need to purchase,
it is noteworthy that the majority of the respondents (82.4%) specified that it
influences them to some extent. This is an interesting finding which has implications
for the distribution strategies of insurance companies. Even with the shift to direct
online purchasing, there is still a demand for the intermediated distribution channel,
such as brokers (as discussed in Section 2.4.2.2). Therefore, short-term insurance
companies that offer personal motor vehicle insurance should look into providing
direct purchase channels and intermediated channels using brokers. Furthermore,
this finding agrees with the literature (see Chapter 4, Section 4.3.1), which explains
how personal motor vehicle insurance can be purchased influences the need to
purchase.
The company’s physical evidence efforts

The respondents were asked to indicate the extent of influence ‘The company’s
physical evidence efforts, for example, how the building looks, employee uniforms and
online physical presence’ have on the need to purchase personal motor vehicle
insurance.
This element’s descriptive findings are illustrated in Figure 6.19 and Table 17.5 in
Appendix C.
239
The company’s physical evidence efforts
Large 14,5%
Moderate 24,3%
Small 23,9%
None 32,0%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.19: The company’s physical evidence efforts (n=678)

Figure 6.19 reveals that 32% of the respondents (217 respondents) indicated that the
company’s physical evidence efforts do not influence their need to purchase, 23.9%
of the respondents (162 respondents) specified that it has a small influence, and
24.3% of the respondents (165 respondents) indicated that it has a moderate
influence. The respondents who indicated that it has a large influence make up 14.5%
of the respondents (98 respondents), and 5.3% (36 respondents) of the respondents
revealed that it has an extremely large influence on their need to purchase.
Even though 32% of the respondents reflected that the company’s physical evidence
efforts do not influence their need to purchase personal motor vehicle insurance,
most respondents (68%) indicated that it influences their need to some extent. This
finding has implications for the branding strategies of insurance companies,
especially the online digital presence, as this might be the first point of access for
most individuals when they search for motor vehicle insurance. Furthermore, this
finding agrees with the literature (see Chapter 4, Section 4.3.1), which explains that
the company’s physical evidence efforts influence the need to purchase.
240
The duration of the process to get a quotation
The respondents were asked to indicate the extent of influence ‘The duration of the
process to get a quotation’ has on the need to purchase personal motor vehicle
insurance. This element’s descriptive findings are illustrated in Figure 6.20 and Table
17.6 in Appendix C.
The duration of the process to get a quotation
Large 34,5%
Moderate 22,6%
Small 14,6%
None 11,5%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.20: The duration of the process to get a quotation (n=678)

that the duration of the process to get a quotation has a large influence on their need
to purchase, and 16.8% of the respondents (114 respondents) revealed that it has an
extremely large influence. The respondents who indicated that it has a moderate
influence, represent 22.6% of the respondents (153 respondents), while 14.6% of the
respondents (99 respondents) specified that it has a small influence, and only 11.5%
of the respondents (78 respondents) indicated that it has no influence.
It is interesting to note that more than half of the respondents (51.3%) revealed that
the duration of the process to get a quotation largely influences their need to
purchase personal motor vehicle insurance. This finding has implications for the
service delivery strategies of insurance companies. Consumers have busy lives and
want to purchase insurance as they need it at that specific time, therefore a degree
241
of automation should be considered by short-term insurance companies.
Furthermore, this finding agrees with the literature (see Chapter 4, Section 4.3.1),
which explains that the duration of the process to get a quotation influences the need
to purchase.
How the employees in the company interact with consumers

The respondents were requested to indicate the extent of influence ‘The way in which
the employees in the company interact with consumers’ has on the need to purchase
personal motor vehicle insurance. This element’s descriptive findings are illustrated in
Figure 6.21 and Table 17.7 in Appendix C.
How the employees in the company interact with consumers
Large 33,9%
Moderate 21,4%
Small 10,3%
None 8,3%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.21: How the employees in the company interact with consumers (n=678)
As shown above, 33.9% of the respondents (230 respondents) specified that how the
company's employees interact with consumers has a large influence on their need to
purchase, and 26.1% (177 respondents) revealed that it has an extremely large
influence. The respondents who indicated it has a moderate influence make up 21.4%
of the respondents (145 respondents), while 10.3% of the respondents (70
respondents) indicated it has a small influence, and only 8.3% of the respondents (56
respondents) specified it does not influence their need to purchase.
242
It is very interesting to note that more than half of the respondents (60%) indicated
that the way the employees in the company interact with consumers has a large
influence on their need to purchase personal motor vehicle insurance. This finding
has implications for the service delivery strategies of insurance companies,
especially how employees threat consumers while they deliver the service.
Insurance companies can benefit from trained and skilled employees who know how
to treat consumers correctly when a policy is sold or when a claim is logged.
Furthermore, this finding agrees with the literature (see Chapter 4, Section 4.3.1),
which explains that the way the employees in the company interact with consumers
influences the need to purchase.
Partnerships or associations with other companies that consumers

trust
The respondents were asked to indicate the extent of influence ‘Partnerships or
associations with other companies that you trust’ have on the need to purchase
personal motor vehicle insurance. This element’s descriptive findings are illustrated in
Figure 6.22 and Table 17.8 in Appendix C.
Partnerships or associations with other companies that

consumers trust
Large 25,4%
Moderate 26,4%
Small 20,8%
None 17,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.22: Partnerships or associations with other companies that you trust
(n=678)
As illustrated in the figure above, 26.4% of the respondents (179 respondents)

indicated that partnerships or associations with other companies they trust have a
243
moderate influence on their need to purchase and 25.4% of the respondents (172
respondents) specified that it has a large influence. Only 10.3% of the respondents
(70 respondents) revealed that it has an extremely large influence. The respondents
who indicated that it has a small influence represent 20.8% of respondents (141
respondents), and 17.1% of the respondents (116 respondents) specified that it has
no influence.
Most respondents (82.9%) thus feel that partnerships or associations with other
companies have an influence (from small to extremely large) on their need to
purchase personal motor vehicle insurance. As stated in the literature (see Chapter
4, Section 4.3.1), more insurance companies focus on partnerships or associations
to drive the growth, create new channels and drive new innovations. It is interesting
to note that consumers and potential consumers also view partnerships and
associations as an aspect which will influence their need to purchase. Therefore,
this finding can motivate insurance companies further in establishing alliances and
partnerships to drive growth, develop new channels and drive innovation.
The influence of sociocultural factors on the need to purchase personal motor

insurance (which is the first step of the decision-making process) is discussed in the
next section.
The sociocultural environment
A positive comment or opinion of a family member

The respondents were requested to indicate the extent of influence ‘A positive
comment or opinion of a family member’ has on the need to purchase personal motor
vehicle insurance. This element’s descriptive findings are illustrated in Figure 6.23 and
Table 18.1 in Appendix C.
244
A positive comment or opinion of a family member
Large 29,6%
Moderate 25,1%
Small 15,9%
None 17,6%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.23: A positive comment or opinion of a family member (n=678)

Figure 6.23 shows that 29.6% of the respondents (201 respondents) specified that a
positive comment or opinion of a family member has a large influence on their need to
purchase, 11.8% of the respondents (80 respondents) revealed that it has an
extremely large influence and 25.1 % of the respondents (170 respondents) indicated
that it has a moderate influence on their need to purchase. Only 15.9% of the
respondents (108 respondents) selected the small influence option, and 17.6% (119
respondents) specified that it does not influence their need to purchase.
Most respondents (82.4%) revealed that a positive comment or opinion of a family

member influences their need to purchase personal motor vehicle insurance to
some extent. This shows the important role of a family member’s positive comment
or opinion in the consumer thought progress which influences their purchase
decision process. The literature (see Chapter 4, Section 4.3.2.1) also provides
evidence that a family member's comment or opinion is a factor from the
sociocultural environment that influences the need to purchase either positively or
negatively.
245
A negative comment or opinion of a family member
The respondents were asked to specify the extent of influence ‘A negative comment
or opinion of a family member’ has on the need to purchase personal motor vehicle
insurance. This element’s descriptive findings are illustrated in Figure 6.24 and Table
18.2 in Appendix C.
A negative comment or opinion of a family member
Large 26,5%
Moderate 22,7%
Small 18,1%
None 14,7%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.24: A negative comment or opinion of a family member (n=678)

As depicted above, 26.5% of the respondents (180 respondents) revealed that a

negative comment or opinion of a family member has a large influence on their need
to purchase, and 17.8% of the respondents (121 respondents) indicated that it has an
extremely large influence. The respondents who specified that it has a moderate
influence make up 22.7% of the respondents (154 respondents), while 18.1% of the
respondents (123 respondents) indicated a small influence, and only 14.7% of the
respondents (100 respondents) specified that it does not influence their need to
purchase.
Interestingly, most respondents (85.3%) revealed that a negative comment or

opinion of a family member influences their need to purchase personal motor vehicle
insurance to some extent. Therefore, a family member's negative comment or
opinion has a slightly higher influence than a family member's positive comment or
opinion (as indicated in Section 6.5.2.1 above) on respondents’ need to purchase
personal motor vehicle insurance. It is interesting to note that a negative comment
246
or opinion from a family member plays a bigger role in the individuals’ thought
process which influences their purchase decision process.
The positive opinion of a reference person or group used as a basis of

comparison
The respondents were asked to indicate the extent of influence ‘The positive opinion
of a reference person or group used as a basis of comparison’ has on the need to
purchase personal motor vehicle insurance. This element’s descriptive findings are
illustrated in Figure 6.25 and Table 18.3 in Appendix C.
The positive opinion of a reference person or group used as a

basis of comparison
Large 26,9%
Moderate 33,6%
Small 18,1%
None 13,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.25: The positive opinion of a reference person or group used as a basis of
comparison (n=678)
As illustrated above, 33.6% of the respondents (228 respondents) specified that the
positive opinion of a reference person or group used as a basis of comparison has a
moderate influence on their need to purchase. Of the respondents, 26.9% (182
respondents) revealed that it has a large influence, while only 8.3% of the respondents
(56 respondents) indicated that it has an extremely large influence. The respondents
who indicated that it has a small influence make up 18.1% of the respondents (123
respondents), and 13.1% of the respondents (89 respondents) specified that it does
not influence their need to purchase.
247
Even though more than half of the respondents (51.7%) specified that a positive
opinion of a reference person or group used as a basis of comparison has a mild
influence on their need to purchase, the majority of respondents (86.9%) revealed
that it influences their need to purchase personal motor vehicle insurance to some
extent. This finding confirms that a positive opinion of a reference group can be used
as a basis of comparison, which will have an impact on their opinion about motor
vehicle insurance. This positive opinion gained from the reference groups influences
their need to purchase. The literature (see Chapter 4, Section 4.3.2.2), also states
that the opinion (positive or negative) of reference groups is a factor from the
sociocultural environment that influences the need to purchase.
The negative opinion of a reference person or group used as a basis

of comparison
The respondents were requested to specify the extent of influence ‘The negative
opinion of a reference person or group used as a basis of comparison’ has on their
need to purchase personal motor vehicle insurance. This element’s descriptive
findings are illustrated in Figure 6.26 and Table 18.4 in Appendix C.
The negative opinion of a reference person or group used as a

basis of comparison
Large 25,8%
Moderate 27,0%
Small 19,0%
None 15,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.26: The negative opinion of a reference person or group used as a basis of
comparison (n=678)
Figure 6.26 shows that 27% of the respondents (183 respondents) indicated that the
negative opinion of a reference person or group used as a basis of comparison has a
248
moderate influence on their need to purchase, while 25.8% of the respondents (175
respondents) revealed that it has a large influence, and 13.1% of respondents (89
respondents) specified that it has an extremely large influence on their need to
purchase. The respondents who indicated that it has a small influence represent 19%
of the respondents (129 respondents), while 15.1% of the respondents (102
The majority of respondents (84.9%) revealed that a negative opinion of a reference

person or group used as a basis of comparison influences their need to purchase
personal motor vehicle insurance to some extent. Therefore, it can be drawn from
this finding that the positive opinion of a reference person or group used as a basis
of comparison (as indicated in Figure 6.25) has a slightly higher influence on the
respondents’ need to purchase. It is noteworthy for insurance companies to realise
that respondents feel that the positive opinion from a reference group plays a slightly
bigger role than the negative opinion from a reference group in their decision to
purchase personal motor vehicle insurance. Positive word-of-mouth is thus an
extremely important aspect to consider when trying to influence consumers to
purchase personal motor-vehicle insurance.
Positive comments from non-commercial sources

The respondents were asked to indicate the extent of influence ‘Positive comments of
a friend, a newspaper article or the views of an expert on a blog or other internet
platforms’ have on the need to purchase personal motor vehicle insurance. This
element’s descriptive findings are illustrated in Figure 6.27 and Table 18.5 in Appendix
C.
249
Positive comments of a friend, a newspaper article or the
views of an expert on a blog or other internet platforms
Large 29,9%
Moderate 29,6%
Small 15,6%
None 15,2%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.27: Positive comments of a friend, a newspaper article or the views of an

expert on a blog or other internet platforms (n=678)
As depicted above, 29.9% of the respondents (203 respondents) indicated that the
positive comments of a friend, a newspaper article or the views of an expert on a blog
or other internet platforms have a large influence on their need to purchase. Only 9.7%
of the respondents (66 respondents) revealed that it as an extremely large influence.
The respondents who specified that it has a moderate influence make up 29.6% of the
indicated that it has a small influence, and 15.2% of the respondents (103
It is interesting to note that most respondents (84.8%) indicated that positive

comments of a friend, a newspaper article or the views of an expert on a blog or
other internet platforms influence their need to purchase personal motor vehicle
insurance to some extent. This just shows the influence positive word-of-mouth from
non-commercial sources can have on the individuals’ decision to purchase personal
motor vehicle insurance. Insurance companies can therefore create positive
conversations about their brands and different offerings. Furthermore, the literature
(see Chapter 4, Section 4.3.2.3), states that positive or negative comments
(personal words and recommendations) from trusted friends and other customers
250
are viewed to be credible during the decision to purchase personal motor vehicle
insurance.
Negative comments from non-commercial sources

The respondents were asked to specify the extent of influence ‘Negative comments of
a friend, a newspaper article or the views of an expert on a blog or other internet
platforms’ have on the need to purchase personal motor vehicle insurance. This
element’s descriptive findings are illustrated in Figure 6.28 and Table 18.6 in Appendix
C.
Negative comments of a friend, a newspaper article or the

views of an expert on a blog or other internet platforms
Large 25,8%
Moderate 26,1%
Small 17,1%
None 15,9%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.28: Negative comments of a friend, a newspaper article or the views of an

expert on a blog or other internet platforms (n=678)
The figure above shows that 26.1% of the respondents (177 respondents) revealed
that the negative comments of a friend, a newspaper article or the views of an expert
on a blog or other internet platforms have a moderate influence on their need to
purchase, while 25.8% of the respondents (175 respondents) specified that it has a
large influence, and only 15.1% of the respondents (102 respondents) indicated that
it has an extremely large influence. The respondents who felt that it had a small
influence represent 17.1% of the respondents (116 respondents), and 15.9% of the
respondents (108 respondents) specified that it does not influence their purchasing
need.
251
The majority of respondents (84.1%) specified that the negative comments of a
friend, a newspaper article or the views of an expert on a blog or other internet
platforms influence their need to purchase personal motor vehicle insurance to some
extent. Thus, indicating that the negative comments of a friend, a newspaper article
or the views of an expert on a blog or other internet have a bigger influence on the
participants need to purchase than the positive comments. There is a very small
difference between the negative and positive influence of comments of a friend, a
newspaper article or the views of an expert on a blog or other internet platforms, but
the positive influence remains important. This finding gives insurance companies
more reason to create positive conversations about their brands and different
offerings.
The positive opinions of individuals in the same social class

The respondents were required to specify the extent of influence ‘The positive opinions
of individuals in the same social class as you’ have on the need to purchase personal
motor vehicle insurance. This element’s descriptive findings are depicted in Figure
The positive opinions of individuals in the same social class
Large 32,0%
Moderate 25,2%
Small 17,0%
None 16,4%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.29: The positive opinions of individuals in the same social class (n=678)
As illustrated above, 32% of the respondents (217 respondents) specified that the
positive opinions of individuals in the same social class have a large influence on their
252
need to purchase, while only 9.4% of the respondents (64 respondents) indicated that
it has an extremely large influence. The respondents who indicated that it has a
moderate influence make up 25.2% of the total (171 respondents), while 17% of the
respondents (115 respondents) revealed that it has a small influence, and 16.4% of
the respondents (111) specified that it does not influence their need to purchase.
The majority of respondents (83.6%) revealed that that the positive opinions of
individuals in the same social class influence their purchasing need to some extent.
This finding illustrates the importance of positive opinions from individuals with the
same values and interests as the personal motor vehicle consumers and potential
consumers. From the literature (see Chapter 4, Section 4.3.2.4), it is also clear that
the opinion (whether it be positive or negative) of members in the consumer’s social
class is a factor from the sociocultural environment that influences the need to
purchase.
The negative opinions of individuals in the same social class

The respondents were asked to indicate the extent of influence ‘The negative opinions
of individuals in the same social class as you’ have on the need to purchase personal
The negative opinions of individuals in the same social class
Large 26,7%
Moderate 22,1%
Small 20,2%
None 17,6%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.30: The negative opinions of individuals in the same social class (n=678)
253
As shown in the figure above, 26.7% of the respondents (181 respondents) indicated
that the negative opinions of individuals in the same social class have a large influence
on their need to purchase, while only 13.4% of the respondents (91 respondents)
revealed that it has an extremely large influence. Of the respondents, 22.1% (150
respondents) specified that it has a moderate influence, 20.2% of the respondents
(137 respondents) indicated that it has a small influence, and 17.6% of the
respondents (119 respondents) revealed that it does not influence their need to
purchase.
The majority of respondents (82.4%) specified that the negative opinions of

individuals in the same social class influence their need to purchase personal motor
vehicle insurance to some extent. Therefore, it is evident that the positive opinions
of individuals (in Figure 6.29) in the same social class have a slightly higher influence
on the respondents’ need to purchase personal motor vehicle insurance. This finding
is significant for insurance companies as this will give them a good reason to create
positive conversations about their brands and different offerings, so that the positive
opinions can be shared amongst individuals in the same social class.
Cultural and subcultural factors

The respondents were requested to indicate the extent of influence ‘Cultural and
subcultural factors, such as language preference, beliefs and values’ have on the need
to purchase personal motor vehicle insurance. This element’s descriptive findings are
presented in Figure 6.31 and Table 18.9 in Appendix C.
The figure below shows that 24.4% of the respondents (165 respondents) indicated
that the cultural and subcultural factors do not influence their need to purchase. The
respondents who specified that it has a small influence make up 20.8% of the
respondents (141 respondents), 22.7% of the respondents (154 respondents)
revealed that it has a moderate influence, while 21.5% of the respondents (146
respondents) indicated that it has a large influence, and only 10.6% of the respondents
(72 respondents) specified that it has an extremely large influence on their need to
purchase.
254
Cultural and subcultural factors, such as language
preference, beliefs and values
Large 21,5%
Moderate 22,7%
Small 20,8%
None 24,4%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.31: Cultural and subcultural factors (n=678)

The majority of respondents (75.6%) specified that cultural and subcultural factors
influence their purchasing need to some extent. As stated in the literature (see
Chapter 4, Section 4.3.2.5), a consumer’s culture is the basic contributing factor of
what he/she wants and how he/she behaves, and subcultures include population
groups, native language, religion and racial groups. It is interesting that this finding
also shows that the cultural and subcultural factors do in fact influence the
respondents’ need to purchase. The cultural and subcultural factors are not
something that can be controlled by insurance companies, but it is important to spot
the cultural shifts in order to offer policies that are wanted by consumers and
potential consumers.
The influence of communication sources (which are the mechanisms that deliver the
marketing mix and sociocultural influences to the consumers) on the need to purchase
personal motor insurance (which is the first step of the decision-making process) is
discussed in the following paragraphs.
Traditional advertisements about motor vehicle insurance

The respondents were asked to specify the extent of influence ‘Traditional
advertisements about motor vehicle insurance (Such as TV and radio advertisements)’
255
have on their need to purchase personal motor vehicle insurance. This element’s
descriptive findings are presented in Figure 6.32 and Table 19.1 in Appendix C.
Traditional advertisements about motor vehicle insurance
Large 8,9%
Moderate 22,4%
Small 33,5%
None 29,6%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.32: Traditional advertisements about motor vehicle insurance (n=678)

As illustrated above, 33.5% of the respondents (227 respondents) revealed that

traditional advertisements about motor vehicle insurance have a small influence on
their need to purchase, and 29.6% of the respondents (201 respondents) indicated
that it does not influence their need to purchase. The respondents who specified that
it has a moderate influence on their decision make up 22.4% of the respondents (152
respondents), while 8.9% of the respondents (60 respondents) indicated that it has a
large influence. Only 5.6% of the respondents (38 respondents) specified that it has
an extremely large influence on their purchasing need.
More than half of the respondents (55.9%) specified that traditional advertisements
about motor vehicle insurance have a mild influence on their need to purchase
personal motor vehicle insurance. Considering that 29.6% of respondents indicated
that it does not influence their decision to purchase at all, it appears that traditional
advertisements do not influence the respondents’ purchasing need to a large extent.
The literature reviewed in Chapter 4, Section 4.3.3.1, states that more companies
are shifting towards digital marketing and online advertising is the focus. This finding
reveals that respondents do not view traditional advertising as an important
256
influencing mechanism when it comes to the decision to purchase personal motor
vehicle insurance.
Online advertisements about motor vehicle insurance

The respondents were required to specify the extent of influence ‘Online
advertisements about motor vehicle insurance (Such as Video advertisements and
banner ads on websites)’ have on the need to purchase personal motor vehicle
insurance. This element’s descriptive findings are depicted in Figure 6.33 and Table
19.2 in Appendix C.
Online advertisements about motor vehicle insurance
Large 9,9%
Moderate 24,6%
Small 31,1%
None 29,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.33: Online advertisements about motor vehicle insurance (n=678)

Figure 6.33 reveals that 31.1% of the respondents (211 respondents) specified that
online advertisements about motor vehicle insurance have a small influence on their
need to purchase, and 29.1% of respondents (197 respondents) indicated that it does
not influence their need to purchase. The respondents who specified that it has a
moderate influence on their need represent 24.6% of the respondents (167
respondents), while 9.9% of the respondents (67 respondents) indicated that it has a
large influence. Only 5.3% of the respondents (36 respondents) specified that it has
an extremely large influence on their purchasing needs.
257
More than half of the respondents (55.7%) specified that online advertisements
about motor vehicle insurance have a mild influence on their need to purchase
personal motor vehicle insurance. Interestingly, 29.1% of the respondents indicated
that it does not influence their need to purchase at all. Still, it appears that online
advertisements do not influence the respondents’ purchasing needs to a large
extent. This is an interesting finding, since the literature reviewed in Chapter 4,
Section 4.3.3.1, states that more companies are shifting towards digital marketing
and online advertising is the focus. Insurance companies will have to investigate
whether the shift to digital marketing is going to be beneficial in terms of advertising
motor vehicle insurance.
Buzz agent
The respondents were requested to indicate the extent of influence ‘Any person that
is a source of a product referral (buzz agent, see Chapter 4 in Section 4.1.2) and tells
you about motor vehicle insurance’ has on the need to purchase personal motor
Buzz agent
Large 8,7%
Moderate 17,7%
Small 29,6%
None 37,2%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.34: Buzz agent (n=678)

As shown above, 37.2% of the respondents (252 respondents) indicated that buzz
agents do not influence their purchasing needs. The respondents who specified that it
258
has a small influence represent 29.6% of the respondents (201 respondents), 17.7%
of the respondents (120 respondents) revealed that it has a moderate influence, while
8.7% of the respondents (59 respondents) indicated that it has a large influence, and
only 6.8% of the respondents (46 respondents) specified that it has an extremely large
influence on their need to purchase.
It is interesting to note that most respondents (66.8%) specified that buzz agents
have a small or no influence on their need to purchase personal motor vehicle
insurance. However, 62.8% of respondents indicated that buzz agents influence
their need to purchase personal motor vehicle insurance to some extent. Therefore,
there are more respondents who feel that buzz agents do not influence their need
to purchase to a large extent. Although it appears that buzz agents are not important,
short-term insurance companies may not be using buzz agents enough. They can
thus increase the use of buzz agents to try and attract more consumers. This is a
finding which has implications for the marketing communication strategies of
insurance companies, as marketing efforts can serve as an effective tool to generate
buzz among individuals in the market (see Chapter 4, Section 4.3.3.2). Insurance
companies can benefit from buzz among consumers and potential consumers about
motor vehicle insurance as this can drive traffic toward their companies and service
offerings.
Customised messages that insurance companies communicate

The respondents were asked to indicate the extent of influence ‘Customised
messages that insurance companies communicate to you about motor vehicle
insurance’ have on the need to purchase personal motor vehicle insurance. This
element’s descriptive findings are presented in Figure 6.35 and Table 19.4 in Appendix
C.
259
Customised messages that insurance companies
communicate
Large 7,8%
Moderate 26,6%
Small 27,4%
None 34,2%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.35: Customised messages that insurance companies communicate (n=678)

As illustrated above, 34.2% of the respondents (232 respondents) specified that

customised messages that insurance companies communicate do not influence their
need to purchase. The respondents who indicated that it has a small influence
represent 27.4% of the respondents (186 respondents), 26.6% of the respondents
(180 respondents) revealed that it has a moderate influence, while 7.8% of the
respondents (53 respondents) indicated that it has a large influence, and only 4% of
the respondents (27 respondents) specified that it has an extremely large influence on
their need to purchase.
It is interesting to note that most respondents (65.8%) indicated that the customised
messages that insurance companies communicate influence their need to purchase
personal motor vehicle insurance to some extent. This finding reveals that
customised messages that insurance companies communicate influence the
respondents’ purchasing need. Taking into consideration that customised messages
communicated to consumers and potential consumers fall under the direct
marketing efforts of insurance companies, this finding has implications for insurance
companies’ marketing communication strategies. As stated in the literature (Chapter
4, Section 4.3.3.3), there is a distinct move toward personalised marketing within
the motor vehicle insurance market as this allows insurers to build a better
260
relationship with consumers. Seeing that the customised messages do influence
respondents’ need to purchase in this study, it is advisable that the customised
messages short-term insurance companies communicate should be personalised to
build better relationships with consumers.
Paid-for social media efforts

The respondents were asked to specify the extent of influence ‘Paid-for social media
efforts (such as Facebook advertisement campaigns) by insurance companies about
motor vehicle insurance’ have on the need to purchase personal motor vehicle
insurance. This element’s descriptive findings are presented in Figure 6.36 and Table
19.5 in Appendix C.
Paid-for social media efforts
Large 5,0%
Moderate 19,3%
Small 31,6%
None 38,8%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0% 45,0%
Figure 6.36: Paid-for social media efforts (n=678)

Figure 6.36 shows that 38.8% of the respondents (263 respondents) specified that
paid-for social media efforts do not influence their purchasing need. The respondents
who indicated that it has a small influence represent 31.6% of the respondents (214
respondents), while 19.3% of the respondents (131 respondents) revealed that it has
a moderate influence, only 5% of the respondents (34 respondents) indicated that it
has a large influence, and 5.3% of the respondents (36 respondents) specified that it
has an extremely large influence on their need to purchase.
261
It is essential to highlight that most respondents (70.4%) indicated that paid-for
social media efforts by insurance companies have a small or no influence on their
need to purchase personal motor vehicle insurance. However, 61.2% of the
respondents revealed that it influences their need to purchase personal motor
vehicle insurance to some extent. It appears that paid-for social media efforts by
insurance companies do not influence the respondents’ purchasing needs to a large
extent, which is contradictory to what has been reviewed in the literature (see
Chapter 4, Section 4.3.3.4). This finding has implications for the marketing
communications strategies of insurance companies. Insurance companies can
investigate whether social media influences will be beneficial to persuade
consumers and potential consumers to purchase personal motor vehicle insurance.
User-generated social media posts

The respondents were requested to stipulate the extent of influence ‘Word-of-mouth,
advice, recommendations on social media posts about motor vehicle insurance’ have
on the need to purchase personal motor vehicle insurance. This element’s descriptive
findings are depicted in Figure 6.37 and Table 19.6 in Appendix C.
Word-of-mouth, advice, recommendations on social media

posts
Large 19,0%
Moderate 27,0%
Small 26,7%
None 17,3%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.37: Word-of-mouth, advice, recommendations on social media posts about

motor vehicle insurance (n=678)
Figure 6.37 shows that 27% of the respondents (183 respondents) indicated that word-
of-mouth, advice, and recommendations on social media posts about motor vehicle
262
insurance have a moderate influence on their need to purchase. Of the respondents,
26.7% (181 respondents) specified that it has a small influence, and 17.3% of the
respondents (117 respondents) revealed that it does not influence their purchasing
needs. The respondents who identified that it has a large influence make up 19% of
the respondents (129 respondents), while only 10% of the respondents (68
respondents) stipulated that it has an extremely large influence on their purchasing
needs.
Interestingly, just over half of the respondents (53.7%) stipulated that word-of-
mouth, advice, recommendations on social media posts about motor vehicle
insurance have a mild influence on their need to purchase personal motor vehicle
insurance. One would have expected that user-generated social media posts would
have a much larger influence on the need to purchase motor vehicle insurance,
since the literature (see Chapter 4, Section 4.3.3.5) suggests that user-generated
content on social media is changing the way consumers purchase products and
services.
The section below presents and discusses the descriptive findings related to the
internal influences.
PROCESS STAGE: INTERNAL INFLUENCES
The item ‘Internal influences’ was measured by looking at the psychological attributes’
elements. Questions were included on each of these elements (motivation, perception,
learning, personality and attitude) relating to the influence of psychological attributes
on the need recognition (which is the first step of the decision-making process). The
scale used for the ‘psychological attributes’ construct elements was a 5-point Likert-
type scale where (1) was no influence, (2) small influence, (3) moderate influence, (4)
large influence and (5) extremely large influence. The descriptive statistics for each of
the elements relating to psychological attributes are presented below.
Psychological attributes: Motivation to purchase personal motor vehicle

insurance
The respondents were requested to specify the level of agreement ‘Your motivation to
purchase personal motor vehicle insurance’ has on the need recognition for personal
263
Motivation to purchase personal motor vehicle insurance
Large 33,9%
Moderate 22,4%
Small 13,0%
None 13,7%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.38: Motivation to purchase personal motor vehicle insurance (n=678)

As revealed in the figure above, 33% of the respondents (230 respondents) indicated
that the motivation to purchase personal motor vehicle insurance has a large influence
on their need recognition for personal motor vehicle insurance, and 17% of the
respondents (115 respondents) specified it has an extremely large influence. The
respondents who selected the moderate influence option make up 22.4% of the
respondents (152 respondents), while 13% of the respondents (88 respondents)
specified it has a small influence, and 13.7% of the respondents (93 respondents)
revealed that it has no influence.
Interestingly, more than half of the respondents (50.9%) specified that the motivation
to purchase personal motor vehicle insurance largely influences their need
recognition for personal motor vehicle insurance. As explained in the literature (see
Chapter 4 in Section 4.4.1.1), the motivation of consumers to purchase insurance is
determined by two main factors, namely risk expectation and risk sensitivity. It is
interesting to note that more than half of the respondents in this study feel that their
risk expectation and risk sensitivity motivate their need to purchase motor vehicle
264
insurance. Insurance companies can develop strategies to convince more
individuals to purchase motor vehicle insurance.
Psychological attributes: Perception about motor vehicle insurance
The respondents were required to indicate the level of agreement ‘Your perception
about motor vehicle insurance’ has on the need recognition for personal motor vehicle
insurance. This element’s descriptive findings are depicted in Figure 6.39 and Table
20.2 in Appendix C.
Perception about motor vehicle insurance
Large 37,3%
Moderate 25,5%
Small 11,1%
None 10,9%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.39: Perception about motor vehicle insurance (n=678)

The figure above shows that 37.3% (253 respondents) of the respondents specified
that their perception about purchasing personal motor vehicle insurance has a large
influence on their need recognition for personal motor vehicle insurance, while 15.2%
of the respondents (103 respondents) indicated that it has an extremely large
influence. The respondents who revealed that it has a moderate influence option make
up 25.5% of the respondents (173 respondents), 11.1% of the respondents (75
respondents) specified it has a small influence and 10.9% of the respondents (74
respondents) indicated that it has no influence.
265
It is essential to highlight that more than half of the respondents (52.5%) specified
that the perception about personal motor vehicle insurance has an important
influence on their need recognition for personal motor vehicle insurance. This finding
shows that how individuals act during the decision-making process is affected by
their perception of motor vehicle insurance. The respondents’ perception of motor
vehicle insurance also plays a role in consumer satisfaction. Consumer satisfaction
relates to the consumer’s perception of the personal motor vehicle policy’s (financial
service) performance compared to the consumer’s expectations (see literature in
Chapter 3, Section 3.2.1.2).
Psychological attributes: Perception about a certain insurance company
The respondents were asked to indicate the level of agreement ‘Your perception about
a certain insurance company’ has on the need recognition for personal motor vehicle
insurance. This element’s descriptive findings are presented in Figure 6.40 and Table
20.3 in Appendix C.
Perception about a certain insurance company
Large 33,0%
Moderate 24,9%
Small 13,7%
None 10,0%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.40: Perception about a certain insurance company (n=678)

As illustrated in the figure above, 33% of the respondents (224 respondents) indicated
that their perception about a certain insurance company has a large influence on need
recognition for personal motor vehicle insurance, and 18.4% of the respondents (125
266
respondents) specified that it has an extremely large influence. The respondents who
revealed that it has a moderate influence make up 24.9% of the respondents (169
respondents), 13.7% of the respondents (92 respondents) specified it has a small
influence and 10% of the respondents (68 respondents) indicated that it has no
influence.
More than half of the respondents (51.4%) specified that the perception about a
certain insurance company has an important influence on their need recognition for
personal motor vehicle insurance. As indicated above, consumer satisfaction during
a previous purchase affects the consumers’ perceptions. Therefore, insurance
companies must ensure that consumers are satisfied with the purchase at the
specific insurance company, as this plays a role in forming their perception about
the specific company and this perception influence their need to purchase personal
Psychological attributes: Learning
The respondents were requested to indicate the level of agreement ‘The process of
how you purchase and consume information, as well as the experience gained’ has
on the need recognition for personal motor vehicle insurance. This element’s
descriptive findings are provided in Figure 6.41 and Table 20.4 in Appendix C.
Learning
Large 33,0%
Moderate 28,8%
Small 13,6%
None 8,8%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.41: Learning (n=678)

267
The figure above shows that 33% of the respondents (224 respondents) specified that
their process of how they purchase and consume information and the experience
gained have a large influence on need recognition for personal motor vehicle
insurance, while 15.8% of the respondents (107 respondents) revealed that it has an
extremely large influence. The respondents who indicated it has a moderate influence
make up 28.8% of the respondents (195 respondents), 13.6% of the respondents (92
respondents) specified it has a small influence, and 8.8% of the respondents (60
respondents) indicated that it has no influence.
The majority of respondents (92.2%) indicated that how they purchase and consume
information and the experience gained have an influence (to some extent) on their
need recognition for personal motor vehicle insurance. Therefore, the
learning/experiences gained from previous motor vehicle insurance purchases does
influence consumers’ future purchase decisions. This finding corresponds with the
literature (Chapter 4 in Section 4.4.1.3) that states the process of how consumers
purchase and consume information, as well as the experience gained (learning) is
an individual factor that has an impact on the consumer’s need recognition, which
is the first step of the decision-making process.
Psychological attributes: Personality
The respondents were asked to specify the level of agreement ‘Your personality
(specific qualities, attributes, traits)’ has on the need recognition for personal motor
vehicle insurance. This element’s descriptive findings are depicted in Figure 6.42 and
268
Personality
Large 29,5%
Moderate 28,3%
Small 16,1%
None 12,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.42: Personality (n=678)

As depicted above, 29.5% of the respondents (200 respondents) indicated that their
personality has a large influence on their need recognition for personal motor vehicle
insurance, while 14% of the respondents (95 respondents) revealed that it has an
extremely large influence. The respondents who indicated it has a moderate influence
make up 28.3% of the respondents (192 respondents), while 16.1% of the respondents
(109 respondents) specified it has a small influence, and 12.1% of the respondents
(82 respondents) indicated that it does not influence their need recognition.
The majority of respondents (87.9%) specified that their personality influences their
need recognition for personal motor vehicle insurance to some extent. As discussed
in Chapter 4 (Section 4.4.1.4), the individual's personality is an individual factor that
impacts the persons’ need recognition, which is the first step of the decision-making
process. Therefore, this finding is consistent with the literature reviewed and
insurance companies can regard consumer and potential consumer personalities as
useful when they analyse the consumer behaviour for certain policies or brand
choices.
269
Psychological attributes: Attitude towards motor vehicle insurance
The respondents were required to indicate the level of agreement ‘Your attitude
towards motor vehicle insurance’ has on the need recognition for personal motor
Attitude towards motor vehicle insurance
Large 36,9%
Moderate 23,9%
Small 13,0%
None 8,1%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.43: Attitude towards motor vehicle insurance (n=678)

As illustrated above, 36.9% of the respondents (250 respondents) indicated that their
attitude towards motor vehicle insurance has a large influence on their need
recognition for personal motor vehicle insurance, and 18.1% of the respondents (123
respondents) revealed that it has an extremely large influence. The respondents who
indicated it has a moderate influence make up 23.9% of the respondents (162
respondents), while 13% of the respondents (88 respondents) specified it has a small
influence, and 8.1% of the respondents (55 respondents) indicated that it does not
influence their need recognition.
It is important to note that more than half of the respondents (55%) indicated that
their attitude towards motor vehicle insurance largely influences their need
recognition for personal motor vehicle insurance. Furthermore, most respondents
(91.9%) specified that their attitude towards motor vehicle insurance has an
influence (to some extent) on their need recognition for personal motor vehicle
insurance. Only 8.1% of the respondents indicated that it does not influence their
270
need recognition and decision at all. Taking into consideration that attitudes are
formed from past purchase experiences (see Chapter 4, Section 4.4.1.5), insurance
companies have to ensure that positive attitudes are formed regarding motor vehicle
insurance as this will influence the individuals’ decision to purchase.
Psychological attributes: Attitude towards a certain short-term

insurance company
The respondents were asked to indicate the level of agreement ‘Your attitude towards
a certain insurance company’ has on the need recognition for personal motor vehicle
insurance. This element’s descriptive findings are provided in Figure 6.44 and Table
20.7 in Appendix C.
Attitude towards a certain short-term insurance company
Large 29,5%
Moderate 22,3%
Small 15,3%
None 12,2%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0%
Figure 6.44: Attitude towards a certain insurance company (n=678)

that their attitude towards a certain insurance company has a large influence on their
need recognition for personal motor vehicle insurance, while 20.7% of the respondents
(140 respondents) revealed that it has an extremely large influence. The respondents
who indicated that it has a moderate influence make up 22.3% of the respondents
(151 respondents), while 15.3% of the respondents (104 respondents) specified it has
a small influence, and 12.2% of the respondents (83 respondents) indicated that it
does not influence their need recognition.
271
More than half of the respondents (50.2%) indicated that their attitude towards a
certain insurance company largely influences their need recognition for personal
motor vehicle insurance. The majority of respondents (87.8%) specified that their
attitude towards a certain insurance company influences their need recognition for
personal motor vehicle insurance to some extent. Only 12.2% of the respondents
indicated that it does not influence their need recognition and decision at all. Looking
at the previous finding (see Section 6.6.6 above), it appears that the attitude about
motor vehicle insurance plays a slightly bigger role in the respondents’ need to
purchase than the attitude towards motor vehicle insurance. However, it is still
important that insurance companies ensure that positive attitudes are formed
regarding their short-term insurance company as this will also influence the
individuals’ decision to purchase motor vehicle insurance.
The next section presents and discusses the descriptive findings related to the
PROCESS STAGE: DECISION-MAKING PROCESS
The item ‘Decision-making process’ was measured by looking at the decision-making

process steps (as illustrated in Figure 4.1). Each of these constructs consists of
various elements. The scale used for the ‘decision-making process’ construct
elements was a 5-point Likert-type scale where (1) no extent, (2) some extent, (3)
moderate extent, (4) large extent and (5) extreme extent. The descriptive statistics for
each of the elements relating to decision-making are presented below.
Decision-making process (Pre-purchase search): Previous experiences
The respondents were requested to specify the extent of use of the following statement
during the decision-making process: ‘In your search for information, you think about
previous experiences’. This element’s descriptive findings are presented in Figure 6.45
and Table 21.1 in Appendix C.
272
Pre-purchase search - Previous experiences
Extreme 21,4%
Large 35,5%
Moderate 19,5%
Some 13,9%
None 9,7%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.45: Pre-purchase search - Previous experiences (n=678)

