CHAPTER FOUR: RESULTS

4.1 Introduction

This chapter contains the findings on the relationship between dietary practices, nutritional

status and school enrollment . The results are presented in the order in which the objectives

are stated in chapter one. Demographic data is presented first followed by results of the

descriptive analysis and finally the relationship between the variables is presented.

4.2 Study Population

The study targeted a sample size of 264 pupils but a sample of 256 was available. Therefore,

the study had a 97% response rate. According to Fleiss, et al. (2013); Fincham, (2015);

Millar et al. (2011), AAPOR, (2011) a response rate of 97% is considered acceptable and is

adequate representation. The results were therefore reported for 256 out of which 8

questionnaires were rejected due to incompleteness.

4.3 Demographic Characteristics of Respondents

Age and gender were important components of the study as they were useful in correlation

of the variable of the study. The study population encountered in class 6 and 7 upper

primary pupils ranged from 11 years to 15 years. Of this population, 55% were male while

45% female as shown in table 4.1. Pupils aged 12 and 13 years had the highest population as

this is the official age of a class 6 and class 7 pupils according to the ministry of education

(NCC, 2014).
Table 4.1 Demographic characteristics of the study population

Demographic characteristics N %

Age

11 years 12 4.7
46.
12 years 118 1
31.
13 years 80 2
10.
14 years 26 2
15 years 20 7.8
Gender
Male 142 55
Female 114 45

4.4 Socioeconomic Status of the Respondents

The socio-economic and demographic characteristics of households of the children were

determined by the education level of the parents, size of the house, type of dwelling and

household asset ownership.

4.4.1 Education Level of the Parents’ Children

Education level of parents does not directly influence dietary practices or nutritional status

of children but it affects the level of decision making regarding the family’s food choices

(UNDP, 2013). The education level may also determine income which may affect a family’s

dietary practices. The survey data of this study revealed that most of the children’s parents

had secondary education, while very few never went to school (Table 4.2).
Table 4.2 Education level of parents of the childrenEducation level

Father Mother

Frequency Percent Frequency Percent

Never went to
school 3 1.2 5 2
Primary education 26 10.2 34 13.2
Secondary
education 127 49.6 87 34
College education 50 19.5 80 31.3
University
education 50 19.5 50 19.5

Education level Father Mother

4.4.2 Type of Dwelling

As revealed in figure 4.5 majority of the children (92%) lived in houses whose wall material

was made of stone followed by 6% who admitted to living in houses whose wall material

was iron sheet. Only 1% of the respondents agreed to be living in grass thatched or houses

made of mud.

Figure 4.1 Type of dwelling

4.4.3 Size of the House

The majority of respondents said they lived in 2 roomed houses (43.4%), while 23%

revealed that they had houses of more than 4 rooms and (19%) lived in one-room houses.

Table 4.3 Size of house

Size of house Frequency Percent

1 room 48 18.8
2 rooms 111 43.4
3 rooms 38 14.8
≥4 rooms 59 23

4.4.4 Asset Ownership

Assets are an important marker of socio-economic position. This study considered the

ownership of six assets: Fridge, television, radio, computer, mobile phone and vehicle. All

which are considered important by urban residents. They were classified and analyzed based

on percentage of assets owned. The assets were arranged in form of a scale and a maximum

of one assigned for each asset, the respondents gave an affirmative answer as owned. The

scale used was classified into quintiles based on percentage of assets owned. Ownership of

24% was classified as lower socio-economic class, 25% to 49% as lower upper, 50% as

middle, 51 to 74% as upper lower and 75 to 100% as upper class (Booyen et al. 2010). This

classification was supported by studies by McKenzie (2014) and Kuppuswamy, 2005)

whereby households with the highestnumber of assets were considered as upper socio-

economic status. It was noted that the number of assets decreased with socio-economic

status as shown in figure 4.2.

 10
9
8
7
6
Mean household
asset score
5
4
3
2
1
0
Lower Upper Lower Middle Lower Upper
Upper
Socio-economic group
Figure 4.2 Mean household asset ownership by socio-economic group

4.5. Children’s Dietary Practices

The first objective of this study was to establish dietary practices of the school-going
children. Dietary practices included the number of meals consumed by the children,
nutrient intake of the children and frequency of consumption of different food groups. The
7-day Food Frequency Questionnaire and 24-dietary recall were used to assess the dietary
practices of children. Approximately 51% of the children reported that the food they
consumed the previous day was the normal amount, 24% said it was more than usual while
25% reported than it was less than usual. This was asked because of daily variations and
irregular eating habits.
Table 4.4 Mean nutrient intake of children

Sex Energy Protein% Fat% CHO% Dietary Vit. A Vit. Iron Zinc
(kcal) fiber( g) (µg) C (mg) (mg)
(mg)
Male 1330.3 11.9 25.8 62.3 16.2 733.8 66.8 6.7 5.9
329.0 3.4 9.2 10.0 6.6 507.7 57.3 2.4 2.8
Female 1266.6 12.6 24.3 63.0 16.0 788.2 71.6 6.5 5.8
320.2 4.6 9.4 11.0 6.3 490.6 57.4 2.2 2.7
Total 1301.9 12.2 25.1 62.7 16.1 758.0 68.9 6.6 5.8
326.0 4.0 9.3 10.5 6.5 499.9 57.3 2.4 2.7

The mean energy intake of children was 1301.9±326 Kcal. The male children consumed
more energy per day than the female, however there was no significant difference in their
mean (Mann-U Whitney test; p≤0.05). With regard to percentage protein and carbohydrate
contribution to daily energy consumption, the mean for girls was slightly higher than boys
(Table 4.4). A similar trend was observed for micronutrients intake. Vitamin A and C intake
was also slightly higher for girls than boys. Iron intake was however lower for girls than
boys.

