COMPILED BY A-MEDIA 09123315386 １

EDU 316
RESEARCH METHODS AND STATISTICS IN EDUCATION

COMPILED BY A-MEDIA

NOTE: Kindly note that this PDF was not made available to replace the textbooks
but rather provided as study guide, in other to help students have more understanding
on the course and to give an insight of how the C.A/EXAM will be. You are expected to
pay attention so that you can understand accordingly. Make sure to share this Study
Guide with your colleagues. Success is yours by God`s grace.

COURSE CONTENTS

 Concept of research
 Case studies method
 Survey research
 Longitudinal studies
 Correctional studies
 Casual comparative studies
 Cross-sectional studies
 Historical research
 Students` test
 Chi-square test

Educational Research
There are probably as many meanings of research as there are researchers. These
definitions underscore certain common characteristics. A few examples are given here

 J.W. Best: That activity which is directed towards the development of the science of
behavior in educational situations to provide knowledge that will permits educators to
achieve their goals by the most effective method.
 Lazarsfeld and Sieber: All effort carried out by the public/private bodies in order to
improve educational methods and educational activity in general, whether involving

１
COMPILED BY A-MEDIA 09123315386 ２

scientific research at a high level or more modern experiments concerning the school
system and educational methods.
 Monroe: The final purpose of educational research is to ascertain principles and
development procedure in the field of education.

Extra: the above are the various definitions of Educational Research and the meaning of
the concept in summary is as below;

Meaning
• It is a systematic manner of understanding educational processes and system properly in
order to improve, make it more effective and grow.
• It is the use of the scientific methods to study education issues, variables and processes.

CONCEPT OF RESEARCH
Re- search means search again/ find out again. It is a search for knowledge. It is a
systematic process of collecting and analyzing information in order to increase our
understanding of the concerned phenomenon. Research is primarily meant to advance
knowledge on an investigated issue/phenomenon.

Re-search is a process by which a person observes a phenomena again and again by

collecting relevant data on the phenomena and draws some conclusions based on the data
collected. It is a systematic process of finding a solution to an identified problem.

Educational research is the formal, systematic application of the scientific method to the
study of educational problems, so as to explain, predict and/or control educational
phenomena

STEPS/PROCESSES INVOLVES IN CARRYING OUT RESEARCH

• Identification of research problem.
• Review of literature.
• Stating the purpose of research.
• Determining specific research questions.
• Formulate hypotheses (if necessary).
• Choice of methodology for data collection.
• Data collection.
• Analysing and interpreting the data.
• Reporting and evaluating the research.
• Communicating the research findings and possible recommendations.

Types of Research
We have the types based on purpose and Method.
Research Types based on purpose

２
COMPILED BY A-MEDIA 09123315386 ３

1. Fundamental or Basic Research is concern with studies that focus on the foundation of
a theory. It discovers broad general principles. It’s aimed at making generalization. It has
no immediate or planned application. It supplies information that has a broad base of
applications. It adds to the already exiting organized body of scientific knowledge. It is
conducted for the purpose of theory development.
2. Applied Research aims at finding solutions to an immediate societal problem. Its
propose is to improve/increase a product or a process (profession practice). It adapts the
theories developed through fundamental research to solve problems.
3. Action Research is a form of applied research whose primary purpose is to increase the
quantity, impact and justice of education for professional practice.It focuses on immediate
application. It places emphasis on a problem in a local setting. It aims at solving immediate
classroom problem or achieving a goal in practice and uses convenient sample which is
immediately available. It uses convenient sample which is immediately available. It focuses
on collection of raw data

Research Types based on Methods

1. Quantitative Research is based on the measurement of quantity or amount. It is
applicable to phenomena that can be expressed in terms of quantity. It is concerned with
testing hypotheses. It utilizing statistical methods in analysis. It data collection methods
rely on random sample. It produces results that are easy to summarize, compare and
generalize.
2. Qualitative Research involves interpretive, naturalistic approach to obtaining and
analysing data on a subject matter. It requires asking broad questions. It is restrictive in
testing hypotheses. It’s linked with the philosophical and theoretical stance of social
constructionism. It involves understanding of human behaviour and reasons that govern
such behaviour which requires asking broad questions and collecting data in the form of
words, images, video e.t.c.
3. Mixed-Method Research is the combination of both quantitative and qualitative
methods in the design of a single study. It is a review of quantitative studies about a
particular phenomenon combined with a review.

Sources of research problems

 Personal experience
 Review of Professional Literatures
 Problem from Theories
 Attending Professional Conference
 Practicing Research Workers
 Technology

Characteristics of a Research problem

 Must be stated clearly and completely.
 Stated in testable format
 Must be expressed in terms of relationship between variables.
３
COMPILED BY A-MEDIA 09123315386 ４

 Should be limited in scope.Should be consistent with known facts.

 Should be stated in simple terms.
 Should be amendable to testing within a reasonable time.
 Should not represent moral or ethical positions.

Forms of Descriptive Research

 Case study
 Survey
 Logitudinal
 Correlational
 Casual-comparative/ex-post fact
 Cross-sectional
 Historical research

CASE STUDY RESEARCH

Case Study Research is useful for investigating trends and specific situations. It is an
intensive data gathering exercise geared towards a thorough understanding of a given
social unit. It is an empirical inquiry that investigate a contemporary phenomenon within
its real-life context; when the boundaries between phenomenon and context are not clearly
evident.

It is a small, inclusive and intensive study of an individual or a systematic gathering of

enough information about a person to permit understanding of how he/she functions as
unit of the society. It is a descriptive research, it combines both quantitative and
qualitative in nature, and the data collected could be mixed method. Generalization may
not be possible in case study.