As shown in the figure above, 35.5% of the respondents (241 respondents) indicated
that they use their previous experiences in the pre-purchase search to a large extent,
while 21.4% of the respondents (145 respondents) use it to an extreme extent. The
respondents who revealed that they use it to a moderate extent represent 19.5% of
the respondents (132 respondents), while 13.9% of respondents (94 respondents) use
it to some extent, and only 9.7% (66 respondents) of the respondents revealed that
they use it to no extent.
Interestingly, more than half of the respondents (56.9%) indicated that they use their
previous experiences in the pre-purchase search to a greater extent, and 90.3% of
respondents specified that they use their previous experiences in the pre-purchase
search to some extent. This finding agrees with the literature (see Chapter 4, Section
4.5) that states that individuals use their previous experiences to perform a pre-
purchase search as part of the decision-making steps. Insurance companies can
therefore deliver customer experiences that inspire loyalty and attract new
consumers who are frustrated and dissatisfied with their current insurers.
Decision-making process (Pre-purchase search): Request quotations

from more than one short-term insurance company
during the decision-making process: ‘In your search for information, you request
273
quotations from more than one short-term insurance company’. This element’s
descriptive findings are depicted in Figure 6.46 and Table 21.2 in Appendix C.
Pre-purchase search - Request quotations from more

than one short-term insurance company
Extreme 19,6%
Large 34,2%
Moderate 22,0%
Some 12,7%
None 11,5%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.46: Pre-purchase search - Request quotations from more than one short-
term insurance company (n=678)
The figure above reveals that 34.2% of the respondents (232 respondents) indicated
that they request quotations from more than one short-term insurance company in the
pre-purchase search to a large extent, and 19.6% of respondents (133 respondents)
use it to an extreme extent. The respondents who revealed that they use it to a
moderate extent represent 22% of the respondents (149 respondents), while 12.7%
of the respondents (86 respondents) use it to some extent, and only 11.5% (78
respondents) of the respondents specified that they use it to no extent.
The results highlight that more than half of the respondents (53.8%) indicated that
they request quotations from more than one short-term insurance company to a
greater extent. Most respondents (88.5%) specified that they requested quotations
from more than one short-term insurance company in the pre-purchase search to
some extent. As the information gathered from quotations is important in the
decision-making process (see Chapter 4, Section 4.5), insurance companies must
ensure that the quotation process is clear, informative, and flawless so that
consumers and potential consumers can obtain all the necessary information before
the alternatives are evaluated.
274
price of the different quotations
during the decision-making process: ‘You compare the price of the different quotations
to measure if there is a policy that will contribute to your need satisfaction and lifestyle’.
This element’s descriptive findings are illustrated in Figure 6.47 and Table 22.1 in
Appendix C.
Evaluation of alternatives - Compare the price of the different

quotations
Extreme 25,2%
Large 35,8%
Moderate 17,9%
Some 10,6%
None 10,5%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.47: Evaluation of alternatives - Compare the price of the different quotations
(n=678)
Figure 6.47 illustrates that 35.8% of the respondents (243 respondents) specified that
they compare the price of the different quotations to measure if there is a policy that
will contribute to your need satisfaction and lifestyle to a large extent, and 25.2% of
the respondents (171 respondents) do it to an extreme extent. The respondents who
revealed that they compare it to a moderate extent represent 17.9% of the
compare it to some extent, and only 10.5% (71 respondents) of the respondents
specified that they do it to no extent.
Interestingly, more than half of the respondents (61%) indicated that they compare
the price of the different quotations to measure if there is a policy that will contribute
to their need satisfaction and lifestyle to a greater extent. Most respondents (89.5%)
275
specified that they compare the price of the different quotations to some extent. It
can be gathered from this finding that individuals consider the price of each policy
as important when choosing between the alternative brands (see Chapter 4, Section
4.5), therefore insurance companies can take steps to ensure that the most
competitive prices are presented to the consumers and potential consumers.

quality/standard of the different quotations
during the decision-making process: ‘You compare the quality of the different
quotations to measure if there is a policy that will contribute to your need satisfaction
and lifestyle’.
This element’s descriptive findings are presented in Figure 6.48 and Table 22.2 in
Appendix C.
Evaluation of alternatives - Compare the quality of the

different quotations
Extreme 26,4%
Large 35,5%
Moderate 18,4%
Some 10,9%
None 8,8%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0%
Figure 6.48: Evaluation of alternatives - Compare the quality of the different

quotations (n=678)
As presented in the figure above, 35.5% of the respondents (241 respondents)

specified that they compare the quality of the different quotations to measure if there
is a policy that will contribute to their need satisfaction and lifestyle to a large extent,
and 26.4% of the respondents (179 respondents) do it to an extreme extent. The
respondents who indicated that they compare it to a moderate extent represent 18.4%
276
of the respondents (125 respondents), while 10.9% the respondents (73 respondents)
compare it to some extent, and only 8.8% (60 respondents) of the respondents
specified that they do it to no extent.
It is interesting to note that more than half of the respondents (61.9%) indicated that
they compare the quality of the different quotations to measure if there is a policy
that will contribute to their need satisfaction and lifestyle to a greater extent, which
is slightly more than the price criterion (61%) above. 91.2% of respondents specified
that they compare the quality of the different quotations to some extent, which is
also slightly more than the price criterion (89.5%) above. This finding demonstrates
the importance of a quality/standard quotation that individuals require when
evaluating the alternatives. Therefore, it is advisable that insurance companies
ensure that they offer quotations of good quality and high standard, as it might
influence the individuals’ decision.

brand list
during the decision-making process: ‘You compare the brand list (list of different
companies) to measure if there is a policy that will contribute to your need satisfaction
and lifestyle’. This element’s descriptive findings are provided in Figure 6.49 on the
next page and Table 22.3 in Appendix C.
As shown below, 27.6% of the respondents (187 respondents) specified that they
compare the brand list to measure if there is a policy that will contribute to their need
satisfaction and lifestyle to a large extent, and 16.1% of the respondents (109
respondents) do it to an extreme extent. The respondents who indicated that they
compare it to a moderate extent represent 24.5% of the respondents (166
respondents), while 17.6% of the respondents (119 respondents) compare it to some
extent, and only 14.3% (97 respondents) of the respondents specified that they do it
to no extent.
277
Evaluation of alternatives - Compare the brand list
Extreme 16,1%
Large 27,6%
Moderate 24,5%
Some 17,5%
None 14,3%
0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%
Figure 6.49: Evaluation of alternatives - Compare brand list (n=678)

Of the respondents, 43.7% specified that they compare the brand list to measure if
there is a policy that will contribute to their need satisfaction and lifestyle to a greater
extent, and 85.7% of respondents indicated that they compare the brand list to some
extent. More than half of the respondents (56.3%) indicated that this criterion is used
to a smaller or no extent, whereas the other two evaluation criteria findings
discussed above (Sections 6.7.3 and 6.7.4) showed: price - 39% and quality –
38.1%. Even though most respondents indicated that they compare the brand list
when evaluating the alternatives, it appears as if this criterion is least used to
evaluate the alternatives. To summarise the last three findings about the criteria
used to evaluate alternatives, the criterion most used is quality/standard, then price
and the criterion least used is brand list.
278
CONCLUSION
This chapter presented and discussed the descriptive data analysis as it was
conducted in this study. In the first section of this chapter, the demographic
composition of respondents was provided, which provided a respondent and
consumer demographic profile. An attempt was made to develop a description of the
average personal motor vehicle insurance consumer and personal motor vehicle driver
in South Arica. The findings of the respondents’ insight on general personal motor
vehicle insurance questions, internal and external influencing factors and decision-
making process were presented and discussed.
Important conclusions that address the research purpose and research objectives
specified in Chapter 1 of this research study were drawn. In the next chapter (Chapter
7), the proposed CPDFPMVI model (as presented in Chapter 4, Figure 4.1) will be
tested.
279
MODEL FIT ANALYSIS
INTRODUCTION
A quantitative research approach was followed to determine consumers (insured) and

potential consumers (non-insured) purchasing decision-making process regarding
personal motor vehicle insurance. This study aimed to develop a unique model that
will assign a probability to each of the possible choices based on the purchase decision
context. The descriptive findings presented in Chapter 6 and the inferential statistical
analysis (reported in this chapter) are provided to address this study's primary and
secondary research objectives, as specified in Table 7.1.
Table 7.1: Primary and secondary research objectives
• To determine the company’s marketing efforts that influence the individual’s

vehicle insurance market in South Africa.
• To assess the demographic variables that influence the individuals’ need to
280
This chapter systematically presents and explains the process followed to develop the
consumer purchasing decision model that illustrates the consumers’ purchasing
decisions in the South African personal motor vehicle insurance market. Figure 7.2
illustrates the conceptual CPDFPMVI model, as presented in Chapter 4. As provided
in this chapter, the inferential statistics determine the model fit of the items listed in
Figure 7.2.
The process underlying the model fit strategy consists of three phases, namely, the
Exploratory Factor Analysis (EFA) phase, Confirmatory Factor Analysis (CFA) phase
and the logistic regression phase. All these were, in essence, part of the process of
informing the logistic regression model. Model fit was just one aspect that was
considered as part of the assessment of the measurements. The model-building
strategy utilised to achieve the final CPDFPMVI model was introduced in Section
5.11.2, and it is illustrated in Figure 7.1 below again for ease of reference purposes.
Figure 7.1: Model-building strategy to achieve CPDFPMVI model

Source: Developed by researcher
281
The steps followed (as illustrated in Figure 7.1) to achieve the final CPDFPMVI model
are discussed in this chapter.
282
Figure 7.2: Conceptual CPDFPMVI model
283
PHASE 1: EXPLORATORY FACTOR ANALYSIS (EFA) WITH
EXPLORATION SAMPLE (N=200)
EFA was introduced and defined in Section 5.11.2. The conceptual model in Figure
7.2 has not been tested before, and therefore, the EFA was considered ideal for
exploring and estimating the factor structure of the current theoretical model. This
statistical method served as being useful to explore the factor structure in previous
consumer purchasing decision-making model-building research (Tendhar, Singh &
Jones, 2017; Wang & Yu, 2017). It is recommended to have a sample of at least 200
and higher when there is a large number of variables and an expected number of
factors (Hair et al., 2019:132-133). Therefore, an EFA with a holdout sample of 200
was performed to explore the factor structure as Phase 1 of this study.
An EFA, using IBM SPSS (Version 27), was performed on the data obtained from the
scales in the data collection instrument to achieve the following targets in terms of this
specific exploration sample: (1) to establish the number of factors underlying the
variation in and the correlation between the items; (2) to state the items that load onto
certain factors; and (3) to identify items for possible exclusion to gain the best
acceptable measurement model (Matsunaga, 2010:98). Furthermore, it is
recommended to perform an EFA if the sample size allows it, especially if the purpose
is to reveal factor structure (Hair et al., 2019:134).
In terms of this study, the overall sample size was 678 and the n=200 exploration
sample seems to be ideal as the conceptual CPDFPMVI model includes a large set of
variables. The EFA method was used as Phase 1 of this study to uncover the
underlying structure of the relatively large set of variables of the conceptual
CPDFPMVI model. The demographic variables in the conceptual model could not be
tested in this phase as factor analysis methods are used to measure constructs using
a scale (items of a scale), and demographic variables are not items of a scales. Thus,
the demographics of individuals will be tested for the first time in the variable selection
step of Phase 3 (see Section 7.4.1).
An EFA was performed using principal component analysis for extraction and varimax
for rotation. Before performing the factor analysis, the researcher first verified that the
data were appropriate for factor analysis using two tests: the Kaiser-Meyer-Olkin
(KMO) test and the Bartlett sphericity test. The KMO measure the sample adequacy if
284
it is 0.7 or above, and the Bartlett sphericity indicates that a significant correlation
exists among variables to proceed if the p-value is below 0.001 (Stehlik-Barry &
Babinec 2017:343; Hair et al., 2019:136). The results of both tests indicated that factor
analysis would be useful given that the KMO was 0.847 and Bartlett sphericity p-value
was 0.00 (see Appendix D, Table D1).
The minimum factor loading criteria was set to 0.40, and taking into consideration that
a sample size of 200 is needed for significance (Hair et al., 2019:152). The
communality of the scale, which indicates the amount of variance in each dimension
(Hair et al., 2019:155), was also assessed to ensure acceptable levels of explanation.
The results show that all the communalities were over 0.40 (see Table D2 in Appendix
D).
Using principal component analysis, the initial factor analysis extracted seven factors
that were evident (Table D3 in Appendix D). All with an eigenvalue greater than one,
indicating good discriminate validity. The seven factors identified explained 59.985%
of the variance among the items in the study. Table 7.2 reveals the factor findings
gained from the EFA and the discussion of the findings follow thereafter.
Table 7.2: Factor findings from EFA
Factor
identified
Constructs Items/variables
from
EFA
1 Psychological Attributes 11.1 - Motivation

(PA) 11.2 - Perception about motor vehicle insurance
11.3 - Perception about a certain insurance
company
11.4 - Learning
11.5 - Personality
11.6 - Attitude towards motor vehicle insurance
11.7 - Attitude towards a certain short-term
insurance company
2 Marketing Efforts (ME) 8.1 - Type of policies offered

8.3 - Price of a policy
8.4 – Distribution channels
8.5 - Physical evidence
8.6 - Process
285
Factor
identified
Constructs Items/variables
from
EFA
8.7 - People
8.8 - Partnerships
3 Pre-purchase search and 12.1- Pre-purchase search considering previous

evaluation of alternatives experiences
steps (Decision-making 12.2 - Pre-purchase search requesting
Process) quotations
(DP) 12.3 - Evaluate alternative prices
12.4 - Evaluate alternative quality/standards
12.5 - Evaluate alternative brands
4 Negative Sociocultural 9.2 - Negative family member

Environment (SE_N) 9.4 - Negative reference person or group
9.6 - Negative non-commercial sources
9.8 - Negative social class
5 Communication Sources B 8.2 - Promotion

(CS_B) 10.3 - Buzz agent
10.4 - Customised messages
10.5 - User-generated social media posts
10.6 - Paid-for social media efforts
6 Positive Sociocultural 9.1 - Positive family member

Environment (SE_P) 9.3 - Positive reference person or group
9.5 - Positive non-commercial sources
9.7 - Positive social class
9.9 - Cultural and subcultural factors
7 Communication Sources A 10.1 - Traditional advertisements

(CS_A) 10.2 - Online advertisements
The seven factors identified as part of the EFA are aligned with the theoretical
proposition in the research. As indicated in Table 7.2 above, Factor 1 includes items
11.1 to 11.7, referring to the psychological attributes. Factor 2 gathers items 8.1, 8.3
to 8.8, which represent the marketing efforts. Factor 3 includes items 12.1 to 12.5,
referring to the pre-purchase search and evaluation alternative steps in the decision-
making process. Factor 4 includes items 9.2, 9.4, 9.6 and 9.8, which represent the
negative sociocultural environment. Factor 5 gathers items 8.2, 10.3 to 10.6, referring
to the communication sources. Factor 6 includes items 9.1, 9.3, 9.5, 9.7 and 9.9, which
286
relates to the positive sociocultural environment. Factor 7 gathers items 10.1 and 10.2,
which represent the communication sources.
These seven factors identified were used in the next phase, called model specification,
using the Confirmatory Factor Analysis (CFA) method. The next section explains
Phase 2.
PHASE 2: CONFIRMATORY FACTOR ANALYSIS (CFA) WITH

VALIDATION SAMPLE (N=478)
EFA is used to understand the theory better. In contrast, CFA is a type of structural
equational modelling (SEM) method used to test how well a pre-specified
measurement theory, composed of measured variables and factors, fits the reality as
captured data (Hair et al., 2019:660). As explained in Section 5.11.2, CFA is also
broadly used to test the factor validity required in this step to ensure the model
specification and identification. Previous studies found it beneficial to use CFA as the
follow-up strategy on EFA (previous stage) to confirm the existence of a specific factor
structure (Nayeem & Casidy, 2015:70). It is important to point out that theory should
always be predominantly considered when making model improvements and
modifications while using the CFA method (Hair et al., 2019:679). Therefore, the
theory was predominantly considered during every step of Phase 2.
CFA requires the use of multiple samples (Hair et al., 2019670). The initial exploration
sample used in the EFA was the first sample set. An additional sample should be
drawn for a CFA. The rule of thumb on minimum sample size for CFA is a minimum of
200, but this can also be too low for complex models (MacCallum & Austin in Kyriazos,
2018:2216). Furthermore, a holdout sample, also known as a validation sample, is
recommended to ensure model validation (Hair et al., 2019:579). Thus, a CFA with a
validation sample of 478 was performed to determine a model fit. In the sections below,
the steps taken during the CFA phase are outlined and discussed.
Model specification
As mentioned above, CFA was used to test the measurement model. To use CFA to
test the measurement theory, the researcher must specify theoretically supported
correspondence rules between indicators and constructs/latent variables (Hair et al.,
2019:670). Phase 1 (EFA) performed above provided insight into the structure of the
287
supported items, and it was helpful in proposing the measurement model. The model
specification in CFA must be firmly grounded in existing theory and previous empirical
research (Hahs-Vaughn, 2017:447).
Thus, the conceptual model (presented in Chapter 4) developed from existing theory,
and the EFA (as explained in the previous section) ensured that the researcher had a
firm substantive and empirical basis to guide the model specification of this CFA. In
other words, the number of latent variables and the patterns on how each item loads
onto certain factors was informed by the results of the EFA and the theoretical
conceptualisation of the constructs as described in the literature review (see Chapters
2 to 4). The latent variables and indicators which explain each latent variable gained
from the EFA are presented in Table 7.3.
Table 7.3: Model specification gained from EFA
Constructs/Latent Indicators (Items from EFA) Factor

variables loadings
gained from
EFA
(See Table D4,
Appendix D)
Psychological Motivation 0.690

Attributes (PA) Perception about motor vehicle insurance 0.758
Perception about a certain insurance company 0.715
Learning 0.626
Personality 0.684
Attitude towards motor vehicle insurance 0.782
Attitude towards a certain short-term insurance 0.653
company
Marketing Efforts Type of policies offered 0.620

(ME) Price of a policy 0.602
Distribution channels 0.644
Physical evidence 0.479
Process 0.711
People 0.679
Partnerships 0.512
Pre-purchase search Pre-purchase search considering previous 0.514

and evaluation of experiences
alternatives steps Pre-purchase search requesting quotations 0.814
(Decision-making
Process) (DP) Evaluate alternative prices 0.811
288
Constructs/Latent Indicators (Items from EFA) Factor
variables loadings
gained from
EFA
(See Table D4,
Appendix D)
Evaluate alternative quality/standards 0.844
Evaluate alternative brands 0.641
Negative Negative family member 0.798

Sociocultural Negative reference person or group 0.859
Environment (SE_N)
Negative non-commercial sources 0.722
Negative social class 0.719
Communication Promotion 0.510

Sources B Buzz agent 0.761
(CS_B) Customised messages 0.690
User-generated social media posts 0.558
Paid-for social media efforts 0.642
Positive Positive family member 0.628

Sociocultural Positive reference person or group 0.604
Environment (SE_P)
Positive non-commercial sources 0.634
Positive social class 0.694
Cultural and subcultural factors 0.446
Communication Traditional advertisements 0.742

Sources A (CS_A) Online advertisements 0.551
Each factor and factor loadings gained from the EFA in Phase 1 was used together
with the theoretical conceptualisation of constructs to generate the latent variables and
the indicators as presented in the table above. Table 7.3 above, also provides the
factor loadings. Each indicator has a loading, and this loading represents the effect
each indicator has on each latent variable. The higher loadings indicate a greater
influence, and the lower loadings indicate less influence.
Within the first latent variable (Psychological Attributes), the indicators scored
relatively high, and it will influence the purchase decision in the following order: (1)
Attitude towards motor vehicle insurance, (2) Perception about motor vehicle
insurance, (3) Perception about a certain insurance company, (4) Motivation, (5)
289
Personality, (6) Attitude towards a certain short-term insurance company and (7)
Learning.
The Marketing Efforts latent variable also scored relatively high, except for the physical
evidence and partnerships indicators, which indicates that these two indicators have
a lower influence on the purchase decision. Pre-purchase search considering previous
experiences and evaluate alternative brands has a lower influence on the purchase
decision than the other indicators in the Decision-making Process latent variable.
All the indicators in the Negative Sociocultural Environment latent variable indicate a
high influence on the purchase decision. The two indicators, promotion and user-
generated social media posts, reveal a lower influence on the purchase decision than
the other indicators in the Communication Sources B latent variable.
In the Positive Sociocultural Environment latent variable, the Cultural and subcultural
factors indicator has a very low loading, indicating a low influence on the purchase
decision. All the other loadings are relatively high in this latent variable and can
therefore be an indication that they have a strong influence on the purchase decision.
In the last latent variable, Communication Sources A, it is evident that online

advertisements have a lower influence on the purchase decision than traditional
advertisements. Once the model is specified, the model identification step is next,
which is discussed below.
Model identification
Once the measurement model is specified, it can be identified. According to Gana and
Broc (2019:100), there are two conditions for identifying a measurement model: a
sufficient number of indicators for the latent variables and identifying the metric for the
latent variable. Regarding the first condition, at least three indicators per latent variable
is a general guideline (Hahs-Vaughn, 2017:449; Hair et al., 2019:671). Even though it
is a general guideline to have three indicators for each latent variable, in some
instances where large sample sizes are in play, it is not uncommon to have fewer than
three indicators per latent variable (Hair et al., 2019:633).
The second condition relates to the metric for the latent variables and seeing that it is
not measured, a latent variable has no metric. Therefore, the researcher must define
the metric of this variable (Gana & Broc 2019:100). Latent variables can be scaled in
290
one of the following two ways: set the variance of the latent variable to 1.0, or set one
parameter (factor loading) relating to the latent variable with an item that can be set to
1.00. This will allow the latent variable to capture the metric of the selected item (Hahs-
Vaughn, 2017:449). Regardless of which option is used to select the scaling, the fit of
the models will be identical, just as long each construct has a value specified (Hair et
al., 2019:672). In this study, option two was implemented as one factor loading of each
latent variable was set as 1. The model estimation and testing are discussed below.
Model estimation and testing
After the model has been identified, the model must be estimated. Model estimation
aims to select a fitting method that is subsequently minimised to obtain the parameter
estimates/indices (Ramlall 2017:57). Multiple estimation methods are available for
SEM models, but the commonly used method specifically for CFA models is Maximum
Likelihood Estimation (MLE). MLE is a computer software generated method that
calculates the parameter estimates that maximise the probability of the observed
matrix sampled from the population (Hahs-Vaughn, 2017:449; Hair et al., 2019:605).
Therefore, the MLE method using Amos was employed when this CFA was performed.
A CFA generates critical information about the instrument's psychometric properties,

which is the reliability and validity of the instrument. Numerous Chi-square tests can
be used for the CFA test, namely CMIN (normed Chi-Squared/df), GFI (Goodness-of-
fit), CFI (Comparative Fit Index) and RMSEA (Root Mean Square Error of
Approximation).
To validate the relationships amongst latent variables and indicators, the model should
present a reasonable fit. There are specific rules that a model has to adhere to, to get
a model fit. Several factors can influence the model fit, namely, sample size, model
complexity, estimation method, the amount and type of misspecification, and the type
and normality of the data (Brown 2015:45). All these factors were covered in the
sections above. Furthermore, there are several indices used to determine model fit. A
short description of each of these indices and their threshold values are presented in
Table 7.4.
291
Table 7.4: Measurements and thresholds to determine model fit
Threshold values proposed and

Measure Index name Description
generally accepted
x2/P-value Model Chi-square Statistical assessment of overall fit and the >0.05
discrepancy between the sample and fitted With large sample sizes, this is not an
covariance matrices. index that is useful as it is rarely more
than 0.05
CMIN/df Normal chi-square, also The minimum discrepancy divided by its degree of Between 1 and 3 (measures less
known as normed chi- freedom. than 5 might still be acceptable)
square
(A)GFI (Adjusted) A measure of fit between the hypothesised model and GFI > 0.90
Goodness of Fit Index. the observed covariance matrix. AGFI > 0.80
Measure the percentage of variances that are (a value close to 1 indicates a perfect
explained by the specified model structure. fit)
The adjusted index (AGFI) corrects the GFI, which is
affected by the number of indicators of each latent
variable.
NFI Normed fit Index NFI is a ratio of difference of the x2 value for the fitted ≥ 0.95
model and the null model divided by the x2 value for ≥ 0.90
the null model.
(a value close to 1 indicates a good
fit)
TLI Tucker Lewis Index TLI is similar to the NFI but varies because it ≥ 0.95
compares the normed chi-square values for the null ≥ 0.90
and specified model. It does take the model
complexity into account to some degree. (a value close to 0.9 indicates a good
fit)
292
Threshold values proposed and
Measure Index name Description
generally accepted
CFI Comparative Fit Index Analyses the model fit by examining the discrepancy >0.90 (1 indicates a very good fit)
between the data and the hypothesised model.
RMSEA Root Mean Square Error of This measure better represents how well the model <0.05 (indicates a close fit).
Approximation fits a population, not just a sample used for A value of 0.0 indicates the exact fit
estimation. This measure explicitly tries to correct for of the model.
both model and ample size by including both in its
calculation. A value of 0.08 or less indicates a
reasonable error of approximation.
Values close to 0 represent a good fit.
Not higher than 1.
PCLOSE p of Close Fit Test that concludes if a model is a close fit. >0,05
Calculated in conjunction with RMSEA, which should
be greater than 0.05.
(S)RMR (Standardised) The square root of the difference between the RMR<0.08 (some sources state that
Root Mean Square residuals of the sample covariance matrix and the it should ideally be <0.05)
Residual hypothesised model. SRMR <0.08 (value of 0.05 to 0.08
A rule of thumb is that an SRMR over 0.1 suggests a indicates a close fit)
problem with the fit.
Sources: Parry (n.d.); Awang (2012:56); Schumacker & Lomax (2016:112); Hair et al. (2019:642)
293
It is essential to highlight that it is not necessary to provide all of the values for reporting
these indices. There is no universal agreement on the number of indices that should
be reported on, but the minimal set would include CMIN/DF, TLI, CFI, RMSEA and
SRMR (Jackson, Gillaspy & Purc-Stephenson, 2009:19; Bandalos & Finney,
2018:115). Therefore, these indices will be reported.
Once the model is correctly specified and estimated, the next step involves the
testing/assessment of the accuracy of the measurement model. As discussed above,
CFA was used to validate and confirm the latent variables, as explored and identified
after the EFA (performed in Phase 1). The first default (baseline) model is presented
in Figure 7.3.
294
Figure 7.3: CFA baseline model for CPDFPMVI
295
As indicated in the Figure 7.3, the predictive model then tested the data fit between
the following purchase decision elements:
• ME (refers to the marketing efforts by insurance companies that influence the need
recognition) consists of:
o Type of policies offered (e7)

o Price of a policy (e6)
o Distribution channels (e5)
o Physical evidence (e4)
o Process (e3)
o People (e2)
o Partnerships (e1)
• SE_P (refers to the positive sociocultural environment factors that influence the
need recognition) consists of:
o Positive family member (e12)

o Positive reference person or group (e11)
o Positive non-commercial sources (e10)
o Positive social class (e9)
o Cultural and subcultural factors (e8)
• SE_N (refers to the negative sociocultural environment factors that influence the
o Negative family member (e16)

o Negative reference person or group (e15)
o Negative non-commercial sources (e14)
o Negative social class (e13)
• CS_A (refers to the communication sources group A that influence the need
o Traditional advertisements (e18)

o Online advertisements (e17)
• CS_B (refers to the communication sources group B that influence the need
296
o Promotion (e23)
o Buzz agent (e22)
o Customised messages (e21)
o User-generated social media posts (e20)
o Paid-for social media efforts (e19)
• PA (refers to the psychological attributes that influence the need recognition)

consists of:
o Motivation (e30)
o Perception about motor vehicle insurance (e29)
o Perception about a certain insurance company (e28)
o Learning (e27)
o Personality (e26)
o Attitude towards motor vehicle insurance (e25)
o Attitude towards a certain short-term insurance company (e24)
• DP (refers to the pre-purchase search and evaluation of alternatives steps in the

decision process) consists of:
o Pre-purchase search considering previous experiences (e35)

o Pre-purchase search requesting quotations (e34)
o Evaluate alternative prices (e33)
o Evaluate alternative quality/standards (e32)
o Evaluate alternative brands (e31)
The model fit was measured between the elements of a purchase decision process
(latent variables) and the fit between the individual scales (indicators) that made up
each purchase decision element. In this way, the model measured what it was set out
to measure.
Table 7.5 presents the model fit statistics of the baseline CPDFPMVI model.
297
Table 7.5: Model fit statistics for CFA baseline CPDFPMVI model
CMIN/DF TLI CFI RMSEA SRMR
Default model 3.737 0.83 0.85 0.076 0.079
Threshold values <5 ≥ 0.90 >0.90 <0.08 <0.08
Poor fit (PF)/ Acceptable fit AF PF PF PF AF

(AF)
The statistics presented in Table 7.5 reveal that the CPDFPMVI model gained from
the EFA has a poor fit against the data. In addition, two of the values (in Table 7.5) fall
outside the prescribed threshold as outlined in Table 7.4. Nevertheless, the above
obtained values indicate that the reliability and validity of the model were evaluated.
See Table 7.6, which presents these statistics.
298
Table 7.6: Reliability and validity of CFA baseline CPDFPMVI model
CR AVE MSV PA ME SE_P SE_N CS_A CS_B DP
Psychological Attributes (PA) 0.911 0.595 0.200 *0.771
Marketing Efforts (ME) 0.791 0.360 0.272 0.447 *0.600
Positive Sociocultural Environment (SE_P) 0.869 0.575 0.466 0.403 0.409 *0.758
Negative Sociocultural Environment (SE_N) 0.913 0.724 0.466 0.288 0.271 0.683 *0.851
Communication Sources A (CS_A) 0.821 0.697 0.587 0.054 0.143 0.321 0.302 *0.835
Communication Sources B (CS_B) 0.811 0.467 0.587 0.005 0.262 0.333 0.331 0.766 *0.684
Decision Process (DP) 0.876 0.595 0.272 0.393 0.522 0.326 0.142 0.063 0.102 *0.771
*square root (SQRT) of average variance extracted (AVE)
299
Table 7.6 provides the reliability and validity statistics generated for the baseline
model. The reliability of a model is reported by considering the CR (Composite
Reliability) value, which indicates the reliability and internal consistency of a latent
variable (Hair et al., 2019:763). The rule of thumb is that the CR value should be higher
than 0.70 to indicate internal consistency, and in exploratory research, 0.60 to 0.70 is
considered acceptable (Hair et al., 2017:122). Thus, the baseline model demonstrates
overall reliability with all the latent variables scoring more than the recommended 0.70.
Regarding the validity in CFA, three types of validity have to be determined. Firstly,
construct validity, which is the main objective of CFA, deals with the measurement's
accuracy. Once a good fit is established, the four additional components of construct
validity should be evaluated (Hair et al., 2019:675). These components include that it
conforms to its conceptual definition, is unidimensional and meets the necessary
levels of reliability (Hair et al., 2019:162). All these components will be evaluated in
Section 7.3.4 when a model fit is achieved.
Secondly, convergent validity assesses the extent to which a measure correlates

positively with others (Civelek 2018:32). Therefore, the AVE (average variance
extracted) value should be higher than 0.5 (Hair et al., 2017:122). The AVE values in
Figure 7.5 above indicate that two latent variables have a value larger than the 0.5
threshold. These latent variables are Marketing Efforts and Communication
Sources_B. On that account, this baseline model does not meet the convergent
validity criterion. However, it is important to note that the AVE values should preferably
be higher than 0.05, but the context of the CR should also be taken into account.
Thirdly, discriminant validity refers to the extent to which a latent variable is distinct
from other latent variables or indicators (Hair et al., 2019:676). The Fornell and Larcker
criterion can be used to assess discriminate validity, and this method compares the
square root of the AVE with the correlation of latent variables. The square root of each
construct’s AVE should have a higher value than the correlations with other latent
variables (Fornell & Larcker 1981; Hair et al., 2019:761).
The baseline model has two discriminant validity concerns. Firstly, the square root of
the AVE for Communication Sources_B is less than one of the correlations with other
latent variables. Plus, the absolute value of the correlations with another factor and
AVE for Communication Sources_B is less than the MSV (Maximum Shared
300
Variance). MSV is the the maximum amount of change that a construct can explain in
another construct (Bhati, 2019:1785). Considering these values, it is evident that the
baseline model does not hold acceptable internal reliability and validity.
To improve the model fit in terms of reliability and validity, some modifications were
made, and a revised model was tested. The revised model is presented and discussed
in the next section.
Model modification
The steps taken to improve the model fit in terms of reliability and validity were to:
1. Covary error terms that were part of the same factor. Only two indicators could be
covaried, namely, Q11.1 (e30) and Q11.2 (e29).
2. Assess and remove indicators with significant standardised residual covariance.

The following indicators were identified with high values regarding the
standardised residual covariance: Q8.5 (e4), Q10.5 (e20), Q8.2 (e23), Q8.4 (e5)
and Q8.8 (e1).
3. Remove the indicators with low loadings. For example, the Q8.1 (e7) and Q12.1
(e35) indicators had factor loadings less than 0.5.
The indicators identified in steps 1 and 2 above were removed after the researcher
assessed whether it would impact the definition of the latent variable. The remaining
indicators in each latent variable were sufficient, and it still explained each latent
variable clearly. After making these modifications to the baseline model, a new revised
model was generated. This model fit is presented in Figure 7.4.
301
Figure 7.4: CFA revised model for CPDFPMVI
302
As shown in Figure 7.4, a few modifications were made to the baseline model to
improve the model fit. The new revised model tested the data fit between the following
purchase decision elements:
• ME (refers to the marketing efforts by insurance companies that influence the need
o Price of a policy (e6)

o Process (e3)
o People (e2)
• SE_P (refers to the positive sociocultural environment factors that influence the
o Positive family member (e12)

o Positive reference person or group (e11)
o Positive non-commercial sources (e10)
o Positive social class (e9)
o Cultural and subcultural factors (e8)
• SE_N (refers to the negative sociocultural environment factors that influence the
o Negative family member (e16)

o Negative reference person or group (e15)
o Negative non-commercial sources (e14)
o Negative social class (e13)
• CS_A (refers to the communication sources group A that influence the need
o Traditional advertisements (e18)

o Online advertisements (e17)
• CS_B (refers to the communication sources group B that influence the need
o Buzz agent (e22)

o Customised messages (e21)
o Paid-for social media efforts (e19)
303
• PA (refers to the psychological attributes that influence the need recognition)
consists of:
o Motivation (e30)
o Perception about motor vehicle insurance (e29)
o Perception about a certain insurance company (e28)
o Learning (e27)
o Personality (e26)
o Attitude towards motor vehicle insurance (e25)
o Attitude towards a certain short-term insurance company (e24)
• DP (refers to the pre-purchase search and evaluation of alternatives steps in the

decision process) consists of:
o Pre-purchase search requesting quotations (e34)

o Evaluate alternative prices (e33)
o Evaluate alternative quality/standards (e32)
o Evaluate alternative brands (e31)
The model fit indices of the revised CPDFPMVI model are given in Table 7.7 below.
Table 7.7: Model fit for CFA revised CPDFPMVI model
CMIN/DF TLI CFI RMSEA SRMR
Default model 3.406 0.89 0.91 0.071 0.046
Threshold values <5 ≥ 0.90 >0.90 <0.08 <0.08
Poor fit (PF)/ Acceptable fit AF AF AF AF AF

(AF)
Taking the indices presented above into consideration, it is evident that the revised
model is an acceptable fit with the data. In comparison to the threshold, most of the
indices reported fall within the accepted norm. The TLI value is 0.89, just under the
0.9 threshold, but a TLI value close to 0.9 can still be deemed acceptable (Schumacker
& Lomax, 2016:112).
Due to the reliability and validity corners raised in the baseline model, it was necessary
to confirm the revised model’s reliability and validity. Table 7.8 on the next page
reveals the reliability and validity results of the revised model. The revised CPDFPMVI
model after the CFA model fit analysis is presented in Figure 7.5 below Table 7.8.
304
Table 7.8: Reliability and validity of the CFA revised CPDFPMVI model
CR AVE MSV PA ME SE_P SE_N CS_A CS_B DP
Psychological Attributes (PA) 0.908 0.586 0.170 *0.765
Marketing Efforts (ME) 0.764 0.520 0.303 0.412 *0.721
Positive Sociocultural Environment (SE_P) 0.869 0.575 0.466 0.406 0.322 *0.758
Negative Sociocultural Environment (SE_N) 0.913 0.723 0.466 0.308 0.173 0.683 *0.850
Communication Sources A (CS_A) 0.821 0.697 0.576 0.063 0.000 0.319 0.302 *0.835
Communication Sources B (CS_B) 0.807 0.582 0.576 -0.048 0.022 0.263 0.270 0.759 *0.763
Decision Process (DP) 0.897 0.687 0.303 0.377 0.550 0.314 0.132 0.060 0.038 *0.829
*square root (SQRT) of average variance extracted (AVE)
305
Figure 7.5: Revised CPDFPMVI model after CFA model fit analysis
306
Table 7.8 provides the reliability and validity of the revised model. As can be seen,
there are no concerns and this revised model may be deemed acceptable. The revised
model can therefore be considered as model fit. The revised CPDFPMVI model after
the CFA model fit analysis is presented in Figure 7.5 on the previous page.
The modifications made to the baseline model were made after careful consideration,
and none of the latent variable meanings changed. Hair et al. (2019:679) suggested
that theory should always be predominantly considered when making model
modifications. The model presented in Figure 7.5 can therefore be regarded as the
acceptable model to use in Phase 3 of the study.
Interpretation of values in CFA revised CPDFPMVI model
Figure 7.5 is the acceptable model that will offer the means to calculate the summative
scores. These scores will reflect the index values of the observed variables that
represent each construct/latent variable. Table 7.9 shows these values.
307
Table 7.9: Index values of the observed variables that represent each
construct/latent variable
Index values
Constructs/Latent
Observed variables gained from
variables
CFA
Marketing Efforts Price of a policy 0.67

(ME) Process 0.71
People 0.78
Positive Positive family member 0.76

Sociocultural Positive reference person or group 0.82
Environment (SE_P)
Positive non-commercial sources 0.84
Positive social class 0.79
Cultural and subcultural factors 0.54
Negative Negative family member 0.84

Sociocultural Negative reference person or group 0.89
Environment (SE_N)
Negative non-commercial sources 0.86
Negative social class 0.82
Communication Traditional advertisements 0.78

Sources A (CS_A) Online advertisements 0.89
Communication Buzz agent 0.76

Sources B (CS_B) Customised messages 0.73
Paid-for social media efforts 0.80
Psychological Motivation 0.72

attributes (PA) Perception about motor vehicle insurance 0.81
Perception about a certain insurance company 0.82
Learning 0.81
Personality 0.68
Attitude towards motor vehicle insurance 0.78
Attitude towards a certain short-term insurance 0.73
company
Pre-purchase search Pre-purchase search requesting quotations 0.80

and evaluation of Evaluate alternative prices 0.91
alternatives steps
(Decision-making Evaluate alternative quality/standards 0.88
Process) (DP) Evaluate alternative brands 0.70
As revealed in the table above, the index values of each observed variable explain the
extent of influence each observed variable has on the construct/latent variable, which
influences the purchase decision of personal motor vehicle consumers. The higher
308
values indicate a greater influence, and the lower values indicate less influence. Within
the Marketing Efforts construct, the index values show that people have the greatest
influence on the purchase decision, and policy price has a lower influence on the
purchase decision.
All the index values in the Positive Sociocultural Environment construct are high,
except for the cultural and subcultural factors, which indicate that this observable
variable does not have such a big influence on the purchase decision as the other
observable variables within this construct.
In the Negative Sociocultural Environment construct, all the index values are high, thus
greatly influencing purchase decisions, especially the negative reference person or
group variable with the highest index score. It is interesting to see that the index values
of the Negative Sociocultural Environment’s observable variables are all higher than
the Positive Sociocultural Environment’s variables. This indicates that the Negative
Sociocultural Environment variables have a larger influence on the purchase decision
than the Positive Sociocultural Environment variables.
Both index values of the Communication Sources A construct scored high, but it is
evident that online advertisements have a slightly larger influence on the purchase
decision than traditional advertisements. Paid-for social media efforts are revealed to
have the largest influence in the Communication Sources B construct, with the other
two observable constructs also scoring high.
The personality variable scored the lowest in the Psychological Attributes construct,
which means that this variable has the least influence on the purchase decision.
Perception about a certain insurance company has the highest index value in this
construct, which indicates that this variable has the greatest influence on the purchase
decision within the Psychological Attributes construct. All the other index values in this
construct are high and indicate that they have a large influence on the purchase
decision. The personality variable has the lowest index value, and therefore the
assumption can be made that this variable has the lowest influence on the purchase
decision within the Psychological Attributes construct.
Lastly, the observable variables in the Decision-making Process construct indicate

high index values. The evaluate alternative prices variable has the greatest influence
309
on the purchase decision, and the evaluate alternative brands variable has the lowest
influence within this construct.
In Phase 3, the next phase, the observed variables and constructs/latent variables
(findings from Phase 2) will form the basis of the variable selection step of the logistic
regression model-building strategy. This strategy is discussed in the next section.
PHASE 3: LOGISTIC REGRESSION MODEL-BUILDING