The data also revealed the frequency of consumption of different foods in the seven days
prior to the survey (Table 4.5). The foods were later categorized into food groups: cereals,
meats, eggs, milk and milk products; legumes, nuts and pulses; green leafy vegetables, other
vegetables; vitamin A rich fruits and other fruits, sweets and sugars and beverages as shown
in Table 4.5.
In general, children consumed food very frequently within the seven days preceding the
survey. The mean cereal consumption for the seven days was approximately 28.9 times per
child; this translates to 4 – 5 times a day. It was discovered that the male children consumed
cereals more frequently (29.9±14.4) than the female children (27.7±14.2). The same goes for
milk and milk products, legumes, nuts and pulses and vitamin A rich fruits. It was however
noted that the girls consumed sweets and sugars, beverages, other fruits and meats more
frequently than boys (Table 4.5) Consumption of eggs was adequate for both boys and girls
with the mean frequency consumption per week being 3.2±3.3 times.
 Table 4.5 Frequency of consumption of different food groups in 7 days

Total
Sex Male Female Average
Cereals 29.9±14.4 27.7±14.2 28.9±14.3
Meats 10.7±7.9 11.3±8.4 10.9±8.1
Milk & Milk
products 12.1±9.7 10.1±7.2 11.2±8.7
Eggs 3.1±3.3 3.3±3.4 3.2±3.3
Legumes, nuts &
pulses 12.4±11.1 11.3±8.8 11.9±10.2
Green Leafy
vegetables 8.1±4.8 9.7±6.1 8.8±5.5
Other Vegetables 9.3±8.4 8.3±6.8 8.8±7.7
Other Fruits 7.9±7.0 9.2±6.8 8.5±6.9
Vitamin A rich
fruits 8.7±6.4 8.5±5.6 8.6±6.0
Sweets & Sugars 14.6±10.3 16.1±9.0 15.3±9.8
Beverages 6.5±6.4 7.4±6.8 6.9±6.6

Qualitative data from FGD (Appendix G) gave an analysis on the theme of dietary practices
and particularly number of meals consumed by the respondents as well as the adequacy of
meals. FGD1 comprising of the male adolescent children responded that they had an average
of two meals a day (lunch and supper) while FGD2 which comprised of female adolescent
children reported an average of 3 meals per day (breakfast, supper and lunch). Of both
genders, those who said they had all the three meals a day highlighted that the amount of
food they received per meal was not always enough and that at times they felt hungry. Of
the interviews conducted during the FGD, majority of the students stated that if allowed,
they would consume certain foods for breakfast, lunch and supper.

Majority of the students stated that if allowed they would choose bread, tea and mandazi for
breakfast. For lunch and dinner, majority of the students stated that if allowed they would
eat chips, indomie (noodles), rice or meat as a meal. None of the students mentioned that if
allowed they would include vegetables, cereals or legumes in their diet. This was an
indication of their dietary practices and their food choices.
4.6 Children’s Nutritional Status

The nutrition status of children was determined by measuring their weight and height. The
BMI for age was then computed for each child and compared against the WHO (2006)
reference standards for girls and boys to get the Z-scores. Table 4.6 and 4.7 shows the
nutrition status of the school going children. The survey data revealed that majority (84%) of
the children were of normal BMI (Z-score of between -1 to 1 SD) with a higher proportion
(88%) of male children in this category than female (80%).

Table 4.6 Nutrition status of children by sex

Nutrition status Male (n=142) Female (n=114) Total (n=256)

Severely underweight 4 1 5
(SD<= -3.0) 2.8% 0.9% 2.0%
Underweight 10 8 18
(SD<= -2.0) 7.0% 7.0% 7.0%
Normal 125 91 216
(SD -1.0 - 1.0) 88.0% 79.8% 84.4%
Overweight 7 15 22
(SD>=2.0) 4.9% 13.2% 8.6%

Only 9% of the children were undernourished with approximately 2% being severely

underweight. Regarding over nutrition, more girls (13%) were overweight (Z-score of ≥
2SD) than (5%) boys (Table 4.6). The mean BMI for age was higher for girls (18.2 ± 2.8)
than boys (17.7 ± 2.3) (Table 4.7). The WHO Z-scores also showed that the mean for boys
was lower than for girls but still within the recommended median range.

Table 4.7 The mean BMI for Age and Z-scores for children

Sex BMI for Age Z-scores

Male (n=142) 17.7 ± 2.3 -0.042 ± 1.06
Female (n=114) 18.2 ± 2.8 0.132 ± 1.17
Total (n=256) 17.9 ± 2.6 0.035 ± 1.11
4.7 Children’s School Enrollment

For this study a 5-point likert scale was used to capture pupil’s responses on school
enrollment . Studies that have used likert scales have either analyzed the items separately or
summed them to create a score for a group of items (Wuensch, 2011; Carifio & Perla, 2011;
Dawes, 2008). The likert scale was ideal for this study as it prevents the respondents from
trying to portray themselves or their families in a light that they believe the researcher or the
society considers more favourable. The responses were scored based on whether they were
positive or negative in terms of involvement. A sure proportion of the pupils were
considered to be highly involved in extra-curricular activities (63%) and participation of
after school games (46%) (Tables 4.8). A few (25%) love sitting in the field during games.
However, approximately 31% reported that they hate involvement in field activities.