Types of Case Study Research

1. Illustrative Case Studies: primarily a descriptive study that utilize one or two
instances to make unfamiliar familiar and give readers a common language about the
topic in question.
2. Explanatory Case Studies: aimed at providing information on the causes underlying a
phenomenon or behaviour.
3. Exploratory (or pilot) Case Study Research: This is condensed case study performed
before implementing a large scale of investigation. The function is to help identify
questions and select types of measurement prior to the main investigation.
4. Cummulative Case Study: Serves to aggregate information from several past studies
collected at different times for greater generalization.
5. Multiple-Case Study Research: This type explore differences within and between
cases. The aim is to replicate findings across cases because it draws comparisons.
6. Critical Instance Case Study Research: It examines one or more sites for either the
purpose of examining a situation of unique interest with little or no interest in
generalizability.
４
COMPILED BY A-MEDIA 09123315386 ５

Process of developing a Case Study

 Define the objectives of the case study;
 Narrowing down of research goals;
 Identifying the number of unit/subject of the study/target group;
 Selection of the unit/subject of the study;
 Field work;
 Data collection and management techniques;
 Evaluation and analysis of the collected data;
 Writing a final case study report.

Strengths of Case Study Research

 It is strong on reality.
 It provide insights into other similar situations and cases, thereby assisting
interpretation of other similar cases.
 It can be undertaken by a single researcher without the need of a full research team.
 It catch unique features that may otherwise be lost in large scale data.
 It is intelligible because it speaks for itself.
 It can embrace and build in unanticipated events and uncontrolled variables.

Weaknesses of Case Study Research

 The results may not be generalizable.
 It is not easily open to cross-checking, hence it may be selective, biased, personal and
subjective.
 It prone to problems of observer bias.
 It is not possible to quantify the feelings, emotions, reactions, values among others.
 It is difficult to test the reliability and validity of the data collected.

SURVEY RESEARCH DESIGN

Survey Research is the study of large populations by selecting and studying samples chosen
from the populations to discover the relative incidence, distributions and interrelations of
sociological and psychological variables.

It is another type of descriptive research. It is an efficient method which involves

systematic collection of data from a broad spectrum of individuals in educational and other
settings. It gives opportunity to collect data from a later samples to achieve a lot of
information . it is popularly used in non experimental studies in quantitative research.

Characteristics of Survey Research

 Efficiency: Has level of variables that can be measured without substantially
increasing the time or cost of research (Data can be collected at relatively low cost).
 Generalizability: It lend itself to probability sampling from large population.
 Versatility: Enhance the understanding of any educational issue.
５
COMPILED BY A-MEDIA 09123315386 ６

Methods of Survey Research

 Mail Survey: Is a written form of survey research.
 Face to face Survey: Involve oral interview and can also be electronic through the use
of video technology (e.g Skype). It could be in office, home, street etc.
 Telephone Survey: This is popular in advance country. It is within the realm of oral
and electronic interviews. a lot of numbers are needed.
 Email survey: This is when email is used. It is very economical and fast. It has platform
for pictures and sound files for attachment. It include power supply issue, network
problem, internet access , require lists of email address etc
 Internet/Intranet: This is predominantly web survey which are rapidly gaining
popularity due to the proliferation of information technology, offering extreme speed,
limited cost and flexibility advantages.

Structuring of Questions in Survey Research

 Open-Ended Questions: Is a form of question which does not give respondents answers
to choose from, but rather respondents are encouraged to react to the questions with a
sentence, paragraph or even a page.
 Closed-Ended Questions: Is a form that limits the participants responses by allowing
them to choose from either a pre-existing set of dichotomous answers such ass yes/no,
true/false or multiple choice or ranking scale response options.

Basic Guidelines in Constructing Survey Questions

 Comprehensiveness
 Clarity
 Acceptability
 Simple wording
 Multipurpose questions (Avoid double barrel questions)
 Neutrality (Avoid bias phrases)
 Order of questions (Arranged from most familiar to the least familiar for participants)

Strengths of OF Survey Research

 Useful in describing characteristics of a large population.
 Large samples are feasible which makes the result statistically significant.
 It is economical as it is relatively inexpensive to conduct.
 It can be administered from remote locations using mail, email or telephone.
 High reliability is easy to obtain by presenting all subjects with a standardized stimulus.
Observer subjectivity is greatly eliminated.

Weaknesses of Survey Research

 It is not flexible as they require the tool and administration of the tool to remain
unchanged throughout the period of data collection.
 Participants are limited by response choice when close ended questions are used.
６
COMPILED BY A-MEDIA 09123315386 ７

 It may have validity issues.

 It may be error-prone due to non-response expected from respondents.
 It tends to be subject dependent.

Ethical Issues in Survey Research

 Invitation letter to participate in a survey should contain sentences describing the
purpose or goal of the survey.
 Researcher must provide respondents with information on the motivation behind the
survey to elicit their participation and reduce rejection.
 Researcher should ensure that responses supplied are confidential.
 Provide contact information in case where respondents have questions about the
survey.
 Ensure that you acknowledge respondents’ participation in advance.
 Participation in survey research should be truly voluntary.

LONGITUDINAL STUDIES
Developmental research is a type of research of non-experimental research that can be
used to study age-related changes in behaviour.

Longitudinal Research Design is a principal type of developmental research design.

Same or different samples within a population is repeatedly studied over an extended
period of time. It repeatedly observed the same variables over a long period of time.

It is used to observe changes or developmental trends in life of an organization, institutions,

or an individual or group of people. It is a type of research that is carried out for a long
period of time for at least a year and up to 20 years.