STRATEGY
Logistic regression was introduced and defined in Section 5.11.2. Logistic regression
is a widely used model-building strategy to investigate the independent effect of
variables on dichotomous outcomes in consumer purchasing decision-making studies
within the global insurance market (Huber et al., 2015; Sin & Chee, 2017; Leiria, Matos
& Rebelo, 2020). This logistic regression method has proved to be useful in similar
studies. It, therefore, provided a reason to use it in the current study.
A sample size of over 400 is recommended to achieve the best results while using the
logistic regression model-building strategy (Hair et al., 2019:557). The overall sample
in this study is over the recommended 400. Therefore, this model-building strategy is
fitting. The steps followed in this logistic regression model-building strategy, using IBM
SPSS (Version 27) for analysis, are discussed in the sections below.
Variable selection with n=200 exploration sample
The independent variables that impact the dependent variable have to be identified
(Hair et al., 2019:552). In this study, the dependent variables are the answers to
Question 3 in the questionnaire (see Appendix A). This question was a dichotomous
question, and it will be the outcome of this model. Respondents were asked to indicate
whether they have personal motor vehicle insurance or not. The independent variables
are the latent variables tested and accepted in Phase 2 (hereon referred to as the
original variables). Also, certain demographic variables, as tested in the questionnaire,
are included in the independent variables/covariances. Therefore, the covariances are
listed below:
• Marketing Efforts (ME)
• Positive Sociocultural Environment (SE_P)
310
• Negative Sociocultural Environment (SE_N)
• Communication Sources A (CS_A)
• Communication Sources B (CS_B)
• Psychological Attributes (PA)
• Decision Process (DP)
• Age
• Gender
• Marital status
• Employment status
Once the dependent variables and the covariances had been identified, a univariate
analysis was performed to identify all the important covariances (Zhang, 2016:2). As
mentioned at the beginning of this chapter, a holdout sample is recommended to
ensure model exploration. Therefore, an exploration sample of 200 was used to
conduct the univariate analysis to identify all the important covariances.
Specific tests were performed as part of the univariate analysis to identify the important
covariances, namely, T-Test, Pearson Correlation, Chi-Square Tests, Kendall's tau-b
and Phi.
The T-Test and Pearson Correlation were used to analyse the original variables (latent
variables from CFA).
A T-Test is a two-independent samples test used to determine a significant difference

between the means of two groups (Ross & Willson, 2017:17). The rule of thumb is that
variables with a p-value less than 0.25 can be considered important to be included in
the logistic regression (Canchola et al., n.d.:1; Zhang, 2016:2).
The Pearson Correlation is utilised when the strength of the relationship between two
continuous variables is explored (Pallant, 2016:122). If the correlation coefficient (r-
value) indicates 0, there is no relationship between the two variables, but the r-value
311
can range from -1 to +1. The sign in front of the r-value indicates a positive or negative
correlation (Saunders et al., 2019:616).
The Chi-Square Test determines whether two categorical variables are independent
or related (Aldrich, 2019:18). The main value a researcher should be interested in in
this test is the Pearson Chi-Square value, which is significant (p-value) when it is less
than 0.05 (Pallant, 2016:240).
Kendall's tau-b is used to test the degree of association between two ordinal variables
(Saunders et al., 2019:616).
The Phi test is used when comparing pairs of dichotomous variables, and where the
strength of the association between two dichotomous variables is measured
(Saunders et al., 2019:612). For both Kendall's tau-b and Phi tests, the p-value should
be less than 0.05.
Table 7.10 presents a summary the statistics of the univariate analysis to identify the
important covariances.
312
Table 7.10: Univariate analysis summarised to identify the important covariances
T-Test Chi-square Include in

Variable Type r-value P-value Test
P-values P-values baseline model
Original variables
Marketing efforts Interval 0.000 0.363 0.000 Yes
Positive sociocultural environment Interval 0.179 0.098 0.168 Yes
Negative sociocultural environment Interval 0.131 0.109 0.123 Yes
Communication sources A Interval 0.000 -0.292 0.000 Yes
Communication sources B Interval 0.000 -0.299 0.000 Yes
Psychological attributes Interval 0.000 0.365 0.000 Yes
Decision process Interval 0.001 0.231 0.001 Yes
Demographic variables
Age Interval 0.000 0.331 0.000 Pearson Yes
Kendall's
Age regrouped
Ordinal 0.000 0.281 0.000 tau-b Yes
Nominal -
Gender
Dichotomous 0.588 -0.048 0.494 Phi No
Nominal -
Marital status regrouped
Dichotomous 0.000 0.348 0.000 Phi Yes
313
T-Test Chi-square Include in
Variable Type r-value P-value Test
P-values P-values baseline model
Kendall's
Highest level of education Ordinal 0.427 0.000 tau-b Yes
Kendall's
Highest level of education regrouped Ordinal 0.000 0.446 0.000 tau-b Yes
Employment status regrouped Ordinal 0.004 Yes
Kendall's
Household income Ordinal 0.486 0.000 tau-b Yes
Kendall's
Household income regrouped Ordinal 0.000 0.658 0.000 tau-b Yes
314
As indicated in the table above, the Positive Sociocultural Environment and Negative
Sociocultural Environment variables’ T-test shows that nothing distinguishes these two
groups from one another. In other words, there is no relationship between these two
variables and the dependent variables, due to the p-values indicating more than 0.05.
However, the rule of thumb is that a p-value less than 0.25 can be regarded as
important to be included in the logistic regression model (Canchola et al., n.d.:1;
Zhang, 2016:2). Therefore, these two variables can be included in the baseline logistic
regression model.
In the demographic variables univariate analysis, a distinction was made between the
original values and regrouped values. The values were regrouped, ensuring that each
group is big enough, taking into consideration that the sample of 200 was used for this
analysis. The marital status and employment status variables were presented only in
the regrouped values as these two variables had cells with very low cell counts. The
statistics presented in Table 7.10 above show that the gender demographic variable
is not important to be included in the baseline logistic regression model.
At the hand of the statistics provided in the table above, it is clear that the underlying
interrelationships between the covariances and the dependent variables were
assessed. Therefore, gender was the only covariance that could not be included in the
baseline logistic regression model. Step 2 of the logistic regression model-building
strategy is discussed below.
Logistic regression estimation with n=478 validation sample
After the variables have been selected, the selected variables can be fitted in a multiple
logistic regression model (Zhang, 2016:3). This step was performed, and the baseline
model was generated in the IBM SPSS (Version 27) software. The baseline logistic
regression model can now be assessed. Again, it is advised to use a holdout sample
to ensure model validation. A holdout sample of 478 was used to conduct the logistic
regression estimation, as logistic regression model building recommends that a
sample size of more than 400 is used to support the estimation of the model (Hair et
al., 2019:555). Thus, the validation sample of 478 was sufficient for model estimation.
The following tests determined the overall fit of a logistic regression model: Omnibus
Tests of Model Coefficients, Cox & Snell R Square and Nagelkerke R Square, Hosmer
and Lemeshow Test, and Wald test (Pallants, 2016:196-198).
315
The Omnibus Tests of Model Coefficients shows the x2/P-value associated with the
overall model (Keith, 2019:237). In other words, this test gives the researcher an
overall indication of how well the model performs. The Sig. value (x2/P-value) should
be less than 0.05 (Pallant, 2016:196).
The Cox & Snell R-Square and the Nagelkerke R-Square values indicate the amount
of variation in the dependent variable explained by the model, where a value between
0 and 1 is accepted (Pallant, 2016:197).
The Hosmer and Lemeshow Test also provide the support that the model is
worthwhile, and this value should be larger than 0.05 (Pallant, 2016:196). The
statistics of the overall model fit for the baseline logistic regression model are
presented in the tables below.
Table 7.11: Omnibus tests of Model Coefficients for baseline logistic regression
model
Sig.
Chi-square df
p-value
Step 1 Step 359.663 13 0.000
Block 359.663 13 0.000
Model 359.663 13 0.000
As can be seen in Table 7.11 above, all the p-values are less than 0.05. Therefore,
the Omnibus Tests of Model Coefficients shows the researcher that the baseline
model performs well overall.
Table 7.12: Cox & Snell R-Square and the Nagelkerke R-Square values for baseline
logistic regression model
Step -2 Log likelihood Cox & Snell R Square Nagelkerke R-Square
1 236.379 0.566 0.755
Table 7.12 above reveals that the Cox & Snell R-Square and the Nagelkerke R-Square
values are between 0 and 1, and can therefore, be accepted. Higher values better
explain the variation in the dependent variable of the model.
316
Table 7.13: Hosmer and Lemeshow Test for Baseline logistic regression model
Sig.
Step Chi-square df
p-value
1 7.157 8 0.520
Table 7.13 above shows that the p-value in the Hosmer and Lemeshow test is more
than 0.05, which supports that the model is worthwhile.
The classification table drawn from the statistics revealed that 89.6% could be
predicted from this baseline logistic regression model (see Table D5 in Appendix D).
Lastly, the Variables in the Equation table provides the information about the
contribution or importance of each of the predictor variables and the Wald test is used
here (Pallant, 2016:198). The variables in the equation table for the baseline logistic
regression model are presented in Table 7.14. The value of the statistic of each
predictor is labelled in the Wald column. The Wald values that scored very low are
indicated in red, and these covariances will be removed, as it suggests that they are
unimportant predictive variables. The Sig. value (x2/P-value) should be less than 0.05.
If there is a p-value that is more than 0.05, that variable does not significantly
contribute to the predictive ability of the model (Pallant, 2016:198). Therefore, any
variables with a p-value of more than 0.05 should be removed from the model.
317
Table 7.14: Variables in the Equation for baseline logistic regression model
95% CI for
Sig. EXP(B)
B S.E. Wald df Exp(B)
p-value
Lower Upper
Marketing
0.710 0.219 10.490 1 0.001 2.033 1.323 3.123
efforts
Positive
Sociocultural -0.967 0.265 13.358 1 0.000 0.380 0.226 0.638
Environment
Negative
Sociocultural 0.405 0.184 4.856 1 0.028 1.499 1.046 2.149
Environment
Communication
0.075 0.200 0.140 1 0.709 1.078 0.728 1.596
Sources A
Communication
-0.953 0.246 15.009 1 0.000 0.385 0.238 0.624
Sources B
Psychological
0.559 0.232 5.826 1 0.016 1.749 1.111 2.754
Attributes
Decision
0.505 0.207 5.939 1 0.015 1.658 1.104 2.489
Process
Age 0.053 0.015 13.321 1 0.000 1.055 1.025 1.086
Marital status 0.386 0.366 1.107 1 0.293 1.470 0.717 3.015
Highest level of
1.118 0.244 21.060 1 0.000 3.058 1.897 4.928
education
Employment
2.859 2 0.239
status
Employment
-1.298 0.772 2.832 1 0.092 0.273 0.060 1.239
status(1)
Employment
-0.392 0.508 0.594 1 0.441 0.676 0.250 1.830
status(2)
Household
1.556 0.285 29.868 1 0.000 4.740 2.713 8.283
income
Constant -9.804 1.383 50.222 1 0.000 0.000
The table above indicates the variables in the equation for the baseline logistic
regression model. As discussed above, the p-values above 0.05 are variables that do
not significantly contribute to the model's predictive ability. The p-values indicated in
318
red are more than 0.05. These covariances will be removed from the model as they
do not significantly contribute to the model's predictive ability, and theoretically, it
makes sense to remove them. The B values provided in the table above are the values
used to calculate the probability of a case falling into a specific category. The positive
or negative sign before the value indicates the direction of the relationship (Pallant,
2016:198). Thus, the negative values will decrease the probability of purchasing
personal motor vehicle insurance, and positive values will increase the probability of
purchasing personal motor vehicle insurance.
The other useful column in Table 7.14 is the Exp(B) column. These are the odds ratios
(OR) for each independent variable (Pallant, 2016:198). The odds ratio refers to the
change in odds of being in one of the outcome categories (yes or no purchase) when
the value of a predictor increases by one unit (Tabachnick & Fidell, 2019:367).
Therefore, the biggest drivers that explain variation of the decision to purchase are
marketing efforts (2.033), highest level of education (3.058) and household income
(4.740).
Furthermore, for each odds ratio Exp(B) indicated in the Variables in the Equation
table, there is a 95% confidence interval (95% CI for EXP(B)) that gives a lower value
and upper value. It indicates that if an infinite number of similar studies were
undertaken and the estimates calculated, in 95% of cases, the true odds will be
included (Pallant, 2016:199). The 95% CI for EXP(B) column in Table 7.14 indicates
the lower and upper confidence levels for the baseline logistic regression model.
After the baseline logistic regression model overall fit analysis, the variables that did
not appear statistically important were eliminated. Thus, the variables that will be used
in the final logistic regression model are as follow:
• Marketing Efforts (ME)
• Positive Sociocultural Environment (SE_P)
• Negative Sociocultural Environment (SE_N)
• Communication Sources B (CS_B)
• Psychological Attributes (PA)
• Decision Process (DP)
• Age
319
Before the final logistic regression model can be tested, the absence of
multicollinearity should be determined, because logistic regression is sensitive to
extremely high correlations between predictor variables (Tabachnick & Fidell,
2019:352). To establish multicollinearity problems, ‘collinearity diagnostics’ need to be
performed via IBM SPSS (Version 27). These results are present in the Coefficients
table, and the Collinearity Statistics column values are particularly important (Pallant,
2016:178).
Table 7.15 presents the Collinearity Statistics of the variables in the final logistic
regression model.
Table 7.15: Collinearity Statistics for final logistic regression variables
Variables in final logistic regression model Collinearity Statistics
Tolerance VIF
Marketing Efforts (ME) 0.686 1.458
Positive Sociocultural Environment (SE_P) 0.511 1.957
Negative Sociocultural Environment (SE_N) 0.604 1.655
Communication Sources B (CS_B) 0.854 1.172
Psychological Attributes (PA) 0.700 1.429
Decision Process (DP) 0.655 1.527
Age 0.938 1.067
Highest level of education 0.597 1.674
Household income 0.554 1.805
The table above presents the variables and collinearity statistics. The two values
indicated are tolerance and VIF. Tolerance is an indicator of how much of the variability
of the independent is not explained by other independent variables in the model. If the
tolerance value is less than 0.1, it shows that there is multicollinearity present.
VIF (Variance Inflation Factor) is the other value that is just the opposite of tolerance,
and a value above ten would be a concern indicating multicollinearity (Pallant,
2016:178). As shown in the table above, none of the tolerance values are less than
320
0.1, and none of the VIF values are above 10. Therefore, the assumption can be made
that there is no multicollinearity amongst the variables for the final logistic regression
model. The next step in the logistic regression model-building strategy is discussed in
the section below.
Check the assumption of linearity
A logistic regression assumes that a linear relationship exists between the continuous
predictors (covariances) and the dependent variable (Tabachnick & Fidell, 2019:351).
Therefore, covariances that have a non-linear relationship have to be removed from
the logistic regression model. The Box-Tidwell procedure is the simplest way to test
for non-linearity. The linearity assumption is tested by including the model interactions
between the covariances and their logs (Hair et al., 2019:556). Table 7.16 illustrates
the variables in the equation to test the linearity assumption.
Table 7.16: Variables in the equation to test linearity assumption
Sig.
Variables in final logistic regression model
p-value
Marketing Efforts (ME) 0.910
Positive Sociocultural Environment (SE_P) 0.423
Negative Sociocultural Environment (SE_N) 0.194
Communication Sources B (CS_B) 0.185
Psychological Attributes (PA) 0.571
Decision Process (DP) 0.591
Age 0.000
Highest level of education 0.000
Household income 0.000
ME by ME_ln 0.818
SE_P by SE_P_ln 0.680
SE_N by SE_N_ln 0.282
CS_B by CS_B_ln 0.439
PA by PA_ln 0.417
DP by DP_ln 0.811
321
The variables in the equation table above (see complete table in Table D6, Appendix
D) indicated no significant interaction amongst the covariance’s logs. The blue
highlighted values indicate no significance, as all these p-values are higher than 0.05.
Therefore, the assumption of linearity can be accepted.
Next, the overall model fit of the final logistic regression model can be assessed, which
is discussed below.
Assessing the overall fit of final logistic regression model
Taking into consideration that there is no multicollinearity amongst the variables in the
final logistic regression model and that the assumption of linearity can be accepted,
the overall fit was assessed next. As discussed in Section 7.4.2 above, specific tests
determine the overall fit of a logistic regression model. These tests are discussed
below.
Firstly, the Omnibus Tests of Model Coefficients are presented in the table below.
Table 7.17: Omnibus Tests of Model Coefficients for final logistic regression model
Sig.
Chi-square df
p-value
Step 1 Step 355.941 9 0.000
Block 355.941 9 0.000
Model 355.941 9 0.000
The table above indicates that all the p-values are less than 0.05. Thus, the Omnibus
Tests of Model Coefficients shows that the final model performs well overall.
Secondly, the Cox & Snell R-Square and the Nagelkerke R-Square values are
presented in Table 7.18 below.
322
Table 7.18: Cox & Snell R-Square and the Nagelkerke R-Square values for final
logistic regression model
Step -2 Log likelihood Cox & Snell R-Square Nagelkerke R-Square
1 240.100 0.562 0.750
Table 7.18 above reveals that the Cox & Snell R-Square and the Nagelkerke R-Square
values are between 0 and 1, and can therefore, be accepted.
Thirdly, the Hosmer and Lemeshow test results are provided in the table below.
Table 7.19: Hosmer and Lemeshow Test for Baseline logistic regression model
Sig.
Step Chi-square df
p-value
1 10.364 8 0.240
In the table above, the p-value in the Hosmer and Lemeshow test is more than 0.05,
which supports that the model is worthwhile.
The classification table drawn from the statistics revealed that 88.9% could be
predicted from this final logistic regression model (see Table 7.21 in Section 7.4.5).
Lastly, the information about the Wald test is available in the Variables in the Equation
table. The Variables in the Equation table for the final logistic regression model are
presented in Table 7.20. There are no Wald values that scored very low, and therefore
all the covariances indicate that they are important predictive variables. The Sig. value
(x2/P-value) should be less than 0.05.
323
Table 7.20: Variables in the Equation for final logistic regression model
95% CI for
Sig. EXP(B)
B S.E. Wald df Exp(B)
p-value
Lower Upper
Marketing
0.707 0.218 10.516 1 0.001 2.027 1.323 3.108
efforts
Positive
Sociocultural -0.978 0.254 14.772 1 0.000 0.376 0.229 0.619
Environment
Negative
Sociocultural 0.423 0.181 5.464 1 0.019 1.526 1.071 2.176
Environment
Communication
-0.932 0.221 17.707 1 0.000 0.394 0.255 0.608
Sources B
Psychological
0.568 0.226 6.308 1 0.012 1.765 1.133 2.749
Attributes
Decision
0.478 0.204 5.484 1 0.019 1.612 1.081 2.405
Process
Age 0.061 0.013 23.270 1 0.000 1.063 1.037 1.090
Highest level of
1.108 0.236 22.011 1 0.000 3.028 1.906 4.810
education
Household
1.570 0.271 33.697 1 0.000 4.808 2.830 8.171
income
Constant -10.101 1.290 61.305 1 0.000 0.000
As can be seen in the table above, there are no p-values more than 0.05. Thus, all the
covariances are variables that significantly contribute to the predictive ability of this
final model. The B values provided in the tables are the values that are used to
calculate the probability of a case falling into a specific category. These values and
their positive or negative relationship are presented in the table above. Only Positive
Sociocultural Environment and Communication Sources B will decrease the
probability of purchasing personal motor vehicle insurance. All the other B values are
positive.
The other useful column in Table 7.20 is the Exp(B) column. The biggest drivers that
affect the decision to purchase are still marketing efforts (2.027), highest level of
education (3.028), and household income (4.808).
324
The strongest predictor of reporting that they have got insurance was Household
income, recording an odds ratio of 4.808. This indicated that respondents who had
personal motor vehicle insurance were over four times more likely to report higher
levels of household income than those who did not have personal motor vehicle
insurance, controlling for all other factors in the model.
The odds ratio of 0.376 for Positive Sociocultural Environment was less than 1,
indicating that respondents who did not have personal motor vehicle insurance were
less likely to report that positive opinions/comments from the sociocultural
environment greatly influenced their need to purchase, than those who had personal
motor vehicle insurance, controlling for other factors in the model.
Furthermore, the 95% CI for EXP(B) column in Table 7.20 indicates the final logistic
regression model's lower and upper confidence levels. After the final logistic
regression model overall fit analysis, all the covariances appear to be statistical
important variables and should be included in the final CPDFPMVI model. The
CPDFPMVI model classification and validation are provided below.
Classification and validation of the final CPDFPMVI model
The section above mentioned the classification of the CPDFPMVI model. However, a
more detailed classification is required to understand how well the model can predict
the correct category for each case (purchase or not purchase).
The model's sensitivity is the percentage of the group with the characteristic of interest
(purchase) that has been accurately identified in the model. Whereas the model's
specificity is the group without the characteristic (no purchase) that has been correctly
identified (Pallant, 2016:197; Tabachnick & Fidell, 2019:372).
Then there is the positive and negative predictive value, which is also essential to
discuss in the classification of the model. The positive predictive value (PPV) refers to
the percentage of cases that has the characteristic (purchase personal motor vehicle
insurance) and is actually observed in this group. On the other hand, the negative
predictive value (NPV) is the percentage of cases that the model classifies as not
having the characteristic (not purchasing personal motor vehicle insurance), and that
is actually part of this group (Pallant, 2016:197).
Table 7.21 presents the classification table of the final CPDFPMVI model.
325
Table 7.21: Classification table of final CPDFPMVI model
Predicted
Do you currently have motor

vehicle insurance for your
personal motor vehicle?
Percentage
No purchase Purchase correct
Do you currently have No purchase 179 24 88.2

motor vehicle insurance
for your personal motor Purchase 24 204 89.5
vehicle?
Overall percentage 88.2 89.5 88.9
The colours in the table above indicate the important classifications for the final
CPDFPMVI model. The final CPDFPMVI model predicts the following:
• Sensitivity: 89.5% of the group that purchase personal motor vehicle insurance
has been accurately identified by the model (the true positives).
• Specificity: 88.2% of the group that do not purchase personal motor vehicle
insurance is correctly identified (the true negatives).
• PPV: 89.5% of cases that the model classifies as purchasing personal motor
vehicle insurance are actually observed in this group.
• NPV: 88.2% of cases predicted by the model that do not purchase personal
motor vehicle insurance are actually observed not to purchase personal motor
vehicle insurance.
Thus, this final CPDFPMVI model can predict 88.9% of the cases. Taking into
consideration that the overall accuracy measure of 88.9% and the sensitivity and
specificity are sufficiently large, it indicates predictive accuracy for each outcome
group. Both the PPV and NPV values are above 80%, which provides reasonable
levels of predictive ability for both positive and negative outcomes. The validation
sample (n=478) used to perform this logistic regression, was particularly useful as it
provides evidence of external validity. These outcomes conclude that the logistic
regression model demonstrated sufficient external validity for complete acceptance of
326
the result, as was found with the overall model fit analysis of the final CPDFPMVI
model.
The final CPDFPMVI model is presented in the figure below. The data obtained in this
study ensured that the research objectives had been achieved. Table 7.22
summarises the outcome of the research objectives, and this table is presented after
Figure 7.6.
327
Figure 7.6: Final CPDFPMVI model
328
Table 7.22: Outcome of research objectives following logistic regression
To develop a model that explains the consumer purchase decision process in the South African personal motor vehicle insurance industry.
As per the final logistic regression model: The final model for complete acceptance of results obtained in this study is depicted in Figure
7.6 above.
Final CPDFPMVI model after

Secondary research objectives As per EFA As per CFA
logistic regression
To determine the company’s ▪ Type of policies offered ▪ Price of a policy ▪ Price of a policy
marketing efforts that influence ▪ Price of a policy ▪ Process ▪ Process
the individual’s need to purchase
▪ Distribution channels ▪ People ▪ People
personal motor vehicle insurance
in South Africa. ▪ Physical evidence
▪ Process
▪ People
▪ Partnerships
To establish the sociocultural Positive sociocultural Positive sociocultural Positive sociocultural

environment factors that influence environment environment environment
the individual’s need to purchase ▪ Positive family member ▪ Positive family member ▪ Positive family member
in South Africa. ▪ Positive reference person or ▪ Positive reference person or ▪ Positive reference person or
group group group
▪ Positive non-commercial ▪ Positive non-commercial ▪ Positive non-commercial
sources sources sources
▪ Positive social class ▪ Positive social class ▪ Positive social class
▪ Cultural and subcultural ▪ Cultural and subcultural ▪ Cultural and subcultural
factors factors factors
329
logistic regression
Negative sociocultural Negative sociocultural Negative sociocultural

environment environment environment
▪ Negative family member ▪ Negative family member ▪ Negative family member
▪ Negative reference person or ▪ Negative reference person or ▪ Negative reference person or
group group group
▪ Negative non-commercial ▪ Negative non-commercial ▪ Negative non-commercial
sources sources sources
▪ Negative social class ▪ Negative social class ▪ Negative social class
To identify the communication Communication sources A Communication sources A Communication sources

sources that influence the ▪ Traditional advertisements ▪ Traditional advertisements ▪ Buzz agent
individual’s need to purchase
personal motor vehicle insurance ▪ Online advertisements ▪ Online advertisements ▪ Customised messages
in South Africa. ▪ Promotion ▪ Promotion ▪ Paid-for social media efforts
Communication sources B Communication sources B

▪ Buzz agent ▪ Buzz agent
▪ Customised messages ▪ Customised messages
▪ User-generated social media ▪ Paid-for social media efforts
posts
▪ Paid-for social media efforts
330
logistic regression
To determine the individual’s ▪ Motivation ▪ Motivation ▪ Motivation

psychological attributes that ▪ Perception about motor ▪ Perception about motor ▪ Perception about motor
influence the need recognition to vehicle insurance vehicle insurance vehicle insurance
purchase personal motor vehicle
▪ Perception about a certain ▪ Perception about a certain ▪ Perception about a certain
insurance in South Africa.
insurance company insurance company insurance company
▪ Learning ▪ Learning ▪ Learning
▪ Personality ▪ Personality ▪ Personality
▪ Attitude towards motor ▪ Attitude towards motor ▪ Attitude towards motor
vehicle insurance vehicle insurance vehicle insurance
▪ Attitude towards a certain ▪ Attitude towards a certain ▪ Attitude towards a certain
short-term insurance short-term insurance short-term insurance
company company company
To investigate the steps of the ▪ Pre-purchase search ▪ Pre-purchase search ▪ Pre-purchase search
decision-making process for the considering previous requesting quotations requesting quotations
personal motor vehicle insurance experiences ▪ Evaluate alternative prices ▪ Evaluate alternative prices
industry in South Africa. ▪ Pre-purchase search ▪ Evaluate alternative ▪ Evaluate alternative
requesting quotations quality/standard quality/standard
▪ Evaluate alternative prices ▪ Evaluate alternative brands ▪ Evaluate alternative brands
▪ Evaluate alternative
quality/standard
To assess the demographic ▪ Age

variables that influence the ▪ Highest level of education
individual’s need to purchase
in South Africa.
331
logistic regression
To assess the consumer Addressed in Chapter 6. Addressed in Chapter 6. Addressed in Chapter 6.

demographic profile within the
South African personal motor
To compile a general Addressed in Chapter 6. Addressed in Chapter 6. Addressed in Chapter 6.

demographic profile that
describes the average personal
motor vehicle driver in South
Africa.
332
As summarised in Figure 7.6 and Table 7.22 above, the research objectives of this
study have been achieved.
CONCLUSION
This chapter discussed the process followed to build the final CPDFPMVI model. The
model-building strategy that was followed consisted of three phases: the EFA phase,
CFA phase, and logistic regression phase. Firstly, the EFA was used to explore and
estimate the factor structure of the conceptual model, as presented in Figure 2.2.
Secondly, the EFA results were used to perform a CFA, which tested how well a pre-
specified measurement theory, composed of measured variables and factors, fits the
reality as captured data. Thirdly, the CFA model fit findings were used in the logistic
regression to describe the relationship between an outcome (dependent) variable and
a set of independent (predictor or explanatory) variables.
These outcomes conclude that the logistic regression model demonstrated sufficient
external validity for complete acceptance of the result, as was found with the overall
model fit analysis of the final CPDFPMVI model. The final CPDFPMVI model was
presented in this chapter (see Figure 7.6).
Important conclusions were drawn which addressed the research aim, objectives and
overall purpose of the study as set out in Chapter 1.
Finally, in the last chapter (Chapter 8), the conclusions of this study will be
summarised, the limitations will be discussed, and recommendations for future studies
will be provided.
333
CONCLUSIONS AND RECOMMENDATIONS
INTRODUCTION
The goal of this chapter is to conclude this study, which aimed to develop a unique
model that explains the purchase decision process of consumers (insureds) and
potential consumers (non-insureds) in the South African personal motor vehicle
insurance industry. Using a conceptual consumer purchase decision for personal
motor vehicle insurance (CPDFPMVI) model obtained from the literature review, this
research aimed to test the conceptual model and develop a final CPDFPMVI model
by utilising a carefully selected model-building strategy as methodology.
This concluding chapter provides a brief overview of the research that has been
conducted, and presents a summary of the objectives, and the research aim that was
addressed during the research process. This chapter draws conclusions from the data
analysis presented in Chapters 6 and 7. Furthermore, the primary and secondary
objectives, as well as the research question of this study are addressed and
concluded; after which, the recommendations for short-term insurance companies and
for future research studies are provided. The limitations of this study will be
highlighted, and lastly, this chapter and research are concluded.
OVERVIEW OF RESEARCH AIMS AND OBJECTIVES
The overall purpose of this study was to determine the purchase decision process of
consumers (insureds) and potential consumers (non-insureds) regarding personal
motor vehicle insurance. The aim was to better understand the consumer purchase
decision process within the personal motor vehicle insurance industry of South Africa.
Each influencing factor and step in the purchase decision process was explored to
determine which specific items contribute to the consumer purchase decision process.
This study aimed to answer the following main research question: Which factors
influence the individuals’ purchase decision in the South African personal motor
vehicle insurance industry?
A CPDFPMVI conceptual model was proposed in Chapter 4 to achieve this overall

purpose of the study. The conceptual model was developed by collecting relevant and
334
previous literature on various consumer decision-making process models, insurance
as a financial service and the purchase behaviour of consumers. The conceptual
model was tested by following three phases, namely the Exploratory Factor Analysis
(EFA), Confirmatory Factor Analysis (CFA) and the logistic regression. The following
secondary objectives supported the primary objective of this study:





The influences and steps in the consumer purchase decision process, as contained in
the conceptual CPDFPMVI model, were generated after several consumer decision-
making models that had been developed by scholars over the years had been
evaluated. All the limitations and expansions of the various consumer decision-making
models (see Chapter 3) were considered in the development of the conceptual
CPDFPMVI model (presented and discussed in Chapter 4). Only the influences and
steps up to the purchase decision were considered in the models that were reviewed.
The aim of the study was to develop a model that explains the purchase decision
process of personal motor vehicle consumers and potential consumers in South Africa.
335
The justification for including all the items in the conceptual CPDFPMVI model can be
found in Table 5.7 in Chapter 5.
Consequently, this study's primary and secondary research objectives were

addressed by designing a questionnaire that permitted the researcher to measure the
purchase decision process for personal motor vehicle insurance consumers and
potential consumers in South Africa. The particular methodology used to collect this
data was discussed comprehensively in Chapter 5.
After the data had been collected, the descriptive statistical analysis (Chapter 6) and
the model fit analysis (Chapter 7) were performed. The reflections and conclusions of
the analysed data will be discussed in the following section. A summary of the
demographic profile of respondents in this study is presented in Table 8.6 (see Section
8.3.6).
REFLECTIONS AND CONCLUSIONS: ADDRESSING THE

RESEARCH OBJECTIVES
This section provides a reflection of each secondary objective and the primary
objective of the study. The link between the literature and findings is provided.
Conclusions in terms of the research objectives will be drawn based on the literature
and empirical findings from the primary data analysis.
The company’s marketing efforts
This secondary objective aimed to determine the marketing efforts the short-term
insurance company performs, which influence the need to purchase personal motor
vehicle insurance. Table 8.1 showcases the company’s marketing efforts from the
secondary and primary data and the conclusions applicable to this objective.
336
Table 8.1: Reflections and conclusions for Objective 1
Objective
To determine the company’s marketing efforts that influence the individual’s need to
Findings from the literature
The marketing mix is viewed as the basic marketing toolkit that the marketing function in
an organisation can use to implement its marketing plans. However, there is a distinct
difference between the product marketing mix and the service marketing mix. Taking into
consideration that personal motor vehicle insurance is a financial service, the service
marketing mix is used to direct short-term insurance companies' marketing efforts
(Section 3.2.1.4). As explained in Section 3.4, marketing efforts are external influencing
factors that can impact an individual’s purchase decision. Each marketing effort and how it
can influence the purchase decision is discussed in Section 4.3.1. The marketing efforts
that short-term insurance companies can perform are, for example, types of policies
offered, promotions, price of policy, distribution channels, physical, process, people and
partnerships.
Findings from primary data analysis
▪ The descriptive findings showed Type of policy (74.2%), Promotion (65.9%) and Price
of policy (61.7%) as the three marketing efforts that have the most influence on the
need to purchase personal motor vehicle insurance.
▪ During the model fit analysis, the inferential findings of the CFA indicated that the
three marketing efforts that influence the need to purchase personal motor vehicle
insurance in South Arica are also Price of the policy, People and Process. However,
the factor loadings revealed that People has the greatest influence, and Price of the
policy has a lower influence on the purchase decision.
▪ The logistic regression findings indicate that these three marketing efforts will increase
the probability of purchasing personal motor vehicle insurance.
Conclusions
▪ The marketing effort elements in the final CPDFPMVI can therefore be identified in the
following order, (1) People, (2) Process and (3) Price of the policy. These items will
increase the probability of purchasing personal motor vehicle insurance. Respondents
who had personal motor vehicle insurance were more likely to report a higher
influence of marketing efforts elements than those who did not have personal motor
vehicle insurance. Thus, marketing efforts by short-term insurance companies do, in
fact, drive the decision to purchase personal motor vehicle insurance.
▪ The way in which employees in short-term insurance companies interact with the
consumer is seen to be the most important marketing effort element amongst the
respondents in this study. This finding has implications for the service delivery
strategies of insurance companies, especially regarding how employees treat
consumers while they deliver the service.
▪ The duration of the process to obtain a quotation is the marketing effort that has the
second-largest influence on the need to purchase personal motor vehicle insurance.
This finding has implications for the service delivery strategies of insurance
companies.
337
▪ Price is the marketing effort that has the third most influence on the need to purchase
personal motor vehicle insurance. This is a meaningful finding which has implications
for the pricing strategies of insurance companies.
The sociocultural environment
The aim of this secondary research objective was to establish the sociocultural
environment factors that influence the individual’s need to purchase personal motor
vehicle insurance in South Africa. Table 8.2 highlights the literature and primary data
together with the conclusions drawn for this objective.
Objective
To establish the sociocultural environment factors that influence the individual’s need to
Sociocultural environment factors are external influences that could impact individuals’
need to purchase. The sociocultural environment consists of family, reference groups,
non-commercial sources, social class, and cultural and subcultural factors that can affect
individuals' purchase decision process or that of potential consumers. Each one of these
factors and how it can influence the purchase decision was discussed in Section 4.3.2.
Each one of these sociocultural factors can have a positive or negative effect on the
individuals’ need to purchase personal motor vehicle insurance.
The descriptive findings indicated that:

▪ A negative comment or opinion from a family member plays a more important role in
the consumers’ thought process than a positive comment or opinion from a family
member, which influences the consumers’ need to purchase personal motor vehicle
insurance.
▪ The positive opinion of a reference person or group used as a basis of comparison
has a slightly larger influence on the respondents’ need to purchase than the negative
opinion of a reference person or group used as a basis of comparison.
▪ The positive comments of a friend, a newspaper article, or an expert's views on a blog
or other internet platforms have a slightly larger influence on the respondents’ need to
purchase than the negative comments of a friend, a newspaper article, or an expert's
views on a blog or other internet platforms.
▪ The positive opinions of individuals in the same social class have a slightly larger
influence on the respondents’ need to purchase personal motor vehicle insurance than
the negative opinions of individuals in the same social class.
The model fit analysis revealed that:
▪ In the CFA, all the variables (Positive family member, Positive reference person or
group, Positive non-commercial sources and Positive social class) in the Positive
Sociocultural Environment construct index values are high, except for the Cultural and
338
subcultural factor, which indicates that it does not have such a big influence on the
purchase decision as the other Positive Sociocultural Environment variables do (see
Table 7.9).
▪ The CFA showed that all the index values (see Table 7.9) are high in the Negative
Sociocultural Environment construct, thus greatly influencing purchase decisions,
especially the Negative reference person or group variable with the highest index
score.
▪ The logistic regression indicated that the Positive Sociocultural Environment variables
would decrease the probability of purchasing personal motor vehicle insurance,
whereas the Negative Sociocultural Environment variables would increase the
probability of purchasing personal motor vehicle insurance.
Conclusions
▪ The descriptive findings show that more of the Positive Sociocultural Environment
variables greatly influence the respondents’ need to purchase personal motor vehicle
insurance than the Negative Sociocultural Environment variables.
▪ After further analysis, the final model fit results indicate that the Negative Sociocultural
Environment variables have a more important influence on the purchase decision than
the Positive Sociocultural Environment variables. One would think that the negative
opinion of a family member, reference person or group, non-commercial sources and
members of the social class would negatively impact the consumers’ need to
purchase personal motor vehicle insurance. In fact, the final CPDFPMVI model
predicts the opposite. The question regarding this construct in the questionnaire was
an influence scale type of question, therefore, it measured the extent of influence of
each statement. The statements in this question asked the positive or negative
comments/opinions of a family member, a reference person/group, a friend, a
newspaper article or the views of an expert on a blog or other internet platforms and
individuals in the same social class in general, and not specifically for comments or
opinions about motor vehicle insurance or short-term insurance companies. Thus, a
general comment from a family member about the increase of stolen motor vehicles in
the area can be regarded as a negative comment, but this comment will highlight the
risk to a consumer and this can persuade the individual to rather purchase insurance
in case his/her motor vehicle is also stolen. Whereas, a positive comment that there is
a decrease in stolen motor vehicles in the area might let the individual feel that his/her
risk is not too high, and that they therefore can take the chance not to purchase
personal motor vehicle insurance. Taken that insurance is a grudge purchase, this
construct revealed that that is exactly how consumers think when they purchase
personal motor vehicle insurance. Therefore, this study found that Positive
Sociocultural Environment variables decrease the probability of purchasing personal
motor vehicle insurance, whereas the Negative Sociocultural Environment variables
increase the probability of purchasing personal motor vehicle insurance. It is important
to point out that these findings are based on the opinions of respondents. The
opinions gathered in this study revealed that the Negative Sociocultural Environment
variables have a more important influence on the purchase decision than the Positive
Sociocultural Environment variables.
▪ Respondents who had personal motor vehicle insurance were more likely to report
that negative opinions/comments from the sociocultural environment greatly
influenced the need to purchase than those who did not have personal motor vehicle
insurance. In contrast, respondents who did not have personal motor vehicle
insurance were less likely to report that positive opinions/comments from the
sociocultural environment greatly influenced their need to purchase than those who
had personal motor vehicle insurance.
339
The communication sources
This secondary objective aimed to identify the communication sources that influence
the individual’s need to purchase personal motor vehicle insurance in South Africa.
Table 8.3 draws attention to the literature and primary data, as well as the conclusions
for this objective.
Objective
To identify the communication sources that influence the individual’s need to purchase
personal motor vehicle insurance in South Africa.
The communication sources refer to the mechanisms that deliver the company’s
marketing efforts and sociocultural influences to the market. The communication sources
are advertising, buzz agents, customised messages, owned or paid-for social media and
user-generated social media posts (as discussed in Section 4.3.3). A message received
by an individual through a communication source can influence the need for personal
motor vehicle insurance, or it can influence individuals not to buy.
The descriptive findings showed that:
▪ Traditional and online advertisements do not influence the respondents’ purchasing

needs to a large extent.
▪ There were more respondents who felt that buzz agents do not influence their need to
purchase to a large extent.
▪ Customised messages that insurance companies communicate have an influence on
most respondents’ purchasing needs.
▪ Just over half of the respondents stipulated that user-generated social media posts
slightly influence their need to purchase personal motor vehicle insurance.
▪ Social media efforts by insurance companies do not influence the respondents’
purchasing needs to a large extent.
▪ The CFA revealed that Online advertisements have a slightly higher influence than
Traditional advertisements. However, the logistic regression model-building strategy
indicates that these two communication sources did not contribute significantly to the
model's predictive ability and they were removed.
▪ After the CFA, only Buzz agents, Customised messages and Paid-for social media
efforts were deemed significant, with Paid-for social media efforts having the largest
influence.
▪ The logistic regression revealed that these three communication sources (Paid-for
social media efforts, Buzz agents and Customised messages) would decrease the
probability of purchasing personal motor vehicle insurance.
340
Conclusions
▪ According to this model, Communication sources decrease the probability of

purchasing personal motor vehicle insurance, which is an interesting finding as one
would think that Customised messages and Paid-for social media efforts by short-term
insurance companies would aim to increase individuals' purchase decisions. This
question in the questionnaire also tested each statement's influence scale. The
descriptive statistics revealed that two (Buzz agents and Paid-for social media efforts)
out of the three Communication sources included in the final model do not influence
the respondents’ purchasing decisions to a large extent. This explains why the
Communications sources identified in the final model decrease the probability of
purchasing.
▪ One would have expected that user-generated social media posts would have a much
more considerable influence on the need to purchase motor vehicle insurance. Social
media is a commonly used platform where consumers express their experiences of
products and services.
▪ Recent research found (discussed in Section 4.3.3) that word-of-month using user-
generated social media is essential in the motor vehicle insurance market to enable
individuals to talk about the company’s motor vehicle insurance products. However, it
is noteworthy that this study found that it does not influence the need to purchase
personal motor vehicle insurance in South Africa to a large extent.
▪ Although it appears that buzz agents do not influence respondents’ need to purchase
to a large extent, short-term insurance companies may not be using buzz agents
enough or correctly.
▪ Globally, customised messages are seen as an effective way of reaching motor
vehicle insurance consumers (discussed in Section 4.3.3). This study’s findings agree
with the global findings that customised messages as an external factor influence the
individual’s need to purchase personal motor vehicle insurance in South Africa.
▪ Even though traditional and online advertising were not significant enough to be
included in the final CPDFPMVI model, it is still crucial for brand awareness. When
consumers see/hear a traditional or online advertisement, they will recall the brand
and consider it when they are in the purchase decision-making process.
▪ Respondents who do not have personal motor vehicle insurance were less likely to
report a large influence of communication sources than those who had personal motor
vehicle insurance. In other words, communication sources by short-term insurance
companies do, in fact, decrease the probability of making a decision to purchase
The psychological attributes
The aim of this secondary objective was to determine the individual’s psychological
attributes that influence the need recognition to purchase personal motor vehicle
insurance in South Africa. Table 8.4 presents the literature and primary data, as well
as the conclusions drawn for this objective.
341
Objective
To determine the individual’s psychological attributes that influence the need recognition
▪ The motivation of individuals to purchase insurance is determined by two main factors,