With regard to involvement in classroom activities, about 67.6% of the pupils reported that
they volunteer to undertake classroom activities while (23%) reported that they rarely or
never volunteer in class (Table 4.8). Most pupils (66%) reported that they respond to
questions in class while (18.4%) answered that they rarely or never participate in responding
to questions in class. Majority (68.4%) of the pupils reported that they completed
assignments in time with (65.7%) agreeing that their class enrollment was good. Less than
half (48.8%) of the pupils stated that they enjoy undertaking class work while a few (21.5%)
were moderately involved as they only enjoyed undertaking class work sometimes. 29.7% of
the pupils stated that they rarely or never enjoy class work.
 Table 4.8 Pupils responses to questions on school enrollment

School Enrollment Always Often Sometimes Rarely Never Totals

Involvement in extra-curricular activities

I’m involved in extra-curricular 15.2% 48.4% 13.4% 23.0% 0.0% 100%
activities
I participate in after school 9.8% 36.3% 14.5% 37.9% 1.6% 100%
games/clubs
I hate involvement in field 8.2% 22.7% 11.7% 42.2% 15.2% 100%
activities
I love sitting in the field during 9.0% 16.4% 9.0% 39.1% 26.6% 100%
games
Involvement in classroom activities
I volunteer to take activities in 36.7% 30.9% 9.4% 15.2% 7.8% 100%
class
I respond to questions in class 29.7% 36.3% 15.6% 13.7% 4.7% 100%
I complete assignments in time 28.9% 39.5% 18.0% 9.8% 3.9% 100%
I have good class enrollment 13.7% 52.0% 19.9% 7.4% 7.0% 100%
I enjoy undertaking class work 23.0% 25.8% 21.5% 18.8% 10.9% 100%
School attendance
I attend school regularly 37.5% 46.9% 7.8% 5.1% 2.7% 100%
I’m punctual to school 30.1% 43.0% 14.5% 7.0% 5.5% 100%
I always remain in school till end 37.1% 34.8% 7.4% 12.1% 8.6% 100%
day
I’m regularly punished for 5.5% 10.2% 8.6% 35.2% 40.6% 100%
absenteeism
I’m regularly punished for 7.0% 4.3% 6.3% 28.9% 53.5% 100%
running away from school
Regarding school attendance, majority (84.4%) of the pupils reported that they attended
school regularly. A big percentage of the students (73.1%) reported that they were punctual
to school and approximately 71.9% stated that they always remained in school till the end
day. On the other hand, about 15.7% reported to have been punished for absenteeism and
11.3% reported to have been punished for regularly running away from school (Table 4.8).

4.8 Relationships between Variables

This section presents information on the associations and relationships between study
variables. Such information gives insight on how the independent and dependent variables
relate to bring out a clearer picture and understanding on the study.
4.8.1 Relationship between Socio-Economic Status and School Enrollment of
Children

Socio-economic status was examined as a confounding variable to children’s school

enrollment . The study deemed that this variable is linked to enrollment regardless of
children’s dietary practices and their nutritional status (Hanscombe et al., 2012). The socio-
economic status of the household was determined by variables like asset ownership such as
fridge, vehicle, radio, television, computer and mobile phones), size of the house, education
level of the mother and father and type of housing. Spearman correlation coefficient was
used to determine the relationship between the non-parametric variables. It was discovered
that there was a weak positive correlation (r s = 0.160; P≤0.05) between household asset
ownership and class attendance (Table 4.9). This implies that the pupils from households
owning more household items had a better class attendance as compared to those with lesser
household items.
Table 4.9 Relationship between household socio-economic status and school
enrollment of children

Social Involvement Involvement School School

Economic in Class in attendance enrollment
Status activities extracurricul
ar activities
Ownership of Correlation -0.009 0.095 .160* 0.078
household Coefficient
items
Sig. (2- 0.887 0.128 0.011 0.213
tailed)
Father’s Correlation 0.054 0.074 0.049 0.072
education Coefficient
level
Sig. (2- 0.389 0.237 0.434 0.252
tailed)
Mother’s Correlation -0.01 .174* .164* .194**
education Coefficient
level
Sig. (2- 0.876 0.014 0.033 0.002
tailed)
Size of the Correlation 0.107 .154* .129* .179**
house Coefficient
Sig. (2- 0.088 0.014 0.04 0.004
tailed)
No. of Correlation 0.101 0.003 -0.013 0.057
children Coefficient
Sig. (2- 0.106 0.958 0.836 0.367
tailed)

Note: *** = Correlation is significant at the 0.05 level and 0.01 level (2-tailed)
respectively.

The data also revealed a weak positive relationship between the size of the house and class
attendance rs = 0.129; P (0.04) < 0.05), involvement in extracurricular activities rs=0.154;
P(0.014) < 0.05) and class attendance rs = 0.179; P (0.004) < 0.01) combined (Table 4.9).
Regarding education level of the parents, the data showed a positive relationship between
mother’s education level and involvement in extra-curriculum activities, school attendance
and with all the variables combined. There was no significant relationship observed between
the number of children/siblings and involvement in class activities, class attendance and
involvement in extracurricular activities.
4.8.2 Relationship between Dietary Practices and Nutritional Status of Children

The relationship between dietary practices and nutrition status of pupils was determined
using spearman’s correlation co-efficient. As shown in Table 4.10, it was found that
consumption of other food groups including other vegetables, other fruits and beverages
exerted a weak positive influence on the nutrition status of children based on their Z-Scores
(rs = 0.164**; P (0.008) < 0.05, rs = 0.164; P (0.008) < 0.05 and rs = 0.148; P (0.018) < 0.05
respectively). This indicated that children who consumed more of these food groups had
higher Z-scores. A similar trend was observed for BMI for Age as shown in Table 4.10.
Consumption of foods from other food groups such as cereals, meat products, milk and milk
products, legumes, sugar, protein, fat and carbohydrates did not show any significant
correlation with either the children Z-scores or BMI for age (Table 4.10).
Table 4.10 Relationship between dietary Practices and Nutritional Status of
children