Types of Longitudinal Research Design

1. Trend Study: Requires taking different samples from a population whose members
may change at different point in time. E.G (Attitude of 300level students to teaching
and learning of EDU 316). The samples are submitted at different point in time and
then selected samples are used to represent all data collected.
2. Cohort Study: It involves identifying and studying sub-populations based on specific
features. It requires large population at the outset. It simply means sub-group instead
of collecting data from different individuals, you must have a large population so that
at each data collection stage, you cover data from the sub group.
3. Panel Study: Here the researcher surveys the same sample of individuals at different
times during the course of survey. This involves collecting data from the same
individuals each year, by giving them some instrument to collect information about the
survey habit. I.e 100, 200, 300l in a 3 years analysis.

Issues Related to Longitudinal Research Design

 Generalizability of the results to others in their environment .
７
COMPILED BY A-MEDIA 09123315386 ８

 Loss to follow-up and missing data of random.

 Repeated measures on individuals are not independent.
 Observations of varying time points since symptom onset should be recorded.
 Assessing the effect of treatment is difficult.

Stengths of Longitudinal Research Design

 It allows assessment of stability and continuity of several attributes of a sample by
repeatedly observing the same participants.
 It allows identification of developmental trends by looking for common attributes that
the subjects share.
 It avoids cohort effects because the researcher examines one group of people over time
rather than comparing several different groups that represent different ages.
 It allows researcher to discuss how a single individual’s behaviour changes with age.

Weakness of Longitudinal Research Design

 It is very time consuming.
 It is very expensive to conduct. Because it takes a longer period of time, the research
outcome may seem trivial.
 Practice effect may threaten the validity of the study.
 Participant attention may create biased and non-representative samples.
 There may be cross-generalizational problem which may limit the conclusions of the
longitudinal to only the participate growing up while the study was in progress.

CORRELATIONAL RESEARCH DESIGN

A correlation study is a descriptive research type that measures association between two
variables in order to establish a statistically corresponding relationship between them. It is
a bivariate Correlational Study. It investigates a number of variable believe to be
correlated. It involves correlating two or more variables (e.g. Self-concept and academic
performance).

It is involved in the collection of 2 sets of data, one of which will be retrospective, which
will be determining the relationship between them. It is a quantitative method of research
in which two or more quantitative variables from the same group of subjects are determine
to find out if there is a relationship between them.

Correlation studies investigates a number of variable believe to be correlated.It

investigates degree of relationship between two or more variables . There is no
manipulation of variables in correlational research. It involves two or more quantitative
variables or data from the same group of subjects. It provides an estimate of how the
variables are related only.

Purpose of Correlation Studies

８
COMPILED BY A-MEDIA 09123315386 ９

 Explanatory studies: clarifies understanding of important phenomena through the

identification of relationships among variables
 Prediction studies: this occur of a relationship of sufficient magnitude exist between
two variables
 Positive correlation occurs when an increase in variable lead to another and also
decrease in variable lead to another
 Negative correlation occurs when an increase in one variable lead to a decrease in
another and vice versa
 No correlation occurs when two variables are not related

Types of Correlational Research

1. Positive Correlational Research: This is a research method involving 2 variables that
are statistically corresponding where an increase or decrease in 1 variable creates a
like change in the other.
2. Negative Correlational Research: This is a research method involving 2 variables
that are statistically opposite where an increase in one of the variables creates an
alternate effect or decrease in the other variable.
3. Zero Correlational Research: Zero correlational research is a type of correlational
research that involves 2 variables that are not necessarily statistically connected. In
this case, a change in one of the variables may not trigger a corresponding or alternate
change in the other variable.

Characteristics of Correlational Research

 Correlational Research is Backward-looking.
 It is non-experimental. It does not involve manipulating variables.
 Correlational Research is Dynamic.

The Correlation Coefficient

A correlation coefficient is a systematic, scale-variant measure of association between two
random variables. It may however involves more than 2 variables. It is an important value
in correlational research that indicates whether the Inter-relationship between 2 variables
is positive, negative or non-existent. It has a possible correlation coefficient ranging from -
1.0 to +1.0.

A perfect positive correlation is indicated by a value of 1.0 coefficient. A perfect negative

correlation is indicated by a value of -1.0 coefficient.A zero correlation is indicated by a
value of 0.0coefficient.

Bivariate Data
 Person product moment correlation coefficient (PPMCC) - Karl Pearson
 Spearman rank order correlation - Charles spearman

Multivariate Data
９
COMPILED BY A-MEDIA 09123315386 １０

 Multiple regression
 Coefficient of pative correlation

Condition for correlation with bivariate data

 When description of association or relationship between variables, it's of interest to
the researchers
 Only two variables are measures and applicable to each and every subject .

PPMCC
PPMCC is used when continuous data are involved and are calculated from active score or
values rather than ordinary position.

Condition for using PPMCC

1. When describing linear relationship
2. When continuous or active data are involved.
3. When large sample is involved.

Rank order or different correlation

It's donated as rho or RHO and used when ordinal data are involved and ranks are given to
different scores or values rather than the active scores.