risk expectation and risk sensitivity, as discussed in Section 4.4.1.1.
▪ The individuals’ perception of a financial service, such as motor vehicle insurance,
also plays a role in consumer satisfaction (see Section 4.4.1.2).
▪ The process of how individuals purchase and consume information, and the
experience gained (learning) is an individual factor that impacts the consumer’s need
recognition, which is the first step of the decision-making process (see Section
4.4.1.3).
▪ Personality is the inner characteristics of an individual that impact the individual’s need
recognition to purchase a financial service, such as personal motor vehicle insurance
(as discussed in Section 4.4.1.4).
▪ As discussed in Section 4.4.1.5, attitudes regarding insurance and a specific
insurance company are mainly formed from past purchase experiences.
The descriptive findings indicated that:
▪ 50.9% of the respondents specified that the motivation to purchase personal motor
vehicle insurance largely influences their need recognition for personal motor vehicle
insurance.
▪ 52.5% of the respondents specified that the perception about personal motor vehicle
insurance largely influences their need recognition for personal motor vehicle
insurance.
▪ 51.4% of the respondents specified that the perception about a certain short-term
insurance company largely influences their need recognition for personal motor
vehicle insurance.
▪ 48.8% of the respondents indicated that how they purchase and consume information
and the experience gained largely influence their need recognition for personal motor
vehicle insurance.
▪ 43.5% of the respondents specified that their personality largely influences their need
recognition for personal motor vehicle insurance.
▪ 55% of the respondents indicated that their attitude towards motor vehicle insurance
largely influences their need recognition for personal motor vehicle insurance.
▪ 50.2% of the respondents indicated that their attitude towards a certain short-term
insurance company largely influences their need recognition for personal motor
vehicle insurance.
The model fit analysis showed that:
▪ Based on the CFA, Personality has the least influence on the purchase decision, and
Perception about a certain short-term insurance company has the greatest influence
342
on the need to purchase personal motor vehicle insurance. All the other items have a
large influence on the purchase decision, as the index values are high.
▪ The logistic regression indicated that all five Psychological Attributes would increase
the probability of purchasing personal motor vehicle insurance.
▪ Respondents who had personal motor vehicle insurance were more likely to report
higher levels of agreement with the psychological attributes’ influence on the need
recognition than those who did not have personal motor vehicle insurance.
Conclusions
▪ More than half of the respondents in this study feel that their risk expectations and
sensitivity motivate their need to purchase motor vehicle insurance.
▪ The findings show that the way a consumer acts during the decision-making process
is affected by their perception of motor vehicle insurance and a certain short-term
insurance company.
▪ The learning/experiences gained from previous motor vehicle insurance purchases do
influence respondents’ future purchase decisions.
▪ The findings revealed that most respondents feel that their attitude towards motor
vehicle insurance does (to some extent) influence their need recognition for personal
▪ The findings confirmed that the attitude towards motor vehicle insurance plays a
slightly bigger role than the attitude towards a certain short-term insurance company in
the respondent’s decision-making process.
The decision-making process
The secondary objective regarding the decision-making process aimed to investigate

the steps of the decision-making process for the personal motor vehicle insurance
industry in South Africa. Table 8.5 presents the reflections on and conclusions drawn
for this secondary objective.
Objective
To investigate the steps of the decision-making process for the personal motor vehicle
insurance industry in South Africa.
▪ As discussed in Section 4.5, the ‘Pre-purchase’ step in the decision-making process

will typically be the step during which individuals think about previous experiences
(internal search), request a few quotations of policies from different companies, and
consider the price of each policy (external search).
▪ During the ‘Evaluation of alternatives’ step (see Section 4.5), individuals who have
obtained information about a financial service will use the criteria to compare the
different quotations, and measure if a product/policy will contribute to their need
satisfaction. The criteria consist of three aspects, namely, price (price of the policy on
quotation), quality (the standard of the quotation) and the brand list (list of different
companies).
343
The descriptive findings showed that:
▪ The descriptive findings revealed that both items under the Pre-purchase search step
(considering previous experiences and request quotations) are used by respondents.
o 56.9% of the respondents indicated that they use their previous experiences in the
pre-purchase search to a greater extent.
o 53.8% of the respondents indicated that they request quotations from more than one
short-term insurance company to a greater extent.
▪ The Evaluation of alternatives criterion most used is quality/standard, followed by
price, and the criterion least used is the brand list.
o 61.9% of the respondents indicated that they compare the quality of the different
quotations to measure if there is a policy that will contribute to their need satisfaction
and lifestyle to a greater extent.
o 61% of the respondents indicated that they compare the price of the different
quotations to measure if there is a policy that will contribute to their need satisfaction
and lifestyle to a greater extent.
o 43.7% of the respondents specified that they compare the brand list to measure if
there is a policy that will contribute to their need satisfaction and lifestyle to a greater
extent.
▪ The CFA indicated that the Pre-purchase search considering previous experiences’
item had a low loading and was not acceptable for inclusion in the CPDFPMVI model.
▪ According to the CFA, the Evaluate alternative prices variable has the greatest
influence on the purchase decision. The Evaluate alternative brands variable has the
lowest influence on the purchase decision.
▪ The logistic regression indicated that all four Decision Process steps would increase
the probability of purchasing personal motor vehicle insurance. Thus, respondents
who had personal motor vehicle insurance were more likely to report a higher extent of
use of the decision-making steps than those who did not have personal motor vehicle
insurance.
Conclusions
▪ This study’s findings confirm that individuals use their previous experiences to perform
a pre-purchase search for personal motor vehicle insurance as part of the decision-
making steps.
▪ Knowing that price is an external influence that influences the need to purchase
personal motor vehicle insurance, it is noteworthy to find that the criterion that
specifies that individuals consider the price of each policy has the greatest influence
on their decision while choosing between the alternative brands. Thus, the findings in
this study confirm the role price plays in selecting and choosing a short-term insurance
company’s motor vehicle insurance policy.
344
The demographic variables
This secondary objective aimed to assess the demographic variables that influence
the individual’s need to purchase personal motor vehicle insurance in South Africa.
Table 8.6 showcases the secondary and primary data, as well as the conclusions
applicable to this objective.
Objective
To assess the demographic variables that influence the individual’s need to purchase
personal motor vehicle insurance in South Africa.
▪ Organisations often segment markets based on demographics as this information is

widely available, and it can also relate to the individuals’ purchase decisions (refer to
Section 3.2.1.3 and 4.4.2).
▪ As discussed in Sections 3.2.1.3 and 4.4.2, the general demographic variables include
age, gender, income, household size, education, ethnic background and family
lifestyle.
A summary of the demographic profile of the respondents as per the descriptive statistics:
▪ The largest group of respondents (25.7%) fell within the age category of 31-40 years.
▪ The largest population group in this study was White, comprising 60% of the
respondents.
▪ 52.5% of the respondents were male, and the other 47.5% of the respondents were
female.
▪ 55.6% of the respondents were married, the single category made up 23.2% of the
respondents, 10.5% of the respondents were living with a partner but were not
married, 7.1% of the respondents were divorced, and 3.6% of respondents were
widowed.
▪ The three highest levels of education categories were: the Grade 12 category was
made up of 25.7% of the respondents, the Bachelor’s Degree category had 13.8% of
the respondents and the Diploma category comprised 13.6% of the respondents.
▪ Most respondents (57.8%) were permanently employed on a full-time basis.
▪ The three provinces in which respondents mostly travel with their personal motor
vehicles were Gauteng (43.4%), Western Cape (12.8%) and KwaZulu-Natal (10.9%).
▪ 75.8% of the respondents had a monthly household income of more than R6 000.
▪ 49.1% of the respondents earned more than R16 000 personal income per month.
The model fit analysis showed that:
▪ Age, the Highest education level, and Household income significantly contributed to
the CPDFPMVI model's predictive ability.
345
▪ All three of these demographic variables will increase the probability of purchasing
Conclusions
▪ These findings show that Age, the Highest level of education, and Household income
indeed relate to and influence the purchase decision of personal motor vehicle
insurance.
▪ The Household income and Highest level of education are also some of the biggest
drivers that influence purchasing personal motor vehicle insurance, according to the
CPDFPMVI model's predictive ability. It is interesting to note that household income
and not personal income is a driver that influences the purchase decision, as the
general questions revealed that most respondents pay for themselves. However,
family finances can be divided, or most families might feel that all the income per
month is seen as a collective income before the expenses are paid.
▪ The respondents that have personal motor vehicle insurance were more likely to
report higher education levels than those who did not have personal motor vehicle
insurance. The highest level of education could link up with insurance knowledge or
literacy, which relates to the financial literacy that is the key to making informed
purchase decisions.
▪ The respondents that have personal motor vehicle insurance were more likely to
report higher household incomes than those who did not have personal motor vehicle
insurance. Household income is the biggest driver that affects the purchase decision
of personal motor vehicle insurance, which corresponds with the importance of price
to consumers. Personal motor vehicle insurance is known to be a grudge purchase,
and it can play a role because if consumers do not have enough disposable income to
cover other monthly expenses, it will affect the purchase decision related to motor
vehicle insurance.
Demographic profile of personal motor vehicle insurance consumers
This secondary objective deals with the demographic profile of respondents who
purchase personal motor vehicle insurance in this study. Table 8.7 presents the
reflections on and conclusions drawn for this secondary objective.
Objective
To assess the consumer demographic profile within the South African personal motor
As discussed in Section 3.2.1.3, a general consumer profile can only be compiled after an
organisation has segmented its market and categorised all the homogenous groups. A
general consumer profile will allow the organisation to select a promising segment that
has the potential to be targeted.
346
The overall findings regarding the demographic profile of respondents that purchase
personal motor vehicle insurance can be used to develop a description of the average
personal motor vehicle insurance consumer in South Africa. The table below presents a
description of the average personal motor vehicle insurance consumer.
Average personal motor vehicle consumer

Construct
according to this study
Population group Likely to be a white consumer.
Gender Equal probability towards gender.
Marital status Likely to be married.
Highest level of education Likely to have a Post Higher Diploma (Masters or

Doctoral Diploma) or higher education level.
Employment status Likely to be permanently employed on a full-time

basis.
Province mostly travelled with Likely to travel most in Gauteng.

personal motor vehicle
Household income Likely to have a household income of more than

R16 001.
Personal income Likely to earn more than R16 001 personally.
Conclusions
Each short-term insurance company has a consumer profile that describes the
demographics of their current consumers. Therefore, the consumer profile of every short-
term insurance company will differ, but this description of the average personal motor
vehicle insurance consumer provides insight into the different characteristics of
consumers in South Africa that specifically purchase personal motor vehicle insurance.
Demographic profile of personal motor vehicle drivers
This secondary objective with regard to the demographic profile of respondents aimed
to compile a general profile that describes the average personal motor vehicle driver
that participated in this study. Table 8.8 presents the reflections on and conclusions
drawn for this secondary objective.
Objective
To compile a general demographic profile that describes the average personal motor
vehicle driver in South Africa.
347
Organisations, such as short-term insurance companies, should segment their market

according to demographics. This information guides them on how to target the
audience/market (Section 3.2.1.3). Section 2.7.2.1 highlighted demographic findings in
international and African studies. In South Africa, the Black African population group is the
population group that owns the most motor vehicles, and a higher percentage of male
drivers own a motor vehicle than females. The provinces with the highest number of
registered motor vehicles are: (1) Gauteng, (2) Western cape, (3) KwaZulu-Natal and (4)
Eastern Cape. No information could be obtained about the age, highest level of education,
marital status, employment status, personal income and household income of drivers in
South Africa.
The overall findings regarding the demographic profile of respondents (see a summary of
the demographic profile of respondents in Table 8.6 above) can be used to develop a
description of the average personal motor vehicle driver. The table below presents a
description of the average personal motor vehicle driver.
Construct Average personal motor vehicle driver

according to this study
Population group Likely to be a white driver.
Gender Likely to be a male driver.
Marital status Likely to be married.
Highest level of education Likely to have a Post-Matric Certificate

(Grade 12) or higher education level.
Employment status Likely to be permanently employed on a

full-time basis.
Province mostly travelled with personal Likely to travel most in Gauteng.

motor vehicle
Household income Likely to have a household income more

than R6 001.
Personal income Likely to earn more than R2 501

personally.
Conclusions
The findings of this study are not similar to the information available in the literature on
previous studies in countries such as the US, the UK, Ghana, Qatar and Denmark. These
demographic profile findings do not mirror the characteristics of South African motor
vehicle drivers, taking into consideration that this study’s respondents consisted of
personal motor vehicle drivers and not all motor vehicle drivers. However, this description
still provides insight into the different characteristics of the average personal motor vehicle
driver.
348
The consumer purchase decision for personal motor vehicle insurance
(CPDFPMVI) model
The primary objective of this study was to develop a model that explains the consumer
purchase decision process in the South African personal motor vehicle insurance
industry. To address and conclude this primary research objective, the final
CPDFPMVI model, as presented in Chapter 7, will be discussed. In this section,
specific attention will be given to the items that have been removed from the CFA
baseline model and logistic regression baseline model. The final CPDFPMVI model is
presented in Table 8.9.
Table 8.9: The CPDFPMVI model
CPDFPMVI model after item

According to the final
Constructs structure was confirmed with
CPDFPMVI model
the EFA
Company’s ▪ Type of policies offered ▪ Price of a policy

marketing efforts ▪ Price of a policy ▪ Process
▪ Distribution channels ▪ People
▪ Physical evidence
▪ Process
▪ People
▪ Partnerships
Sociocultural Positive sociocultural Positive sociocultural

environment environment environment
factors
▪ Positive family member ▪ Positive family member
▪ Positive reference person or ▪ Positive reference person or
group group
▪ Positive non-commercial ▪ Positive non-commercial
sources sources
▪ Positive social class ▪ Positive social class
▪ Cultural and subcultural ▪ Cultural and subcultural
factors factors
Negative sociocultural Negative sociocultural
environment environment
▪ Negative family member ▪ Negative family member
▪ Negative reference person or ▪ Negative reference person
group or group
▪ Negative non-commercial ▪ Negative non-commercial
sources sources
▪ Negative social class ▪ Negative social class
349
CPDFPMVI model after item
According to the final
Constructs structure was confirmed with
CPDFPMVI model
the EFA
Communication Communication sources A ▪ Buzz agent

sources ▪ Customised messages
▪ Traditional advertisements
▪ Online advertisements
▪ Promotion
Communication sources B
▪ Buzz agent
▪ Customised messages
▪ User-generated social media
posts
Psychological ▪ Motivation ▪ Motivation

attributes that ▪ Perception about motor ▪ Perception about motor
influence the need vehicle insurance vehicle insurance
recognition
▪ Perception about a certain ▪ Perception about a certain
short-term insurance short-term insurance
company company
▪ Learning ▪ Learning
▪ Personality ▪ Personality
▪ Attitude towards motor ▪ Attitude towards motor
vehicle insurance vehicle insurance
▪ Attitude towards a certain ▪ Attitude towards a certain
short-term insurance short-term insurance
company company
Decision-making ▪ Pre-purchase search ▪ Pre-purchase search

process considering previous requesting quotations
experiences ▪ Evaluate alternative prices
▪ Pre-purchase search ▪ Evaluate alternative qualities
requesting quotations
▪ Evaluate alternative prices
▪ Evaluate alternative qualities
Demographic ▪ Age
variables ▪ Highest level of education
The items in the final CPDFPMVI model are tabulated in the last column of Table 8.9
on the right, and are illustrated in Figure 8.1 below.
350
Figure 8.1: Final CPDFPMVI model
351
The findings from the data collected in this study indicated that the following items in
Table 8.10 did not contribute to the final CPDFPMVI model.
Table 8.10: Items that were discarded from the final CPDFPMVI model
Constructs Items removed
Company’s marketing efforts ▪ Type of policies offered

▪ Distribution channels
▪ Physical evidence
▪ Partnerships
Sociocultural environment factors No items were discarded
Communication sources Communication sources A

▪ Traditional advertisements
▪ Online advertisements
▪ Promotion
Communication sources B
▪ User-generated social media posts
Psychological attributes that influence No items were discarded

the need recognition
Decision-making process ▪ -Pre-purchase search considering

previous experiences
Demographic variables ▪ Marital status

▪ Employment status
It is recommended that further research be conducted to explore if these items would

contribute to the purchase decision process if another population and/or other
financial service is used.
The next section addresses the main research question.
ADDRESSING THE RESEARCH QUESTION
The research question was raised in Chapter 1. After this study's primary data
collection and analysis phases, the researcher could address the main research
question successfully. The final CPDFPMVI model was accepted by following a unique
model-building strategy (explained in Section 7.1) in the model fit analysis. This final
352
CPDFPMVI model points out all the factors that influence the individuals’ purchase
decisions in the South African personal motor vehicle insurance industry. Table 8.9
and Figure 8.1 above indicate all the factors (internal and external) which influence the
individuals’ purchase decisions in the South African personal motor vehicle insurance
industry. According to the final CPDFPMVI model, the items under the Company’s
marketing efforts, Sociocultural environment factors, Communication sources,
Psychological attributes and Demographic variables influence the need recognition,
and are all factors that influence the individuals’ purchase decision in the South African
personal motor vehicle insurance industry (see Table 8.9, according to the final
CPDFPMVI model column and presented in Figure 8.1).
As explained in the literature review, there are distinct differences between the internal
and external factors. The external factors that influence the purchase decision are the
Company’s marketing efforts, Sociocultural environment factors, and Communication
sources. The internal factors are Psychological attributes and Demographic variables.
The following section deliberates on this study's contribution to the body of knowledge.
CONTRIBUTION TO THE BODY OF KNOWLEDGE
This study has made a modest contribution to the knowledge exploring the evolving
issue of individuals’ purchase behaviour in insurance. In the present case, only the
perceptions and opinions of motor vehicle drivers in South Africa were used to develop
a CPDFPMVI model. However, during this research process, the study achieved some
additional contributions by:
• Closing the gap in knowledge about motor vehicle insurance consumers in South
Africa, specifically the personal motor vehicle insurance consumers. Thus far, very
little literature has been available on South African motor vehicle drivers.
Consequently, the researcher could find no literature that specifically focuses on
the personal motor vehicle insurance consumers and potential consumers in South
Africa. Therefore, in this study, an attempt was made to describe the average
personal motor vehicle driver by analysing and interpreting the demographic data
collected.
353
• Creating new understandings of the existing individual/internal factors that
influence the individual’s purchase decision of personal motor vehicle insurance,
specifically.
• Merging the previously established marketing concepts from the literature on

consumer behaviour in such a way that the purchase decision aspects can be
viewed from the individual’s perspective instead of from an organisational point of
view.
• Merging the field of business management with specific reference to consumer

behaviour and marketing by developing an acceptable CPDFPMVI model that
explains the purchase decision process of personal motor vehicle insurance
consumers in South Africa.
• Demonstrating originality by focusing on the personal motor vehicle insurance

industry. The reason why a consumer purchases personal motor vehicle insurance
is different from the reasons why consumers purchase motor vehicle insurance for
business purposes. Most of the current purchase behaviour research and literature
in insurance is based on medical and life insurance. While a few studies have been
conducted on purchase behaviour for motor vehicle insurance, none are based in
South Africa. Taking into consideration that motor vehicle insurance in South Africa
is not compulsory as it is in most other countries, the individuals’ need for personal
motor vehicle insurance is determined by internal and external factors, as found in
this study.
• Field-testing the newly developed CPDFPMVI conceptual model. Therefore, a

unique consumer purchase decision model for the personal motor vehicle
insurance industry contributes to the South African insurance knowledge.
• Formulating and using a unique model-building strategy as the methodology in the

model fit analysis of this study to generate the accepted CPDFPMVI model can
also be regarded as a contribution.
The following section briefly summarises the recommendations made to the short-term
insurance companies in South Africa.
354
RECOMMENDATIONS FOR COMPANIES IN THE SHORT-TERM
INSURANCE INDUSTRY
It is important for short-term insurance companies, specifically, to have insight into and
understanding of the purchase behaviour amongst consumers and potential
consumers in terms of personal motor vehicle insurance within South Africa. However,
academics can also benefit from this understanding. Short-term insurance companies
are the policymakers that offer and market motor vehicle insurance products to
consumers and potential consumers. Therefore, these companies should consider the
main recommendations as summarised in the paragraphs below.
Short-term insurance companies can benefit from trained and skilled employees who
treat consumers acceptably and according to their satisfaction when a policy is sold
or when a claim is lodged. Furthermore, a degree of automation during the quotation
process will be highly productive in offering individuals in the market what they want,
when they want it. Also, by incorporating pricing strategies that will still draw
consumers even during these difficult financial times, can enable short-term insurance
companies to be proactive in adapting to the changing market environment. In
addition, cultural and subcultural factors are part of the market environment of short-
term insurance companies. They cannot control it, but it is important to spot the cultural
shifts to offer policies that individuals in the market want. Moreover, short-term
insurance companies will have to investigate whether the shift to digital marketing will
be beneficial in terms of advertising motor vehicle insurance, and the use of buzz
agents can also be investigated.
Strategies can be developed to convince more individuals to purchase motor vehicle

insurance and to change their perception regarding a certain short-term insurance
company. Likewise, short-term insurance companies must ensure that consumers are
satisfied with the purchase from these companies, as this might influence their future
purchase decisions with that company. Additionally, short-term insurance companies
can consider personalities as useful when analysing consumer behaviour in terms of
certain policies or brand choices, and positive attitudes regarding motor vehicle
insurance and the specific short-term insurance company should be formed.
Short-term insurance companies must ensure that the quotation process is

straightforward, informative and flawless so that individuals in the market can obtain
355
all the necessary information before the alternatives are evaluated. This is especially
important as the information gained from the quotations can be used to perform a pre-
purchase search for more products at a later stage. Furthermore, short-term insurance
companies should take the necessary steps to ensure that the most competitive prices
are presented to the market. They should also ensure that good quality and high
standard quotations providing sufficient information are offered, and that their brand
can be distinguished from their competitors.
Considering that this study revealed that Age, Highest level of education, and
Household income influence the purchase decision of individuals, short-term
insurance companies must ensure that the market is segmented accordingly to fit the
different motor vehicle insurance product offerings. Lastly, the demographic
segmentation information available in each short-term insurance company can be
used in marketing efforts to target the correct audience.
The following section presents the recommendations for future research.
RECOMMENDATIONS FOR FUTURE RESEARCH
The empirical findings of this study of the consumer purchase behaviour in the
personal motor vehicle insurance industry of South Africa have introduced a new set
of questions. As expected, new information comes to light as the research process
unfolds, or researchers learn something new about the research topic. By answering
this study's research question and objectives, potential research areas for future
research were identified.
• This study was conducted at the beginning of the COVID-19 pandemic in South
Africa. Therefore, a recommendation is that further research be conducted to
determine what influence a cheaper policy price will have on personal motor
vehicle consumers now that the COVID-19 pandemic has impacted the disposable
income of consumers across the country.
• The brand awareness aspect of the marking efforts can be further researched to
establish if short-term insurance companies' marketing efforts play a role when
individuals in the market recall brands when searching for information during the
356
• Further research can be conducted to determine what individuals in the market
would like to see/hear in communication sources, like buzz agents, customised
messages and paid-for social media efforts to increase the probability of
purchasing personal motor vehicle insurance.
• Further research can be conducted on how the learning attribute contributes to

future purchase decisions for insurance products.
• Further research can be conducted in the purchasing behaviour of another type of

insurance such as medical or household insurance. For example, to establish if
previous experiences in the Pre-purchase Search step influence consumers’
purchase decisions, and to determine if age, highest level of education, and
household income will also influence individuals’ purchase decisions.
• It is recommended that further research be conducted to explore if the items

discarded from the final CPDFPMVI model would contribute to the purchase
decision process if another population and/or other financial service is used.
• The conceptual CPDFPMVI model presented in this study can be tested on

different insurance products, and the possible differences can be investigated. In
other words, future research can apply this study to other types of insurance.
• The post-purchase behaviour step in the purchase decision-making process can

be investigated, since this was not covered in this research.
• Lastly, the theory of planned behaviour can be investigated to determine

individuals' purchase intention towards motor vehicle insurance for personal use.
The limitations of this study are presented in the following section.
LIMITATIONS OF THE STUDY
This research was conducted after considering the quality requirements and
thoroughness of the research design and methodologies. However, the following
limitations need to be considered along with the findings.
• The low response rate can be regarded as a limitation. A larger sample would have
been preferred, considering that the live motor vehicle population in South Africa
is in the region of twelve to thirteen million. Some technical difficulties were
experienced with the distribution of the survey, which the researcher had no control
357
over. The survey was first distributed to the AA database, and the researcher could
only obtain the number of respondents much later than what was planned for at
the beginning. The number of respondents who completed the survey from the AA
database was not the number initially anticipated. Therefore, the researcher had
to think of innovative ways on how to invite more individuals to the study. Hence,
the extended period to complete the data-collection phase. The researcher used
more than one sampling method (convenience and snowball sampling) via
Facebook to increase the response rate, ensuring that the largest sample possible
was invited to partake in the study. After all these attempts to invite respondents
on Facebook, there was still a very low response rate. Then the researcher decided
to distribute hard copy questionnaires in the different regions of the country by
means of the snowball sampling method. A large number of hard copy
questionnaires were not returned or could not be used due to incompleteness. For
this reason, the study only reached a response rate of 678 responses that could
be successfully used.
• Generalisability could be limited due to the sampling size and characteristics of the
research sample, as well as the sampling methods used. Though the sample
comprised of n=678, it was sufficient to perform statistical analysis that could lead
to significant conclusions. However, a larger response rate would have guaranteed
findings that could confidently be generalised. Furthermore, the characteristics of
the research sample do not mirror the characteristics of drivers in South Africa.
Finally, the sampling methods used in this study were not the researcher’s first
choice. The lack of complete contact details on, and accessibility constraints to the
Road Traffic Management Cooperation (RTMC) database forced the researcher to
follow non-probability sampling methods. Nevertheless, the findings still provide
valuable new insight for future replication studies.
• The findings are based on data from one industry, namely, the personal motor
vehicle insurance industry, which is only a part of the larger motor vehicle
insurance industry of South Africa. The statistics currently available regarding the
market penetration of motor vehicle insurance include all three motor vehicle
insurance lines (personal, commercial and corporate business). As this study
focused on the personal motor vehicle insurance line, the findings cannot be
compared to the current statistics available.
358
These limitations do not prevent the study from contributing to the body of knowledge
on the individuals’ purchase behaviour in the personal motor vehicle insurance
industry of South Africa. However, they do limit the extent to which the study can
contribute. The limitations of this study may serve as future research direction for other
research studies.
RESEARCH CONCLUSION
It is complex and challenging to gain an understanding of the market individuals’

purchase decision process, as there are internal and external factors that could
influence the purchase decision. This study aimed to determine the purchasing
decision process of consumers and potential consumers regarding personal motor
vehicle insurance in South Africa.
The motivation behind choosing this research topic was two-fold. Firstly, to develop a
model that explains the consumer purchase decision process within the personal
motor vehicle insurance industry of South Africa. Secondly, to contribute to practical
relevant research by determining the factors that influence the individual’s purchase
decision in the South African personal motor vehicle insurance industry.
The research set out to expand the body of knowledge in terms of purchase behaviour,
and made a contribution to the personal motor vehicle insurance industry of South
Africa. A quantitative research approach was followed to gain an understanding of the
individual’s purchase decision process for personal motor vehicle insurance. This
research confirmed the factors that influence the purchase decision of individuals in
the South African personal motor vehicle insurance industry. This thesis has provided
recommendations for short-term insurance companies in South Africa and future
studies.
359
LIST OF REFERENCES
Acevedo, C. 2018. Consumer behaviour and advertising management. London: ED-Tech Press.
Adamova, M., Boudet, J., Kalaoui, H. & Segev, I. 2018. Personalization is the future of insurance
marketing. Available: https://www.mckinsey.com/industries/financial-services/our-insights/
how-traditional-insurance-carriers-can-disrupt-through-personalized-marketing# [Accessed
2021, March 24].
Adams, K.A. & Lawrence, E.K. 2019. Research methods, statistics, and applications. 2nd ed.
Thousand Oaks: SAGE Publications.
Adams, J., Khan, H.T.A. & Raeside, R. 2014. Research methods for business and social science
students. 2nd ed. New Delhi: SAGE Publications.
Advani, V. 2021. What is Artificial Intelligence? How does AI work, applications and future? [Blog]
Great Learning, 19 October 2021. Available: https://www.mygreatlearning.com/blog/what-is-
artificial-intelligence/ [Accessed 2021, October 24].
Africa Check. 2017. Factsheet: South Africa’s crime statistics for 2016/17. Available:
https://africacheck.org/factsheets/south-africas-crime-statistics-201617/ [Accessed 2021,
November 05].
Africa Check. 2021. South Africa’s crime statistics for 2019/20. Available: https://africacheck.org/
fact-checks/factsheets/factsheet-south-africas-crime-statistics-201920 [Accessed 2021,
September 08].
Agarwal, M. 2020. Importance of user generated content as a part of social media marketing that
drives customer’s brand awareness and purchase intentions. The International Journal of
Analytical and Experimental Modal Analysis, 7(2): 3071–3089.
Aghazadeh, H. 2016. Principles of marketology, volume 1: Theory. England: Palgrave Macmillan.
Aghazadeh, H. 2017. Principles of marketology, volume 2. Tehran: Palgrave Macmillan.
Agyei, J., Sun, S., Abrokwah, E., Penney, E.K. & Ofori-Boafo, R. 2020. Influence of trust on
customer engagement: Empirical evidence from the insurance industry in Ghana. SAGE
Open, 10(1): 21582440198.
Ahmed, A. 2019a. Importance of mission vision in organizational strategy. Available:

http://smallbusiness.chron.com/importance-mission-vision-organizational-strategy-
16000.html [Accessed 2021, October 07].
Ahmed, A. 2019b. Benefit of studying consumer behavior. Available: https://bizfluent.com/facts-

5949659-benefit-studying-consumer-behaviour.html [Accessed 2021, October 20].
360
Ajzen, I. 1985. From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann
(Eds.). Action control: From cognition to behavior, pp. 11–39. Berlin, Heidelberg: Springer.
Ajzen, I. 2015. Consumer attitudes and behavior: The theory of planned behavior applied to food
consumption decisions. Rivista di Economia Agraria, (2): 121–138.
Alatinga, K.A. & Williams, J.J. 2016. Mixed Methods Research for Health Policy Development in
Africa: The Case of Identifying Very Poor Households for Health Insurance Premium
Exemptions in Ghana. Journal of Mixed Methods Research, 13(1): 1–16.
Aldrich, J.O. 2019. Using IBM SPSS statistics: An interactive hands-on approach. 3rd ed.
Thousand Oaks: Sage.
Alfreds, D. 2016. SA business “unprepared” for cybercrime. Fin24. 9 June. Available:

http://www.fin24.com/Tech/Cyber-Security/sa-business-unprepared-for-cybercrime-
20160609 [Accessed 2021, October 15].
De Allegri, M., Sanon, M., Bridges, J. & Sauerborn, R. 2006. Understanding consumers’
preferences and decision to enrol in community-based health insurance in rural West Africa.
Health Policy, 76: 58–71.
Allen, M. 2017. The SAGE encyclopedia of communication research methods. Thousand Oaks:
SAGE Publications.
American Marketing Association. 2021. Definition of marketing. Available:

https://www.ama.org/AboutAMA/Pages/Definition-of-Marketing.aspx [Accessed 2021,
October 17].
Amica. 2021. Company facts and history. Available: https://www.amica.com/en/about-

us/company-facts-and-history.html [Accessed 2021, October 01].
Andreasen, A.R. 1965. Attitudes and consumer behaviour: A decision model. In L.E. Preston,
(Ed.). New research in marketing. California: University of California.
Ang, S.H. 2021. Research projects for business & management students. 2nd ed. London: SAGE
Publications.
Aon Risk Solutions. 2014. Autonomous vehicle: The risks and rewards of the future of personal
transportation. Available: https://www.aon.com/attachments/risk-services/Automotive-
Practice-Autonomous-Vehicles.pdf [Accessed 2021, October 15].
Appel, G., Grewal, L., Hadi, R. & Stephen, A.T. 2020. The future of social media in marketing.
Journal of the Academy of Marketing Science, 48:79–95.
361
Araujo, M. & Cheng, M. 2020. Do you need an all-risk insurance policy? Available:
https://www.thebalance.com/what-is-all-risk-insurance-policy-4155325 [Accessed 2021,
August 16].
Ardé, A. 2016. Does telematics benefit drivers? IOL, 17 February 2016. Available:
http://www.iol.co.za/personal-finance/does-telematics-benefit-drivers-1985693 [Accessed
2021, October 13].
Ardé, A. 2018. You don’t really need extra funeral cover. Sowetan Live, 24 May 2018. Available:
https://www.sowetanlive.co.za/business/money/2018-05-24-you-dont-really-need-extra-
funeral-cover/ [Accessed 2021, July 23].
Ardé, A. 2020. Third-party car insurance will benefit all. Sowetan Live, 5 March 2020. Available:
https://www.sowetanlive.co.za/business/money/2020-03-05-third-party-car-insurance-will-
benefit-all/ [Accessed 2021, October 07].
Armstrong, G., Adam, S., Denize, S. & Kotler, P. 2015. Principles of marketing. 6th ed. Melbourne:
Pearson Australia.
Arrive Alive. 2009. What you should know about car & vehicle insurance. Moneyweb, 19 March
2009. Available: https://www.moneyweb.co.za/archive/what-you-should-know-about-car-
vehicle-insurance/ [Accessed 2017, April 20].
Arrive Alive. 2021a. Speed as a contributory factor to road traffic crashes. Available:
https://www.arrivealive.co.za/Speed-Factor [Accessed 2021, October 01].
Arrive Alive. 2021b. Car insurance and road safety. Available: https://www.arrivealive.co.za/Car-
Insurance-and-Road-Safety [Accessed 2021, October 04].
Arrive Alive. 2021c. An overview of road safety issues globally in South Africa and in the Western
Cape. Available: https://www.arrivealive.co.za/An-Overview-of-Road-Safety-Issues-Globally-
in-South-Africa-and-in-The-Western-Cape [Accessed 2021, October 01].
Arunangshu, G., Pradip, P. & Satakshi, C. 2021. Retail management: Text and cases. Delhi: PHI
Learning.
Atadil, H.A., Sirakaya-Turk, E., Meng, F. & Decrop, A. 2018. Exploring travelers’ decision-making
styles. International Journal of Contemporary Hospitality Management, 30(1): 618–636.
Atombo, C., Wu, C., Tettehfio, E.O. & Agbo, A.A. 2017. Personality, socioeconomic status,
attitude, intention and risky driving behavior. Cogent Psychology, 4(1): 1–20.
Auto & General Insurance. 2015. The top 5 reasons people don’t buy car insurance. [Blog] Auto
& General, 7 October 2015. Available: https://www.autogen.co.za/blog/motor/the-top-5-
reasons-people-dont-buy-car-insurance/ [Accessed 2021, February 02].
362
Auto Accident. 2019. Old drivers vs young drivers - Who is the worst driver? Available:
https://www.pmtxlaw.com/old-drivers-vs-young-drivers/ [Accessed 2021, November 13].
Automobile Association of South Africa. 2020. Don’t take the insurance bait - cheaper is not
always better. Available: https://www.aa.co.za/insights/dont-take-the-insurance-bait-cheaper-
is-not-always-better [Accessed 2021, October 04].
Awang, Z. 2012. Research methodology and data analysis. 2nd ed. Selangor: UiTM Press.
Babbie, E. 2016. The practice of social research. 14th ed. Boston: Cengage Learning.
Babbie, E.R. 2021. The practice of social research. 15th ed. Boston: Cengage Learning.
Babin, B.J. & Zikmund, W.G. 2016a. Exploring marketing research. 11th ed. Boston: Cengage
Learning.
Babin, B.J. & Zikmund, W.G. 2016b. Essentials of marketing research. 6th ed. Boston: Cengage
Learning.
Babin, B.J., Carr, J.C., Griffin, M. & Quinlan, C. 2015. Business research methods. Andover:
Cengage Learning.
Bae, T., Kim, C. & Kulperger, R.J. 2009. Securitization of motor insurance loss rate risks.
Insurance: Mathematics and Economics. 44(1): 48–58.
Bagus, U., de Girancourt, F.J., Mahmood, R. & Manji, Q. 2020. How insurance in Africa can find
success. Available: https://www.mckinsey.com/featured-insights/middle-east-and-africa/
africas-insurance-market-is-set-for-takeoff [Accessed 2021, July 23].
Baines, P., Fill, C. & Rosengren, S. 2017. Marketing. 4th ed. New York: Oxford University Press.
Baines, P., Whitehouse, S., Antonetti, P. & Rosengren, S. 2021. Fundamentals of marketing. 2nd
ed. New York: Oxford University Press.
Baione, F., Biancalana, D. & De Angelis, P. 2020. A quantile regression approach for the analysis
of the diversification in non-life premium risk. Soft Computing, 24: 8523–8534.
Baker, M.J. 2014. Marketing strategy and management. 5th ed. London: Palgrave Macmillan.
Baker, M.J. & Saren, M. 2016. Marketing theory: A student text. 3rd ed. London: SAGE
Publications.
Balasubramanian, R., Libarikian, A. & McElhaney, D. 2021. Insurance 2030-The impact of AI on