BMI for Age Z-scores

Adequacy r -0.076 -0.048
P value 0.226 0.447
Cereals r 0.014 0.006
P value 0.828 0.918
Meats, fish & Chicken r 0.094 0.12
P value 0.132 0.055
Milk and milk Products r -0.001 0.019
P value 0.986 0.76
Green Leafy Vegetables r 0.086 0.11
P value 0.17 0.078
Legumes, nuts & pulses r 0.107 0.108
P value 0.088 0.084
Other Vegetables r .147* .164**
P value 0.019 0.008
Other Fruits r .135* .164**
P value 0.031 0.008
Vitamin A rich Fruits r 0.108 0.118
P value 0.086 0.058
Sugar r -0.004 0.022
P value 0.944 0.726
Beverages r .142* .148*
P value 0.023 0.018
Energy Intake (kcal) r -0.104 -0.042
P value 0.098 0.502
% Protein Intake r 0.038 0.071
P value 0.547 0.26
% Fat Intake r -0.023 -0.004
P value 0.716 0.951
% CHO Intake r 0.003 -0.026
P value 0.963 0.675
Note: *, ** = Correlation is significant at the 0.05 level and 0.01 level (2-
tailed), respectively.
4.8.3 Relationship between Dietary Practices and School Enrollment of Children

Spearman correlation coefficient was used to determine the relationship between dietary
practices and school enrollment of children. The results showed that food adequacy had a
positive correlation (rs = 0.132*; P (0.035) < 0.05) with enrollment of extracurricular
activities in school (Table 4.11). This implies that the children who reported to consume
sufficient amount of food also performed extracurricular activities in school. A similar trend
was observed for consumption of cereals (r s = .184**; P (0.003) < 0.05, meats, fish and
chicken (rs = .159*; P (0.011) < 0.05, other vegetables (rs*.128; P (0.04) < 0.05, vitamin A rich
fruits (rs = .163**; P (0.009) < 0.05, other fruits (rs = .147*; P (0.019) < 0.05 and beverages (rs =
.164**; P (0.008) < 0.05 (Table 4.11). Higher consumption of these food groups by the pupils
seemed to exert a positive influence on enrollment of extracurricular activities in school. Energy
intake (Kcal) (rs = .005; P (0.934) seemed to exert influence on overall school enrollment
(attendance, class and extra-curriculum).

Consumption of other fruits and beverages showed a weak positive influence on school
attendance, enrollment and involvement in extracurricular activities combined (Table 4.11).
However, there was no significant correlation between consumption of other food groups
and pupils school enrollment . Likewise, consumption of other fruits and beverages revealed
a weak positive influence on school attendance, enrollment and involvement in
extracurricular activities combined (Table 4.11). Nonetheless, there was no significant
correlation between consumption of other food groups and pupils school enrollment .
Table 4.11 Relationship between dietary practices and school enrollment
of children
Class School Extracurricul School
activities attendance ar activities enrollment
Adequacy r -0.088 0.032 .132* -0.005
P value 0.159 0.607 0.035 0.934
Cereals r 0.059 -0.021 .184** 0.093
P value 0.351 0.738 0.003 0.138
Meats, fish & Chicken r -0.021 -0.082 .159* 0.003
P value 0.733 0.192 0.011 0.965
Milk and milk Products r 0.006 -0.021 0.113 0.023
P value 0.928 0.734 0.071 0.712
Green Leafy r 0.00 0.027 0.117 0.051
Vegetables
P value 0.994 0.666 0.062 0.412
Legumes, nuts & pulses r -0.073 -0.03 0.113 -0.016
P value 0.247 0.629 0.072 0.796
Other Vegetables r 0.032 0.049 .128* 0.107
P value 0.606 0.432 0.04 0.086
Other Fruits r 0.056 0.069 .147* .134*
P value 0.374 0.271 0.019 0.033
Vitamin A rich Fruits r 0.06 -0.008 .163** 0.113
P value 0.335 0.903 0.009 0.072
Sugar r 0.002 0.071 0.044 0.067
P value 0.978 0.26 0.479 0.287
Beverages r 0.077 0.042 .164** .135*
P value 0.221 0.502 0.008 0.031
Table 4.12 Relationship between dietary practices and school enrollment of
children
Class School Extracurricula School
activities attendance r activities enrollment
Energy Intake (kcal) r 0.041 -0.048 0.005 .139*
P value 0.517 0.448 0.934 0.925
% Protein Intake r 0.046 0.09 0.054 0.098
P value 0.459 0.153 0.389 0.117
% Fat Intake r 0.074 0.01 0.112 0.073
P value 0.238 0.877 0.074 0.242

% CHO Intake r -0.095 -0.04 -0.111 -0.101

P value 0.128 0.524 0.078 0.107

4.8.4 Relationship between Nutritional Status and School Enrollment of Children

The relationship between nutrition status in terms of BMI for Age and Z-scores and school

enrollment was determined using Spearman’s correlation co-efficient. A weak positive

correlation (rs = 0.133; P ≤ 0.05) was observed between BMI for Age and pupils’

completion of school assignment in time (Table 4.12). This implies that the higher the BMI

for Age, the higher the chances of a pupil completing assignments in time. Z-scores on the

other hand, showed a weak positive correlation with inactivity during field games. The

children/pupils who reported that they love sitting in the field during games are more likely

to have a higher Z-scores (meaning that there nutrition status was inclining towards