Condition for using rank order method

1. When describing monotonic relationship
2. When categorical data are involved
3. When involving smaller sample

Using SPSS : PPMCC method

 Enter the data
 Data analysis : click analysis
 Select corelate
 Click on bivariate
 More lebels in the variables box
 Click on Pearson box
 Click ok

Using SPSS : rank order method

 Enter the data
 Data analysis : click analysis
 Select correlate
 Click on bivariate
 More labels in the variables box
 Click on Pearson box
 Click ok
１０
COMPILED BY A-MEDIA 09123315386 １１

Basic Steps in Carrying out Correlation Study

 Selection of research problem or topic
 Review relevant literature based on the identified variables and sub-variables in the
research topic.
 State your research questions and research hypothesis.
 Determine and select participants based on the locale of the study and characteristics
of the respondents.
 Select measuring instrument.
 Go to the field and collect data.
 Analyze collected data based on the research questions raised and the hypothesis
postulated stated using appropriate statistical technique.
 Interpret the result of analysis carried out.

Advantages of Correlational Research

 Correlational research can be used to determine the relationship between 2 variables.
 The correlation coefficient helps you to determine the degree of statistical relationship
that exists between variables.
 It is less time-consuming and less expensive than experimental research.Correlation
coefficient only reflects the linear relationship between 2 variables.
 It allows collection of much more data than experiments.
 The result is more applicable to everyday life because the studies usually takes place
outside the laboratory.
 It open up a great deal of further research to other scholars.

Disadvantages of Correlational Research

 The information received from correlational research is limited.
 It does not show what causes the relationship.
 It does not capture non-linear relationships and cannot separate dependent and
independent variables.
 It does not account for cause and effect between 2 variables.
 It doesn’t highlight which of the 2 variables is responsible for the statistical pattern
that is observed.
 Correlational research allows the researcher to carry out shallow data gathering.It does
not provide a conclusive reason for why there’s a relationship.

CAUSAL COMPARATIVE STUDIES(EX-POST FACTO)

Causal-comparative research is otherwise called Ex post facto research design (after the
fact). It is an attempt to identify a causative relationship between an independent variable
and a dependent variable. Causal-comparative compare groups to see whether
independent variable(s) has caused a change in a dependent variable after an action has
already occurred. The event cannot be controlled or manipulated by the researcher.

１１
COMPILED BY A-MEDIA 09123315386 １２

It seeks to find relationship between the independent and the dependent variables after an
action or event has already occurred. It investigates possible cause-and-effect relationships
by observing existing condition. It is ex-post facto because both the effect and the alleged
cause have already occurred and must be studied as past facts. It is a non-experimental
research design in which pre-existing groups are compared searching back plausible causal
factor.

It test possible antecedents of events that have happened and cannot be controlled, or
manipulated by the investigations. When it is not possible to manipulate an independent
variable, it is advisable employ casual comparative study.

Types of Causal-comparative Study

1. Retrospective Causal-comparative Study: This type of causal-comparative study
requires the researcher to start investigation with effects and end up to causes after
occurrence of incidence e.g. “Lecturers’ teaching experience and job satisfaction”.
2. Prospective Causal-comparative Study: This type of causal-comparative study
requires the researcher to start investigation with causes and end up to effects after
occurrence of incidence e.g. “Lecturers’ teaching experience and job satisfaction”.

Steps in Conducting Causal-comparative Research

1. Selection of research problem (Topic),e.g. “Lecturers’ teaching experience and job
satisfaction”.
2. Identify and define the variables in the topic, e.g.“Lecturers’ teaching experience” as the
Independent variable. Or “Job satisfaction” as the dependent variable.
3. Review relevant literature based on the identified variables in the research topic.
4. State your research hypothesis usually in a null form, e.g. Relationship “HO: No
significant relationship exist between lecturers’ teaching experience and job
satisfaction”.OR Prediction “HO: Lecturers’ teaching experience as a determinant of job
satisfaction”.
5. Determine and select participants based on the locale of the study and characteristics of
the respondents.
6. Select measuring instrument.
7. Go to the field and collect data.
8. Analyze collected data based on the way hypothesis was stated using appropriate
statistical technique.
9. Interpret the result of analysis carried out.

Advantages of Causal-Comparative Study

 It is used where rigorous experimental approach cannot be used.
 It can be used where the researcher want to know nature of a phenomena.
 It is useful where simple cause-and-effect relationship are been explored.
 It can give a sense of direction and provide a fruitful source of hypothesis testing.

１２
COMPILED BY A-MEDIA 09123315386 １３

Disadvantages of Causal-Comparative Study

 There is the problem of lack of control group in the study which reduces the
manipulation of the independent variable.
 The researcher cannot determine the actual number of factors is the cause.
 Having data in dichotomous form can be problematic.
 It is regarded by some researchers as too simple research design.
 It lacks confirmation unlike experimental study

CROSS- SECTIONAL STUDIES

Cross-Sectional Studies is the investigations that are conducted across different age, status,
classes e.t.c and compared on some set of variables. It is a developmental research that
assumes that growth has taken place in is sample by relying on available data due to its
growth. It does not follow up the same sample over-time but depends on the existing
characteristics that are observables a result of assured growth over a given period. The
design allows collection of data at one point in time from a different specific group.

It is called cross-sectional because the information gathered at cross sections represent

what goes on at a particular point in time. It involves the examination of different ages at
the same time, hence observed differences between groups can be assumed to be
attributed to age or class differences rather than to other variables.

In contrast to longitudinal studies that look at a group of people over a long period of time,
things happening at the moment are the concern of cross- sectional studies.

Strengths of Cross-Sectional Studies

 It is time saving.
 It is not affected by drop out.
 There is sample stability.
 It is cheap to run since it requires little financial implication compared to longitudinal
study.
 It allows for more representation of target population.
 It can be used as secondary data analysis.

Weakness of Cross-Sectional Studies

 Data are distorted by the faulty recollection of the respondent.
 It does not allow the researcher to investigate and interpret causes of a variable.
 There might be errors resulting from simultaneous investigation of variables.