the future of insurance. Available: https://www.mckinsey.com/industries/financial-
services/our-insights/insurance-2030-the-impact-of-ai-on-the-future-of-insurance [Accessed
2021, October 13].
363
Bandalos, D.L. & Finney, S.J. 2018. Factor Analysis: Exploratory and Confirmatory. In G.R.
Hancock, L.M. Stapleton & R.O. Mueller (Eds.). The reviewer’s guide to quantitative methods
in the social sciences, pp. 98–122. 2nd ed. New York: Routledge.
Bandara, W.W.M.C. 2014. Consumer decision-making styles and local brand biasness:
Exploration in the Czech Republic. Journal of Competitiveness, 6(1): 3–17.
Barendse, R. 2015. Outsurance judgment - insurance laws should ultimately be aligned with the
Consumer Protection Act. Available: https://www.cliffedekkerhofmeyr.com/en/news/press-
releases/2015/dispute-resolution/outsurance-judgment-insurance-laws-should-ultimately-be-
aligned-with-the-consumer-protection-act-.html [Accessed 2021, October 16].
Barker, S. 2020. How social media influencers can impact your customers’ purchase decisions.
Available: https://www.socialmediatoday.com/news/how-social-media-influencers-can-
impact-your-customers-purchase-decisions/569813/ [Accessed 2021, March 24].
Barry, A.E., Chaney, E.H., Stellefson, M.L. & Chaney, J.D. 2011. So you want to develop a survey:
Practical recommendations for scale development. American Journal of Health Studies, 26(2):
97–105.
Barysevich, A. 2020. How social media influence 71% consumer buying decisions. Available:
https://www.searchenginewatch.com/2020/11/20/how-social-media-influence-71-consumer-
buying-decisions/ [Accessed 2021, October 27].
Batra, S.K. & Kazmi, S.H.H. 2008. Consumer behaviour: Text and cases. 2nd ed. New Delhi:
Excel Books India.
Beachey, A. 2020. How driverless car insurance is changing the sector. Available:
https://www.raconteur.net/finance/insurance/driverless-car-insurance-impact/ [Accessed
2021, October 15].
Beer, M., Finnström, M. & Schrader, D. 2016. Why leadership training fails—and what to do about
it. Available: https://hbr.org/2016/10/why-leadership-training-fails-and-what-to-do-about-it
[Accessed 2021, October 12].
De Beer, A. & Rossouw, D. 2018. Focus on operational management: A generic approach. 4th
ed. Cape Town: Juta.
Beik, J.I. 2018. Health insurance today: A practical approach. 6th ed. Missouri: Elsevier.
Belch, G.E. & Belch, M.A. 2015. Advertising and promotion: An integrated marketing
communications perspective. 10th ed. New York: McGraw-Hill.
Bell, E., Bryman, A. & Harley, B. 2019. Business research methods. 5th ed. New York: Oxford
University Press.
364
Bellis, M. 2018. Biography of Karl Benz. Available: https://www.thoughtco.com/karl-benz-and-
automobile-4077066 [Accessed 2021, October 01].
Berger-Grabner, D. 2021. Strategic retail management and brand management: Trends, tactics,
and examples. Boston: Walter de Gruyter GmbH.
Berkowitz, E.N. 2017. Essentials of health care marketing. 4th ed. Burlington: Jones & Bartlett
Publishers.
Bernoulli, D. 1954. Exposition of a new theory on the measurement of risk. Econometrica. 22(1):
23–36.
Bernstein, J.S. 2014. Standing room only: Marketing insights for engaging performing arts
audiences. 2nd ed. New York: Palgrave Macmillan.
Betz, S. & Payne, C. 2021. How insurers can transform by adopting public cloud. Available:
https://www.ey.com/en_gl/financial-services-emeia/how-insurers-can-transform-by-adopting-
public-cloud [Accessed 2021, October 14].
Bhati, N.S. 2019. Validation of customers’ perceived determinants: A study of online shoppers e-
service quality. International Conference on Sustainable Computing in Science, Technology
& Management: 1782–1787.
Biemer, P.P., De Leeuw, E.D., Eckman, S., Edwards, B., Kreuter, F., Lyberg, L.E., Tucker, C.N.
& West, B.T. 2017. Total survey error in practice. New Jersey: John Wiley & Sons.
Bilgin, Y. 2018. The effect of social media marketing activities on brand awareness, brand image
and brand loyalty. Business & Management Studies: An International Journal, 6(1): 128–148.
Binder, S., Klais, P. & Mußhoff, J. 2021. Global insurance pools statistics and trends: An overview
of life, P&C, and health insurance. Available: https://www.mckinsey.com/industries/financial-
services/our-insights/global-insurance-pools-statistics-and-trends-an-overview-of-life-p-and-
c-and-health-insurance [Accessed 2021, June 24].
Bolhaar, J., Lindeboom, M. & Van der Klaauw, B. 2015. Insurance search, switching behaviour
and the role of group contracts. De Economist, 163:25–60.
Bolia, B., Jha, S. & Jha, M.K. 2016. Cognitive dissonance: A study of post purchase behavior of
consumers in the context of financial products. International Journal of Innovative Research &
Development, 5(3): 148–153.
Bonorchis, R. 2014. South African insurers urge mandatory auto insurance. Insurance Journal,
16 October 2014. Available: http://www.insurancejournal.com/news/international/2014/10/
16/343336.htm [Accessed 2016, February 04].
365
Booms, B.H. & Bitner, M.J. 1981. Marketing strategies and organisation structures for service
firms. In Marketing of Services, American Marketing Association. Chicago, pp. 47–51.
Boone, L.E. & Kurtz, D.L. 2014. Essentials of contemporary business. New York: John Wiley &
Sons.
Boone, L.E. & Kurtz, D.L. 2016. Contemporary marketing. 17th ed. Boston: Cengage Learning.
Borda, M. & Jędrzychowska, A. 2012. Attitudes and decisions of the motor insurance buyers in
Poland. Business and Economic Horizons, 8:23–32.
Bornstein, J., Rakow, J. & Clemente, D. 2017. Development of driverless vehicles in the United
Kingdom. Deloitte University Press.
Borscheid, P. & Haueter, N.V. 2012. World insurance: The evolution of a global risk network.
Oxford: Oxford University Press.
Bose, D.C. 2012. Principles of management and administration. 2nd ed. Delhi: PHI Learning
Private Limited.
Botha, T. 2018. Entrepreneurship and how to establish your own business. 6th ed. Cape Town:
Juta.
Bracher, P. 2017. Social media and insurance. Social Media Law Bulletin, 18 May 2017. Available:
http://www.socialmedialawbulletin.com/2017/05/social-media-insurance/ [Accessed 2021,
October 14].
Bradley, N. 2013. Marketing research: Tools and techniques. 3rd ed. Oxford: Oxford University
Press.
Bray, J.P. 2008. Consumer behaviour theory: Approaches and models. Unpublished discussion
paper. Bournemouth University, Bournemouth, UK. Available: http://bit.ly/162wa5o [Accessed
2021, August 30].
Briggs, R.A. 2014. Normative theories of rational choice: Expected utility. In Stanford
Encyclopedia of Philosophy. Available: https://plato.stanford.edu/archives/spr2017/entries/
rationality-normative-utility/ [Accessed 2021, October 21].
Brinberg, D. & Lutz, R.J. 1986. Perspectives on methodology in consumer research. New York:
Springer-Verlag New York.
Britton, N. & Van der Vyver, E. 2021. South Africa: Publication of proposed amendments to the
policyholder protection rules. Clyde & Co. Available: https://www.clydeco.com/
en/insights/2021/08/south-africa-publication-of-proposed-amendments-to [Accessed 2021,
November 14].
366
Brown, M. 2017. London’s first motor crashes. Available: https://londonist.com/2015/06/londons-
first-motor-crashes [Accessed 2021, October 01].
Brown, T.A. 2015. Confirmatory factor analysis for applied research. 2nd ed. New York: The
Guilford Press.
Bryman, A. 2016. Social research methods. 5th ed. New York: Oxford University Press.
Bryman, A. & Bell, E. 2015. Business research methods. 4th ed. New York: Oxford University
Press.
Bryman, A., Bell, E., Hirschsohn, P., Dos Santos, A., Du Toit, J., Masenge, A., Van Aardt, I. &
Wagner, C. 2014. Research methodology: Business and management contexts. 5th ed. Cape
Town: Oxford University Press.
Bubear, R. 2016. Here’s how many cars on SA roads are uninsured. CAR magazine, 21 June
2016. Available: http://www.carmag.co.za/news/how-many-cars-on-sa-roads-are-uninsured-
this-many/ [Accessed 2018, October 05].
Buchholz, K. 2020. Americans get driver’s licenses later in life. Available:

https://www.statista.com/chart/18682/percentage-of-the-us-population-holding-a-drivers-
license-by-age-group/ [Accessed 2021, October 12].
Burghardt, J. & Bodansky, A.N. 2021. Why Psychology Needs to Stop Striving for Novelty and
How to Move Towards Theory-Driven Research. Frontiers in Psychology, 12: 609802. DOI:
10.3389/FPSYG.2021.609802.
Burrow, J.L. & Fowler, A.R. 2016. Marketing. 4th ed. Boston: Cengage Learning.
BUSINESSTECH. 2019. Here’s how car insurance excess works. Available:

https://businesstech.co.za/news/motoring/337285/heres-how-car-insurance-excess-works/
Buthelezi, L. 2020. It’s part of our business, says Santam as claims soar. News24, 5 March 2020.
Available: https://www.news24.com/fin24/companies/financial-services/its-part-of-our-
business-says-santam-as-claims-soar-20200305 [Accessed 2021, July 23].
Buthelezi, L. 2021. Evolving risks leave insurance customers exposed as reinsurers reconsider
cover – Santam research. News24, 9 September 2021. Available: https://www.news24.com/
fin24/companies/evolving-risks-leave-insurance-customers-exposed-as-reinsurers-
reconsider-cover-santam-research-20210908 [Accessed 2021, October 16].
Buys, D. 2015. Sustainability of motor insurance. Available: https://www.cover.co.za/iisa-

overview-of-motor-insurance-dawie-buys/ [Accessed 2016, August 30].
Cahill, D.J. 2006. Lifestyle market segmentation. New York: Haworth Press.
367
Cambridge Dictionary. 2021. Meaning of legal person in the English. Available:
http://dictionary.cambridge.org/dictionary/english/legal-person [Accessed 2021, November
01].
Campbell, C.J., Goldberg, L. & Rai, A. 2003. The impact of the European Union insurance
directives on insurance company stocks. The Journal of Risk and Insurance, 70(1): 125–167.
Canchola, J.A., Otero-Sabogal, R., Owens, D. & Tabnak, F. n.d. Logistic regression model-
building strategies for predicting regular mammography screening adherence among
uninsured immigrant women. Available: https://www.lexjansen.com/wuss/
2003/DataAnalysis/c-logistic_regression_model-building_strategies.pdf [Accessed 2021, July
14].
Canorea, E. 2021. Technology trends that will shape the insurance industry. Available:
https://www.plainconcepts.com/insurtech-trends-insurance-industry/ [Accessed 2021,
October 14].
Cant, M. 2020. Essentials of marketing. 6th ed. Cape Town: Juta.
Cape Business News. 2017. On the financial wish list for 2017. Cape Business News, 13 January
2017. Available: https://www.cbn.co.za/component/k2/on-the-financial-wish-list-for-2017
Carter, S.M. 2018. Social media may make you overspend and it’s not just because of ads.
Available: https://www.cnbc.com/2018/03/15/social-media-may-make-you-overspend-and-
its-not-just-because-of-ads.html [Accessed 2021, October 27].
Casquilho, J.P. 2015. Combining expected utility and weighted Gini-Simpson index into a non-
expected utility device. Theoretical Economics Letters, 05(02): 185–195.
Chaffey, D. & Smith, P.R. 2017. Digital marketing excellence: Planning, optimizing and integrating
online marketing. 5th ed. New York: Taylor & Francis.
Chakar, M. 2020. Top risks for the global insurance industry: Could COVID-19’s second wave
shake up the sector? Available: https://www.spglobal.com/marketintelligence/en/
documents/top-risks-for-the-global-insurance-industry.pdf [Accessed 2021, October 16].
Chapman, H. 2016. Insurance industry perspectives: 2016 Review. Available:

https://www.hoganlovells.com/~/media/hogan-lovells/pdf/publication/2016/11383_c1_global-
ins-review_bro_e.pdf?la=en [Accessed 2017, May 31].
Chebat, J.-C. & Oumlil, A. Ben. 2015. Proceedings of the 1998 Multicultural Marketing
Conference. New York: Springer International Publishing.
368
Chen, J. 2021. Expected utility definition. Available: https://www.investopedia.com/
terms/e/expectedutility.asp [Accessed 2021, October 22].
Cherry, K. 2020a. Becoming an informed consumer of psychology. Available:

https://www.verywellmind.com/becoming-a-consumer-of-psychology-2795611 [Accessed
2021, October 13].
Cherry, K. 2020b. What Is cognitive dissonance? Available: https://www.verywell.com/what-is-

cognitive-dissonance-2795012 [Accessed 2021, October 21].
Cherunilam, F. 2016. Business environment text and cases. 25th ed. New Delhi: Himalaya
Publishing House.
Chilisa, B. 2019. Indigenous research methodologies. Thousand Oaks: SAGE Publications.
Chitiga-Mabugu, M., Henseler, M., Mabugu, R. & Maisonnave, H. 2021. Economic and
distributional impact of COVID-19: Evidence from macro-micro modelling of the South African
economy. South African Journal of Economics, 89(1): 82–94.
Choudhary, A. 2020. The impact of AI on the insurance industry in the short and long term. [Blog]
Finextra, 17 August 2020. Available: https://www.finextra.com/blogposting/19215/the-impact-
of-ai-on-the-insurance-industry-in-the-short-and-long-term [Accessed 2021, October 14].
Chutel, L. 2016. South Africa doesn’t have enough developers or coders to build a digital
economy. Available: https://qz.com/africa/714662/south-africa-doesnt-have-enough-
developers-to-build-a-digital-economy/ [Accessed 2021, October 13].
Civelek, M.E. 2018. Essentials of structural equation modeling. University of Nebraska-Lincoln

Libraries.
Clark, T., Foster, L. & Bryman, A. 2019. How to do your social research project or dissertation.
Oxford: Oxford University Press.
Claro, D.P. & Bortoluzzo, A.B. 2015. Profiling the buzz agent: Product referral and the study of
social community and brand attachment. Brazilian Administration Review, 12(2): 209–228.
Clatworthy, S.D. 2019. The experience-centric organization: How to win through customer
experience. Sebastopol: O’Reilly Media.
Coetzee, R. 2021. Make sure this option is ticked on your car insurance policy as civil unrests rise
in SA. News24, 15 July 2021. Available: https://www.news24.com/wheels/news/
guides_and_lists/make-sure-this-option-is-ticked-on-your-car-insurance-policy-as-civil-
unrests-rise-in-sa-20210715-2 [Accessed 2021, September 27].
369
Coggins, B., Robinson, K. & Wilkie, J. 2020. The great consumer shift: Ten charts that show how
US shopping behavior is changing. Available: https://www.mckinsey.com/business-
functions/marketing-and-sales/our-insights/the-great-consumer-shift-ten-charts-that-show-
how-us-shopping-behavior-is-changing [Accessed 2021, October 17].
Collis, J. & Hussey, R. 2021. Business research : A practical guide for students. 5th ed. England:
Macmillan Education Limited.
Colman, A.M. 2015. A dictionary of psychology. 4th ed. Oxford: Oxford University Press.
Conner, M. & Armitage, C.J. 1998. Extending the theory of planned behavior: A review and
avenues for further research. Journal of Applied Social Psychology, 28(15): 1429–1464.
Conrad, C.F. & Serlin, R.C. 2011. Research design in qualitative quantitative/mixed methods. 2nd
ed. Thousand Oaks: SAGE Publications.
Conway, J. 2017. Risks of unlawful social media content. Insurance Getaway, 17 February 2017.
Available: http://www.insurancegateway.co.za/ShorttermProfessionals/PressRoom/View
Press/Irn=14914&URL=Risks+of+unlawful+social+media+content+1#.WQXdmPl96pp
Cooper, D.R. 2018. Business research: A guide to planning, conducting, and reporting your study.
Thousand Oaks: SAGE Publications.
Cooper, D.R. & Schindler, P.S. 2014. Business research methods. 12th ed. McGraw Hill.
Cooper, D.R., Schindler, P.S. & Sharma, J.K. 2019. Business research methods. 12th ed.
Chennai: Mc Graw Hill (India) Private Limited.
Coutts, A., Matthew, Q., Pillay, S., Van Wyk, G., Blaauw, A., Wild, P., Rankoe, V. & Lomberg, J.
2019. Santam Insurance barometer report. Available:
https://www.santam.co.za/media/2685167/santam-insurance-barometer-2019-v11.pdf
[Accessed 2021, February 01].
Cox-Steib, J. 2020. Does car color affect insurance? Available:

https://www.coverage.com/insurance/auto/does-car-color-affect-insurance/ [Accessed 2021,
October 07].
Cox, S.A. 2014. Managing information in organizations: A practical guide to implementing an

information management strategy. London: Palgrave Macmillan.
Coxon, M. 2012. Cognitive psychology. London: SAGE.
Creswell, J.W. 2014. Research design: Qualitative, quantitative, and mixed methods approaches.
4th ed. London: SAGE Publications.
370
Creswell, J.W. & Creswell, J.D. 2017. Research design: Qualitative, quantitative, and mixed
methods approaches. 5th ed. London: SAGE Publications.
Crime Stats SA. 2015. Crime stats simplified. Available:

http://www.crimestatssa.com/national.php [Accessed 2016, February 04].
Crouth, G. 2016. Social media dangers for insurance cheats. Personal Finance, 14 November
2016. Available: http://www.iol.co.za/news/south-africa/social-media-dangers-for-insurance-
cheats-2089752 [Accessed 2021, October 14].
Cunningham, N. 2018. Introduction to marketing: A South African perspective. Pretoria: Van

Schaik Publishers.
Danckwerts, M., Naran, K. & Swana, A. 2020. The South African insurance industry survey 2020.
Available: https://home.kpmg/content/dam/kpmg/za/pdf/pdf2020/south-african-insurance-
survey-2020.pdf [Accessed 2021, June 24].
Daniele, D. 2016. Social media overload. Social Media Law Bulletin, 19 April. Available:
http://www.socialmedialawbulletin.com/2016/04/social-media-overload/ [Accessed 2021,
October 14].
Das, S.C. & Rao, S. 2017. Consumer buying behaviour and satisfaction towards motor insurance
policies: Experience from Indian general insurance industry. International Journal on
Customer Relations, 5(1): 5–13.
Daskal, T. 2021. Top 16 digital transformation trends in insurance in 2021. [Blog] EasySend, 26
November 2020. Available: https://www.easysend.io/blog/top-16-digital-transformation-
trends-in-insurance-in-2021 [Accessed 2021, October 16].
De Mooij, M. 2019. Consumer behavior and culture: Consequences for global marketing and
advertising. Thousand Oaks: SAGE Publications.
De Vries, F. 2017. Big data: What’s the big deal? Moonstone, 17 March 2017. Available:
http://www.moonstone.co.za/big-data-whats-the-big-deal/ [Accessed 2021, October 14].
Deetjen, U., Kaniyar, S., Methner, N. & Münstermann, B. 2019. IoT insurance industry
opportunities. Available: https://www.mckinsey.com/industries/financial-services/our-insights/
digital-ecosystems-for-insurers-opportunities-through-the-internet-of-things [Accessed 2021,
October 14].
Deffree, S. 2019. Karl Benz drives the first automobile, July 3, 1886. [Blog] EDN Network, 3 July
2019. Available: http://www.edn.com/electronics-blogs/edn-moments/4376656/Karl-Benz-
drives-the-first-automobile--July-3--1886 [Accessed 2021, October 01].
371
Delaporte, T. & Bastid, V. 2016. World insurance report 2016. Available: http://riskinfo.com.au/
news/files/2016/03/WIR_2016.pdf [Accessed 2021, October 15].
Deloitte. 2020. The economic value of insurance broking. Available: https://www2.

deloitte.com/content/dam/Deloitte/au/Documents/Economics/deloitte-au-dae-niba-economic-
value-of-insurance-brokers-151020.pdf [Accessed 2021, August 16].
Demissie, S. 2014. Assessment of customer satisfaction on motor insurance services: the case
of Ethiopian insurance corporation. St. Mary’s University College. Available:
http://repository.smuc.edu.et/bitstream/123456789/109/1/SOLOMON DEMISSIE.pdf
Desrosiers, M. 2012. How individuals purchase insurance: Going beyond expected utility theory.
Casualty Actuarial Society E-Forum, 2: 1–18.
DeVault, G. 2021. Consumer profile basics: Defining your ideal customer. Available:
https://www.thebalance.com/consumer-profile-defining-the-ideal-customer-2296932
Dharani, S., Isherwood, T., Mattone, D. & Moretti, P. 2018. Telematics: Poised for strong global
growth. Available: https://www.mckinsey.com/industries/automotive-and-assembly/our-
insights/telematics-poised-for-strong-global-growth [Accessed 2021, October 13].
Dialdirect. 2020. 9 factors that affect your vehicle premiums. [Blog] dialdirect, 11 September 2020.
Available: https://www.dialdirect.co.za/blog/motoring-travel/9-factors-that-affect-your-vehicle-
premiums/ [Accessed 2021, October 04].
Dictionary.com. 2021. Utility. Available: http://www.dictionary.com/browse/utility [Accessed 2021,

October 21].
Diers, D., Eling, M. & Marek, S.D. 2012. Dependence modeling in non-life insurance using the
Bernstein copula. Insurance: Mathematics and Economics, 50(3): 430–436.
Dikko, M. 2016. Establishing construct validity and reliability: pilot testing of a qualitative interview
for research in Takaful (Islamic Insurance). Qualitative Report, 21(3): 521–528.
Dillard, T. 2017. To like or not to like – navigating the pitfalls of using social media in underwriting.
Insurance focus, (4): 3–5.
Dinnie, D. 2015. Product liability, the Consumer Protection Act and insurance. Available:
https://www.financialinstitutionslegalsnapshot.com/2015/07/product-liability-the-consumer-
protection-act-and-insurance/ [Accessed 2021, October 16].
372
Dinnie, D., Bracher, P. & Patricks, C. 2013. Ten things to know: Insurance regulation in South
Africa. Mondaq. 26 February 2013. Available: http://www.mondaq.com/southafrica/
x/222724/Insurance/Insurance+regulation+in+South+Africa [Accessed 2021, August 01].
Discovery. 2021a. Does your car colour affect your insurance premium? Available:
https://www.discovery.co.za/portal/individual/insurance-news-aug16-ceo-update [Accessed
2021, October 07].
Discovery. 2021b. Our business. Available: https://www.discovery.co.za/corporate/our-business

Discovery. 2021c. Intelligent technology to keep you safe. Available: https://www.discovery.co.za/

car-and-home-insurance/driving-technology [Accessed 2021, October 13].
Discovery Insure. 2014. New driving app set to rev up driving standards in South Africa. Available:
http://www.mynewsdesk.com/za/discovery-holdings-ltd/pressreleases/new-driving-app-set-
to-rev-up-driving-standards-in-south-africa-1001019 [Accessed 2021, October 13].
Dolnicar, S., Grün, B. & Leisch, F. 2018. Market segmentation analysis: Understanding it, doing
it, and making it useful. Singapore: Springer.
Doyle, L. 2021. Short-term insurance – Why it’s a must and vitally important. Available:
https://www.morebo.co.za/short-term-insurance-why-its-a-must/ [Accessed 2021, October
07].
Drennan, L.T., McConnell, A. & Stark, A. 2015. Risk and crisis management in the public sector.
2nd ed. New York: Routledge.
Droppa, D. 2021. Expect car price hikes in 2021 despite stronger rand. Available:
https://www.businesslive.co.za/bd/life/motoring/2021-06-17-expect-car-price-hikes-in-2021-
despite-stronger-rand/ [Accessed 2021, September 27].
Dunn, S. 2014. Skills development issues and challenges for SA’s short term insurance industry.
FA NEWS, 6 November 2014. Available: https://www.fanews.co.za/article/careers-training-
learnerships/42/general/1196/skills-development-issues-and-challenges-for-sa-s-short-term-
insurance-industry/17024 [Accessed 2021, October 13].
Duwalcoe Financial Consultants. 2021. What is long term insurance? Available:

https://www.duwalcoe.co.za/index.php/long-term [Accessed 2021, November 09].
East, R., Singh, J., Wright, M. & Vanhuele, M. 2017. Consumer behaviour: Applications in
marketing. 3rd ed. London: SAGE Publications.
373
Ebert, J.F., Huibers, L., Christensen, B. & Christensen, M.B. 2018. Paper-or web-based
questionnaire invitations as a method for data collection: Cross-sectional comparative study
of differences in response rate, completeness of data, and financial cost. Journal of Medical
Internet Research, 20(1).
Edelheim, J.R. 2015. Ontology, epistemology and axiology issues. In D. Dredge, D. Airey & M.J.
Gross (Eds.). The Routledge handbook of tourism and hospitality education, pp. 30-42.
Abingdon: Routledge.
EFInA. 2012. EFInA access to financial services in Nigeria 2012 survey key findings. Available:
https://www.efina.org.ng/wp-content/uploads/2018/12/EFInA-Access-to-Financial-Services-
in-Nigeria-2012-surveyKey-Findings.pdf [Accessed 2016, February 06].
Efron, S.E. & Ravid, R. 2019. Writing the literature review: A practical guide. New York: The
Guilford Press.
Einstein, M. 2017. Advertising: What everyone needs to know? New York: Oxford University
Press.
Ekanayake, A. 2021. Word of mouth as a marketing strategy in motor insurance sector in Sri
Lanka: Accomplishing marketing needs. Global Scirntific Journals, 9(3): 231–241.
Eling, M. & Jung, K. 2020. Risk aggregation in non-life insurance: Standard models vs. internal
models. Insurance: Mathematics and Economics, 95: 183–198.
Emilien, G., Weitkunat, R. & Lüdicke, F. 2017. Consumer perception of product risks and benefits.
Cham: Springer.
eNaTIS. 2021. Live vehicle population as per the as per the National Traffic Information System.
Available: file:///C:/Users/marxn/Downloads/LiveVehPopPerClassProv20210831.pdf
[Accessed 2021, September 30].
Encyclopedia.com. 2018. State Farm Mutual automobile insurance company. Available:

https://www.encyclopedia.com/social-sciences-and-law/economics-business-and-labor/
businesses-and-occupations/state-farm-mutual-automobile-insurance-company [Accessed
2021, October 01].
Engel, J.F., Kollat, D.T. & Blackwell, R.D. 1968. Consumer behavior. New York: Holt, Rinehart
and Winston.
Epetimehin, F.M. & Fatoki, O. 2011. The empirical analysis of the impact of inflation on the Nigeria
insurance industry. Journal of Emerging Trends in Economics and Management Sciences,
2(6): 454–460.
374
Epic MSLGROUP. 2017. The need to move closer to our brokers and customers. Insurance
Gateway. 27 February 2017. Available: http://www.insurancegateway.co.za/
ShorttermProfessionals/PressRoom/ViewPress/Irn=14997&URL=The+need+to+move+close
r+to+our+brokers+and+customers#.WQTEn_mGOpp [Accessed 2021, October 13].
Epperson, S. 2014. Annuities for retirement: Good or bad idea? CNBC. 24 October 2014.
Available: http://www.cnbc.com/2014/10/24/annuities-for-retirement-good-or-bad-idea.html
[Accessed 2020, June 18].
Epstein, A. 2021. Powerful mission and vision statements: The why, what and how. [Blog] Cella,
29 June 2021. Available: http://www.cellaconsulting.com/blog/three-things-you-cant-do-
without-mission-vision-and-values-statements/ [Accessed 2021, October 07].
Erasmus, B.J., Strydom, J.W. & Rudansky-Kloppers, S. 2019. Introduction to business

management. 11th ed. Cape Town: Oxford University Press Southern Africa.
Erev, I. 1992. The effect of explicit probability estimates on violations of subjective expected utility
theory in the allais paradox. In J. Geweke (Ed.). Decision making under risk and uncertainty,
pp. 117–124. Dordrecht: Springer Netherlands.
Ernest-Jones, S. 2019. Why smart insurance companies are going direct to consumer. [Blog]
Gwi., 6 June 2019. Available: https://blog.gwi.com/marketing/direct-to-consumer-insurance/
[Accessed 2021, August 16].
Ernst & Young. 2013. The journey toward greater customer centricity. Available:
https://www.thepowerscompany.com/wp-content/uploads/2019/12/Customer_Centricity_
Paper_29_April_Final_US.pdf [Accessed 2021, October 13].
Esau, E.Y.R. 2015. Factors affecting consumer purchase decision on insurance product in PT
Prudential Life Assurance Manado. Jurnal EMBA, 491(3): 491–501.
European Commission. 1990. A single insurance market: State of play and future work. Available:
http://europa.eu/rapid/press-release_MEMO-90-9_en.htm [Accessed 2017, February 16].
Expatica. 2021. Car insurance in South Africa: A how-to guide. Available:

https://www.expatica.com/za/finance/insurance/car-insurance-south-africa-105930/
Facebook IQ. 2019. Understanding the journey of the connected insurance consumer. Available:
https://www.facebook.com/business/news/insights/understanding-the-journey-of-the-
connected-insurance-consumer [Accessed 2021, March 19].
Fahy, J. & Jobber, D. 2019. Foundations of marketing. 6th ed. London: McGraw Hill Higher
Education.
375
Fallon, M. 2016. Writing up quantitative research in the social and behavioral sciences.
Rotterdam: Sense Publishers.
Farley, J.U. & Ring, L.W. 1970. An empirical test of the Howard-Sheth model of buyer behavior.
Journal of Marketing Research. 7(4): 427–438.
Faurie, J. 2014a. A look into the future of the insurance industry. FA NEWS, 29 September 2014.
Available: https://www.fanews.co.za/article/short-term-insurance/15/general/1217/a-look-
into-the-future-of-the-insurance-industry/16783 [Accessed 2019, May 05].
Faurie, J. 2014b. Increased civil disobedience prompts review of insurance covers. FA NEWS,
10 March 2014. Available: https://www.fanews.co.za/article/short-term-insurance/15/general/
1217/increased-civil-disobedience-prompts-review-of-insurance-covers/15538 [Accessed
2017, April 29].
Faurie, J. 2019a. South Africa’s 2019 best performing short term insurers are…. FA NEWS, 26
June 2019. Available: https://www.fanews.co.za/article/surveys-reports-and-ratings/34/
general/1195/south-africa-s-2019-best-performing-short-term-insurers-are/27029 [Accessed
2021, October 13].
Faurie, J. 2019b. Imagine underwriting a claim using social media. FA NEWS, 22 July 2019.
Available: https://www.fanews.co.za/article/legal-affairs/10/general/1120/imagine-
underwriting-a-claim-using-social-media/27188 [Accessed 2021, October 14].
Faurie, J. 2019c. Cyber crime affects us all. FA NEWS, 23 April 2019. Available:
https://www.fanews.co.za/article/short-term-insurance/15/general/1217/cyber-crime-affects-
us-all/26597 [Accessed 2021, October 15].
Fennis, B.M. & Stroebe, W. 2021. The psychology of advertising. 3rd ed. Abingdon: Routledge.
Fenny, A.P., Yates, R. & Thompson, R. 2018. Social health insurance schemes in Africa leave
out the poor. International Health, 10: 1–3.
Ferenzy, D., Silverberg, K., French, C. & van Liebergen, B. 2016. Innovation in insurance: How
technology is changing the industry. Available: https://www.iif.com/Publications/ID/1246/
Innovation-in-Insurance-How-Technology-is-Changing-the-Industry [Accessed 2021, October
1].
Ferreira, E.J. 2011. Fundamentals of office management. Cape Town: Juta.
Ferrell, O.C. & Hartline, M. 2014. Marketing strategy, text and cases. 6th ed. Mason: Cengage
Learning.
Ferrell, O.C., Hartline, M. & Hochstein, B.W. 2021. Marketing strategy: Text and cases. 8th ed.
Mason: Cengage Learning.
376
Festinger, L. 1957. A theory of cognitive dissonance. Stanford: Stanford University Press.
Fichardt, C. 2015. Industry leaders: SA car insurance in crisis. Wheels24, 6 August 2015.
Available: http://www.wheels24.co.za/News/Industry-leaders-SA-car-insurance-in-crisis-
20150806 [Accessed 2016, February 06].
Field, F. 2011. The welfare state - Never ending reform. BBC, 10 March 2011. Available:
http://www.bbc.co.uk/history/british/modern/field_01.shtml [Accessed 2017, May 05].
Field, G. 2017. Technology’s hand in reshaping insurance. Mail & Guardian. 20 April 2017.
Available: https://pressoffice.mg.co.za/amp/content/KPNG878KbN3v4mwD [Accessed 2021,
October 14].
Fielding, N.G., Lee, R.M. & Blank, G. 2017. The SAGE handbook of online research methods.
2nd ed. Thousand oaks: SAGE Publications.
Filiz, C. 2020. Can autonomous vehicles prevent traffic accidents? Available:

https://www.intechopen.com/chapters/72643 [Accessed 2021, October 15].
Financial Sector Conduct Authority. 2021. About us. Available: https://www.fsca.co.za/

Pages/Vision-and-Mission.aspx [Accessed 2021, October 13].
Financial Services Board. 2014. 17th Annual report of the registrar on the results of the short-
term insurance industry for the period ending. Available:
https://www.fsb.co.za/Departmhttps://www.fsca.co.za/Annual Reports/17th Annual Report of
the Short Term Insurance 2014.pdf [Accessed 2021, October 04].
Financial Services Board. 2016. The do’s and don’t’s of DOFA enquiries. FAIS Newsletter, 19
May 2016. Available: https://www.fsca.co.za/Regulated Entities/Regulated Entities
Documents/Newsletter Volume 20.pdf [Accessed 2017, June 20].
Financial Services Board. 2017. Annual report. Available: https://www.fsca.co.za/Annual%

20Reports/FSB%20Annual%20Report%202017.pdf [Accessed 2022, January 26].
Financial Services Board. n.d. Long-term insurance. Available: https://www.fsb.co.za/

feedback/Documents/Long Term Insurance.pdf [Accessed 2017a, March 02].
Financial Services Board. n.d. Short-term insurance. Available: https://www.fsb.co.za/

Departments/consumerEducation/Documents/All you need to know about Short-term
Insurance.pdf [Accessed 2017b, March 07].
Fishbein, M. & Ajzen, I. 1975. Belief, attitude, intention, and behavior: An introduction to theory
and research. Massachusetts: Addison-Wesley.
Fishburn, P.C. 1981. Subjective expected utility: A review of normative theories. Theory and
Decision, 13(2): 139–199.
377
Fisher-French, M. 2010. How an annuity works in retirement. Mail & Guardian, 14 September.
Available: https://mg.co.za/article/2010-09-14-how-an-annuity-works-in-retirement [Accessed
2018, May 15].
Fisher-French, M. 2016. Is online insurance cheaper? City Press, 19 May. Available: http://city-
press.news24.com/Personal-Finance/is-online-insurance-cheaper-20160513 [Accessed
2017, May 01].
Flache, A., Mäs, M., Feliciani, T., Chattoe-Brown, E., Deffuant, G., Huet, S. & Lorenz, J. 2017.
Models of social influence: Towards the next frontiers. Journal of Artificial Societies and Social
Simulation, 20(4).
Fleisher, C.S. & Bensoussan, B.E. 2015. Business and competitive analysis: Effective application
of new and classic methods. 2nd ed. New Jersey: Pearson Education.
Fornell, C. & Larcker, D.F. 1981. Evaluating structural equation models with unobservable
variables and measurement error. Journal of Marketing Research, 18(1): 39–50.
Fortis. 2001. Fortis and China Insurance group to launch landmark life insurance partnership in
China, 19 October 2001. Available: http://www.cntaiping.com/upload/cms/en/201307/
11144910363c.pdf [Accessed 2017, February 16].
Fourie, C. 2013a. Five factors that influence motor insurance premiums. Available:
https://www.fanews.co.za/article/short-term-insurance/15/motor/1023/five-factors-that-
influence-motor-insurance-premiums/13397 [Accessed 2021, October 07].
Fourie, C. 2013b. APPS to play increasingly important role in short-term insurance industry. FA
NEWS, 1 April 2013. Available: https://www.fanews.co.za/article/magazine-archives-fanews-
fanuus/60/short-term/1317/apps-to-play-increasingly-important-role-in-short-term-insurance-
industry/15393 [Accessed 2021, October 13].
Foxall, G. 2004. Consumer psychology in behavioural perspective. Washington: Beard Books.
Frankenfield, J. 2021. Churn rate definition. Available: https://www.investopedia.com/terms/

c/churnrate.asp [Accessed 2021, September 27].
Gana, K. & Broc, G. 2019. Structural equation modeling with lavaan. Hoboken: John Wiley &
Sons.
Gazette, G. 1998. Short-term Insurance Act No 53. Cape Town: Republic of South Africa.
Available: https://www.gov.za/sites/default/files/gcis_document/201409/a53-98.pdf
[Accessed 2017, March 30].
378
Gazette, G. 2008. Short-term Insurance Amendment Act No 53. Republic of South Africa.
Available: https://www.fsb.co.za/Departments/insurance/archives/Documents/Short-term
Insurance Act 1998 Publication of proposed amendment of Regulations made under section
70 for public comment.pdf [Accessed 2017, March 30].
Gbadamosi, A. 2019. Contemporary issues in marketing: Principles and practice. London: SAGE
Publications.
Gbadamosi, A., Bathgate, I.K. & Nwankwo, S. 2013. Principles of marketing: A value-based
approach. Laxenburg: Palgrave Macmillan.
Geisst, C.R. 2006. Encyclopaedia of American business history. New York: Facts On File.
Geldenhuys, J. 2020a. Reintroducing third party insurance. Moonstone, 12 March 2020.

Available: https://www.moonstone.co.za/reintroducing-third-party-insurance-welcomed-by-
most-insurers/ [Accessed 2021, June 25].
Geldenhuys, J. 2020b. SA’s short-term insurance industry – Rise in niche risk solutions amidst
COVID uncertainty. Available: https://www.moonstone.co.za/sas-short-term-insurance-
industry-rise-in-niche-risk-solutions-amidst-covid-uncertainty/ [Accessed 2021, July 23].
George, J. 2017. How to accomplish social media compliance in financial services. Insurance
Gateway. 23 January 2017. Available: http://www.insurancegateway.co.za/Shortterm
Professionals/PressRoom/ViewPress/Irn=14713&URL=How+to+accomplish+social+media+
compliance+in+financial+services#.WQXdmPl96pp [Accessed 2021, October 14].
Geospatial World. 2020. The top 5 technologies impacting the insurance industry. [Blog]
GEOSPATIAL WORLD, 18 June 2020. Available: https://www.geospatialworld.net/blogs/the-
top-5-technologies-impacting-the-insurance-industry/ [Accessed 2021, October 14].
Ghauri, P., Grønhaug, K. & Strange, R. 2020. Research methods in business studies. 5th ed.
New York: Cambridge University Press.
Gibbs, S. 2016. Self-driving cars: who’s building them and how do they work? Available:
https://www.theguardian.com/technology/2016/may/26/self-driving-cars-whos-building-them-
and-how-do-they-work [Accessed 2021, October 01].
Gibson, K. 2021. Schedule exemption of easy pay (Pty) Ltd and its agents (STIA). Available:
https://www.fsca.co.za/Notices/FSCA INS Notice 5 of 2021.pdf [Accessed 2021, November
14].
Gliner, J.A., Morgan, G.A. & Leech, N.L. 2017. Research methods in applied settings: An
integrated approach to design and analysis. 3rd ed. New York: Routledge.
379
Goldby, A. 2020. COVID-19: The effect on motor insurance. Available: https://www.thefloow.com/
wp-content/uploads/The-Floow-Ltd-COVID-19_-The-Effect-on-Motor-Insurance-Andy-
Goldby-April-2020.pdf [Accessed 2021, June 24].
Goldsmith, E.B. 2021. Consumer economics: Issues and behaviors. 4th ed. Abingdon: Routledge.
Govender, S. 2017. Junk status will make your car insurance more expensive. Times LIVE, 5
April. Available: http://www.timeslive.co.za/local/2017/04/05/Junk-status-will-make-your-car-
insurance-more-expensive [Accessed 2021, October 15].
Grable, J.E., Archuleta, K.L. & Nazarinia, R.R. 2011. Financial planning and counseling scales.
New York: Springer Science & Business Media.
Gray-Parker, J. 2015. A word of advice on home insurance. Private Property, 3 July 2015.
Available: https://www.privateproperty.co.za/advice/property/articles/a-word-of-advice-on-
home-insurance/3411 [Accessed 2016, August 02].
Gray, D.E. 2020. Doing research in the business world. 2nd ed. London: SAGE Publishing.
Guinalíu, M. & Díaz De Rada, V. 2020. Combining sources of information to increase survey
response rates. Spanish Journal of Marketing: 2444–9709.
Hackett, P.M.W. 2019. Quantitative research methods in consumer psychology: Contemporary

and data driven approaches. New York: Routledge.
Hahs-Vaughn, D.L. 2017. Applied multivariate statistical concepts. Abingdon: Taylor & Francis.
Haider, T. & Shakib, S. 2017. A Study on the influences of advertisement on consumer buying
behavior. Business Studies Journal. 9(1). Available: https://www.abacademies.org/articles/a-
study-on-the-influences-of-advertisement-on-consumer-buying-behavior-7177.html
[Accessed 2022, January 26].
Hair, J.F., Black, W.C., Babin, B.J. & Anderson, R.E. 2019. Multivariate data analysis. 8th ed.
Andover: Cengage Learning.
Hair, J.F., Page, M. & Brunsveld, N. 2020. Essentials of business research methods. 4th ed. New
York: Routledge.
Hair, J.F.J., Hult, G.T.M., Ringle, C. & Sarstedt, M. 2017. A primer on partial least squares
structural equation modeling (PLS-SEM). 2nd ed. Thousand Oaks: SAGE Publications.
Haldane, M. 2020. Important things to know about endowments. Moneyweb. 23 October 2020.
Available: https://www.moneyweb.co.za/financial-advisor-views/important-things-to-know-
about-endowments/ [Accessed 2021, July 23].
380
Hall, R. 2020. Mixing methods in social research: Qualitative, quantitative and combined methods.
London: SAGE Publications.
Hanger, S., Linnerooth-Bayer, J., Surminski, S., Nenciu-Posner, C., Lorant, A., Ionescu, R. & Patt,
A. 2018. Insurance, public assistance, and household flood risk reduction: A comparative
study of Austria, England, and Romania. Risk Analysis, 38(4): 680–693.
Hansson, S.O. 2005. Decision theory: A brief introduction. Stockholm: Royal Institute of
Technology.
Harrington, F. 2011. The beginning and history of insurance. Available: http://www.insurance

comparison.net/230/beginning-history-insurance/ [Accessed 2017, February 09].
Harrison, G.W. & Rutström, E.E. 2009. Expected utility theory and prospect theory: One wedding
and a decent funeral. Experimental Economics, 12(2): 133–158.
Havlik, P. 2016. Big data a must for insurance industry. COVER Publications. 23 February 2016.
Available: https://www.cover.co.za/big-data-a-must-for-insurance-industry/ [Accessed 2021,
October 14].
Hawthorne, R. & Grzybowski, L. 2019. Fixed lines needed to bridge the digital divide. UCT News.
20 November 2019. Available: https://www.news.uct.ac.za/article/-2019-11-20-fixed-lines-
needed-to-bridge-digital-divide [Accessed 2021, October 13].
Head, T. 2020. Revealed: The average salary earned in South Africa - it’s gone up. The South
African, 5 April 2020. Available: https://www.thesouthafrican.com/news/finance/average-
wage-in-south-africa-2020-how-much/ [Accessed 2021, October 12].
Healey, A. 2021. When did car insurance become mandatory? Available: https://getjerry.
com/insights/when-did-car-insurance-become-mandatory [Accessed 2021, October 01].
Health24. 2011. Medical schemes - A history. Health24, 21 July 2011. Available: http://www.
health24.com/Medical-schemes/About-medical-schemes/Medical-schemes-a-history-
20120721 [Accessed 2017, February 23].
Hearst Autos Research. 2021. Does the color of a car affect insurance? Available: https://www.
caranddriver.com/car-insurance/a36331591/does-the-color-of-a-car-affect-insurance/
Heathfield, S.M. 2020. What is long term and short term disability insurance? The Balance, 11
September 2020. Available: https://www.thebalance.com/what-is-long-term-disability-
insurance-1918178 [Accessed 2021, October 11].
381
Hesse, M. 2014. How personal liability cover protects you. IOL. 27 September 2014. Available:
http://www.iol.co.za/business-report/how-personal-liability-cover-protects-you-1756679
[Accessed 2017, March 07].
Hesse, M. 2015a. The benefits of an insurance broker. IOL, 15 August 2015. Available:
http://www.iol.co.za/personal-finance/the-benefits-of-an-insurance-broker-1900259
[Accessed 2017, May 01].
Hesse, M. 2015b. How insurance companies risk-assess you. IOL. 27 June 2015. Available:
http://www.iol.co.za/personal-finance/how-insurance-companies-risk-assess-you-1877103
[Accessed 2017, April 25].
Hesse, M. 2015c. 3rd-party cover for your car. Personal Finance, 4 July 2015. Available:
http://www.iol.co.za/personal-finance/3rd-party-cover-for-your-car-1880315 [Accessed 2021,
October 07].
Hesse, M. 2016. How good is credit-card travel insurance? IOL, 1 June 2016. Available:
http://www.iol.co.za/personal-finance/my-money/banking/how-good-is-credit-card-travel-
insurance-2028694 [Accessed 2017, March 07].
Hillebrand, M. 2014. Technology is changing the insurance industry, driven by outside influencers,
competition. Available: http://www.propertycasualty360.com/2014/11/24/technology-is-
changing-the-insurance-industry-driv [Accessed 2021, October 21].
Hinkelman, E.G. 2008. Dictionary of international trade. 8th ed. World Trade Press.
Hogg, A. 2015. Let’s find incentives to address SA’s R300bn pa, 38 daily road deaths. BizNews,
22 April 2015. Available: http://www.biznews.com/thought-leaders/2015/04/22/lets-find-
incentives-to-address-sas-r300bn-pa-38-daily-road-deaths/ [Accessed 2018, September 16].
Holden, J. 2014. Why it’s time for government to tackle compulsory third-party motor property
insurance. FA News, 10 September 2014. Available: https://www.fanews.co.za/article/short-
term-insurance/15/motor/1023/why-it-s-time-for-government-to-tackle-compulsory-third-
party-motor-property-insurance/16698 [Accessed 2021, October 07].
Hollensen, S. 2019. Marketing management: A relationship approach. 4th ed. Amsterdam:

Pearson Benelux.
Hoosen, N. 2016. 2016 Top global risks. Accountancy SA, 1 April 2016. Available:
https://www.accountancysa.org.za/special-feature-insurance/ [Accessed 2021, October 13].
Horner, S. & Swarbrooke, J. 2016. Consumer Behaviour in Tourism. 3rd ed. New York: Routledge.
Hosmer, D.W., Lemeshow, S. & Sturdivant, R.X. 2013. Applied logistic regression. 3rd ed. New
Jersey: John Wiley & Sons.
382
Howard, J.A. & Sheth, J.N. 1969. The theory of buyer behavior. New York: Wiley.
Howard, J.A. & Sheth, J.N. 1970. The theory of buyer behavior. Journal of the American Statistical
Association, 65(331): 1406.
Hoyer, W.D., MacInnis, D.J. & Pieters, R. 2018. Consumer behavior. 7th ed. Boston: Cengage
Learning.
Huber, C., Gatzert, N. & Schmeiser, H. 2015. How does price presentation influence consumer
choice? The case of life insurance products. The Journal of Risk and Insurance. 82(2): 401–
431.
Hunt, S.D. 1983. General theories and the fundamental explananda of marketing. Journal of
Marketing, 47(4): 9–17.
Hunt, S.D. & Pappas, J.L. 1972. A crucial test for the Howard-Sheth model of buyer behavior.
Source Journal of Marketing Research, 9101(3): 346–348.
Ilumba, A. 2015. The history of car insurance. Available: http://www.dmv.org/articles/history-of-

car-insurance/ [Accessed 2017, April 12].
Independent Financial Consultants. 2021. Medical aid vs Health insurance: Everything you need
to know. Available: http://www.medicalaid-quotes.co.za/articles/medical-aid-vs-health-
insurance-everything-you-need-know [Accessed 2021, July 23].
Information Resources Management Association. 2017. Fashion and textiles: Breakthroughs in

research and practice. Hershey: IGI Global.
Innovation Group. 2021. Vehicle value added products and services: The key to keeping
customers engaged. Available: https://www.innovation.group/country/south-africa/ [Accessed
2021, November 03].
INSETA. 2014. Status of skills in the insurance industry, Part 2: Focus on the short-term insurance
sector. Available: https://www.inseta.org.za/wp-content/uploads/2019/01/Part-2-Focus-on-
the-Short-Term-Insurance-Industry.pdf [Accessed 2021, October 12].
INSETA. 2016. Sector skills plan. Available: https://financedocbox.com/Insurance/85473765-

Foreword-inseta-board-chairman.html [Accessed 2021, October 12].
Insight Survey. 2016. Will SA’s short-term insurance industry prosper? Bizcommunity, 30 May
2016. Available: http://www.bizcommunity.com/Article/196/363/145492.html [Accessed 2017,
June 28].
Institute Insurance Information. 2021. Facts + Statistics: Auto insurance. Available:

https://www.iii.org/fact-statistic/facts-statistics-auto-insurance [Accessed 2021, February 01].
383
Insurance Europe. 2014. European insurance in figures. Brussels. Available:
http://www.biztositasiszemle.hu/files/201501/statisticsno50europeaninsuranceinfigures.pdf
[Accessed 2017, January 30].
International Monetary Fund. 2010. Detailed assessment of implementation. Available:

http://www.treasury.gov.za/publications/other/imf/2010/2010 03 RPT ROSC AI AN.pdf
[Accessed 2017, February 23].
INtgrty. 2016. Research paradigms: Postmodernism. Available: https://www.intgrty.co.za/

2016/09/20/research-paradigms-postmodernism/ [Accessed 2021, October 26].
IOL. 2014. Hail batters SA insurers’ outlook. IOL, 27 July 2014. Available:
http://www.iol.co.za/business/companies/hail-batters-sa-insurers-outlook-1.1718098#.Vg5v7
PmqpBc [Accessed 2016, February 04].
Isaac, T. 2017. What homeowners need to know about insurance. Available:

https://www.privateproperty.co.za/advice/property/articles/what-homeowners-need-to-know-
about-insurance/5375 [Accessed 2021, August 16].
Ismagilova, E., Dwivedi, Y.K., Slade, E. & Williams, M.D. 2017. Electronic word of mouth (eWOM)
in the marketing context: A state of the art analysis and future directions. Cham: Springer.
Ismagilova, E., Slade, E., Rana, N.P. & Dwivedi, Y.K. 2019. The effect of characteristics of source
credibility on consumer behaviour: A meta-analysis. Journal of Retailing and Consumer
Services, 53: 1–40.
iWYZE. 2021a. Choose your cover. Available: https://www.iwyze.co.za/products/car-general-car-

insurance [Accessed 2021, November 01].
iWYZE. 2021b. About Old Mutual. Available: https://www.iwyze.co.za/about-old-mutual

Jackson, D.L., Gillaspy, J.A. & Purc-Stephenson, R. 2009. Reporting practices in confirmatory
factor analysis: An overview and some recommendations. American Psychological
Association, 14(1): 6–23.
Jacobs, E. 2020a. Car insurance shopping behavior survey. Available: https://www.reviews.com/

insurance/car/shopping-behavior-survey/ [Accessed 2021, October 27].
Jacobs, M. 2020b. Regulations regarding autonomous vehicles and their use in South Africa.
Unpublished Master’s dissertation. North-West University, Potchefstroom.
Jacoby, J. 1978. Consumer research: How valid and useful are all our consumer behavior
research findings? Journal of Marketing, 42(2): 87–96.
384
Jankensgard, H. & Kapstad, P. 2021. Empowered enterprise risk management: Theory and
practice. Chichester: John Wiley & Sons.
Jeffreys, A. 2012. The report: South Africa 2012. Oxford Business Group.
Jenkins, R.D. 2015. Business insurance (simply) explained. LinkedIn, 30 March 2015. Available:
https://www.linkedin.com/pulse/business-insurance-simply-explained-robert-d-jenkins
Jensen, K. 2019. What are the examples of intermediary business? CHRON., 4 February 2019.
Available: https://smallbusiness.chron.com/examples-intermediary-business-20244.html
Joffe, D. 2011. Consumer Protection Act: Impact on the insurance industry. FA NEWS, 1 February
2011. Available: https://www.fanews.co.za/article/magazine-archives-fanews-fanuus/60/
legal/1312/consumer-protection-act-impact-on-the-insurance-industry/11225 [Accessed
2021, October 16].
Johansmeyer, T. 2021. Cybersecurity insurance has a big problem. Available:

https://hbr.org/2021/01/cybersecurity-insurance-has-a-big-problem?registration=success
Johnson, R.B. & Christensen, L. 2017. Educational research: Quantitative, qualitative, and mixed
approaches. 6th ed. Thousand Oaks: SAGE Publications.
Johnson, S., Minie, M., Bam, W. & Ransom, L. 2016. Insurance through challenging times
insurance industry analysis. Available: http://www.pwc.co.za/en/assets/pdf/insurance-
industry-analysis-2016pdf.pdf [Accessed 2021, October 13].
Johnston, L. & Gamon, J. 2013. Enterprise resource planning: Concepts, methodologies, tools,
and applications. IGI Global.
Jooste, C. 2014. Customer decision making. In C. Boshoff (Ed.). Services marketing : A

contemporary approach, p. 404. 2nd ed. Cape Town: Juta.
Juska, J.M. 2018. Integrated marketing communication: Advertising and promotion in a digital
world. New York: Routledge.
Kaesler, S., Leo, M., Varney, S. & Young, K. 2020. How insurance can prepare for the next
distribution model. Available: https://www.mckinsey.com/industries/financial-services/our-
insights/how-insurance-can-prepare-for-the-next-distribution-model [Accessed 2021, August
16].
Kagan, J. 2021a. Accumulation period definition. Available: https://www.investopedia.com/

terms/a/accumulationperiod.asp [Accessed 2021, July 23].
385
Kagan, J. 2021b. Personal liability insurance definition. Available: https://www.investopedia.com/
terms/c/comprehensive-personal-liability.asp [Accessed 2021, August 16].
Kahn, K.B. 2015. Product Planning Essentials. 2nd ed. New York: Routledge.
Kahneman, D. & Tversky, A. 1979. Prospect Theory: An Analysis of Decision under Risk. Journal
of the Econometric Society, 47(3): 263–291.
Kahneman, D. & Tversky, A. 1984. Choices, values, and frames. American Psychologist, 39(4):
341–350.
Kaigorodova, G., Mustafina, A., Pyrkova, G., Grzebyk, M. & Belinskaja, L. 2021. Digitalization of
the insurance business: Systematization of net effects through the example of Russia.
Insurance Markets and Companies, 12(1): 32–42.
Kane, A. 2015. When did car insurance become mandatory? [Blog] Everquote. 20 July 2015.
Available: https://www.everquote.com/blog/questions/when-did-car-insurance-become-
mandatory/ [Accessed 2017, April 12].
Kanno, M. 2016. The network structure and systemic risk in the global non-life insurance market.
Insurance: Mathematics and Economics, 67(2016): 38–53.
Kara, H. 2019. Use your questionnaire data: Little quick fix. Thousand Oaks: Sage.
Kardes, F., Cronley, M. & Cline, T. 2015. Consumer behavior. 2nd ed. Stamford: Cengage
Learning.
Karni, E., Maccheroni, F. & Marinacci, M. 2015. Ambiguity and nonexpected utility. In Handbook
of game theory. Elsevier. 910–943. Available: https://www.sciencedirect.com/topics/
economics-econometrics-and-finance/prospect-theory [Accessed 2021, October 22].
Kaser, K. 2013. Advertising and sales promotion. Mason: Cengage Learning.
Katrodia, A. 2021. A study of indentity consumer purchasing behavior and factors that influence
consumer purchse decision: With reference to Durban. Journal of the Research Society of
Pakistan, 58(3): 60–71.
Kaže, V. 2010. The impact of customer values on purchasing behaviour: A case of Latvian
insurance market. Economics and Management, 15: 605–611.
Kearney, L. 2016. Funeral cover: What to look for. IOL. 2 April 2016. Available:
http://www.iol.co.za/business-report/funeral-cover-what-to-look-for-2004168 [Accessed 2019,
June 22].
Keith, T.Z. 2019. Multiple regression and beyond : An introduction to multiple regression and
structural equation modeling. New York: Routledge.
386
Khanna, S.R. & Hanspal, S. 2019. Consumer affairs and customer care. [eBook] New Delhi:
Prowess Publishing.
Khartit, K. 2021. Customer definition. Available: https://www.investopedia.com/terms/c/

customer.asp [Accessed 2021, October 20].
Killam, L. 2013. Research terminology simplified: Paradigms, axiology, ontology, epistemology

and methodology. Sudbury: Laura Killam.
Klein, B. 2001a. The Great Fire of London. Available: https://www.irmi.com/articles/expert-

commentary/the-great-fire-of-london [Accessed 2017, January 26].
Klein, B. 2001b. The world’s first insurance company. Available: https://www.irmi.com/articles/

expert-commentary/the-worlds-first-insurance-company [Accessed 2017, January 26].
Klenke, K. 2016. Qualitative research in the study of leadership. 2nd ed. Berlin, Heidelberg:
Emerald Group Publishing.
Knoesen, M. 2017. Cost of damages in millions. FA News, 25 October 2017. Available:

https://www.fanews.co.za/article/talked-about-features/25/straight-talk/1146/cost-of-
damages-in-millions/23205 [Accessed 2018, October 05].
Knoesen, M. 2018. Can social media validate a claim? FA NEWS. 25 September 2018. Available:
https://www.fanews.co.za/article/short-term-insurance/15/general/1217/can-social-media-
validate-a-claim/25470 [Accessed 2021, October 14].
Knoesen, M. 2020. What customers want… are we getting this right? FA NEWS, 29 January
2020. Available: https://www.fanews.co.za/article/intermediaries-brokers/7/general/1227/
what-customers-want-are-we-getting-this-right/28192 [Accessed 2021, October 13].
Knowler, W. 2020. Is linking your car insurance premium to mileage becoming the new normal?
Times LIVE, 3 June 2020. Available: https://www.timeslive.co.za/news/consumer-live/2020-
06-03-is-linking-your-car-insurance-premium-to-mileage-becoming-the-new-normal/
Kollewe, J. 2016. Insurer launches UK’s “first driverless car policy”. The Guardian, 7 June 2016.
Available: https://www.theguardian.com/business/2016/jun/07/uk-driverless-car-insurance-
policy-adrian-flux [Accessed 2021, October 01].
Koontz, H., Weihrich, H. & Cannice, M. V. 2020. Essentials of management: An international,

innovation, and leadership perspectives. 11th ed. Chennai: Mc Graw Hill.
Kosinski, M., Matz, S.C., Gosling, S.D., Popov, V. & Stillwell, D. 2016. Facebook as a research
tool. American Psychological Association, 47(3): 70.
Kotler, P. & Armstrong, G. 2018. Principles of marketing. 17th ed. London: Pearson.
387
Kotler, P. & Keller, K.L. 2012. Marketing Management. 14th ed. New Jersey: Prentice Hall.
Kotler, P., Burton, S., Deans, K., Brown, L. & Armstrong, G. 2013. Marketing. 9th ed. Frenchs
Forest: Pearson Education.
KPMG. 2015a. The South African insurance industry survey 2015. Available: https://assets.kpmg/
content/dam/kpmg/pdf/2016/06/2015-KPMG-Insurance-Survey.pdf [Accessed 2016, January
10].
KPMG. 2015b. Automobile insurance in the era of autonomous vehicles. Available: https://home.
kpmg/cn/en/home/insights/2015/06/automobile-insurance-in-the-era-of-autonomous-
vehicles-o-201506.html [Accessed 2017, August 15].
KPMG. 2016. Business unusual. Available: https://assets.kpmg/content/dam/kpmg/pdf/

2016/07/2016-Insurance-Survey-v2.pdf [Accessed 2021, October 04].
KPMG. 2017. Envolve intelligent insurance. Available: https://assets.kpmg/content/dam/kpmg/za/

pdf/2017/08/2017-insurance-survey.pdf [Accessed 2021, October 14].
KPMG. 2019. The South African insurance industry survey 2019. Available:
https://cisp.cachefly.net/assets/articles/attachments/79692_south-african-insurance-survey-
2019.pdf [Accessed 2021, June 25].
Kraus, E. 2016. Churn: The churn rate formula and definition. [Blog] MergeLane, 22 September
2016. Available: https://blog.mergelane.com/2016/09/22/churn-the-churn-rate-formula-and-
definition/ [Accessed 2016, December 01].
Krieger, N. 2011. Epidemiology and people’s health: Theory and context. New York: Oxford
University Press.
Kriel, G. 2017. Personal insurance gets a revamp. News24. 24 January 2017. Available:
https://www.news24.com/fin24/finweek/featured/personal-insurance-gets-a-revamp-
Krige, S. & Laskov, H. 2018. Bye bye FSB, hello FSCA. Available: https://www.werksmans.
com/legal-updates-and-opinions/bye-bye-fsb-hello-fsca/ [Accessed 2021, November 10].
Kruger, L. 2018. Exploring antecedents of consumer behaviour within the South African life
insurance industry. Unpublished Master’s dissertation. North-West University. Available:
https://repository.nwu.ac.za/bitstream/handle/10394/30997/Kruger_LJ.pdf [Accessed 2021,
July 01].
Kumar, R. 2019. Research methodology: A step-by-step guide for beginners. 5th ed. London:
SAGE Publications.
388
Kunreuther, H.C. & Pauly, M. V. 1982. Market equilibrium with private knowledge: An insurance
example. Laxenburg. Available: http://pure.iiasa.ac.at/id/eprint/1958/ [Accessed 2021,
October 21].
Kyriazos, T.A. 2018. Applied psychometrics: Sample size and sample power considerations in
factor analysis (EFA, CFA) and SEM in general. Psychology, 09(08): 2207–2230.
Lagomarsino, G., Garabrant, A., Adyas, A., Muga, R. & Otoo, N. 2012. Moving towards universal
health coverage: Health insurance reforms in nine developing countries in Africa and Asia.
The Lancet, 380:933–943.
Laing, A. 1988. Change and development in short-term insurance in South Africa 1950–1985.
South African Journal of Economic History, 3(1): 33–44.
Lamb, C.W., Hair, J.F. & McDaniel, C. 2011. Marketing. 11th ed. Mason: South-Western Cengage
Learning.
Lamb, C.W., Hair, J.F. & McDaniel, C. 2013. Marketing. 12th ed. Cengage Learning.
Lamb, C.W., Hair, J.F. & McDaniel, C. 2017. MKTG. 10th ed. Boston: Cengage Learning.
Lamb, C.W., Hair, J.F. & McDaniel, C. 2018. MKTG. 11th ed. Boston: Cengage Learning.
Lambert, R. 2014. Mission critical cloud is on the rise in the insurance industry. [Blog] Microsoft
Enterprise. 8 December 2014. Available: https://cloudblogs.microsoft.com/industry-
blog/financial-services/2014/12/08/mission-critical-cloud-is-on-the-rise-in-the-insurance-
industry/ [Accessed 2021, October 14].
Lamprecht, I. 2014. Hailstorms, car insurance and SMS warnings. Moneyweb, 10 January 2014.
Available: https://www.moneyweb.co.za/archive/hailstorms-car-insurance-and-sms-warnings/
Lamprecht, I. 2016. Do I really need dread disease and disability cover? Moneyweb, 15 April
2016. Available: https://www.moneyweb.co.za/mymoney/moneyweb-financial-planning/
really-need-dread-disease-disability-cover/ [Accessed 2019, August 10].
Langley, G. 2014. Who was Bismarck? Available: http://projectm-online.com/social-security/who-

was-bismarck-and-discover-facts-about-the-first-pension/ [Accessed 2017, May 10].
Lantos, G.P. 2011. Consumer behavior in action: Real-life applications for marketing managers.
New York: Armonk.
Lappeman, J., Egan, P., Rightford, G. & Ramogase, T. 2021. Marketing to South African
consumers. Cape Town: UCT Liberty Institute of Strategic Marketing & UCT Libraries.
Available: https://openbooks.uct.ac.za/uct/catalog/view/29/43/1430-1 [Accessed 2021,
October 16].
389
Laubenstein, C. 2018. What’s a good click-through rate (CTR) for PPC? [Blog] WordStream, 27
September 2018. Available: https://www.wordstream.com/blog/ws/2010/04/26/good-click-
through-rate [Accessed 2019, June 04].
Layon, K. 2014. Digital product management: Design websites and mobile apps that exceed
expectations. New Jersey: New Riders.
Leavy, P. 2017. Research design: Quantitative, qualitative, mixed methods, arts-based, and
community-based participatory research approaches. New York: Guilford Publications.
Legal & General. 2021. What is life insurance? Available: https://www.legalandgeneral.com/life-

cover/articles-and-guides/life-insurance-guides/what-is-life-insurance/ [Accessed 2021,
November 10].
Leiria, M., Matos, N. & Rebelo, E. 2020. Non-life insurance cancellation: A systematic quantitative
literature review. Geneva Papers on Risk and Insurance: Issues and Practice. Available:
https://www.researchgate.net/publication/343835760_Non-life_insurance_cancellation_a_
systematic_quantitative_literature_review [Accessed 2021, November 10].
Letshwiti, V., Stanway, R.A. & Mokonyama, M. 2003. Strategic overview and analysis of vehicle
ownership trends in South Africa with an emphasis on light passenger vehicles. In
Proceedings of the 22nd Southern African Transport Conference (SATC2003). Pretoria. 1–9.
Lewandowski, M. 2017. Prospect theory versus expected utility theory: Assumptions, predictions,
intuition and modelling of risk attitudes. Central European Journal of Economic Modelling and
Econometrics, 9: 275–321.
Lewarne, J. 2017. Big Data: The key to eradicating insurance fraud. Insurance Gateway, 6 March
2017. Available: http://www.insurancegateway.co.za/ShorttermProfessionals/PressRoom/
ViewPress/Irn=15052&URL=Big+Data+The+Key+to+Eradicating+Insurance+Fraud+2#.WQ
Xdf_l96pp [Accessed 2021, October 14].
Liakopoulos, A. 2016. Human capital trends in the insurance industry. Available: https://www2.
deloitte.com/content/dam/Deloitte/no/Documents/financial-services/human-capital-trends-in-
the-insurance-industry-2016.pdf [Accessed 2021, October 07].
Livingston, A. 2020. Improve customer retention with the power of APIs. Coverager, 1 June 2020.
Available: https://coverager.com/improve-customer-retention-with-the-power-of-apis/
Llewellyn, D. 2012. Annual report 2011/2012. Available: https://www.moonstone.co.za/upmedia/

uploads/library/Moonstone Library/MS Reports/FAIS OMBUD 2011-2012 ANNUAL
REPORT.pdf [Accessed 2021, October 13].
Loacker, L.D. 2015. Informed insurance choice? Cheltenham: Edward Elgar Publishing.
390
Loader, I., Goold, B. & Thumala, A. 2015. Grudge spending: The interplay between markets and
culture in the purchase of security. Sociological Review, 63(4): 858–875.
Long, D. 2020. What are the different types of consumers and how do I reach them? Available:
https://toughnickel.com/industries/Types-Of-Consumers [Accessed 2021, October 20].
Lovells, H. 2016. Driverless cars in the distance. Lexology, 15 September 2016. Available:
http://www.lexology.com/library/detail.aspx?g=ade37a02-3d1f-4318-8205-6d0608af58c5
[Accessed 2017, April 30].
Lutkevich, B. 2019. What are self-driving cars and how do they work? Available: https://search
enterpriseai.techtarget.com/definition/driverless-car [Accessed 2021, October 15].
Lutz, R.J. & Resek, R.W. 1972. More on testing the Howard-Sheth model of buyer behavior.
Journal of Marketing Research, 9: 344–353.
Machi, L.A. & McEvoy, B.T. 2016. The literature review: Six steps to success. 3rd ed. California:
Corwin Press.
Mackay, N., Mostert, P.G. & Petzer, D.J. 2015. The interrelationship between service quality,
relational benefits, customer satisfaction and behavioural intentions in the South African short-
term insurance industry. The Retail and Marketing Review, 15(1): 44–60.
Madden, T.J., Ellen, P.S. & Ajzen, I. 1992. A comparison of the theory of planed behavior and the
theory of reasoned action. Personality and Social Psychology Bulletin, 18(1): 3–9.
Madzingira, N. 2021. The effects of climate change on South Africa’s insurance industry. FA
NEWS, 21 September 2021. Available: https://www.fanews.co.za/article/short-term-
insurance/15/general/1217/the-effects-of-climate-change-on-south-africa-s-insurance-
industry/32939 [Accessed 2021, October 16].
Maharaj, Y. 2016. Will SA’s short-term insurance industry prosper? Bizcommunity, 30 May 2016.
Available: https://www.bizcommunity.com/Article/196/363/145492.html [Accessed 2017,
March 09].
Mahmood, C.K. & Baharun, R. 2019. Consumer purchase decision models: A review of financial
services context. In The European Proceedings of Social & Behavioural Sciences, 476–489.
Available: https://www.researchgate.net/publication/333176948_Consumer_Purchase_
Decision_Models_A_Review_Of_Financial_Services_Context [Accessed 2021, October 25].
Maluleke, R. 2020. Mid-year population estimates 2020. Available:

http://www.statssa.gov.za/publications/P0302/MYPE 2020 Presentation.pdf [Accessed 2021,
November 03].
391
Mankum, Y. 2019. Purchase influences in motor vehicle insurance. Applied research project, Wits
Business School. Available: http://wiredspace.wits.ac.za/bitstream/handle/10539/28438/ARP
YM 28032019 FV Sakai.pdf?sequence=1&isAllowed=y [Accessed 2020, March 31].
Manning, J. 2020. How new technology is revolutionising motor insurance. Available:

https://www.fleeteurope.com/en/safety/europe/features/how-new-technology-revolutionising-
motor-insurance?a=JMA06&t%5B0%5D=Insurance&t%5B1%5D=Artificialintelligence&
curl=1 [Accessed 2021, October 14].
Del Marmol, T. & Feys, B. 2015. PESTLE analysis: Understand and plan for your business
environment. 50Minutes.com. Available: https://books.google.co.za/books?id=h_uQCg
AAQBAJ&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false
Mårtenssona, P., Fors, U., Wallinc, S.-B., Zander, U. & Nilsson, G.H. 2016. Evaluating research:
A multidisciplinary approach to assessing research practice and quality. Research Policy, 45:
593–603.
Martinussen, L.M., Møller, M., Prato, C.G. & Haustein, S. 2017. How indicative is a self-reported
driving behaviour profile of police registered traffic law offences? Accident Analysis and
Prevention, 99: 1–5.
Masterson, R., Phillips, N. & Pickton, D. 2021. Marketing: An introduction. 5th ed. London: SAGE.
Mastinu, G. & Ploechl, M. 2014. Road and off-road vehicle system dynamics handbook. Boca
Raton: CRC Press.
Masuku, I. 2018. Car insurance premiums: It’s mostly about you. Sowetan Live. 21 June.
Available: https://www.sowetanlive.co.za/business/money/2018-06-21-car-insurance-
premiums-its-mostly-about-you/ [Accessed 2021, October 07].
Mathew, S. & Mash, R. 2019. Exploring the beliefs and attitudes of private general practitioners
towards national health insurance in Cape Town, South Africa. African Journal of Primary
Health Care & Family Medicine, 11(1): 1–10.
Matsunaga, M. 2010. How to factor-analyze your data right: Do’s, don’ts, and how-to’s.
International Journal of Psychological Research, 3(1): 97–110.
Mau, S., Pletikosa, I. & Wagner, J. 2018. Forecasting the next likely purchase events of insurance
customers: A case study on the value of data-rich multichannel environments. International
Journal of Bank Marketing, 36(6): 1125–1144.
Maylor, H., Blackmon, K. & Huemann, M. 2017. Researching business and management. 2nd ed.
London: Palgrave.
392
Mazhar, S.S., Rehman, A. & Din, S.U. 2015. Impact of demographic variables on purchase of E-
insurance in urban areas in India. International Journal of Management, Innovation &
Entrepreneurial Research, 1(1): 7–10.
McDonald, M. & Dunbar, I. 2012. Market segmentation: How to do it and how to profit from it. 4th
ed. New York: John Wiley & Sons.
McDonnell, J.J. & Bartlett, J.L. 2009. Marketing to change public opinion on climate change: A
case study. The International Journal of Climate Change, 1(3): 63–73.
McFarlane, A. 2010. How the UK’s first fatal car accident unfolded. BBC News, 17 August.
Available: http://www.bbc.com/news/magazine-10987606 [Accessed 2017, May 11].
McIntee, H. 2014. Understading the consumer. In Introduction to Marketing. Cape Town: Juta,
64–85.
McLeod, S. 2018. Cognitive dissonance. Available: https://www.simplypsychology.org/cognitive-

dissonance.html [Accessed 2021, October 21].
Meral, H. 2019. The behavioral science on buying insurance. Available:

https://www.insurancethoughtleadership.com/the-behavioral-science-on-buying/ [Accessed
2021, March 25].
Mertens, D.M. 2019. Research and evaluation in education and psychology. 4th ed. Thousand
Oaks: SAGE Publications.
Merviö, M.M. 2015. Management and participation in the public sphere. Hershey: IGI Global.
Middlebrook, S.B. 2020. Personality and related characteristics that affect consumer buying
behavior. Available: https://toughnickel.com/industries/Buyer-Characteristics-and-Consumer-
Behavior [Accessed 2021, October 15].
Milner, T. & Rosenstreich, D. 2013. A review of consumer decision-making models and

development of a new model for financial services. Journal of Financial Services Marketing,
18(2): 106–120.
MiWay. 2019, 6 Factors that influence your car insurance premium. [Blog] Miway, 14 May 2019.
Available: https://www.miway.co.za/blog/short-term-insurance/6-factors-that-influence-your-
car-insurance-premium [Accessed 2021, October 07].
Mlambo, S. 2020. R176 billion - that’s the cost of road crashes to South African economy. IOL.
26 August 2020. Available: https://www.iol.co.za/news/south-africa/gauteng/r176-billion-thats-
the-cost-of-road-crashes-to-south-african-economy-436031d2-1e6c-43a5-9c8c-
fdcf9db54eb9 [Accessed 2021, September 08].
393
Moffatt, M. 2019. Economic utility: The pleasure of products. Available: https://www.thoughtco.
com/definition-of-utility-1148048 [Accessed 2021, October 21].
Mohamed, M. & Bromfield, N.F. 2017. Attitudes, driving behavior, and accident involvement
among young male drivers in Saudi Arabia. Transportation Research Part F, 47: 59–71.
Available: http://dx.doi.org/10.1016/j.trf.2017.04.009 [Accessed 2021, January 28].
Moig, C. 2015. SA car insurance in crisis. News24, 13 August 2015. Available:

http://www.news24.com/SouthAfrica/Local/South-Coast-Fever/SA-car-insurance-in-crisis-
Mojekwu, J.N. & Ibekwe, U.A. 2015. Socio-economic indicators affecting life insurance business
in Nigeria. Journal of Emerging Trends in Economics and Management Sciences, 6(8): 340–
344.
Mokhlis, S. & Salleh, H.S. 2009. Consumer decision making styles in malaysia: An exploratory
study of gender differences. European Journal of Social Sciences, 10(4): 574–584.
Mokhothu, M. 2020. The top five reasons why people don’t like to pay to insure their car. The
South African, 19 April 2020. Available: https://www.thesouthafrican.com/news/finance/car-
insurance-the-top-five-reasons-why-people-dont-like-to-pay-to-insure-their-ride/ [Accessed
2021, February 02].
Momentum. 2021. Vision and mission statements. Available: https://www.momentum.co.za/

for/you/about-momentum/vision-and-mission [Accessed 2021, October 07].
Moodley-Isaacs, N. 2013. How to ... cut premiums for young drivers. News24, 30 June 2013.
Available: http://www.news24.com/Archives/City-Press/How-To-Cut-premiums-for-young-
drivers-20150429 [Accessed 2017, April 25].
Moodley, A.J. 2019. Digital transformation in South Africa’s short-term insurance sector:
Traditional insurers’ responses to the internet of things (IoT) and insurtech. The African
Journal of Information and Communication (AJIC). 24:1–16.
Moodley, C. 2021. Post pandemic consumer behaviour and what it means for your insurance and
personal financial planning. [Blog] FIA, 25 March 2021. Available:
https://www.fia.org.za/blog/post-pandemic-consumer-behaviour-and-what-it-means-for-your-
insurance-and-personal-financial-planning/ [Accessed 2021, October 16].
Moodley, C., Forbes, S. & Olivier, E. 2020. Top trends that will shape the insurance sector in the
next decade. IOL, 5 March. Available: https://www.iol.co.za/personal-finance/insurance/top-
trends-that-will-shape-the-insurance-sector-in-the-next-decade-44100469 [Accessed 2021,
August 16].
394
Moon, C. 2021. Personal liability insurance: How much does it cost? Available: https://www.
valuepenguin.com/homeowners-insurance/personal-liability-insurance [Accessed 2021,
August 16].
Moorcraft, B. 2021. The proven ROI for business text messaging in insurance. Insurance
Business America. 27 July. Available: https://www.insurancebusinessmag.com/us/news/
technology/the-proven-roi-for-business-text-messaging-in-insurance-292546.aspx [Accessed
2021, November 02].
Mora, M., Gelman, O., Steenkamp, A. & Raisinghani, M.S. 2012. Innovations and Philosophies in
Software Systems Engineering and Information Systems. Hershey: IGI Global.
Moyo, A. 2016. Insurers to miss R115bn digital opportunity. ITWeb, 1 August 2016. Available:
https://www.itweb.co.za/content/JKjlyrvw9AZMk6am [Accessed 2021, October 16].
Mpinganjira, M., Dos Santos, M.A.O., Botha, E., Du Toit, D., Erasmus, A.C., Maree, T. & Mugobo,
V.V. 2013. Consumer behaviour: South African psychology and marketing applications. Cape
Town: Oxford.
Muia, F. 2017. Effect of competitive strategies on the performance of insurance companies in

Kenya. Unpublished Master’s dissertation. United States International University-Africa,
Nairobi, Kenya.
Murugesan, G. 2012. Principles of management. New Delhi: Laxmi Publications.
Nagel, E. 2021. Does the colour of your car affect your insurance premium? [Blog] Gumtree, 23
July 2021. Available: https://blog.gumtree.co.za/does-the-colour-of-your-car-affect-your-
insurance-premium-2/ [Accessed 2021, October 07].
Naghi, L.E. 2015. Impact of financial crisis on the decision making process for insurance
consumers. Theoretical and Applied Economics, 22(1): 283–292.
Naik, T. 2020. The benefits of using an insurance broker. Available: https://www.primesure.co.nz/

benefits-insurance-broker/ [Accessed 2021, August 16].
Nash, L., Boehmer, G., Wireman, M. & Hillaker, A. 2017. Addressing cybersecurity challenges in
the future of mobility: Addressing cyber risk in self-driving cars and beyond. Available:
https://www2.deloitte.com/content/dam/Deloitte/be/Documents/strategy/Securing The Future
Of Mobility.pdf [Accessed 2021, October 15].
Natarajan, R. 2018. Direct-to-customer models in the insurance industry. [Blog] Mindtree, 22 June
2018. Available: https://www.mindtree.com/insights/blog/direct-customer-model-changing-
insurers-business-ways [Accessed 2021, October 13].
395
Nayeem, T. & Casidy, R. 2015. Australian consumers’ decision-making styles for everyday
products. Australasian Marketing Journal. 23(1): 67–74.
Von Neumann, J. & Morgenstern, O. 1944. Theory of games and economic behavior. New Jersey:
Princeton University Press.
Nguyen, T. 2019. History of self-driving cars. Available: https://www.thoughtco.com/history-of-

self-driving-cars-4117191 [Accessed 2021, October 01].
Nicol, E., Hanmer, L.A., Mukumbang, F.C., Basera, W., Zitho, A. & Bradshaw, D. 2021. Is the
routine health information system ready to support the planned national health insurance
scheme in South Africa? Health Policy and Planning, 36(5): 639–650.
Nicosia, F.M. 1966. Consumer decision processes: Marketing and advertising implications. New
Jersey: Prentice-Hall.
Nieuwenhuizen, C. 2015. Business magagement for entrepreneurs. 3rd ed. Cape Town: Juta.
Niranjan, N. & Rohita Kumar, M. 2014. Evolution of insurance sector: A global perspective.
Management Review : An International Journal, 9(2): 4.
Noonan, K. 2019. What does the future hold for self-driving cars? Available:
https://www.fool.com/investing/what-does-the-future-hold-for-self-driving-cars.aspx
North, E. 2021. 4 ways personal short-term insurance may evolve in 2021. IOL, 15 January 2021.
Available: https://www.iol.co.za/personal-finance/insurance/4-ways-personal-short-term-
insurance-may-evolve-in-2021-1edab377-0144-4bf2-a8dd-b3939c633034 [Accessed 2021,
October 04].
Nulty, D.D. 2008. The adequacy of response rates to online and paper surveys: What can be
done? Assessment & Evaluation in Higher Education, 33(3): 301–314.
O’Very, P. 2018. 3 Ways self-driving cars will affect the insurance industry. Available:
https://venturebeat.com/2018/01/26/3-ways-self-driving-cars-will-affect-the-insurance-
industry/ [Accessed 2021, October 15].
Oliveira, A. 2007. A discussion of rational and psychological decision-making theories and

models: The search for a cultural-ethical decision-making model. EJBO Electronic Journal of
Business Ethics and Organization Studies, 12(2): 12–17.
Omarjee, L. 2016. Insurance sector vital for sustainable economy - Treasury. News24, 25 July
2016. Available: http://www.fin24.com/Economy/insurance-sector-vital-for-sustainable-
economy-treasury-20160725 [Accessed 2021, October 04].
396
Omotoyinbo, C., Worlu, R. & Ogunnaike, O. 2017. Consumer behaviour modelling: A myth or
heuristic device? Perspectives of Innovations, Economics & Business, 17(2): 101–119.
Onafalujo, A.K., Abass, O.A. & Dansu, S.F. 2011. Effects of risk perception on the demand for
insurance: Implications on Nigerian road users. Journal of Emerging Trends in Economics and
Management Sciences, 2(4): 285–290.
OECD (Organisation for Economic Co-operation and Development). 2015. New approaches to
SME and entrepreneurship financing: Broadening the range of instruments. Istanbul.
Available: https://www.oecd.org/cfe/smes/New-Approaches-SME-full-report.pdf [Accessed
2021, November 16].
OECD. 2020. Global insurance market trends 2020. Available: https://www.oecd.org/

daf/fin/insurance/Global-Insurance-Market-Trends-2020.pdf [Accessed 2021, June 24].
Oxford Business Group. 2021. South African insurance sector undergoes a period of
transformation. Available: https://www.oxfordbusinessgroup.com/overview/opportunity-and-
challenge-sector-midst-far-reaching-transformations [Accessed 2021, November 01].
Oxford English and Spanish Dictionary. 2021. Meaning of risk in English. Available:
https://en.oxforddictionaries.com/definition/risk [Accessed 2021, October 21].
Painter, M. 2016. Human capital trends in the insurance industry. Available:

https://www2.deloitte.com/content/dam/Deloitte/no/Documents/financial-services/human-
capital-trends-in-the-insurance-industry-2016.pdf [Accessed 2017, April 29].
Pallant, J. 2016. SPSS survival manual. 6th ed. New York: Open University Press.
Panwar, D., Anand, S., Ali, F. & Singal, K. 2019. Consumer decision making process models and
their applications to market strategy. International Management Review, 15(1): 36–44.
Parkin, N., Jooste, C. & Thorpe, B. 2019. Social mania: The impact of likes, posts, and swipes on
insurance. RGA, 17 July 2019. Available: https://www.rgare.com/knowledge-center/media/
articles/social-mania-the-impact-of-likes-posts-and-swipes-on-insurance [Accessed 2021,
October 13].
Parry, S. n.d. Fit indices commonly reported for CFA and SEM. Available: https://dokumen.tips/
documents/fit-statistics-commonly-reported-for-cfa-and-sem-parry-kline-suggests-that-at-
a.html [Accessed 2021, May 12].
Parumasur, S.B. & Roberts-Lombard, M. 2014. Consumer behaviour. 3rd ed. Cape Town: Juta.
Pasian, B. 2015. Designs, methods and practices for research of project management. Farnham:
Gower Publishing Company.
397
Patten, M.L. 2017. Proposing empirical research: A guide to the fundamentals. 5th ed. Abingdon:
Taylor & Francis.
Patten, M.L. & Newhart, M. 2018. Understanding research methods: An overview of the
essentials. 10th ed. New York: Taylor & Francis.
Pearson, V. 2016a. SAIA Bulletin May 2016. Available: https://www.saia.co.za/index.php?id=232