overweight).
 46

Table 4.13 Relationship between Nutritional Status and School Enrollment

of children

School Enrollment BMI for Age Z-scores

I’m Involved in extra-curricular activities r -0.025 -0.032
P - value 0.69 0.608
I participate in after school games/clubs r -0.022 -0.067
P - value 0.732 0.282
I love sitting in the field during games r 0.108 .158*
P - value 0.085 0.011
I volunteer to take activities in class r -0.013 0.048
P - value 0.84 0.443
I respond to questions in class r 0.055 0.022
P - value 0.379 0.726
I complete assignments in time r .133* 0.118
P - value 0.033 0.059
I have good class enrollment r -0.088 -0.069
P - value 0.161 0.27
I enjoy undertaking my class work r -0.014 -0.03
P - value 0.827 0.634
I attend school every day r 0.035 -0.071
P - value 0.576 0.255
I’m punctual to school r -0.104 -0.09
P - value 0.097 0.153
I always remain in school till end day r -0.012 -0.02
P - value 0.849 0.756
I’m regularly punished for absenteeism r 0.056 0.24
P - value 0.398 0.762
I’m regularly punished for running r -0.013 -0.03
away from school

P - value 0.865 0.568

Note: = Correlation is significant at the 0.05 level and 0.01 level (2-tailed) respectively.
No significant relationship was observed between BMI for Age and Z-scores with the
other variables
 CHAPTER FIVE: DISCUSSION, CONCLUSIONS AND

RECOMMENDATIONS

5.1 Introduction

This chapter gives the interpretation and explanation of the findings presented in

chapter four. It also shows how the findings of the present study relate with those of

other studies. Lastly, this chapter gives a summary of the study as well as the

conclusions and recommendation of the study

5.2 Socio-economic Characteristics of the Respondents

Education level of parents, asset ownership, number of children, size and type of

dwelling are some of the most important socio-economic indicators that determine

children’s dietary practices and nutritional status. Majority of the respondents

revealed that their parents had secondary school education while very few (1%) had

never been to school. According to UNDP (2013), Kenya urban cities have a high

population of educated persons. This study found that there was a positive

relationship between mother’s education level and their children’s school

enrollment . These findings are consistent with those of NAEP, 2015; Heath, 2013;

Mann & Truswell, 2012) who found that highly educated mothers have a greater

success in providing their children with skills that contribute to success in school.

The parent’s education to a large extent determines the type of occupation they

engage in; people with post-secondary training

 48

often have entry into formal employment compared to those with secondary

education or less (Dauda, 2010).

Evidently, the study revealed that there is positive relationship between household

asset ownership and class attendance as well as a positive relationship between the

size of the house and class attendance and involvement in extracurricular.

Wealthier households have more income that can be converted to household assets

and they have been seen to play a major role in determining the different aspects of

school enrollment . Studies conducted by Buchman (2010) and Jiensen and Nielson

(2008) found that household assets had a positive effect on school attendance.

Moreover, Destin and Oyserman (2009) found that when children believed that

they had adequate household and financial assets then they tend to have higher

grades and put more effort into their school work.

Consequently, for this study socio-economic factor was viewed as a confounding

variable as evidence showed that various aspects of socio-economic status

influenced children’s school enrollment .

5.3 Dietary Practices of Children

The first objective of this study was to find out the dietary practices of adolescent

children within public primary schools in Wajir County. The food composition and

the frequency as well as the amount of food are important in terms of school

enrollment . At ages between eleven and fifteen years, the children are in their
 49

developmental state of puberty. During this state of brain development, there are

greater requirements for energy and nutrients. Deficiencies in nutrients and energy

could result in impaired development of the structure of the brain. The effects of

the impaired development are long-term for the mental functioning of the children

(Mann & Truswell, 2012).

This study found out that male children consumed more energy per day than the

female. Barooah, 2012 states that male adolescent children have higher energy

requirements than females and therefore they tend to consume more food than their

female counterparts. Contrary to this, the study found out that proteins and

specifically meats and eggs as well as carbohydrate had the highest contribution to

the daily energy consumption for girls. Subsequently, the mean consumption of

these food groups was higher for girls than for boys. It was also noted that

adolescent girls had a higher consumption of sweets, sugars and beverages than

their counterparts. Overall consumption of protein, fats and carbohydrates was

higher for girls than for boys (Table 4.4).

Zarborskis et al. (2012) had conducted a cross-sectional study on eating habits

among school-age children aged between 11 and 15 years in Lithuania. Similar to

this study, it was found that school aged children had low intakes of fruits and

vegetables. The study also found that there was increased intake of high sugar
snacks, sweets, chocolates, biscuits and soft drink beverages among the adolescent

school-aged children and in

 50

particular girls. High consumption of meat, high sugar foods and high fat foods can

be attributed to feeding practices where children are primarily allowed to choose

what they want to eat, as well as financial situations and work demands of parents

that prevent them from actively participating in planning and organizing their

children’s diet (Neumark-Sztainer, 2006). Besides, high intake of fat and sugar

foods can be linked to many upcoming fast food establishments that sell food near

schools and target school children (Onyiriuka, et al., 2013)

The mean cereal consumption for the seven days was approximately 28.9 times;

this translates to 4 – 5 times a day and the study further discovered that male

children consumed cereals more frequently (29.9±14.4) than the female children

(27.7±14.2). The same goes for milk and milk products, legumes, nuts and pulses

and vitamin A rich fruits. Generally, male adolescents showed higher intake of

energy and dietary fiber. Consequently, Iron intake and Zinc was also higher for

male adolescents than female. A similar trend goes for Micronutrient intake

involving Vitamin A intake which was also slightly higher for boys than girls.

Consumption of eggs, vegetables and fruits was notably low for both boys and

girls.