HISTORICAL RESEARCH
Historical Research is an attempt of examining past events or combination of events to
arrive at an account of what has happened in the past. It is a process of systematically
examining past educational events to make us to learn from the past experience and what

１３
COMPILED BY A-MEDIA 09123315386 １４

kinds of solutions we have used in the past, how it can be transformed for the future
decisions in the field of education.

It shows perpetual structure that occurred in the past and over a period of time.It involves
no manipulation and no control of variables like there is in experimental research. It may
be concerned with an individual, a group, an idea, a movement or an institution.

The method comprises of the use of historical sources such as manuscripts, autobiographic
newspapers, records, corporate documents, correspondence, diaries, photographs,
artefacts, patterns, building, artworks, crafts, e.t.c The various historical guidelines
commonly used in the study are external criticism, internal criticism and synthesis.

Nature of Historical Research

 Macro-level ( broad in scope) which focus on the entire country, society, system
 Meso-level (medium level) like the district or senatorial level.
 Micro level (narrow in scope) which focus on a selective set of people, institutions or
events of interest.

Purpose of Historical Research

 To uncover the unknown.
 To answer questions about the past.
 To identify the relationship that the past has with the present.
 To record and evaluate the accomplishment of individuals, agencies or institutions.
 To assists in understanding the culture in which we live.
 To inform people about what has happened in the past so that they can learn from the
past failure and success.

Types of Historical Research

1. Historical Event Research: Examines particular event or process that occurred over
short spans of time.
2. Historical Process Research: Which traces a sequence of events over a number of
years and such studies will cover many periods (e.g The fall of Oyo Empire).
3. Cross-sectional Comparative Historical Research: Is a study that compares two or
more social settings or groups (usually countries) at one particular point in time.
4. Comparative Historical Research: Combines historical process research and cross-
sectional comparative research to understand causal processes at work within
particular groups and to identify general historical patterns across groups. E.g Gangl
(2004) compared welfare state policies and unemployment in Germany and the United
States.
5. Narrative Historical Research: Is a form of qualitative research. It researches a story
involving specific actors and other events occurring at the same time or one that takes
account of the position of actors and events in a unique historical context.

１４
COMPILED BY A-MEDIA 09123315386 １５

6. Interpretive Historical Research: Focuses on culture, tries to see through the eyes
of those being studied, reconstructs the lives of the people studied, and examines
particular individual or groups.

Historical Research Methodology Steps

 Identification of research topic and formulation of research problems or question
(issues in education, accomplishments of an individual, an education policy or
relationship between events)
 Literature review or data collection
 Evaluation of historical research materials (external criticism or internal criticism)
 Data synthesis
 Report preparation or preparation of the narrative exposition.

Strengths of Historical Research

 It is interesting and entertaining to read.
 It gives an overall feel for life in a different era.Make people identify emotionally with
people in the past
 It surrounds Individual and events with mix of many aspects of social reality.
 It can illuminate the effects of key interactions within a culture or sub-culture
 It allows for revaluation of data in relation to selected theories and generalizations

Weakness of Historical Research

 It hides causal theories and concepts.
 It uses rhetoric, ordinary language and common sense logic to persuade and therefore
is subject to logical fallacies of semantic distortion.
 It does little to create general knowledge.
 It tends to be overly individualistic.
 Objects of research will never really be available for observation.
 It gives a fractional view of the past, its knowledge is never complete.
 It depends on surviving records of a limited number of past events.

QUASI EXPERIMENTAL RESEARCH

Quasi Experimental Research means non-equivalence or non-randomized experimental
research (i.e the subject are not randomly assigned to a group). It is applicable in a
situation whereby subjects cannot completely assign to a group. It shares similar
characteristics with true experimental design but not all sources of threats to internal
validity are completely controlled.

Types of Quasi Experimental Research

1. Non-equivalent Control Post test Only Design : This involve two groups of subjects.
It involves administration of experimental treatment to one group of subjects and then
post tested (Experimental group) and another group of subjects is given post test only.
This design is limited in scope since it does not involve randomization of subject (i.e
１５
COMPILED BY A-MEDIA 09123315386 １６

subjects are not assign to group equally). Hence differences in post test may be due to
the initial group differences.
2. Non-equivalent Group Pre Test, Post Test Only Design: This involve the use of
intact groups that is assumed to be equivalent on all variables, which may affect the
dependent hence subjects were not randomly assigned to experimental and control
groups. Pre test is conducted for the two groups to determine initial differences and
the threats to internal validity to the barest minimal.
3. Time Series Design: The selected subjects will be measured on dependent variable
over a periods of time using the same test. It involve taking a series of measurement
before and after the introduction of same treatment condition. This design take effect
when it is difficult to include control group in an experiment.

Advantages of Quasi Experimental Research

 It is easier to set up than true experimental design.
 It minimizes threats to external validity as natural environment do not suffer from the
same problems in artificial one.
 It allows for generalization about population since it is natural experiment.
 It is efficient in longitudinal research that involves longer period of time which can be
followed up in different environment.

Disadvantages of Quasi Experimental Research

 It does not use random sampling in grouping into experimental and control
 It has potential for low internal validity.

EXPERIMENTAL RESEARCH
Experimental Research Design is a process that provides a systematic and logical
procedure of identifying and evaluating the relationship between variables that create a
particular state of affairs under controlled condition. It involves manipulation of one or
more causal factor while keeping all other possible causal factors constant.

The purpose is to determine whether the manipulated factor has an effect on the
phenomenon being studied. In essence experimental research take place when a researcher
attempts to determine a cause-and –effect relationship.