[Accessed 2021, September 30].
Pearson, V. 2016b. The role that the short-term insurance industry can play in economic
transformation in South Africa. COVER Publications, 24 May 2016. Available:
https://www.cover.co.za/the-role-that-the-short-term-insurance-industry-can-play-in-
economic-transformation-in-south-africa/ [Accessed 2017, March 09].
Pearson, V. 2017. SAIA adds its voice to concerns on S&P ratings downgrade. MoneyMarketing,
4 April 2017. Available: https://moneymarketing.co.za/saia-adds-its-voice-to-concerns-on-sp-
ratings-downgrade/ [Accessed 2017, April 30].
Pearson, R. & Yoneyama, T. 2015. Corporate forms and organisational choice in international
insurance. Oxford: Oxford University Press.
Peng, M.W. 2017. Global business. 4th ed. Boston: Cengage Learning.
Percy, L. & Rosenbaum-Elliott, R.H. 2016. Strategic advertising management. 5th ed. New York:
Oxford University Press.
Perera, R. 2017. The PESTLE Analysis. Nerdynaut.
Perkmann, M. & Salter, A. 2012. How to create productive partnerships with universities.
Available: https://sloanreview.mit.edu/article/how-to-create-productive-partnerships-with-
universities/ [Accessed 2021, November 16].
Perreault, W.D. & McCarthy, E.J. 2002. Basic marketing: A global−managerial approach. 14th ed.
New York: McGraw Hill.
Peyper, L. 2021. 70% of South African motorists have no insurance. News24, 2 July 2021.
Available: https://www.news24.com/citypress/business/70-of-south-african-motorists-have-
no-insurance-20210702 [Accessed 2021, September 27].
Phillip, X. 2020. South Africa: 'Retail is in for a tough period’ as consumers pinch pennies. The
Africa Post, 31 July 2020. Available: https://www.theafricareport.com/35785/south-africa-
retail-is-in-for-a-tough-period-as-consumers-pinch-pennies/ [Accessed 2021, June 25].
Phillips, S. 2011. Men and women Drivers: The gender divide. [Blog] PsychCentral, 27 July 2011.
Available: https://blogs.psychcentral.com/healing-together/2011/07/men-and-women-drivers-
the-gender-divide/ [Accessed 2017, April 25].
398
Piacentini, M.G. & Cui, C.C. 2012. Multicultural perspectives in customer behaviour. New York:
Routledge.
Pirgmaier, E. 2021. The value of value theory for ecological economics. Ecological Economics,
179(C).
Du Plessis, S. 2017. Short-term insurance claim trends and unforeseen risks. Insurance Gateway,
16 January 2017. Available: http://www.insurancegateway.co.za/ShorttermProfessionals/
PressRoom/ViewPress/Irn=14660&URL=Shortterm+Insurance+claim+trends+and+unforese
en+risks+1#.WQZmh_l96pq [Accessed 2021, October 16].
Du Plessis, L. & Roberts-Lombard, M. 2013. Customer loyalty in the South African long-term
insurance industry. Acta Commercii, 13(1): 1–8.
Du Plessis, P.J., Strydom, J.W. & Jooste, C.J. 2012. Marketing management. 6th ed. Cape Town:
Juta.
Porral, C.C. & Stanton, J.L. 2017. Principles of marketing. Madrid: ESIC Editorial.
Prakash, G., Singh, P.K. & Yadav, R. 2018. Application of consumer style inventory (CSI) to
predict young Indian consumer’s intention to purchase organic food products. Food Quality
and Preference, 68:90–97.
Prasad, R.K. & Jha, M.K. 2014. Consumer buying decisions models: A descriptive study.
International Journal of Innovation and Applied Studies, 6(3): 335-351.
Preston, K. 2017. Big Data transforming insurance. ITWeb, 1 March 2017. Available:
http://www.itweb.co.za/index.php?option=com_content&view=article&id=159781 [Accessed
2021, October 14].
Pride, W.M. & Ferrell, O.C. 2021. Foundations of marketing. 9th ed. Boston: Cengage Learning.
Pride, W.M., Ferrell, O.C., Lukas, B.A., Schembri, S. & Niininen, O. 2015. Marketing principles.
2nd ed. Melbourne: Cengage Learning Australia.
Pride, W.M., Hughes, R.J. & Kapoor, J.R. 2017. Foundations of business. 5th ed. Boston:
Cengage Learning.
Pride, W.M., Ferrell, O.C., Lukas, B.A., Schembri, S., Niininen, O. & Casidy, R. 2018. Marketing
principles with student resource access 12 months. 3rd ed. Victoria: Cengage Learning.
Progressive. 2021. How Progressive began & evolved. Available:

https://www.progressive.com/about/history/ [Accessed 2021, October 01].
PwC. 2014. Africa insurance trends. Available: https://www.pwc.co.za/en/assets/pdf/south-

african-insurance-2014.pdf [Accessed 2021, October 15].
399
PwC. 2015. Insuring African growth: Insurance industry analysis. Available: https://www.pwc.
co.za/en/assets/pdf/insurance-industry-analysis-2015.pdf [Accessed 2016, February 06].
PwC. 2016. Top issues: An annual report. Available: https://www.pwc.com/us/en/ insurance/

publications/assets/pwc-top-issues-the-insurance-industry-2016.pdf [Accessed 2017, June
14].
PwC. 2018. Ready and willing African insurance industry poised for growth. Available:
https://www.pwc.co.za/en/publications/insurance-survey-south-africa.html [Accessed 2018,
October 19].
RAC Foundation. 2021. General facts and figures about roads and road use. Available:
https://www.racfoundation.org/motoring-faqs/mobility#a10 [Accessed 2021, October 12].
Raghvan, A. 2021. Research and evaluation in education and psychology. Burlington: Society
Publishing.
Ragunathan, V.M. 2016. A study on consumer behavior towards life insurance products with
refernce to Idbi Fortis Company Limited, Dindigul District. Indian Journal of Applied Research,
6(8): 182–185.
Rahman, A. 2011. Insurance, profit companies, the principle basic, denial of insurance and
history. [Blog] Insurance Dot Com, 2011. Available: http://insurancedotcom2012.
blogspot.com/2012/04/insurance-profit-companies-principle.html [Accessed 2021, December
12].
Raiguru, S. & Misra, N.R. 2017. An exploratory research on buying behavior model in insurance
sector. Indian Journal of Applied Research. 7(2): 663–665.
Rajagopal. 2019. Contemporary marketing strategy: Analyzing consumer behavior to drive

managerial decision making. Cham: Springer.
Ralph, M. & Rankin, W. 2021. Decoder: The slave insurance market. Available:
http://foreignpolicy.com/2017/01/16/decoder-slave-insurance-market-aetna-aig-new-york-life/
[Accessed 2021, November 11].
Ramaswamy, V.S. & Namakumari, S. 2013. Marketing management. 5th ed. New Delhi: McGraw-
Hill Education.
Ramlall, I. 2017. Applied structural equation modelling for researchers and practitioners: Using R
and Stata for behavioural research. Bingley: Emerald Group Publishing.
Ranft, F., Adler, M., Diamond, P., Guerrero, E. & Laza, M. 2016. Freeing the road: Shaping the
future for autonomous vehicles. London. Available: http://policynetwork.org/wp-content/
uploads/2017/09/Freeing-the-road.pdf [Accessed 2021, October 15].
400
Ranger, S. 2020. What is the IoT? Everything you need to know about the Internet of Things right
now. Available: https://www.zdnet.com/article/what-is-the-internet-of-things-everything-you-
need-to-know-about-the-iot-right-now/ [Accessed 2021, October 14].
Rankoe, V. 2014. Are we ready for the driverless car? FA NEWS, 1 October 2014. Available:
https://www.fanews.co.za/article/magazine-archives-fanews-fanuus/60/short-term/1317/are-
we-ready-for-the-driverless-car/19369 [Accessed 2021, October 04].
Rašković, M., Ding, Z., Hirose, M., Žabkar, V. & Fam, K.S. 2020. Segmenting young-adult
consumers in East Asia and Central and Eastern Europe – The role of consumer
ethnocentrism and decision-making styles. Journal of Business Research, 108: 496–507.
Reader, G. 2016. The power of alliances: Partnering for growth in the insurance sector. Available:
https://assets.kpmg/content/dam/kpmg/pdf/2016/03/power-of-alliance.pdf [Accessed 2021,
March 18].
Regassa, M.G. 2018. Assessment of performance of motor insurance: A case of Ethiopian

insurance corporation. Unpublished Master’s dissertation. St. Mary’s University, Addis Ababa,
Etiopia.
Reilly, A. 2021. Introducing fashion theory: From androgyny to zeitgeist. London: Bloomsbury
Publishing.
Reyes, M. 2020. Consumer behavior and marketing. London: IntechOpen.
Reynolds, C. & Faye, M. 2015. Roots and developing structure of critical illness insurance.
PartnerRe. 28 July 2015. Available: https://partnerre.com/opinions_research/roots-and-
developing-structure-of-critical-illness-insurance/.
Richter, A., Schiller, J. & Schlesinger, H. 2014. Behavioral insurance: Theory and experiments.
Journal of Risk and Uncertainty, 48(2): 85–96.
Road Accident Fund. 2016. RAF annual report 2015/16. Available: online.anyflip.com/
dqjy/dyjv/#p=4 [Accessed 2016, September 15].
Road Accident Fund. 2021. Profile. Available: http://www.raf.co.za/about-us/pages/profile.aspx

[Accessed 2021, June 25].
Roberts, J. 2016a. Short-term insurance - the broker model is alive and well. MoneyMarketing.
20 April 2016. Available: https://www.moneymarketing.co.za/short-term-insurance-the-broker-
model-is-alive-and-well/ [Accessed 2017, May 01].
Roberts, J. 2016b. Weak rand poses real threat for the insurance industry. MoneyMarketing, 13
May 2016. Available: https://www.b2bcentral.co.za/weak-rand-poses-real-threat-for-the-
insurance-industry/ [Accessed 2021, October 15].
401
Robinson, B. 2011. London’s burning: The Great Fire. BBC. 29 March 2011. Available:
http://www.bbc.co.uk/history/british/civil_war_revolution/great_fire_01.shtml [Accessed 2017,
January 26].
Roller, M.R. & Lavrakas, P.J. 2015. Applied qualitative research design: A total quality framework
approach. New York: The Guilford Press.
Rosenberg, A. 2018. Philosophy of social science. 5th ed. New York: Routledge.
Ross, A. & Willson, V.L. 2017. Basic and advanced statistical tests: Writing results sections and
creating tables and figures. Rotterdam: Sense Publishers.
Ruel, E., Wagner, W.E. & Gillespie, B.J. 2016. The practice of survey research: Theory and
applications. Thousand Oaks: SAGE Publications.
S&P Global. 2021. How COVID-19 has changed the insurance industry. Available: https://www.
spglobal.com/en/research-insights/featured/how-covid-19-has-changed-insurance [Accessed
2021, June 24].
Sagiroglu, S. & Sinanc, D. 2013. Big Data: A review. In Collaboration Technologies and Systems
(CTS), 2013 International Conference. San Diego: IEEE Xplore, 42–47. Available:
https://ieeexplore.ieee.org/document/6567202 [Accessed 2017, June 06].
Salleh, F., Noor Hazzarita Ag Kassim, D., Shukri Yazid, A. & Rashid, N. 2018. Consumer
behaviour and insurance claim fraud in Malaysia. International Journal of Academic Research
in Business and Social Sciences, 8(12): 586–598.
Samson, A. 2016. The behavioral economics guide 2016. London: Behavioral Science Solutions.
Available: http://eprints.lse.ac.uk/66934/ [Accessed 2021, October 21].
Sandell, K. 2019. Olfactory cues and purchase behavior: Consumer characteristics as

moderators. European Journal of Marketing, 53(7): 1378–1399.
Sanders, J. & Tyson, A. 2014. 5 Reasons people change auto insurance carriers. InsuraMatch,
17 November 2014. Available: https://www.insuramatch.com/learning-center/5-reasons-
people-change-auto-insurance-carriers [Accessed 2021, October 13].
Santam. 2014. Car insurance. Available: https://www.santam.co.za/products/personal/car-

insurance/ [Accessed 2021, June 29].
Santam. 2015. The problem with car insurance. Moneyweb, 11 December 2015. Available:
http://www.moneyweb.co.za/news/economy/the-problem-with-car-insurance/ [Accessed
2021, October 13].
402
Santenac, I., Majkowski, E., Manchester, P. & Peters, G. 2021. 2021 Global insurance outlook.
Available: https://assets.ey.com/content/dam/ey-sites/ey-com/en_gl/topics/insurance/
insurance-pdfs/ey-2021-global-insurance-outlook.pdf [Accessed 2021, June 24].
Santora, J. 2021. 100 Influencer marketing statistics for 2021. Influencer Marketing Stats. 24
February 2021. Available: https://influencermarketinghub.com/influencer-marketing-statistics/
Saria SOC Ltd. n.d. About Sasria. Available: https://www.sasria.co.za/ [Accessed 2021,
September 27].
Saunders, M., Lewis, P. & Thornhill, A. 2012. Research methods for business students. 6th ed.
London: Pearson Education.
Saunders, M.N.K., Lewis, P. & Thornhill, A. 2019. Research methods for business students. 8th
ed. New York: Pearson Education.
Saurav, S. 2020. Understanding needs, wants and demands in marketing world. LinkedIn. 2 April
2020. Available: https://www.linkedin.com/pulse/understanding-needs-wants-demands-
marketing-world-sumit-saurav/ [Accessed 2021, October 18].
Savage, L.J. 1954. The foundations of statistics. New York: John Wiley & Sons.
Schaerer, M., Wanner, H. & Grinyer, C. 2011. Transforming the insurance industry to increase
customer relevance. Available: http://www.cisco.com/c/dam/en_us/about/ac79/docs/fs/
Insurance_Customer_Relevance_0301_IBSG.pdf [Accessed 2021, November 11].
Schiffman, L.G. & Wisenblit, J. 2015. Consumer behavior. 11th ed. New Jersey: Pearson
Education.
Schiffman, L.G. & Wisenblit, J. 2019. Consumer behavior. 12th ed. Essex: Pearson Education.
Schiffman, L., O’Cass, A., Paladino, A. & Carlson, J. 2014. Consumer Behaviour. 6th ed. Frenchs
Forest: Pearson Higher Education.
Schiffman, L.G., Kanuk, L.L. & Hansen, H. 2012. Consumer behaviour: A European outlook. 2nd
ed. New Jersey: Pearson Education.
Schliesser, M. 2014. The impact of technology in the short-term insurance sector. Finweek, 30
April 2014. Available: https://www.discountmags.com/au/magazine/finweek-english-april-30-
2014-digital [Accessed 2021, October 14].
Schmidt, A.P. 2019. The impact of cognitive style, consumer demographics and cultural values
on the acceptance of Islamic insurance products among American consumers. International
Journal of Bank Marketing, 37(2): 492–506.
403
Schoeman, A. 2016. A comprehensive look at car insurance in South Africa. [Blog] Automart, 19
January 2016. Available: http://blog.automart.co.za/2016/01/19/a-comprehensive-look-at-car-
insurance-in-south-africa/ [Accessed 2021, October 04].
Schumacker, R.E. & Lomax, R.G. 2016. A beginner’s guide to structural equation modeling. 4th
ed. New York: Routledge.
Segovia-Gonzalez, M.M., Guerrero, F.M. & Herranz, P. 2009. Explaining functional principal
component analysis to actuarial science with an example on vehicle insurance. Insurance:
Mathematics and Economics, 45(2): 278–285.
Sekaran, U. & Bougie, R. 2016. Research methods for business: A skill building approach. 7th
ed. Chichester: John Wiley & Sons.
Sekaran, U. & Bougie, R. 2020. Research methods for business: A skill building approach. 8th
ed. Danvers: John Wiley & Sons.
Senapati, P.S. 2015. Business environment and strategy. New Delhi: Educreation Publishing.
Sethna, Z. & Blythe, J. 2016. Consumer Behaviour. 3rd ed. London: SAGE Publications. DOI:
9781473919129.
Sethna, Z. & Blythe, J. 2019. Consumer behaviour. 4th ed. London: SAGE Publications.
Seymour, K. 2019. Social media: A double-edged sword. Pacific Life Re (London). 8 May 2019.
Available: https://www.pacificlifere.com/home/library/marketing-materials/thought-leadership.
html [Accessed 2021, November 12].
Shinn, L. 2021. Pros and cons of buying car insurance online. Available: https://www.thebalance.
com/pros-and-cons-of-online-car-insurance-527046 [Accessed 2021, August 16].
Sibanda, S. & Sibanda, T. 2016. Financial Education in South Africa: Overview of key initiatives
and actors. Available: https://www.ilo.org/wcmsp5/groups/public/---ed_emp/documents/
publication/wcms_545803.pdf [Accessed 2021, November 03].
Silvia, P. 2020. Select a sample: Little quick fix. Los Angeles: Sage.
Simms, S. 2004. State farm, George Mecherle, and Farm Bureau early history timeline. Available:
http://np1.ilfb.org/history/simms/simms.html [Accessed 2021, October 01].
Simon, H.A. 1959. Theories of decision-making in economics and behavioral science. The
American Economic Review, 49(3): 253–283.
Sin, T.S. & Chee, L.C. 2017. A preliminary study on the relationship between psychographic
factors and the purchase of life insurance. International Journal of Management Studies.
24(1): 1–22.
404
Singh, H. 2021. South African enterprises can’t ignore the risk of cyber attacks: The threat is on
the rise. Available: https://theconversation.com/south-african-enterprises-cant-ignore-the-
risk-of-cyber-attacks-the-threat-is-on-the-rise-166133 [Accessed 2021, October 15].
Sinha, S. 2019. What is an endowment policy and when should you go for it. The Economic Times,
21 May 2019. Available: https://economictimes.indiatimes.com/tdmc/your-money/what-is-an-
endowment-policy-and-when-should-you-go-for-it/articleshow/48465113.cms [Accessed
2021, November 11].
Sithole, S. 2021. The Protection of Personal Information Act – new responsibilities for insurers.
FA NEWS, 17 March 2021. Available: https://www.fanews.co.za/article/compliance-
regulatory/2/general/1082/the-protection-of-personal-information-act-new-responsibilities-for-
insurers/31457 [Accessed 2021, October 16].
Slovic, P. 1987. Perception of risk. Science, 236: 280–285.
Slovic, P., Fischhoff, B. & Lichtenstein, S. 1977. Behavioral decision theory. Annual Review of
Psychology, 28: 1–39.
Smith, A. 2019. Consumer behaviour and analytics. New York: Routledge.
Smith, J. 2021. 8 Things to know about the history of auto insurance. [Blog] USAgencies, 3 June
2021. Available: https://www.usagencies.com/blog/car-insurance/auto-insurance-history/
Snook, A. 2019. 3 Ways to use social media in insurance fraud investigations. Available:
https://www.i-sight.com/resources/3-ways-to-use-social-media-in-insurance-fraud-
investigations/ [Accessed 2021, October 14].
Snyders, R. 2016. 2016: The insurance industry at a glance. Insurance Gateway, 17 March 2016.
Available: https://insurancegateway.co.za/ShorttermConsumers/PressRoom/ViewPress/Irn=
12395&URL=2016+The+insurance+industry+at+a+glance+1#.YY9a_2BByUk [Accessed
2021, October 13].
Sodiq, B.O. 2020. Factors influencing the buying behaviour of millennials towards insurance
products in the Nigerian insurance market. Journal of Economics and Management Research.
9:86–101.
Soliman, A., Alhajyaseen, W., Alfar, R. & Alkaabi, I. 2018. Changes in driving behavior across
age cohorts in an Arab culture: The case of State of Qatar. In Procedia Computer Science, V
130: 652–659.
Solomon, M., Russell-Bennett, R. & Previte, J. 2013. Consumer behaviour: Buying, having, being.
3rd ed. Frenchs Forest: Pearson Australia.
405
Solomon, M.R., Hogg, M.K., Askegaard, S. & Bamossy, G. 2019. Consumer behaviour: A
European perspective. 7th ed. Harlow: Pearson Higher Education.
Sonkiya, S. 2020. Vision, mission, objectives, strategies and tactics. [Blog] iZenBridge, 4
December 2020. Available: https://www.izenbridge.com/blog/vision-mission-objectives-
strategies-and-tactics/ [Accessed 2021, October 07].
Souiden, N. & Jabeur, Y. 2015. The impact of Islamic beliefs on consumers’ attitudes and
purchase intentions of life insurance. International Journal of Bank Marketing, 33(4): 423–441.
South African Insurance Association. 2021. About us. Available: http://www.saia.co.za/ [Accessed
2021, October 04].
South African Insurance Association. n.d. Home owners insurance: A guide to help you purchase
the right cover. Available: https://www.saia.co.za/index.php?id=2&search=household+
insurance&simplesearch_offset=10 [Accessed 2021a, August 16].
South African Insurance Association. n.d. Motor insurance: A guide to help you purchase the right
cover. Available: file:///C:/Users/marxn/Downloads/motinsz (1).pdf [Accessed 2021, October
07].
Spaan, E., Mathijssen, J., Tromp, N., Mcbain, F., Ten Have, A. & Baltussen, R. 2012. The impact
of health insurance in Africa and Asia: a systematic review. Bulletin of the World Health
Organization, 90: 685–692.
Sproles, G.B. & Kendall, E.L. 1986. A methodology for profiling consumers’ decision-making
styles. The Journal of Consumer Affairs. 20(2): 267–279.
Squaremouth. 2021. What does travel insurance cover? Available: https://www.squaremouth.

com/travel-advice/what-does-travel-insurance-cover [Accessed 2021, August 16].
Staffaroni, S. 2019. How to define customer satisfaction and measure it. Available:
https://www.getfeedback.com/resources/cx/how-to-define-customer-satisfaction-and-
measure-it/ [Accessed 2021, October 19].
Stankevich, A. 2017. Explaining the consumer decision-making process: Critical literature review.
Journal of International Business Research and Marketing, 2(6): 7–14.
Statistics South Africa. 2017a. Victims of crime survey 2016/17. Available: http://www.statssa.
gov.za/publications/P0341/P03412016.pdf [Accessed 2018, October 05].
Statistics South Africa. 2017b. Victims of crime survey: 2016/17. Available: https://www.statssa.
gov.za/publications/P0341/P03412016.pdf [Accessed 2021, October 12].
Stehlik-Barry, K. & Babinec, A.J. 2017. Data Analysis with IBM SPSS Statistics. Birmingham:
Packt Publishing.
406
Still, L. & Stokes, G. 2016. Short term insurance in South Africa 2016. Johannesburg: S and S
Analytica.
Stokes, G. 2010. A sensible approach to third party insurance. FA NEWS July 2010. Available:
https://www.fanews.co.za/article/short-term-insurance/15/motor/1023/a-sensible-approach-
to-third-party-insurance/8138 [Accessed 2017, April 19].
Stokes, G. 2012a. Another take on the direct insurer debate. FA NEWS, 5 September 2012.
Available: http://www.fanews.co.za/article/intermediaries-brokers/7/general/1227/another-
take-on-the-direct-insurer-debate/12488 [Accessed 2021, October 13].
Stokes, G. 2012b. Insurer of strike perils braces for flood of claims. FA NEWS,16 October 2012.
Available: https://www.fanews.co.za/article/short-term-insurance/15/general/1217/insurer-of-
strike-perils-braces-for-flood-of-claims/12693 [Accessed 2021, October 13].
Stokes, G. 2012c. A different take on short-term insurance risk. FA NEWS. 7 February 2012.
Available: https://www.fanews.co.za/article/compliance-regulatory/2/saia-south-african-
insurance-association/1081/a-different-take-on-short-term-insurance-risk/11427 [Accessed
2021, October 13].
Stokes, G. 2021a. Insurers weigh in on the car parts debate. IOL, 13 April 2021. Available:
https://www.iol.co.za/personal-finance/insurance/insurers-weigh-in-on-the-car-parts-debate-
232bfaf2-9a6c-4606-ba96-9fc8f330843d [Accessed 2021, September 27].
Stokes, G. 2021b. Insurance Ombudsman featured article: Your clients get another layer of
protection. FA NEWS. 20 July 2021. Available: https://www.fanews.co.za/article/compliance-
regulatory/2/insurance-ombudsman/1424 [Accessed 2021, October 04].
Stokes, G. 2021c. A decade perfecting behaviour-based insurance. FA NEWS. 2 June 2021.

Available: https://www.fanews.co.za/article/talked-about-features/25/featured-story/1147/a-
decade-perfecting-behaviour-based-insurance/32024 [Accessed 2021, October 13].
Stokes, P. & Wall, T. 2014. Research methods. London: Palgrave Macmillan.
Stoner, J.A.F. & Wankel, C. 2012. Managing climate change business risks and consequences.
New York: Macmillan Publishers Limited.
Strydom, J. 2015. Entrepreneurship and how to establish your own business. 5th ed. Cape Town:
Juta.
Strydom, J.W. 2014. Introduction to marketing. 5th ed. Cape Town: Juta.
Strydom, J.W. 2018. Principles of business management. 4th ed. Cape Town: Oxford University
Press.
407
Strydom, S., Naidoo, E. & Botha, L. 2014. New horizons: 2014 Annual review. Available:
https://www.readkong.com/page/2014-annual-review-new-horizons-5819695 [Accessed
2021, October 16].
Suter, J., Duke, C., Harms, A., Joshi, A., Rzepecka, J., Lechardoy, L., Hausemer, P., Wilhelm,
C., et al., 2017. Study on consumers’ decision making in insurance services: A behavioural
economics perspective. European Commission, Luxembourg. Available:
https://ec.europa.eu/info/sites/default/files/finalreport_en.pdf [Accessed 2021, October 27].
Sutherland, C. 2015. Cybercrime in South Africa on the rise. FA NEWS. 25 February 2015.
Available: https://www.fanews.co.za/article/short-term-insurance/15/general/1217/cybercrime
-in-south-africa-on-the-rise/17479 [Accessed 2017, April 30].
Tabachnick, B. & Fidell, L. 2019. Using multivariate statistics. Boston: Pearson Education.
Tadman, J. 2017. PoPI: Reinforcing the value of digital ethics? Insurance Gateway, 6 March 2017.
Available: http://www.insurancegateway.co.za/ShorttermProfessionals/PressRoom/View
Press/Irn=15055&URL=PoPI+reinforcing+the+value+of+digital+ethics+1#.WUjT5-t96po
Takemura, K. 2014. Behavioral decision theory. Tokyo: Springer Japan.
Takemura, K. 2021. Escaping from bad decisions: A behavioral decision-theoretic perspective.

Cambridge: Elsevier.
Taylor, D. 2021. What is Big Data? Introduction, types, characteristics and examples. Available:
https://www.guru99.com/what-is-big-data.html [Accessed 2021, October 14].
Taylor, J. & Gutman, J. 1974. A reinterpretation of Farley and Ring’s test of the Howard-Sheth
model of buyer behavior. ACR North American Advances, 1: 438–446.
Techasurin, W., Nuangjamnong, C. & Dowpiset, K. 2020. Factors influencing consumer

purchasing decision of used car insurance in Bangkok. SSRN Electronic Journal, 1–10.
Teixeira, N., Pereira, H.G. & Dionísio, P. 2018. Online consumer generated content it’s for real!
The rise of social influence marketing. The Business and Management Review, 9(3): 358–
366.
Tendhar, C., Singh, K. & Jones, B.D. 2017. Using the domain identification model to study major
and career decision-making processes. European Journal of Engineering Education, 43(2):
235–246.
The Economic Times. 2021. Definition of “Reinsurance”. Available: http://economictimes.

indiatimes.com/definition/reinsurance [Accessed 2021, July 23].
408
Thomas, K.J. & Loughran, T.A. 2018. Rational choice and prospect theory. Encyclopedia of
Criminology and Criminal Justice. Available: https://link.springer.com/referenceworkentry/
10.1007%2F978-1-4614-5690-2_399 [Accessed 2021, October 22].
Thompson, W. 2021. How insurance began: 3000 Years of history. Available: http://wsrinsurance.
com/how-insurance-began-3000-years-of-history/ [Accessed 2021, November 11].
Thoyts, R. 2010. Insurance theory and practice. New York: Routledge.
Toews, R. 2016. The biggest threat facing connected autonomous vehicles is cybersecurity.
TechCrunch, 25 August 2016. Available: https://techcrunch.com/2016/08/25/the-biggest-
threat-facing-connected-autonomous-vehicles-is-cybersecurity/ [Accessed 2021, October 15].
Tomczak, T., Reinecke, S. & Kuss, A. 2018. Strategic marketing: Market-oriented corporate and
business unit planning. Wiesbaden: Springer Gabler.
Trussell, M. & Reader, G. 2015. A new world of opportunity. Available: https://assets.kpmg.com/

content/dam/kpmg/pdf/2016/01/the-insurance-innovation-imperative.pdf [Accessed 2017,
May 31].
Tsai, Y.-L. & Honka, E. 2019. Non-informational advertising informing consumers: How
advertising affects consumers’ decision-making in the U.S. auto insurance industry. Available:
https://www.msi.org/wp-content/uploads/2020/06/MSI_Report_19-128-1.pdf [Accessed 2021,
July 01].
Tshidi, D. & Sithole, A. 2016. 2016 Annual report. Available: https://www.masthead.co.za/wp-

content/uploads/2016/11/FSB-Annual-Report-2016.pdf [Accessed 2021, October 04].
Tversky, A. & Kahneman, D. 1991. Loss aversion in riskless choice: A reference dependent
model. The Quarterly Journal of Economics, 106: 1039–1061.
Upreti, V., Adams, M. & Jia, Y. 2021. Risk management and the cost of equity: Evidence from the
United Kingdom’s non-life insurance market. The European Journal of Finance: 1–20.
Van der Kolk, D.-J. 2021. Motor vehicles. Available: https://www.vantraa.nl/en/expertise/our-

expertise/insurance-liability/motor-vehicles/ [Accessed 2021, June 29].
Van der Walt, C. 2016. Everything you need to know about car insurance. W24, 30 March 2016.
Available: https://www.news24.com/w24/work/legal/the-low-down-on-getting-car-insurance-
Van Eeden, J. 2017. Technology is disrupting Africa’s insurance industry. Mail & Guardian, 27
February 2017. Available: https://mg.co.za/article/2017-02-24-00-technology-is-disrupting-
africas-insurance-industry [Accessed 2017, June 08].
409
Van Rensburg, H. 2018. Think hard before you blow your cover. IOL, 7 June 2018. Available:
https://www.iol.co.za/personal-finance/think-hard-before-you-blow-your-cover-15344517
Van Wyk, C. 2016. Innovation in transport calls for equally innovative change in vehicle insurance.
Bizcommunity, 8 November 2016. Available: http://www.bizcommunity.com/Article/196/582/
153456.html [Accessed 2017, May 01].
Vazeer, A. 2016a. Climate change disrupts short-term insurance industry. FA NEWS, 22 February
2016. Available: https://www.fanews.co.za/article/short-term-insurance/15/general/1217/
climate-change-disrupts-the-short-term-insurance-industry/19811 [Accessed 2021, October
16].
Vazeer, A. 2016b. Weak rand poses real threat for the insurance industry. FA NEWS, 12 May
2016. Available: https://www.fanews.co.za/article/company-news-results/1/lion-of-africa/
1248/weak-rand-poses-real-threat-for-the-insurance-industry/20281 [Accessed 2022,
January 21].
Venter, I. 2017. 39% of South Africans interested in self-driving cars. Engineering News, 14 March
2017. Available: https://www.engineeringnews.co.za/print-version/39-of-south-africans-
interested-in-self-driving-cars-deloitte-2017-03-14 [Accessed 2021, October 15].
Viksne, K., Salkovska, J., Gaitniece, E. & Puke, I. 2016. Comparative analysis of customer
behaviour models. In 2016 International Conference Economic Science for Rural
Development. Jelgava, pp. 231–238. Available: https://llufb.llu.lv/conference/economic
_science_rural/2016/Latvia_ESRD_43_2016-231-238.pdf [Accessed 2021, October 25].
Vivian, R. & Mushai, A. 2020. General principles of insurance: Volume 1. University of the
Witwatersrand.
Vivian, R.W. & Todd, P. 2011. Mutual & Federal, a thoroughly South African company, turns 180.
FA NEWS, 15 March 2011. Available: https://www.fanews.co.za/article/company-news-
results/1/mutual-federal/1070/mutual-federal-a-thoroughly-south-african-company-turns-
180/9249 [Accessed 2021, December 11].
Viviers, S., Van Greunen, C. & Venter, E. 2012. Variables influencing unethical decision-making
and behaviour in the life insurance sector in South Africa : A theoretical framework.
Management Dynamics. 21(2): 34–50.
Vohra, T. 2019. Why don’t people buy car insurance? [Blog] Insurify, 15 April 2021. Available:
https://insurify.com/blog/car-insurance/5-reasons-drivers-dont-buy-a-car-insurance-policy/
[Accessed 2021, November 11].
Wagner, C., Kawulich, B. & Garner, M. 2012. Doing social research: A global context. Berkshire:
McGraw Hill Higher Education.
410
Walsh, G., Wayne-Mitchell, T., Wayne-Mitchell, V. & Wiedmann, K.-P. 2001. Consumers’
decision-making style as a basis for market segmentation Thorsten Hennig-Thurau. Journal
of Targeting, Measurement and Analysis for Marketing, 10(2): 117–131.
Wang, Y. & Yu, C. 2017. Social interaction-based consumer decision-making model in social
commerce: The role of word of mouth and observational learning. International Journal of
Information Management, 37(3): 179–189.
Ward, S. 2020. Target marketing and market segmentation. Available: https://www.thebalances

mb.com/target-marketing-2948355 [Accessed 2021, October 19].
Wassink, K.B. 2021. Are your insurance solutions in tune with new consumer needs? Available:
https://www.ey.com/en_gl/insurance/are-your-insurance-solutions-in-tune-with-new-
consumer-needs [Accessed 2021, August 16].
Weedige, S.S., Ouyang, H., Gao, Y. & Liu, Y. 2019. Decision making in personal insurance:
Impact of insurance literacy. Sustainability, 11(6795): 1–24.
West, D.C., Ford, J. & Ibrahim, E. 2015. Strategic marketing: Creating competitive advantage. 3rd
ed. New York: Oxford University Press.
Wheels24. 2015. Car parts prices: Vehicle repairs cost SA billions. Wheels24, 13 November 2015.
Available: http://www.wheels24.co.za/News/car-parts-prices-vehicle-repairs-cost-sa-billions-
Wheels24. 2016a. First car in SA: Happy 120th anniversary to the horseless carriage! Wheels24,
11 July 2016. Available: http://www.wheels24.co.za/news/classic_cars/first-car-in-sa-happy-
120th-anniversary-to-the-horseless-carriage-20160711 [Accessed 2021, October 01].
Wheels24. 2016b. Insurance claims: Know the “no-claim bonus” fine print. Wheels24, 14 October
2016. Available: http://www.wheels24.co.za/News/Guides_and_Lists/insurance-claims-know-
the-no-claim-bonus-fine-print-20160914 [Accessed 2021, October 07].
Whittaker, G. 2007. Actuarial Involvement in road accident fund claims in South Africa.
International News, 43: 29–32. Available: https://silo.tips/download/article-from-international-
news-november-2007-issue-no-43 [Accessed 2021, October 04].
Willan, M. 2021. The essentials of contemporary marketing. Fairford: Bloomsbury Publishing.
Williamson, B.C. 1990. Workers Compensation Law in NSW: Workers benefits and the role of
rehabilitation. Australian Occupational Therapy Journal, 37(2): 95–101.
Williamson, K. & Johanson, G. 2018. Research methods: Information, systems, and contexts. 2nd
ed. Cambridge: Chandos Publishing.
411
Wills, M.A. 1986. An examination of the progression towards no-fault motor vehicle insurance,
with particular reference to the Republic of South Africa. Unpublished Master’s dissertation.
University of Natal, Durban. Available: https://researchspace.ukzn.ac.za/xmlui/handle/
10413/5234?show=full [Accessed 2021, October 01].
Wilson, J. 2014. Essentials of business research: A guide to doing your research project. 2nd ed.
London: SAGE.
Woolard, C. 2015. Smarter consumer communications. Available: https://www.fca.org.uk/

publication/discussion/dp15-05-smarter-consumer-communications.pdf [Accessed 2017,
June 28].
Wooster, M.M. 2014. History of the welfare state written in the past. Philanthropy Daily, 28 April
2014. Available: https://www.philanthropydaily.com/history-of-the-welfare-state-written-in-the-
past/ [Accessed 2021, November 11].
WP Motors. 2021. Why the colour of your car’s paint influences the amount you pay to insure it.
Available: https://www.wpmotors.co.za/why-the-colour-of-your-cars-paint-influences-the-
amount-you-pay-to-insure-it/ [Accessed 2021, October 07].
Xian, H. & Meng-Lewis, Y. 2018. Business research methods for Chinese students: A practical
guide to your research project. London: SAGE Publications.
Xu, X. & Fan, C.K. 2019. Autonomous vehicles, risk perceptions and insurance demand: An
individual survey in China. Transportation Research Part A: Policy and Practice, 124: 549–
556.
Zavadil, T. 2014. Do the better insured cause more damage? Testing for asymmetric information
in car insurance. Journal of Risk and Insurance, 889(2015): 865–889.
Zeiler, L. 2012. The history of insurance throughout the world. Available:

http://thez.zeiler.com/the-history-of-insurance-throughout-the-world [Accessed 2017,
February 09].
Zerbst, F. 2011. What you need to know about insurance claims and excess. Mail & Guardian, 1
June 2011. Available: https://mg.co.za/article/2011-06-01-what-you-need-to-know-about-
insurance-claims-and-excess [Accessed 2017, April 25].
Zhang, Z. 2016. Model-building strategy for logistic regression: Purposeful selection. Annals of
Translational Medicine, 4(6): 1–7.
Zikmund, W.G., Babin, B.J., Carr, J.C. & Griffin, M. 2013. Business research methods. 9th ed.
Mason: South-Western Cengage Learning.
412
Zikmund, W.G., D’Allesandro, S., Lowe, B., Winzar, H. & Babin, B.J. 2017. Marketing research.
4th ed. Sydney: Cengage Learning.
Zingwevu, E. & Sibindi, A.B. 2014. Is compulsory third party motor insurance the panacea for the
South African insurance industry? Corporate Ownership & Control, 11(4): 657–669.
413
APPENDIX A
- DATA COLLECTION INSTRUMENT -
414
Dear prospective participant,
You are invited to participate in a survey conducted by Ms Nadia van Huyssteen, under the
supervision of Prof. Sharon Rudansky-Kloppers, a professor in the Department of Business
Management, towards a PhD degree at the University of South Africa.
The survey you have received has been designed to study the reasons why South African
consumers purchase (or do not purchase) personal motor vehicle insurance in South Africa.
You were selected to participate in this survey because you are a registered or licensed motor
vehicle owner and/or driver in South Africa. By completing this survey, you agree that the
information you provide may be used for research purposes, including dissemination through
peer-reviewed publications and conference proceedings.
It is anticipated that the information we gain from this survey will help us to understand the
South African motor vehicle insurance market and the South African motorists’ purchasing
habits of personal motor vehicle insurance. You are, however, under no obligation to complete
the survey and you can withdraw from the study prior to submitting the survey. The survey is
developed to be anonymous, meaning that we will have no way of connecting the information
that you provide to you personally. Consequently, you will not be able to withdraw from the
study once you have clicked the send button based on the anonymous nature of the survey.
If you choose to participate in this survey it will take up no more than 20 minutes of your time.
You will not benefit from your participation as an individual, however, it is envisioned that the
findings of this study will be used in a PhD dissertation. It could also be shared in conference
presentations, the publication of academic articles and other communication. If you would like
feedback regarding the results of this study you are more than welcome to contact the main
researcher. We do not foresee that you will experience any negative consequences by
completing the survey. The researcher(s) undertake to keep any information provided herein
confidential, not to let it out of our possession and to report on the findings from the perspective
of the participating group and not from the perspective of an individual.
The records will be kept for five years for audit purposes where after it will be permanently
destroyed (electronic versions will be permanently deleted from the hard drive of the
computer). You will not be reimbursed or receive any incentives for your participation in the
survey.
The study fulfils all the requirements as set out in the Unisa Policy on Research Ethics and
the research was reviewed and approved by the Department of Business Management
Research Ethics Review Committee. A copy of the approval letter can be obtained from the
researcher if you so wish. The primary researcher, Ms Nadia van Huyssteen, can be contacted
during office hours at marxn@unisa.ac.za or 012 429 8014. The study leader, Prof Sharon
Rudansky-Kloppers, can be contacted during office hours at rudans@unisa.ac.za or 012 429
8014. Should you have any questions regarding the ethical aspects of the study, you can
contact the chairperson of the Department of Business Management Research Ethics Review
Committee, at vissed@unisa.ac.za or 012 429 2113. Alternatively, you can report any serious
unethical behaviour at the University’s Toll Free Hotline 0800 86 96 93.
I agree to the recording of the online questionnaire and by clicking on “next” I am giving my
consent to participate in this study.
Please answer all the questions by placing a tick in the specific block (there are no right
or wrong answers).
Next
415
Q1. Do you drive a motor vehicle, which includes a motor car/sedan/hatchback/station
wagon/SUV/single or double cab bakkie?
1 2
Yes No
If you ticked YES please continue If you ticked NO, please stop
with QUESTION 2 below. completing the questionnaire. Thank
you for completing the survey. Your
assistance is appreciated.
Q2. Are you personally responsible for the decision to purchase your personal motor
vehicle’s insurance cover? Please note that this is only for personally financed or
owned motor vehicles.
1 2
Yes No
If you ticked YES please continue If you ticked NO, please stop
with QUESTION 3 below. completing the questionnaire. Thank
you for completing the survey. Your
assistance is appreciated.
Q3. Do you currently have motor vehicle insurance for your personal motor vehicle? Please
note that this is only for personally financed or owned motor vehicles.
1 2
Yes No
If you ticked YES please continue with If you ticked NO, please continue
QUESTION 4 below, SKIP QUESTION 7 with QUESTION 7.
and CONTINUE with QUESTION 8.
Q4. Which type of personal motor vehicle insurance policy do you currently have for your
motor vehicle? (Only select one option)
1
Comprehensive
2
Third-party fire and theft
3
Third-party only
4
Pay As You Drive
5
Not certain
416
Q5. Who pays for the motor insurance on your personal motor vehicle? (Only select one
option)
1
I pay for it myself
2
My family member pays for it
3
My employer pays for it as part
of my employment package
Q6. Why do you currently have personal motor vehicle insurance? (Tick all the applicable
options)
Not willing to take the chance that I cannot cover the costs when in an
accident
Replacement value of my motor vehicle is high
Motor insurance is a necessity
I know that I can benefit from having motor vehicle insurance
I have motor vehicle insurance because my father/mother has/had it
Due to the high crime rate in the country
Other (please specify)
Q7. Why do you currently not have personal motor vehicle insurance? (Tick all the
applicable options)
Insurance companies refused coverage for personal motor vehicles
I do not believe that I can benefit from having insurance
I am dissatisfied with previous insurance policy or insurance company
I do not need insurance
I cannot afford insurance
The replacement value of my motor vehicle is low
I do not know what motor insurance is
417
Section A: External influences
Q8. Below are statements which relate to the influence which marketing efforts by short-term insurance companies have on your need
to purchase personal motor vehicle insurance. Use the following scales to indicate the extent of the influence of each statement.
Statement No Small Moderate Large Extremely
influence influence influence influence large
influence
Short-term insurance company’s marketing efforts
8.1 Type of personal motor vehicle insurance policies offered at a