These findings concurs with a study in Nigerian by Onyiriuka et al. (2013) which

focused on assessing the eating habits of adolescent school girls. It was found that

adolescent girls have a high consumption of daily starch and more worrisome was
that they have low consumption of fruits and vegetables. Onyiriuka et al. (2013)

also
 51

revealed a high consumption of soft drink beverages among adolescent girls despite

the fact that high consumption of soft drinks has been linked to poor bone

formation and increased risk of bone fractures (Erikson, 2006). This findings on

high intake of sweets, sodas, chocolates and beverages high in sugar among school

girls is consistent across many previous researches studies carried out (Stang, et al.,

2007; Onyiriuka, et al., 2013;Warriach, et al., 2009). These foods have high

glycemic index making them quite unhealthy (Brands, 2011).

In general, children had a high frequency of consumption of food within the seven

days preceding the survey. This was proof that eating of snacks in between meals is

a common phenomenal among the adolescent children in public primary schools in

Wajir County. Qualitative data from the FGD also showed that the children

preferred fatty and processed foods (Mandazi (ngumu), rice, chips and Indomie

(noodles). Majority of the children did not consider vegetables as a core part of

their diet. This is in line with other studies such as Liu et al. (2006), where

adolescents increasingly indulge in nibbling non-nutritious foods (snacks mainly)

in between meals which contributes to their overall energy intake. Similarly,

Zarborskis et al. (2012) found that both sexes had high intake of sweets, chocolates,

biscuits , soft drinks and pastries while Lindsay et al. (2012) attributed the high

intake of sweets and other refined foods by adolescents to mothers enticing their

children with unhealthy foods as a way of compensating their absence brought by

work demands.
 52

Further, from the findings of this study it can be presumed that female adolescents

showed a higher risk of obesity as compared to males due to their increased

consumption of foods rich in fats and processed sugars (Table 4.2.2). This

contradicts the study done by Srivastava et al. (2012) where findings showed that

under-nutrition was significantly more prevalent in girls than boys. Perhaps this is

because the study by Srivastava et al. (2012) was carried out in a rural setting and

the dietary practices of children differ from those in an urban setting. However, in

support of this study’s findings are Law et al. (2007) in the journal of public health

who observed that there were differences in obesity prevalence by both age and

gender and that adolescent females were at higher risk of obesity than males.

5.3.1 Dietary Practices and School Enrollment

The dietary practices in this study were assessed using the 24-hour food recall and

the 7-day food frequency. Using the 24-hour food recall and the 7-day food

frequency, the study found that 51% of the children who were interviewed

consumed the normal amount of food in the day preceding the data collection. The

study also showed that 24% of the children consumed an amount which was more

than the usual while 25% consumed an amount that was less than usual in the day

preceding the data collection. With regards to energy intake, the study found that

the average intake for the male respondents was 1330.3 kilocalories per day while

that for. The average intake for the female respondents was 1266.6 kilocalories per

day.
 53

In reference to the journal of the American Dietetic Association, the energy needs

of adolescent are influenced by their activity level, metabolic rates as well as the

increased need to support pubertal growth. As such, adolescent males require

higher caloric requirements due to their variability in growth in comparison to

females. Even though the males require more energy compared to their female

counterparts, the mean energy intakes for both the male and female members in the

study were found to be lower than the recommended daily allowances for the ages

studied. The children used in this study were between the ages of eleven and fifteen

years. The recommended daily energy intake for adolescent females aged between

nine and thirteen years is 2,071 kilocalories (Story & Stang 2005) while those aged

between fourteen to eighteen years is 2,368 kilocalories.

Based on these recommended energy intakes, it is evident that the mean energy

intake for the female respondents was significantly lower than the recommended

daily intake for the adolescents in the age bracket in which the female respondents

were derived. Similarly, Story and Stang (2005) recommend daily intakes for male

adolescents aged between nine years and thirteen users as 2,279 kilocalories while

those aged between fourteen and eighteen years is 3,152 kilocalories. At 1330.3

kilocalories per day as the average intake for the male respondents according to our

study is far lower than that of the recommended daily intakes. The mean energy

intake of children was 1301.9±326 Kcal. This average energy intake is still lower
than the recommended daily intakes for the children aged between eleven and

fifteen years (Story & Stang 2005).

 54

A diet low in energy can lead to delayed puberty and growth retardation among

adolescents (Stang et al., 2007). Insufficient intake of energy may occur because of

certain factors. The most common being inadequate monetary resources to

purchase food, restrictive dieting or chronic illness (Bandura, 2006).

The results showed that food adequacy had a positive correlation (r s = 0.132;

P≤0.05) with enrollment of extracurricular activities in school. This implies that

the children who reported to consume sufficient amount of food performed better in

extracurricular activities in school. Adolescent children who received adequate

amounts of food had better school enrollment . These findings also parallel those by

Belachew et al. (2011) conducted in Jimma zone in Southwest Ethiopia where

adolescent participants aged between 13-17 years who were food insecure had

higher cases of absenteeism and lower educational attainment in comparison to

their more food secure peers.

Also noted was the higher consumption of cereals, meats, other vegetables, vitamin

A rich fruits, other fruits and beverages (Table 4.11) which seemed to exert a

positive influence on enrollment of extracurricular activities in school. The study

also found that the consumption of other fruits and beverages showed a weak

positive influence on school attendance and involvement in extracurricular

activities (Table 4.11).

 55

During childhood, the brain development includes among others the frontal lobes of

the brain. The development of the frontal lobes is greatly impacted by nutrition.

