COURSE INFORMATION:

• Course Title: Educational Research and Statistics

• Course Unit: 2
• Course Status: c
• Semester: First
• Required Study Hour: Two hours per week
• Lecturer in charge: Mr. L.O AMINU

COURSE OBJECTIVES:
On successful completion of this course, you should be able to:
i. explain the concepts and processes of educational research
ii. discuss the different types of research
iii. identify researchable problems in education
iv. demonstrate skills in literature review
v. formulate research questions and hypotheses
vi. explain the different types of research designs
vii. differentiate between population and samples
viii. apply the different methods of data collection
ix. compare the different methods of data collection
x. discuss the properties of a good instrument
xi. organize and present data using different methods
xii. demonstrate proficiency in the use of descriptive statistics
xiii. use inferential statistics to test hypotheses
xiv. write reports of research projects.
MODULE 1: RESEARCH PROCESS
Unit 1: The research Process I
Unit 2: The research Process II
Unit 3: The Research Problem
Unit 4: The Review of Literature
Unit 5: Population and Sample

MODULE 2: RESEARCH DESIGNS

Unit 1: Research Design

MODULE 3: TECHNIQUES AND INSTRUMENTS FOR DATA COLLECTION

Unit 1: Techniques for Data Collection
Unit 2: Validity and Reliability of Instruments

MODULE 4: STATISTICS METHODS

Unit 1: Introduction to Statistics
Unit 2: Organisation and Representations of data
Unit 3: Measure of Central Tendency
Unit 4: Measures of Variability
Unit 5: Measure of Association and Correlation
Unit 6: The Testing of Hypothesis

MODULE 5: WRITING RESEARCH REPORTS

Unit 1: Writing Research Reports
 THE RESEARCH PROCESS I
Introduction
The world which we live is full of many unknown and undiscovered facts. Perhaps, it was because
of this circumstances that has made man to be inquisitive i.e., always trying to find answer to
perplexing and unanswered questions. The process of making enquiries into these unanswered
questions has led to the birth of research. The birth of science however brought about a significant
change in the approach adopted in conducting such enquiries. Science came with a formalized and
systematized approach to enquiring into the unknown in our environment, an approach referred to
as the scientific method. In this unit, you will study what research and education research are, roles
of educational research as well as characteristics of educational research.

Meaning of Research
Many writers have described research in various ways, and according to the type and the area. We
can say that research is the process of arriving at dependable solutions to problems through a
planned and systematic collection, analysis and interpretation of data. You may have to note that
the key words in this description are planned and systematic. It means that research is considered
as a logical scientific thinking. This implies that a study or an activity cannot be regarded as
research unless it is carried into the higher levels of vigorous and scientific reasoning.
We can also look at research as a process of seeking answers to hypothetical questions using
scientific methods of inquiry to produce valid information. It means therefore, that when you use
research methods to solve problems, you are more efficient than either trial-and-error attempts or
“short-run” approaches. This is because you will direct all efforts towards a particular goal based
on a sound hypothetical solution. According to Leedy (1997), research is the systematic process
of collecting and analyzing information (data) in order to increase our understanding of the
phenomenon with which we are concerned or interested. He goes on to say that research is not:
i. mere information gathering;
ii. mere transformation of facts from one location to another;
iii. merely rummaging for information;
iv. a catch-word used to get information.
Best and Kahn (1995) agreed that research is the systematic and objective analysis and recording
of controlled observations that may lead to the development of generalizations, principles or
theories, resulting in prediction and possibly, ultimate control of events.
From these and other definitions, you have to note that one of the most reliable ways of ascertaining
that decisions are correctly made is to use a scientific approach to arrive at those decisions.
Therefore, research is needed in order to arrive at objective conclusions. Thus, the use of data to
quantify statements and very assumptions with empirical evidence becomes imperative.

Educational Research
Educational research is implied when research activities are geared towards unraveling educational
problems or bringing about improvement in teaching and learning. According to Ary et. al.
(1972:21), educational research is “the way in which one acquires dependable and useful
information about the educative process”. While Travers (1964:5) says that educational research
is “an activity directed towards the development of an organised body of scientific knowledge
about the events with which educators are concerned”.
Ali (1996) described educational research as those activities or processes which allow one to
systematically test and/or obtain a body of information, data or knowledge about teaching/ learning
or conditions which affect teaching and learning. This means that research in education is a
systematic attempt to define and investigate pertinent problems involved in teaching and learning.
This can take place within or outside the school setting. It can take place at various levels of
education, such as early childhood, primary, secondary or tertiary levels.
In his own definition, Kerlinger (1964) says that educational research is a systematic, controlled,
empirical and critical investigation of hypothetical propositions about the presumed relations
among natural phenomenon. Education is mainly concerned with the processes which deal with
deliberate change in the behaviour of people through the acquisition of knowledge, skills, attitudes,
interests and appreciation. Therefore, the goal of educational research is to discover general
principles on which interpretations, predictions, explanations and control of behaviour can be
based since educational research is concerned with the study of the problems of teaching and
learning. It means that any research in this area should contribute to some aspect related directly
or indirectly to the teaching-learning situation. Therefore, the topic selected for investigation
should be truly educational, preferably a problematic topic whose findings will add to the
knowledge-bank of education and to lead to some solution of a societal problem.

The roles of Educational Research

Research, in general, plays important roles in human existence. It is an undeniable fact that the
improved conditions of man today have been possible by the application of the finding of scientific
research. In the field of education, research plays a very important role as is the case in other
disciplines. Some of those roles are:
1. The Advancement of Knowledge: Through research, the frontiers of knowledge in the
discipline of education are extended.
2. Increase understanding of educational phenomena: Research in education plays a vital
role in increasing our awareness and understanding of educational phenomena.
3. Providing solution to educational problems: One of the fundamental roles educational
research plays is to help in solving educational problem.
4. Improvement of educational practices: research leads to better and improved practice in
education. Results of educational research reveal those instructional materials, techniques
and methods that prove to be more effective and at the same time indict those that seem to
be ineffective.
5. Bringing about overall development and progress: in the long run, educational research
brings about overall development and progress in the field of education.