1 2 3 4 5
company.
8.2 Advertisements or promotional activities performed by a short-term

1 2 3 4 5
insurance company.
8.3 Price of a policy. 1 2 3 4 5
8.4 Way in which you can purchase (through a broker/ask the company
to call you back on their website/email the company directly) personal 1 2 3 4 5
8.5 The company’s physical evidence efforts. For example, how building
1 2 3 4 5
looks, employee uniforms and online physical presence.
8.6 The duration of the process to get a quotation. 1 2 3 4 5
8.7 The way in which the employees in the company interact with
1 2 3 4 5
consumers.
8.8 Partnerships or associations with other companies that you trust. 1 2 3 4 5
418
Q9. Below are statements which relate to the effect that the sociocultural environment has on your need to purchase personal motor
vehicle insurance. Use the following scales to indicate the extent of the influence of each statement.
Statement No Small Moderate Large Extremely
influence
Sociocultural/non-commercial influences
9.1 A positive comment or opinion of a family member. 1 2 3 4 5
9.2 A negative comment or opinion of a family member. 1 2 3 4 5
9.3 The positive opinion of a reference person or group used as a basis

1 2 3 4 5
of comparison.
9.4 The negative opinion of a reference person or group used as a basis

1 2 3 4 5
of comparison.
9.5 Positive comments of a friend, a newspaper article or the views of

1 2 3 4 5
an expert on a blog or other internet platforms.
9.6 Negative comments of a friend, a newspaper article or the views of

1 2 3 4 5
an expert on a blog or other internet platforms.
9.7 The positive opinions of individuals in the same social class as you. 1 2 3 4 5
9.8 The negative opinions of individuals in the same social class as you. 1 2 3 4 5
9.9 Cultural and subcultural factors, such as your language

1 2 3 4 5
preference, beliefs and values.
419
Q10. Below are statements which relate to the communication sources (methods or platforms used to communicate a message) that
influence your need to purchase personal motor vehicle insurance. Use the following scales to indicate the extent of the influence of
each statement.
No Small Moderate Large Extremely
Statement influence
10.1 Traditional advertisements about motor vehicle insurance (Such as

1 2 3 4 5
TV and radio advertisements).
10.2 Online advertisements about motor vehicle insurance (Such as

1 2 3 4 5
Video advertisements and banner ads on websites).
10.3 Any person that is a source of a product referral (buzz agent) and tells
1 2 3 4 5
you about motor vehicle insurance.
10.4 Customised messages that insurance companies communicate to

1 2 3 4 5
you about motor vehicle insurance.
10.5 Word-of-mouth, advice, recommendations on social media posts

1 2 3 4 5
about motor vehicle insurance.
10.6 Paid-for social media efforts (Such as Facebook advertisement

1 2 3 4 5
campaigns) by insurance companies about motor vehicle insurance.
420
Section B: Internal influences
Q11. Below are statements which relate to the psychological attributes that have an influence on your need recognition for personal
motor vehicle insurance. Use the following scales to indicate your level of agreement with each statement.
No Small Moderate Large Extremely
Statement influence
Psychological attributes that influence your desire to recognise a need for motor vehicle insurance
11.1 Your motivation to purchase personal motor vehicle insurance. 1 2 3 4 5
11.2 Your perception about motor vehicle insurance. 1 2 3 4 5
11.3 Your perception about a certain insurance company. 1 2 3 4 5
11.4 The process of how you purchase and consume information, as

1 2 3 4 5
well as the experience gained.
11.5 Your personality (specific qualities, attributes, traits). 1 2 3 4 5
11.6 Your attitude towards motor vehicle insurance. 1 2 3 4 5
11.7 Your attitude towards a certain short-term insurance company. 1 2 3 4 5
421
Section C: Decision-making process
Q12. Below are statements which relate to the decision-making process to purchase personal motor vehicle insurance. Use the following
scales to indicate the extent to which you use the following in the decision-making process.
No Some Moderate Large Extreme

Extent Extent Extent Extent Extent
Statement
Decision-making process
12.1 In your search for information, you think about previous

1 2 3 4 5
experiences.
12.2 In your search for information, you request quotations from more
1 2 3 4 5
than one short-term insurance company.
12.3 You compare the price of the different quotations to measure if

there is a policy that will contribute to your need satisfaction and 1 2 3 4 5
lifestyle.
12.4 You compare the quality of the different quotations to measure if

there is a policy that will contribute to your need satisfaction and 1 2 3 4 5
lifestyle.
12.5 You compare the brand list (list of different companies) to

measure if there is a policy that will contribute to your need 1 2 3 4 5
satisfaction and lifestyle.
422
Section D: Demographic profile
Q13. What is your age? Please type your age on the space provided below.
__________
Q14. How would you describe yourself in terms of population group?
1
Asian
2
Black African
3
Coloured
4
Indian
5
White
6
Q15. What is your gender?
1 2
Male Female
Q16. What is your marital status?
1
Single
2
Married
3
Living with partner, but not married
4
Divorced
5
Widowed
423
Q17. What is your highest level of education?
1
Less than Grade 12
2
Grade 12 (matric)
3
Post Matric Certificate
4
Higher Certificate
5
Diploma
6
Higher Diploma
7
Post Higher Diploma (Masters or Doctoral Diploma)
8
Bachelors Degree
9
Honours Degree
10
Master’s Degree
11
Doctorate Degree
12
Q18. What is your employment status?
1
Student
2
Unemployed
3
Temporarily employed (including Fixed Term Contracts)
4
Permanently employed on a full-time basis
5
Permanently employed on a part-time basis
6
Retired
424
Q19. In which province do you travel the majority of time with your personal motor vehicle?
1
Gauteng province
2
Western Cape province
3
KwaZulu-Natal province
4
Northern Cape province
5
Free State province
6
Limpopo province
7
Mpumalanga province
8
Eastern Cape province
9
North West province
425
Q.20 Which of the following income categories best describe the monthly income of your
immediate family from all sources (including your income), before tax and other
deductions? (If some members of your immediate family have a weekly income, please
estimate what this typically would be for a month).
R0 – R800 1
R801 – R1 200 2
R1 201 – R2 500 3
R2 501 – R6 000 4
5
R6 001– R16 000
R16 001 – R26 000 6
R26 001 – R36 000 7
R36 001 – R46 000 8
R46 001 – R56 000 9
10
R56 001 or more
Prefer not to say/Don’t know/NA 99
Q21. Which of the following income categories best describe your monthly personal
income from all sources, before tax and other deductions? (If you earn a weekly income,
please estimate what this typically would be for a month.
R0 – R800 1
2
R801– R1 200
R1 201 – R2 500 3
R2 501 – R6 000 4
5
R6 001– R16 000
426
R16 001 – R26 000 6
R26 001 – R36 000 7
R36 001 – R46 000 8
R46 001 – R56 000 9
10
R56 001 or more
Prefer not to say/Don’t know/NA 99
Thank you for completing the survey. Your assistance is appreciated.
427
APPENDIX B
- ETHICAL CLEARANCE CERTIFICATE -
428
429
430
APPENDIX C
- FREQUENCY TABLES AND DESCRIPTIVE STATISTICS -
431
Table C1.1: Age categories (in years)
Age categories(in
years) Frequency Valid Percent
18-30 128 18.90
31-40 174 25.70
41-50 134 19.70
51-60 128 18.90
over 60 114 16.80
Total 678 100
Table C1.2: Age mean (in years)
Age provided
(in years) Frequency Valid Percent Sum of years
18 6 0.9 108
19 3 0.4 57
20 3 0.4 60
21 6 0.9 126
22 3 0.4 66
23 3 0.4 69
24 4 0.6 96
25 5 0.7 125
26 8 1.2 208
27 14 2.1 378
28 12 1.8 336
29 17 2.5 493
30 44 6.5 1320
31 27 4.0 837
32 24 3.5 768
33 13 1.9 429
34 19 2.8 646
432
35 22 3.2 770
36 8 1.2 288
37 17 2.5 629
38 9 1.3 342
39 18 2.7 702
40 17 2.5 680
41 10 1.5 410
42 14 2.1 588
43 10 1.5 430
44 13 1.9 572
45 19 2.8 855
46 7 1.0 322
47 17 2.5 799
48 16 2.4 768
49 14 2.1 686
50 14 2.1 700
51 13 1.9 663
52 11 1.6 572
53 10 1.5 530
54 11 1.6 594
55 21 3.1 1155
56 17 2.5 952
57 8 1.2 456
58 11 1.6 638
59 11 1.6 649
60 15 2.2 900
61 8 1.2 488
62 8 1.2 496
63 10 1.5 630
64 5 0.7 320
65 7 1.0 455
66 6 0.9 396
433
67 6 0.9 402
68 4 0.6 272
69 8 1.2 552
70 4 0.6 280
71 8 1.2 568
72 4 0.6 288
73 5 0.7 365
74 6 0.9 444
75 4 0.6 300
76 4 0.6 304
77 2 0.3 154
78 2 0.3 156
79 3 0.4 237
80 2 0.3 160
83 1 0.1 83
85 1 0.1 85
87 2 0.3 174
88 3 0.4 264
91 1 0.1 91
Total 678 100.0 30736
Table C2: Population group
Population group Frequency Valid Percent
Asian 3 0.5
Black African 173 26.1
Coloured 58 8.8
Indian 28 4.2
White 397 60.0
Other 3 0.5
Total 662 100.0
434
Table C3: Gender
Gender Frequency Valid Percent
Male 356 52.5
Female 322 47.5
Total 678 100.0
Table C4: Marital status
Marital status Frequency Valid Percent
Single 157 23.2
Married 377 55.6
Living with partner, but not married 71 10.5
Divorced 48 7.1
Widowed 25 3.6
Total 678 100.0
Table C5: Highest level of education
Highest level of education Frequency Valid Percent
Less than Grade 12 45 6.7
Grade 12 (matric) 174 25.7
Post Matric Certificate 34 5.0
Higher Certificate 29 4.3
Diploma 92 13.6
Higher Diploma 51 7.5
Post Higher Diploma (Masters or Doctoral Diploma) 13 1.9
Bachelors Degree 93 13.8
Honours Degree 62 9.2
Master’s Degree 60 8.9
Doctorate Degree 23 3.4
Total 676 100.0
435
Table C6: Employment status
Employment status Frequency Valid Percent
Student 17 2.5
Unemployed 42 6.2
Temporarily employed (including Fixed Term Contracts) 71 10.5
Permanently employed on a full-time basis 392 57.8
Permanently employed on a part-time basis 35 5.2
Retired 121 17.8
Total 678 100.0
Table C7: Province mostly traveled with personal motor vehicle
Province mostly traveled with

personal motor vehicle Frequency Valid Percent
Gauteng 294 43.4
Western Cape 87 12.8
KwaZulu-Natal 74 10.9
Free State 69 10.2
North West 39 5.8
Eastern Cape 38 5.6
Mpumalanga 33 4.9
Limpopo 25 3.7
Northern Cape 19 2.8
Total 678 100.0
436
Table C8: Household income
Household income Frequency Valid Percent
R0 – R800 24 3.5
R801 – R1 200 8 1.2
R1 201 – R2 500 12 1.8
R2 501 – R6 000 44 6.5
R6 001– R16 000 145 21.4
R16 001 – R26 000 77 11.4
R26 001 – R36 000 68 10.0
R36 001 – R46 000 49 7.2
R46 001 – R56 000 41 6.0
R56 001 or more 134 19.8
Prefer not to say/Don’t know/NA 76 11.2
Total 678 100.0
Table C9: Personal income
Personal income Frequency Valid Percent
R0 – R800 77 11.4
R801 – R1 200 22 3.2
R1 201 – R2 500 27 4.0
R2 501 – R6 000 54 8.0
R6 001– R16 000 102 15.0
R16 001 – R26 000 89 13.1
R26 001 – R36 000 78 11.5
R36 001 – R46 000 39 5.8
R46 001 – R56 000 42 6.2
R56 001 or more 85 12.5
Prefer not to say/Don’t know/NA 63 9.3
Total 678 100.0
437
Table C10: Drive a motor vehicle
Drive a motor vehicle Frequency Valid Percent
No 1 0.1
Yes 705 99.9
Total 706 100.0
Table C11: Personally responsible for the decision to purchase personal motor
vehicle insurance
Personally responsible for the

decision to purchase personal
motor vehicle insurance Frequency Valid Percent
No 27 3.8
Yes 678 96.2
Total 705 100.0
Table C12: Currently have/do not have personal motor vehicle insurance
Decision to purchase Frequency Valid Percent
Yes 379 55.9
No 299 44.1
Total 678 100.0
438
Table C13: Type of personal motor vehicle insurance
Type of personal motor

vehicle insurance Frequency Percent Valid Percent
Comprehensive 362 53.4 95.5
Third-party fire and theft 5 0.7 1.3
Third-party only 5 0.7 1.3
Pay As You Drive 1 0.1 0.3
Not certain 6 0.9 1.6
Total 379 55.9 100.0
Table C14: Payment party
Payment party Frequency Valid Percent
I pay for it myself 341 90.0
My family member pays for it 33 8.7
My employer pays for it as part of my 5 1.3

employment package
Total 379 100.0
Table C15.1: Why you have insurance - Not willing to take the chance that I
cannot cover the costs when in an accident
Not willing to take the chance that I

cannot cover the costs when in an
accident Frequency Valid Percent
Not selected 68 17.9
Selected 311 82.1
Total 379 100.0
439
Table C15.2: Why you have insurance - Replacement value of my motor vehicle
is high
Replacement value of my motor

vehicle is high Frequency Valid Percent
Selected 173 45.6
Total 379 100.0
Table C15.3: Why you have insurance - Motor insurance is a necessity
Motor insurance is a necessity Frequency Valid Percent
Selected 217 57.3
Total 379 100.0
Table C15.4: Why you have insurance - I know that I can benefit from having
motor vehicle insurance
I know that I can benefit from having motor vehicle

insurance Frequency Valid Percent
Selected 104 27.4
Total 379 100.0
Table C15.5: Why you have insurance - I have motor vehicle insurance because
my father/mother has/had it
I have motor vehicle insurance because my

father/mother has/had it Frequency Valid Percent
Selected 10 2.6
Total 379 100.0
440
Table C15.6: Why you have insurance - Due to the high crime rate in the country
Due to the high crime rate in the country Frequency Valid Percent
Selected 150 39.6
Total 379 100.0
Table C15.7: Why you have insurance - Other
Other Frequency Valid Percent
Selected 24 6.3
Total 379 100.0
Table C16.1: Why you do not have insurance - Insurance companies refused
coverage for personal motor vehicle
Insurance companies refused

coverage for personal motor vehicle Frequency Valid Percent
Selected 23 7.7
Total 299 100.0
Table C16.2: Why you do not have insurance - I do not believe that I can benefit
from having insurance
I do not believe that I can benefit

from having insurance Frequency Valid Percent
Selected 103 34.4
Total 299 100.0
441
Table C16.3: Why you do not have insurance - I am dissatisfied with previous
insurance policy or insurance company
I am dissatisfied with previous

insurance policy or insurance
company Frequency Valid Percent
Selected 39 13.0
Total 299 100.0
Table C16.4: Why you do not have insurance - I do not need insurance
I do not need insurance Frequency Valid Percent
Selected 45 15.1
Total 299 100.0
Table C16.5: Why you do not have insurance - I cannot afford insurance
I cannot afford insurance Frequency Valid Percent
Selected 174 58.2
Total 299 100.0
Table C16.6: Why you do not have insurance - The replacement value of my
motor vehicle is low
The replacement value of my motor

vehicle is low Frequency Valid Percent
Selected 98 32.8
Total 299 100.0
442
Table C16.7: Why you do not have insurance - I do not know what motor
insurance is
I do not know what motor insurance

is Frequency Valid Percent
Selected 26 8.7
Total 299 100.0
Table C16.8: Why you do not have insurance - Other
Other Frequency Valid Percent
Selected 6 2.0
Total 299 100.0
Table C17.1: Company’s marketing efforts - Type of personal motor vehicle

insurance offered
Type of personal motor vehicle

insurance offered Frequency Valid Percent
None 175 25.8
Small 124 18.3
Moderate 161 23.7
Large 149 22.0
Extremely large 69 10.2
Total 678 100.0
443
Table C17.2: Company’s marketing efforts - Advertisements or promotional
activities performed
Advertisements or promotional activities performed Frequency Valid Percent
None 231 34.1
Small 219 32.3
Moderate 140 20.6
Large 66 9.7
Total 678 100.0
Table C17.3: Company’s marketing efforts - Price of a policy
Price of a policy Frequency Valid Percent
None 48 7.1
Small 75 11.1
Moderate 136 20.1
Large 240 35.4
Total 678 100.0
Table C17.4: Company’s marketing efforts - Way in which personal motor

vehicle insurance can be purchased
Way in which personal motor vehicle insurance can

be purchased Frequency Valid Percent
None 119 17.6
Small 135 19.9
Moderate 188 27.7
Large 173 25.5
Total 678 100.0
444
Table C17.5: Company’s marketing efforts - The company’s physical evidence
efforts
The company’s physical evidence efforts Frequency Valid Percent
None 217 32.0
Small 162 23.9
Moderate 165 24.3
Large 98 14.5
Total 678 100.0
Table C17.6: Company’s marketing efforts - The duration of the process to get a
quotation
The duration of the process to get a quotation Frequency Valid Percent
None 78 11.5
Small 99 14.6
Moderate 153 22.6
Large 234 34.5
Total 678 100.0
Table C17.7: Company’s marketing efforts - The way in which the employees in
the company interact with consumers
The way in which the employees in the company

interact with consumers Frequency Valid Percent
None 56 8.3
Small 70 10.3
Moderate 145 21.4
Large 230 33.9
Total 678 100.0
445
Table C17.8: Company’s marketing efforts - Partnerships or associations with
other companies that consumers trust
Partnerships or associations Frequency Valid Percent
None 116 17.1
Small 141 20.8
Moderate 179 26.4
Large 172 25.4
Total 678 100.0
Table C18.1: Sociocultural environment - A positive comment or opinion of a

family member
A positive comment or opinion of a family member Frequency Valid Percent
None 119 17.6
Small 108 15.9
Moderate 170 25.1
Large 201 29.6
Total 678 100.0
Table C18.2: Sociocultural environment - A negative comment or opinion of a

family member
A negative comment or opinion of a family member Frequency Valid Percent
None 100 14.7
Small 123 18.1
Moderate 154 22.7
Large 180 26.5
Total 678 100.0
446
Table C18.3: Sociocultural environment - The positive opinion of a reference
person or group used as a basis of comparison
The positive opinion of a reference person or group

used as a basis of comparison Frequency Valid Percent
None 89 13.1
Small 123 18.1
Moderate 228 33.6
Large 182 26.8
Total 678 100.0
Table C18.4: Sociocultural environment - The negative opinion of a reference

person or group used as a basis of comparison
The negative opinion of a reference person or group

used as a basis of comparison Frequency Valid Percent
None 102 15.0
Small 129 19.0
Moderate 183 27.0
Large 175 25.8
Total 678 100.0
Table C18.5: Sociocultural environment - Positive comments of a friend, a

newspaper article or the views of an expert on a blog or other internet platforms
Positive comments of a friend, a newspaper article or

the views of an expert on a blog or other internet
platforms Frequency Valid Percent
None 103 15.2
Small 106 15.6
Moderate 200 29.5
Large 203 29.9
Total 678 100.0
447
Table C18.6: Sociocultural environment - Negative comments of a friend, a
newspaper article or the views of an expert on a blog or other internet platforms
Negative comments of a friend, a newspaper article or the

views of an expert on a blog or other internet platforms Frequency Valid Percent
None 108 15.9
Small 116 17.1
Moderate 177 26.1
Large 175 25.8
Total 678 100.0
Table C18.7: Sociocultural environment - The positive opinions of individuals in

the same social class
The positive opinions of individuals in the same social class Frequency Valid Percent
None 111 16.4
Small 115 17.0
Moderate 171 25.2
Large 217 32.0
Total 678 100.0
Table C18.8: Sociocultural environment - The negative opinions of individuals

in the same social class
The negative opinions of individuals in the same social class Frequency Valid Percent
None 119 17.6
Small 137 20.2
Moderate 150 22.1
Large 181 26.7
Total 678 100.0
448
Table C18.9: Sociocultural environment - Cultural and subcultural factors
Cultural and subcultural factors Frequency Valid Percent
None 165 24.3
Small 141 20.8
Moderate 154 22.7
Large 146 21.5
Total 678 100.0
Table C19.1: Communication sources - Traditional advertisements about motor

vehicle insurance
Traditional advertisements Frequency Valid Percent
None 201 29.6
Small 227 33.5
Moderate 152 22.4
Large 60 8.8
Total 678 100.0
Table C19.2: Communication sources - Online advertisements about motor

vehicle insurance
Online advertisements Frequency Valid Percent
None 197 29.1
Small 211 31.1
Moderate 167 24.6
Large 67 9.9
Total 678 100.0
449
Table C19.3: Communication sources - Buzz agent
Buzz agent Frequency Valid Percent
None 252 37.2
Small 201 29.6
Moderate 120 17.7
Large 59 8.7
Total 678 100.0
Table C19.4: Communication sources - Customised messages that insurance

companies communicate
Customised messages Frequency Valid Percent
None 232 34.2
Small 186 27.4
Moderate 180 26.5
Large 53 7.8
Total 678 100.0
Table C19.5: Communication sources – Paid-for social media efforts
Paid-for social media efforts Frequency Valid Percent
None 263 38.8
Small 214 31.6
Moderate 131 19.3
Large 34 5.0
Total 678 100.0
450
Table C19.6: Communication sources - Word-of-mouth, advice,
recommendations on social media posts
Word-of-mouth, advice,
recommendations on social media
posts Frequency Valid Percent
None 117 17.3
Small 181 26.7
Moderate 183 27.0
Large 129 19.0
Total 678 100.0
Table C20.1: Psychological attributes - Motivation to purchase personal motor

vehicle insurance
Motivation Frequency Valid Percent
None 93 13.7
Small 88 13.0
Moderate 152 22.4
Large 230 33.9
Total 678 100.0
Table C20.2: Psychological attributes - Perception about motor vehicle

insurance
Perception about motor vehicle

insurance Frequency Valid Percent
None 74 10.9
Small 75 11.1
Moderate 173 25.5
Large 253 37.3
Total 678 100.0
451
Table C20.3: Psychological attributes - Perception about a certain insurance
company
Perception about a certain

insurance company Frequency Valid Percent
None 68 10.0
Small 92 13.6
Moderate 169 24.9
Large 224 33.0
Total 678 100.0
Table C20.4: Psychological attributes - Learning
Learning Frequency Valid Percent
None 60 8.8
Small 92 13.6
Moderate 195 28.8
Large 224 33.0
Total 678 100.0
Table C20.5: Psychological attributes - Personality
Personality Frequency Valid Percent
None 82 12.1
Small 109 16.1
Moderate 192 28.3
Large 200 29.5
Total 678 100.0
452
Table C20.6: Psychological attributes - Attitude towards motor vehicle
insurance
Attitude towards motor vehicle insurance Frequency Valid Percent
None 55 8.1
Small 88 13.0
Moderate 162 23.9
Large 250 36.9
Total 678 100.0
Table C20.7: Psychological attributes - Attitude towards a certain short-term

insurance company
Attitude towards a certain short-term insurance company Frequency Valid Percent
None 83 12.2
Small 104 15.3
Moderate 151 22.3
Large 200 29.5
Total 678 100.0
Table C21.1: Pre-purchase search - Previous experiences
Previous experiences Frequency Valid Percent
None 66 9.7
Some 94 13.9
Moderate 132 19.5
Large 241 35.5
Extreme 145 21.4
Total 678 100.0
453
Table C21.2: Pre-purchase search - Request quotations from more than one
short-term insurance company
Request quotations from more than

one short-term insurance company Frequency Valid Percent
None 78 11.5
Some 86 12.7
Moderate 149 22.0
Large 232 34.2
Extreme 133 19.6
Total 678 100.0
Table C22.1: Evaluation of alternatives - Compare the price of the different

quotations
Compare the price of the different

quotations Frequency Valid Percent
None 71 10.5
Some 72 10.6
Moderate 121 17.8
Large 243 35.8
Extreme 171 25.2
Total 678 100.0
454
Table C22.2: Evaluation of alternatives - Compare the quality of the different
quotations
Compare the quality of the different

quotations Frequency Valid Percent
None 60 8.8
Some 73 10.8
Moderate 125 18.4
Large 241 35.5
Extreme 179 26.4
Total 678 100.0
Table C22.3: Evaluation of alternatives - Compare the brand list
Compare the brand list Frequency Valid Percent
None 97 14.3
Some 119 17.6
Moderate 166 24.5
Large 187 27.6
Extreme 109 16.1
Total 678 100.0
455
APPENDIX D
- TABLES AND INFERENTIAL STATISTICS -
456
Table D1: KMO and Bartlett's Test for EFA
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. 0.847
Bartlett's Test of Approx. Chi-Square 4485.424

Sphericity
df 595
Sig. 0.000
Table D2: Communalities for EFA
Communalities
Initial Extraction
8.1 [Accessed MARKETING EFFORTS] Type of personal motor vehicle 0.533 0.462
insurance policies offered at a company.
8.2 [Accessed MARKETING EFFORTS] Advertisements or promotional 0.513 0.476

activities performed by a short-term insurance company.
8.3 [Accessed MARKETING EFFORTS] Price of a policy. 0.546 0.490
8.4 [Accessed MARKETING EFFORTS] Way in which you can purchase 0.528 0.490
8.5 [Accessed MARKETING EFFORTS] The company’s physical 0.565 0.500

evidence efforts.
8.6 [Accessed MARKETING EFFORTS] The duration of the process to 0.625 0.630
get a quotation.
8.7 [Accessed MARKETING EFFORTS] The way in which the employees 0.611 0.573
in the company interact with consumers.
8.8 [Accessed MARKETING EFFORTS] Partnerships or associations with 0.534 0.441

other companies that you trust.
9.1 [Accessed SOCIOCULTURAL] A positive comment or opinion of a 0.704 0.653

family member.
9.2 [Accessed SOCIOCULTURAL] A negative comment or opinion of a 0.701 0.687

family member.
9.3 [Accessed SOCIOCULTURAL] The positive opinion of a reference 0.750 0.619

person or group used as a basis of comparison.
457
9.4 [Accessed SOCIOCULTURAL] The negative opinion of a reference 0.833 0.801
person or group used as a basis of comparison.
9.5 [Accessed SOCIOCULTURAL] Positive comments of a friend, a 0.733 0.676

newspaper article or the views of an expert on a blog or other internet
platforms.
9.6 [Accessed SOCIOCULTURAL] Negative comments of a friend, a 0.759 0.657

newspaper article or the views of an expert on a blog or other internet
platforms.
9.7 [Accessed SOCIOCULTURAL] The positive opinions of individuals in 0.786 0.748

the same social class as you.
9.8 [Accessed SOCIOCULTURAL] The negative opinions of individuals in 0.764 0.651

the same social class as you.
9.9 [Accessed SOCIOCULTURAL] Cultural and subcultural factors, such 0.466 0.396
as your language preference, beliefs and values.
10.1 [Accessed COMMUNICATION] Traditional advertisements about 0.649 0.761

10.2 [Accessed COMMUNICATION] Online advertisements about motor 0.645 0.596

vehicle insurance.
10.3 [Accessed COMMUNICATION] Any person that is a source of a 0.622 0.597

product referral (buzz agent) and tells you about motor vehicle insurance.
10.4 [Accessed COMMUNICATION] Customised messages that 0.589 0.535

insurance companies communicate to you about motor vehicle insurance.
10.5 [Accessed COMMUNICATION] Word-of-mouth, advice, 0.553 0.459

recommendations on social media posts about motor vehicle insurance.
10.6 [Accessed COMMUNICATION] Paid-for social media efforts by 0.568 0.451

insurance companies about motor vehicle insurance.
11.1 [Accessed PSYCHOLOGICAL] Your motivation to purchase 0.731 0.684

11.2 [Accessed PSYCHOLOGICAL] Your perception about motor vehicle 0.737 0.687
insurance.
11.3 [Accessed PSYCHOLOGICAL] Your perception about a certain 0.701 0.669

insurance company.
11.4 [Accessed PSYCHOLOGICAL] The process of how you purchase 0.682 0.618
and consume information, as well as the experience gained.
11.5 [Accessed PSYCHOLOGICAL] Your personality (specific qualities, 0.520 0.526

attributes, traits).
458
11.6 [Accessed PSYCHOLOGICAL] Your attitude towards motor vehicle 0.672 0.675
insurance.
11.7 [Accessed PSYCHOLOGICAL] Your attitude towards a certain short- 0.599 0.501
term insurance company.
12.1 [Accessed PRE-PURCHASE] [Accessed In your search for 0.582 0.492

information, you think about previous experiences.
12.2 [Accessed PRE-PURCHASE] In your search for information, you 0.752 0.737
request quotations from more than one short-term insurance company.
12.3 [Accessed ALTERNATIVES] You compare the price of the different 0.749 0.756
quotations to measure if there is a policy that will contribute to your need
12.4 [Accessed ALTERNATIVES] You compare the quality of the different 0.778 0.802
quotations to measure if there is a policy that will contribute to your need
12.5 [Accessed ALTERNATIVES] You compare the brand list (list of 0.580 0.498
different companies) to measure if there is a policy that will contribute to
your need satisfaction and lifestyle.
Table D3: Total Variance Explained for EFA
Total Variance Explained
Extraction Sums of Squared Rotation Sums of Squared

Factor Initial Eigenvalues Loadings Loadings
% of Cumulative % of Cumulative % of Cumulative

Total Variance % Total Variance % Total Variance %
1 10.540 30.115 30.115 10.152 29.005 29.005 4.315 12.327 12.327
2 3.584 10.241 40.357 3.184 9.096 38.102 3.513 10.037 22.364
3 3.146 8.988 49.344 2.799 7.996 46.098 3.353 9.580 31.944
4 2.056 5.875 55.220 1.703 4.866 50.964 3.172 9.062 41.007
5 1.967 5.620 60.840 1.546 4.416 55.380 3.044 8.696 49.703
6 1.239 3.540 64.380 0.873 2.494 57.874 2.430 6.943 56.645
7 1.109 3.169 67.549 0.739 2.112 59.986 1.169 3.340 59.986
8 0.953 2.722 70.271
9 0.822 2.349 72.620
459
10 0.761 2.174 74.794
11 0.716 2.046 76.840
12 0.704 2.010 78.851
13 0.609 1.739 80.590
14 0.590 1.686 82.276
15 0.514 1.469 83.745
16 0.495 1.414 85.159
17 0.476 1.360 86.519
18 0.462 1.320 87.839
19 0.437 1.248 89.087
20 0.423 1.208 90.295
21 0.393 1.123 91.418
22 0.339 0.967 92.386
23 0.333 0.953 93.339
24 0.329 0.940 94.279
25 0.282 0.806 95.085
26 0.267 0.762 95.847
27 0.241 0.689 96.536
28 0.233 0.666 97.202
29 0.192 0.549 97.751
30 0.173 0.495 98.246
31 0.164 0.469 98.715
32 0.141 0.402 99.117
33 0.131 0.375 99.492
34 0.098 0.280 99.772
35 0.080 0.228 100.000
460
Table D4: Rotated Factor Matrixa for EFA
Rotated Factor Matrixa
Items Factor
1 2 3 4 5 6 7
8.1 [Accessed MARKETING EFFORTS] Type of 0.620

personal motor vehicle insurance policies offered at
a company.
8.2 [Accessed MARKETING EFFORTS] 0.416 0.473

by a short-term insurance company.
8.3 [Accessed MARKETING EFFORTS] Price of a 0.602

policy.
8.4 [Accessed MARKETING EFFORTS] Way in 0.644

which you can purchase personal motor vehicle
insurance.
8.5 [Accessed MARKETING EFFORTS] The 0.479 0.448

company’s physical evidence efforts.
8.6 [Accessed MARKETING EFFORTS] The 0.711

duration of the process to get a quotation.
8.7 [Accessed MARKETING EFFORTS] The way in 0.679

which the employees in the company interact with
consumers.
8.8 [Accessed MARKETING EFFORTS] 0.512

Partnerships or associations with other companies
that you trust.
9.1 [Accessed SOCIOCULTURAL] A positive 0.300 0.628

comment or opinion of a family member.
9.2 [Accessed SOCIOCULTURAL] A negative 0.798

comment or opinion of a family member.
9.3 [Accessed SOCIOCULTURAL] The positive 0.363 0.604

opinion of a reference person or group used as a
basis of comparison.
9.4 [Accessed SOCIOCULTURAL] The negative 0.859

opinion of a reference person or group used as a
basis of comparison.
9.5 [Accessed SOCIOCULTURAL] Positive 0.314 0.634

comments of a friend, a newspaper article or the
views of an expert on a blog or other internet
platforms.
461
9.6 [Accessed SOCIOCULTURAL] Negative 0.722
comments of a friend, a newspaper article or the
views of an expert on a blog or other internet
platforms.
9.7 [Accessed SOCIOCULTURAL] The positive 0.694

opinions of individuals in the same social class as
you.
9.8 [Accessed SOCIOCULTURAL] The negative 0.719

opinions of individuals in the same social class as
you.
9.9 [Accessed SOCIOCULTURAL] Cultural and 0.446

subcultural factors, such as your language
preference, beliefs and values.
10.1 [Accessed COMMUNICATION] Traditional 0.397 0.742

advertisements about motor vehicle insurance.
10.2 [Accessed COMMUNICATION] Online 0.510 0.551

advertisements about motor vehicle insurance.
10.3 [Accessed COMMUNICATION] Any person 0.761

that is a source of a product referral (buzz agent)
and tells you about motor vehicle insurance.
10.4 [Accessed COMMUNICATION] Customised 0.690

messages that insurance companies communicate
to you about motor vehicle insurance.
10.5 [Accessed COMMUNICATION] Word-of- 0.558

mouth, advice, recommendations on social media
posts about motor vehicle insurance.
10.6 [Accessed COMMUNICATION] Paid-for social 0.642

media efforts by insurance companies about motor
vehicle insurance.
11.1 [Accessed PSYCHOLOGICAL] Your motivation 0.690

to purchase personal motor vehicle insurance.
11.2 [Accessed PSYCHOLOGICAL] Your 0.758

perception about motor vehicle insurance.

perception about a certain insurance company.
11.4 [Accessed PSYCHOLOGICAL] The process of 0.626

how you purchase and consume information, as
well as the experience gained.

personality (specific qualities, attributes, traits).
462
11.6 [Accessed PSYCHOLOGICAL] Your attitude 0.782
towards motor vehicle insurance.
11.7 [Accessed PSYCHOLOGICAL] Your attitude 0.653

towards a certain short-term insurance company.
12.1 [Accessed PRE-PURCHASE] [Accessed In 0.348 0.514

your search for information, you think about
previous experiences.
12.2 [Accessed PRE-PURCHASE] In your search 0.814

for information, you request quotations from more
than one short-term insurance company.
12.3 [Accessed ALTERNATIVES] You compare the 0.811

price of the different quotations to measure if there
is a policy that will contribute to your need

quality of the different quotations to measure if there
is a policy that will contribute to your need

brand list (list of different companies) to measure if
there is a policy that will contribute to your need
a Extraction Method: Principal Axis Factoring.

Rotation Method: Varimax with Kaiser Normalization.
Table D5: Classification Table for baseline LogReg model
Observed Predicted
Do you currently have

motor vehicle
insurance for your
personal motor
vehicle?
No Percentage
purchase Purchase Correct
Step 1 Do you currently No 181 22 89.2

have motor purchase
vehicle insurance
for your personal Purchase 23 205 89.9
motor vehicle?
Overall Percentage 89.6
463
Table D6: Variables in the equation to test linearity assumption
B S.E. Wald df Sig. Exp(B)
Step 1a ME 0.237 2.099 0.013 1 0.910 1.268
SE_P -1.920 2.395 0.643 1 0.423 0.147
SE_N 2.401 1.851 1.684 1 0.194 11.039
CS_B -2.323 1.753 1.756 1 0.185 0.098
PA -1.269 2.241 0.321 1 0.571 0.281
DP 0.945 1.757 0.289 1 0.591 2.572
Age 0.061 0.013 23.064 1 0.000 1.063
Highest level of 1.156 0.246 22.117 1 0.000 3.177

education
Household income 1.593 0.279 32.531 1 0.000 4.921
ME by ME_ln 0.225 0.978 0.053 1 0.818 1.253
SE_P by SE_P_ln 0.492 1.196 0.170 1 0.680 1.636
SE_N by SE_N_ln -0.990 0.920 1.156 1 0.282 0.372
CS_B by CS_B_ln 0.737 0.953 0.598 1 0.439 2.090
PA by PA_ln 0.850 1.049 0.657 1 0.417 2.340
DP by DP_ln -0.199 0.833 0.057 1 0.811 0.820
Constant -7.423 3.646 4.145 1 0.042 0.001
a. Variable(s) entered on step 1: ME, SE_P, SE_N, CS_B, PA, DP, 21. Age,
25. Highest level of education, 28. Household income, ME * ME_ln , SE_P * SE_P_ln ,
SE_N * SE_N_ln , CS_B * CS_B_ln , PA * PA_ln , DP * DP_ln .
464
APPENDIX E
- PROFESSIONAL EDITING CERTIFICATE -
465
466

Thesis Van Huyssteen N

Uploaded by

Copyright:

Available Formats

Thesis Van Huyssteen N

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thesis Van Huyssteen N

Uploaded by

Copyright:

Available Formats

A CONSUMER PURCHASE DECISION MODEL

FOR THE SOUTH AFRICAN PERSONAL MOTOR VEHICLE

NADIA VAN HUYSSTEEN

submitted in accordance with the requirements

for the degree of

DOCTOR OF PHILOSOPHY IN MANAGEMENT STUDIES

UNIVERSITY OF SOUTH AFRICA

Supervisor: Prof. S. Rudansky-Kloppers

Name: Nadia van Huyssteen

Student number: 5084-289-7

Degree: PhD Management Studies

(Mrs Nadia van Huyssteen)

My heartfelt thanks goes to the following people:

My supervisor, Prof. Sharon Rudansky-Kloppers, thank you for your commitment to

Retha Burger for editing my dissertation.

Purchase behaviour; purchase decision process; external decision-making influences;

Die lae penetrasiekoers van motorvoertuigversekering in Suid-Afrika plaas enorme

Koopgedrag; koopbesluitproses; eksterne besluitnemingsinvloede; korttermyn-

Izinga lokungena eliphansi lomshwalense wezimoto eNingizimu Afrika libeka

Kubalulekile ukuthi izinkampani zomshwalense wesikhathi esifushane zithole

Idatha yobuningi iqoqwe ngohlu lwemibuzo oluku-inthanethi. Ucwaningo lulandele isu

komshwalense kanye nabathengi abangase babe ngabathengi siqukethe imithelela

Indlela yokuziphatha kokuthengwayo, inqubo eyisinqumo sokuthenga, amathonya

Figure 1.1: Layout of the thesis .............................................................................................22

AVE Average variance extracted

CEO Chief executive officer

CFA Confirmatory Factor Analysis

CFI Comparative Fit Index

CMIN Normed Chi-Squared/df

CPDFPMVI Consumer purchase decision for personal motor vehicle

ECU Electric control unit

EFA Exploratory Factor Analysis

FIA Financial Intermediaries Association of Southern Africa

FPI Financial Planning Institute of South Africa

FSB Financial Services Board

FSCA Financial Sector Conduct Authority

GDP Gross domestic product

IFRS International Financial Reporting Standards

ILASA Institute of Loss Adjusters of Southern Africa

INSETA Insurance Sector Education and Training Authority

IoT Internet of Things

IRMSA Institute of Risk Management South Africa

LDV Light delivery vehicle

MSV Maximum Shared Variance

NPV Negative predictive value

OEM Original equipment manufacturer

PPV Positive predictive value

RMSEA Root Mean Square Error of Approximation

RTMC Road Traffic Management Corporation

SAIA South African Insurance Association

SAICB South African Insurance Crime Bureau

Sasria South African Special Risks Insurance Association

SAUMA South African Underwriting Managers Association

SEM Structural equational modelling

SEU Subjective expected utility

SQRT Square root

TLI Tucker Lewis Index

US United States of America

BACKGROUND TO THE STUDY