The adequate provision of nutrients and energy causes higher cognitive functions

that are controlled by the frontal lobes. Such functions include the inhibition of

irrelevant stimulation and focusing attention (Bryan et al., 2015). Additionally,

various nutrients affect the integrity and structures of the brain cells. Nutrients also

affect the neurotransmission, brain energy supply, signal transduction and

metabolism (Schmitt, Benton & Kallus, 2005). However, for this study there was

no significant correlation between consumption of the different food groups and

pupils involvement in classroom activities such as completion of assignments and

direct participation in class. This aspect of the research posed stern challenges.

According to Brands, (2011) research into the influence of particular nutrients on

mental development is particularly challenging. The challenge arises from the

specification and characterization of the interaction that occurs between a single

nutrient in the diet and cognitive development and mental enrollment in school

going children. Given this hurdle in the research, it was not possible to quantify the

effect of the various food groups on the classroom activity enrollment of the

student beyond the correlation. A review of studies examining the exact relation of

diet and classroom enrollment in children has only revealed inconclusive results

(Ells et al., 2008, Erickson, 2006).

 56

5.4 Nutritional Status of Children

The nutrition status of children was determined by measuring their weight and

height. The BMI for age was then computed for each child and compared against

the WHO (2006) reference standards for girls and boys to get the Z-scores.

Nutritional status is an important factor in the enrollment of school going children.

Numerous studies have shown the effect that malnutrition has on the cognitive

development. Grantham-McGregor (2001) argued that chronic under nutrition

results in low enrollment levels, especially for school going children. In order to

achieve the full educational potential, optimal nutrition and good health are

required.

According to Ernesto (1990) and UN/ACC/SCN (1990), nutrition affects the

learning ability and intellectual development of children. Many researchers have

explored the effect of nutrition on school enrollment . Many of the researchers have

reported a significant correlation between the nutritional status of the school going

children and their school enrollment and cognitive tests. For instance, Pollitt

(2001) and WHO (2008) show that school going children who are supplied with

adequate nutrition show higher scores in the tests of factual knowledge when

compared to children who do not have optimal nutrition.

The study data revealed that majority (84%) of the children had normal Z-score of

between -1 to 1 SD with a higher proportion (88%) of male children in this

category than female (80%). Only 7% of the children were undernourished with
approximately 2% being severely underweight. Regarding over-nutrition the study

found that 8.6% of

 57

children were overweight with girls having a higher prevalence than boys. These

findings concur to some extent with the findings by KDHS (2014) which indicates

that underweight prevalence in urban children is at 7% while overweight among

children is at 5.7% and rising. Similarly, the findings in this study relates to the

study by Kimani (2014) who found that the double burden of malnutrition which

comprises of under- and over nutrition occurring simultaneously is a reality in

Wajir County. Kyallo et al. (2013) also conducted a similar study and found that

the prevalence of overweight children in public schools was at 11.5 %, with girls

being at higher risk. The differences in outcome of the two studies may be due to a

difference in methodology but it is worth noting that both findings pose a public

health concern.

The mean BMI for age was higher for girls (18.2 ± 2.8) than boys (17.7 ± 2.3)

(Table 4.7). WHO Z-scores also showed that the mean for boys was lower than for

girls but still within the recommended median range. The findings are relatable to

findings by Kyallo et al. (2013) who conducted a study on overweight and obesity

among public and private primary school children in Wajir. The study found that

the prevalence of overweight and obesity was higher among girls (19%) than boys

(16%), which is indicative of a public health problem. The study by Kyallo et al.

(2013) just like this study showed a trend of increasing overweight/obesity within

Wajir County.
Moreover, the findings of this study showed that 7% of the children were

undernourished with approximately 2% being severely underweight. Hence, there

is a
 58

need to ensure that this number is reduced sustainably. It is also worth noting that

the percentage of underweight children (7%) who were undernourished is similar to

that by KDHS (2014) which reported a comparable figure for children in urban like

Wajir. However, it is significantly lower compared to the prevalence rates reported

by other studies. For instance, Chesire et al. (2008) carried out a cross-sectional

descriptive study in Kawangware peri-urban slum in Wajir, Kenya to establish the

determinants of under-nutrition among the school going children between ages six

and twelve. Their study showed that the number of underweight children were

14.9% while 4.5% were wasted. The study by Chesire et al. (2008) concluded that

there was evidence of under-nutrition among the school going children. The

recommendations given were for increased awareness among the parents, children

and teachers on the daily recommended intakes for the boys and the girls as well as

the need for policy development by the Ministry of Education and Ministry of

Health with the aim of alleviating under-nutrition among the school going children.

The differences between the findings of Chesire et al. (2008) and those of this study

may be because their study was limited to Kawangware peri-urban slum while this

study concentrated on the whole of Wajir County. The findings of this study may

also be indicative of the improvements that have been made since then and the need

for sustained efforts in order to reduce the prevalence of under-nutrition among

school going children. Similarly, Mwaniki and Makokha (2013) conducted a

descriptive cross-sectional study in Dagoretti Division in Wajir, Kenya to assess the

nutrition status and

 59

the associated factors for the children in public schools. Their findings revealed that

the prevalence of underweight children was 14.9% while 9.7% of the sampled

children were wasted. Even though the study by Mwaniki and Makokha (2013) and

Chesire et al. (2008) are five years apart, it is worth noting that the setting of the

two studies have similar socio-economic characteristics. Both settings comprise of

unplanned settlements while this study focused on different sub-counties within

Wajir. This shows concern that the nutritional status of the school going children

still requires concerted efforts.

Even though only 7% of the children were undernourished with approximately 2%

being severely underweight, the findings of this study offer no reprieve in terms of

improvements in the nutrition status of the children. This is because the socio-

economic characteristics of the parents of the children on whom this study was

performed are different from the social-economic settings of the families in the

studies by Mwaniki and Makokha (2013) and Chesire et al. (2008).