Some Basic Terms in Experimental Research

 Independent Variable – This is the variable that will be manipulated, the “cause” or
treatment variable.
 Dependent Variable – This variable is the effect or outcome of manipulating the
independent variable.
 Experimental Group – The group that receives treatment being investigated.
 Control Group – The group that remains the same (conventional group) for
comparison.
 Randomization: Equal assignment of subjects to group
１６
COMPILED BY A-MEDIA 09123315386 １７

 Pretest: Measure before treatment is administered

 Post test: Measure after treatment is administered

Types of Experimental Research

There are three basic types of experimental research designs. They are:
1. Pre-Experimental Research Design
2. True Experimental Research Design
3. Quasi- Experimental Research Design
Pre-Experimental Research Design
A single or multiple groups of subjects are observed after some intervention or treatment
presumed to cause change. They are “pre-”, indicating preparatory to true experimental. No
pretest, no control group and no randomization. Examples of Pre-Experimental Design are:

 One-Shot Case Design: here, a single experimental group is exposed to a treatment and
observations are made after the measure of that treatment. No randomization and
control group.
 One-Group Pretest-Posttest Design: where only the experimental group is selected as
the study subjects.

True Experimental Research Design

It is characterised by the random selection of participants and the random assignment of
the participants to groups in a study because the researcher has complete control over the
extraneous variables. Types of True Experimental Research Design are:

1) Pre test- post test control Group Design: An experimental group and a control group
are carefully selected through appropriate randomization procedures.

The experimental group is observed (pre test), subjected to the experimental treatment
and observed once again. The control group is isolated from any influences of the
experimental treatment, it is simply observed both at the beginning and at the end of the
experiment. This design solve two major problems:
 We can determine whether change takes place after the treatment
 We can eliminate other possible explanations in terms of confounding variables and
why the change has taken place.

2) Solomon Four- Group Design: This design involves the following steps:
Random assignment of subjects to four groups: one experimental group and three control
groups.

The first group receives the pre test, treatment and post test. The second group receives
the pre test and post test, but not the experimental treatment. The third group receives the
experimental treatment and the post test, but not the pre test. The fourth group receives
only the post test
１７
COMPILED BY A-MEDIA 09123315386 １８

This design is very strong in that the demand for comparison is well satisfied. The groups
are statistically equivalent since subjects are assigned at random. Extreneous variables
such as history, maturation, or regression are controlled. The main setback of this design is
that it requires a rather large sample and much researcher effort.

3) Post Test Only Control Group Design: In this design, a group of subjects is randomly
assigned to experimental and the control groups. The experimental group receives the
treatment while the control group receives no treatment. The two groups are then post
tested and the scores of both groups on dependent variable are compared. The design
is recommended when it is not possible to locate a suitable pre-test or when there is a
possibility that the pre-test may have an effect on the experimental treatment. The
design relies primarily on random assignment to equate the experimental and the
control group.

4) Factorial Design: The researcher manipulates two or more IVs(factors)

simultaneously to observe their effects on the DV. For instance, if a researcher intends
to test effects of two different teaching methods on the achievement of a particular
subject. This design allows for experimenting the two hypotheses in a single research
in order to examine main and interaction effects of the teaching methods.

Validity of Experimental Research

 Internal validity of a research is the extent to which the outcome of the research is a
function of the treatments administered in the specific experiment under consideration.
Extraneous variables which affect internal validity are:

 History: Specific events other than the independent variable that occur between the
first and second measurement of the dependent variables.
 Maturation: Refers to changes in internal conditions of the individuals that occur as a
result of passage of time (such changes include age, learning, fatigue, boredom e.t.c).
 Testing: If the pre-test and post-test are similar, subject may show an improvement
simply as an effect of their experience with the pre-test (i.e becoming test-wise) .
 Instrumentation: A learning gain may be observed from pre-test to post-test because
the nature of the measuring instrument has changed.
 Experimental mortality (Attrition): Refers to the differential loss or subject drop out
from various comparison groups after the research is underway and which may affect
the findings of the investigation.
 Selection-maturation interaction: When the experimental and control groups have
the same pre-test score, some other differences between them may cause one of them
to get higher post-test scores rather than the experimental treatment.
 Expectancy: Human research involves an interaction between an experimenter and
subject, each of which brings some expectation into the experiment which may affect
the actual outcome of the study.
１８
COMPILED BY A-MEDIA 09123315386 １９

 Sequencing Effect: Occurs when participation in one condition affects the response
the subject will make in a subsequent treatment condition.
 Differential Selection: A selection bias exist when a differential selection procedure is
used for placing subjects in the various comparison groups or when intact classes are
employed as experimental or control groups.
 Statistical regression: Refers to a phenomenon when extreme scores in a particular
distribution tend to move or regress towards the mean of the distribution as a result of
repeated testing.

 External validity of a research is the generalization or representativeness of the

research findings, which can be threatened by the following:
 Interaction effects of selection bias and experimental treatment.
 Reactive or interactive effect of pretesting.
 Reactive effect of experimental procedures.
 Inadequate treatment interference.
 Hawthorne effect (Refers to the fact that one’s performance in an experiment is
affected by knowledge of the fact that one is in an experiment).

Ways of confounding variables that affect validity of experimental research are:

 Keeping some factors constant include a control group.
 Randomly assign people to groups.
 Assess equivalence before the treatment with one or more pre-tests.
 Expose participants to all experimental conditions.
 Statistical control for confounding variables (such as Partial correlation, Analysis of
covariance, Structural equation model).

REVIEW OF RELATED LITERATURE

Literature Review is evaluation report of information found in the previous studies related
to one’s selected area of study. It is an essential preliminary task in order to acquaint
yourself with the available body of knowledge in your area of interest/ intended research.
It is an exercise in which the researcher tries to identify, locate, read and evaluate previous
studies, observations, opinions and comments related to his intended research.

It is an effective evaluation of selected document on a research topic. It includes critical

analysis of a segment published body of knowledge through summary, classification, and
comparison of prior research studies, literature and theoretical articles.

Phases of Reviewing of literature

 Identifying all the relevant published materials in the problem area and relating that
part of it which are thoroughly familiar.
 Writing out this foundation of ideas into a chapter two of project.

Major Sources of Literature

１９
COMPILED BY A-MEDIA 09123315386 ２０

The sources of literature could be classified into:

 Primary Source: which are usually original records, historical events, eye witness
account, firsthand information or original thinking relevant to an event.
 Secondary Source: are accounts of events which were created well after the event
occurred. It include articles in newspaper, magazines books etc.

Sources of Literature Review

 Books and textbooks material.
 Periodicals
 Abstracts.
 ERIC (Educational Research Information Centre)
 Research project report, dissertations and theses.
 Newspaper
 Journals

Procedure for conducting review of literature

 Choosing interest area: that is the researcher’s main topic and the variables involved
in thee topic.
 Networking: Technique of gaining information by using references in the recent
articles to find other related articles or information. Basic tools used to locate
references includes Education Index, Reference book, e.t.c.
 Searching: identification of key terms and using them in locating materials related to
the investigation at hand.
 A visit to the library, internet and other educational resource centre such as ERIC,
could help, in searching for published articles, unpublished articles and dissertations.
 Analysis: Examination of previous studies relatedness and differences and the
present study.
 Documentation / Reporting: Involve summarizing in brief to provide a global picture
of whole knowledge of the studies. AVOID disjointed information.
 Paragraphs should be properly link.

Important guidelines on how to present review of literature:

 Introduction: gives a quick idea of the topic in the review of the literature such as the
central theme and presentation of subthemes. Ensure coverage of the study’s specific
purpose
 Arrange the sub-headings or subtopics for review in line with the major variables in
the study.
 It should involve a well organised, systematic and logical presentation of both
empirical and theoretical information related to the study.
 Empirical information should carry the surname of the author (s), years of the study,
topic of investigation, subject used for the study, methods adopted, results obtained
and conclusions drawn that is related to the present study.

２０
COMPILED BY A-MEDIA 09123315386 ２１

 A brief comment can be made at the end of each review indicating how it agrees or
disagrees with earlier studies and highlighting the pitfalls and strength of the study.
 Materials should be organised chronologically i.e earlier work stated before later dates.
 Paraphrase the review reports in the past tense.
 Enclose the short quotation with quotation marks but for large quotations (more than
40 words) should be indent without quotation marks .
 There should be appraisal of review literature.(Gap identification)

References
 This is the list of all the sources that were made use of in conducting the study.
 It consist of all documents, including journal articles, books magazines, newspaper and
unpublished works mentioned in the text.
 Reference section should not be confused with a bibliography.
 Bibliography contains everything in the reference section plus other useful publication
not cited in the manuscripts.
 Citation Styles includes American Psychological Association (APA), Modern Language
Association (MLA), American Medical Association (AMA) etc
 References should be written consistently in the APA format.
 It should be arranged in alphabetical order by the last names (Surname) and initials of
the authors.
 Each reference should be written in single spacing format and a double space should be
left between references.
 List of reference should not be numbered.

Functions of Literature Review

 Bring clarity and focus to the research problem.
 Broadening the researcher’s knowledge based on the research area.
 Improve your methodology.
 Contextualize the research findings.
 Identifying the gaps and areas that need to be further investigated.

STATISTICS
Statistics is divided into two; Descriptive statistics and inferential statistics. Descriptive
statistics include; percent, means, standard deviation. While inferential include z-score, t-
test, chisquare etc.

Inferencial statistics is divided into two ; parametric and non parametric.

1. Parametric inlude; T-Test, PPMC, One way Anowa, Two way Anova, Ancova,
Regression etc.
2. Non-parametric include; Chi-square, Spearman rank order, etc.

INFERENTIAL STATISTICS
２１
COMPILED BY A-MEDIA 09123315386 ２２

 Inference is a logical conclusion based on evidence and prior knowledge.

 Generalization is a logical conclusion based on many observations and data

Inferential statistics allow inference to be drawn on population parameters based on

empirical sample results(empirically obtained quantities/statistics. The unknown
theoretical or population quantities is called parameters. It is CONSTANT.

Statistics is the empirical calculated sample result from data collected. It varies from one
sample to another. In order to make inference, hypothesis is formulated about theoretical
or population parameter and tested on the basis of the obtained sample statistic.

 Hypothesis is a tentative statement that can only be accepted or rejected when tested it
only.
 Sample is a portion of population selected that actually have the characteristics of all
the needed information.

Types of inferential statistics

1) Parametric: statistical techniques that depends on population parameters
2) Non- parametric: statistical techniques that do NOT depends on population
parameters

Decision Rule
Traditional
a. If the calculated value is greater than the table value, hypothesis will be REJECTED, but
b. If the calculated value is less than the table value, hypothesis will be NOT REJECTED.
Modern
a.DO NOT REJECT the null hypothesis if the calculated significant value (p-value) is greater
than 0.05
b.REJECT the null hypothesis if the calculated significant value (p-value) is less than 0.05

Alpha level/error chance/confidence level (0.05, 0.01 etc).

Computation of appropriate statistics ( t-test, ANOVA etc).