5.4.1 Nutritional Status and School Enrollment

There has been a lot of focus on the effect of under-nutrition on the enrollment of

school going children. This study, in addition to assessing the effect of under-

nutrition on the enrollment of school going children also explored the effect of

over-nutrition on the enrollment of school going children. It is worth noting that

enrollment was operationally defined to include school attendance, involvement in

classroom activities and involvement in extra-curricular activities. The relationship

between nutrition status

 60

in terms of BMI for Age and Z-scores and school enrollment was determined using

Spearman’s correlation co-efficient. A weak positive correlation (r s = 0.133; P ≤

0.05) was observed between BMI for Age and classroom enrollment in particular

pupils’ completion of school assignment in time (Table 4.12). This implies that the

higher the BMI for Age, the higher the chances of a pupil completing assignments

in time. Z-scores on the other hand, showed a weak positive correlation with

inactivity in during field games. The children/pupils who reported that they love

sitting in the field during games were more likely to have a higher Z-scores

(meaning that there nutrition status was inclining towards overweight).

According to the reviewed literature by Tobin (2013) and Bowman et al. (2004) the

growth of the fast-food industry and busy lifestyles means that children are

consuming more empty calories on a daily basis. Similarly, Shore et al. (2008)

found that the GPAs of non-overweight students were about 11% higher than those

of the overweight students. In the seventh grade 75% percent of all non-overweight

students participated in at least one school-based athletic team, whereas only 61%

of at-risk and 33% of overweight students do. The study by shore et al. (2008) was

a pointer that children who are overweight perform less well compared to those

with normal weight in relation to physical enrollment . The findings by Bowman et

al. (2004) and Shore et al. (2008) are relatable to the findings of this study where it

was found out that that the girls consumed more sweets and sugars, beverages and
that more of the girls than boys were overweight (13.2%). Just like the reviewed

study by Shore et al. (2008) the findings showed that the

 61

children/pupils who reported that they love sitting in the field during games were

more likely to have a higher Z-scores and were inclining towards overweight/obese.

5.5 Recommendations

The findings have shown that malnutrition (under and over nutrition) co-exists

among adolescent school going children within Wajir county. They also show that

various dietary practices directly influence children’s nutritional status and school

enrollment (involvement in class activities, involvement in extra-curriculum

activities and school attendance). The following recommendations are made for

parents:

Recommendations for Parents

i. Children’s dietary practices are directly influenced by their

parents/guardians and if the parents are well informed then, they might offer

positive dietary influence on their children. Nutrition education should be

provided in schools. The school setting provides a good avenue of reaching

large numbers of parents. The need for education is founded on the findings

that have established that adolescent school children who eat adequate

quantities of food have better school enrollment . This recommendation is

based on the fact that the findings have showed that adolescent boys and

adolescent girls have various eating preferences that may need to be

encouraged or discouraged depending on their effect on health. For

instance, it was noticed that girls have higher consumption of sweets and

sugars.
 62

ii. The basis of nutrition for children begins at home where parents are directly

involved in planning, purchasing and preparation of the food. Findings from

this study showed that what children eat influence their enrollment in extra-

curriculum activities, class activities as well as attendance. Therefore,

parents and caregivers should be educated to become aware of the foods

that significantly affect their children’s brain development which directly

affect their school enrollment . Hence, the parents should be encouraged to

take an active role in teaching their children so as they make healthy food

choices as a way of life

Recommendations for Schools

i. Schools have a role to play in encouraging parents to provide support for

their children. This recommendation is based on the findings that adolescent

school going children who are provided with adequate amounts of food

have better enrollment more so school attendance. Schools should conduct

awareness forums on the importance of nutrition/diet on their children’s

education.

ii. From the findings it is also clear that schools also have a role to play in

educating school children; the importance of nutrition, physical activity and

more so the importance of making healthy food choices.

 63

iii. Schools should engage in programmes and activities aimed at providing

nutrition information to children/parents/communities for purposes of

raising healthy children with gained advantage of performing well in

school.

Recommendations for Government Practice

This study has showed that both over and under nutrition are existent within

Wajir County and that they affect children’s enrollment which according to

the ministry of education has been deteriorating for the last years (NCC,

2014). As such, some action plan is needed. Therefore, the researcher

recommends that:

i. The ministry of education should ensure that primary schools have

professional nutritionists attached to the schools. At least three per county

whose work will entail providing nutrition education and counselling to the

children, parents, teachers as well as working with them to ensure that

children meet their nutritional needs.

ii. The government should launch programmes to train and empower citizens

on nutrition information such as use of USDA’s my plate, consumer

education on point of purchase, provisional of recipes as well as information

on eating healthy on a lean budget. When such information is accessible to

the general public/communities, then it will trickle down to school going

children.
iii. The government may liaise with private organizations to provide basic

meals (Breakfast & Lunch) to school-children especially those from the

very poor and food insecure households.

 64

Recommendations for Policy

i. The government should come up with policies safeguarding children’s

school enrollment by offering greater financial assistance in scaling up of

school feeding programmes throughout the country.

5.6 Suggestions for Further Research

The following recommendations for further studies are made in view of the

limitations of the current study

i. Future studies can gather data on parental role in children’s dietary practices

ii. Future researchers can also investigate class-teachers perception on

children’s nutritional status and school enrollment

iii. This study was conducted in primary schools in Wajir County. Future

researchers can use a nationally representative sample to study dietary

practices on nutritional status and school enrollment .

iv. Future studies can also employ an experimental design to examine dietary

practices on nutritional status and aspects of school enrollment .