T-Test
The T-Test can only analyse two groups; the T-Test/T-Distribution. It was named after a
statistician called William S. Gosset but with appellation “STUDENT”. It is a parametric
statistical test that compares two sample means It tries to find out the probability that the
difference between the means is a real (dependable)difference or a chance (undependable)
difference

General Assumptions of the T-Test

 Data collected must be at interval or ratio scale which must be from the same
population;
２２
COMPILED BY A-MEDIA 09123315386 ２３

 Sample of data must be randomly and independently selected from the chosen
population; Data obtained or to be used must be in a continuous form; and
 Data must be normally distributed in the targeted population

Types of T-test
Basically there are two types of t- tests:
1). One sample t-test; and
2). Two sample t-test: This type is sub-divided into
a. Independent/uncorrelated t-test
b. b.Non-independent/correlated/paired/repeated t- test

One sample t-test

Assumptions of One Sample t-Test
 Random selection of subjects/participants from the population.
 One sample of data(one data set) is collected from the population.
 It compares sample scores(mean) within a known population mean.
 Use to know whether the sampled mean come from a defined population.
 The population from which the sample is taken should be normally distributed
 It cannot be used for a skewed data or for data with outlier

Two sample t-test

It determines difference between the means of two independent samples. It could be sub-
divided into two:
a)Uncorrelated/ unpaired /independent t-test
b)Correlated/ paired/dependent/Repeated Measures/non- independent t-test

Assumptions of Independent t-Test

 Two groups are unrelated to one another;
 The participants/respondents/students should appear in the two groups
 The population from which the samples are taken should be normally distributed
 The scale of measurement is usually at the interval or ratio level
 The subjects/participants must be randomly selected from the population being used
for the study.
 There may be unequal participants in each group.
 It requires large sample

CHI-SQUARE TEST
Chi-square is a statistical computation with categorical variable. It requires no assumptions
about the shape of the population distribution from which a sample is drawn. This method
does not have any dependence on the population of interest and can therefore be referred
to as distribution free or non-parametric method.

Assumptions of Chi-Square
２３
COMPILED BY A-MEDIA 09123315386 ２４

 Observations are randomly selected from a large population else generalizing the
result should be done with caution
 Independence of observations which ensures that each observation should be
generated by a different subject and no subject is counted twice
 The number of expected observation in any given category should be large with at least
five (5) observations in each category in any given analysis

Conditions for using Chi-Square

 When a researcher uses nominal data for comparison as such variable under study
categoricals When the observations are not limited by sample size, this makes the test
useful in rare situations and by their nature contains few observations
 When research questions involves collecting data on a single variable used for
comparison based on how the observations are categorized
 The population is at least 10 times as large as the sample.

Computation procedures for Chi-square

 Determine whether the data were collected using nominal or ordinal
 State the null (H0) and the alternative hypothesis (H1)
 Perform the analysis using statistics and determine whether to reject or retain the
hypothesis
 Interpret the results in line with the stated hypothesis

Types of chi-square
1. Chi square goodness of fit test is appropriate when a researcher attempts to fit a
statistical model to the observed data if he/she wants to determine how close are the
observed values to those which would be expected under the fitted model. It is
commonly used to test association of variables in a two way tables where the assumed
model of independence is evaluated against the observed data

2. Chi-square Test of Independence is used when there are two categorical variables. It
is useful for testing whether variables are dependent or Independent of each other i.e.
the absence of association between two cross-tabulated variables. It is used to test if
the percentage distributions of the dependent variable within each category of the
independent variable are identical.

RESEARCH DESIGN

Research Design is a blueprint/plan for collection, measurement and analysis of data. It is

the conceptual structure within which research would be conducted. It is the arrangement
of conditions for collection and analysis of data in a manner that aims to address the
purpose of the investigation/research.

２４
COMPILED BY A-MEDIA 09123315386 ２５

It is a plan, structure and strategy of investigation conceived so as to obtain answers to

research questions and to control variance. The types of research will determine the
structure of design the research will use for data collection.

Purpose of Research Design

 Serves as a plan of action to any research studies.
 Helps researchers to obtain answers to research problems and issues involved in the
research.
 Help researchers to understand the procedure for data collection.
 Guide researchers on tool for Data Collection.
 Guide researchers about the statistical techniques to be used for analysis.
 Assists the researchers to organise his ideas and thoughts.
Instrumentation refers to all process by which researcher’s attempt to measure variables
or items of interest in the data collection process. It involves development and validation of
a research instrument and how it will be used to obtain the relevant information. E.G

The instrument may be adapted, adopted or self-constructed.

 Adapted instruments are those instrument constructed by other researchers which is
modified by other researcher before use.
 Adopted instruments are those constructed by other researchers which another person
can use without any modification.
 Self-constructed are those developed by the researcher.

Classification of Data Collection Instruments:

 Inquiry forms
 Questionnaire
 Checklist
 Rating scale
 Attitude scale
 Observation
 Interview technique
 Sociometric devices
 Psychological Tests
 Achievement test
 Aptitude test
 Intelligent test
 Interest inventory
 Personality inventory/Test

THE END!!!

２５
COMPILED BY A-MEDIA 09123315386 ２６

REFERENCE: Educational Research Design and Online lectures on EDU316.

DON’T USE THIS STUDY MATERIAL AS A REPLACEMENT TO THE TEXTBOOK.

THANKS.

Make use of this material very well and share with your colleagues. If there is any
thing confusing, message A-Media on Whats-app. BEST OF LUCK IN YOUR
ACADEMIC ACTIVITIES.

WRITTEN AND COMPILED BY A-MEDIA : 09123315386.

２６