Characteristics of Educational Research

Before any research in any field can be said to be scientific it must possess certain characteristics.
All scientific investigations are:
1. Systematic: This means that the steps followed have to be sequential and logical. The
procedures used can be repeated by another research to verify the findings.
2. Empirical: They involve the collection of data which provide the basis for drawing
conclusions. Collections are not based on what the researcher feels or think, but on concrete
evidence gathered from data collected of the phenomena being investigated.
3. Theoretical: Any scientific research aims at producing a set of inter – related propositions
which can explain certain phenomena among variables.
 4. Cumulative: Each scientific investigation tries to build upon the existing facts and theories
and helps in refining and extending the existing principle.
5. Objective: This means that the findings are reported as they are. The biases, prejudices,
beliefs or interests of the researcher and/or the society are not allowed to interfere with the
research procedure and/or results.
6. Replicable/Verifiable: Results or findings of a research can be verified by the researcher
or any other person interested in the findings. This can be done by going through the data
used in the study or the processes involved. The research can be replicated by re-
administering the same instrument or similar instrument of data collection to the same
subjects or similar group of subjects. This replication can help one to justify the
authenticity or otherwise of an earlier conclusion.

Exercise
1. Distinguish between Research and Educational Research?
2. Discuss any four characteristics of Educational Research according
3. Identify any four roles which educational research play
UNIT 2: THE RESEARCH PROCESS II

Introduction
Educational research process is an adoption of the basic research process which utilizes the
methodology of scientific thinking and necessarily goes about its business of problem analysis in
a series of stages and steps of activities. The ultimate goal of these series of activities is to search
for dependable, valid and reliable answers as solutions to some defined questions or problems in
the field of education. In this unit you will also learn types of educational research.

Steps in the Research Process

One of basic characteristic of any scientific activity is that it is systematic. These characteristics
follow a sequential step and these steps constitute the educational research process. Let us now
look at the sequence recommended for any educational research:
1. Identifying and defining the problem: In caring out a research project in Education, the
first step is to identify a problem that is worthy of investigation. These problems or
questions must deal with issues of sufficient consequence to warrant investigation and
should be the one that can be solved through scientific investigation. Examples of
educational problems are: What are the causes of Mathematics failure in Nigerian
Secondary School? Effect of teacher teaching strategies on students` learning outcome in
Mathematics.
2. Review of Related Literature: Having identified and stated the problem in concise and
clear terms, the researcher now embarks on a review of the pertinent literature. This will
help the researcher to gain better knowledge in the area of study. It will also help you to:
i. eliminate duplication of what has been done already;
ii. guide your formulation of research questions or hypotheses;
iii. sharpen your focus of the study as well as given you insight as to whether the
problem is researchable or not; and
iv. the type of problems you may encounter in the process and how to take care of
them.
3. Formulating research questions and/or hypotheses: this stage involves the conjecturing
of the relationship between the concepts and variables identified in the problem. The
 research questions and hypotheses provide clues to the researcher as to where to search for
the solution of the problem. It serves as a tentative answer to the problem. It can be from
the result of employing logical processes of deduction and induction to formulate an
expectation of the outcome of the study.
4. Selecting the Research Design: A research design can be a specification of the operations
for the testing or verification of the hypotheses under a given set of conditions and of
procedures for measuring variables. It involves the selection of persons or things to be
studied. Examples can be descriptive survey, experimental, quasi-experimental, factorial
designs etc.
5. Collection of Data: The stage involves the development of relevant instruments which
have satisfactory properties and administrating these to the subjects involved in the study.
It involves the construction and administration of the measuring instruments. Prominent
among the techniques are questionnaires, observations, interviews, documentary sources
etc.
6. Data Analysis and Interpretations: The data collected in the preceding stage are
subjected to appropriate statistical analysis in order to answer the research questions and
test the hypotheses. This will enable us to generate some research findings from which
conclusions and generalizations are drawn.
7. Discussion of Research Findings: Research findings from the analysis of data are
discussed to justify, interpret, explain and further the development of theories for
knowledge. This discussion is done in the context and direction of the information gathered
in the process of literature review.
8. Drawing necessary inferences/conclusion: At this stage, you are expected to draw
conclusions from your research findings and to make generalizations of your findings as
generated from your samples to the larger population. You will have to bear in mind that
the generation of valid findings, making of correct generalisation and useful
recommendations for possible applications to the field of education help to bring about
progress that may be registered in the development and practice of education.
9. Writing the Research Report: This is the final stage of activities in the process of
conducting research in education. You have the responsibility of making your procedures,
 findings, conclusions and recommendations available to others in an intelligible form. This
involves a clear, concise presentation of the steps in the study through a research report.

Types of Research
Educational research can be classified into different types. These classifications may depend on
the goals, methods of investigation, kind of evidence and analysis used. These classifications are:

On the Basis of Goal Typology

These are basic research and applied research.
(a) Basic Research
This type of research is also called pure research and is concerned with obtaining empirical data
that can be used to formulate, expand or evaluate theory. These are studies carried out in natural
sciences and mathematics for the advancement of knowledge. Its main aim is to extend the frontiers
of knowledge with no regards to the practical application. The experiments are conducted based
on hypotheses and if the hypotheses are confirmed they become laws.
(b) Applied Research
This is the application of theories and principles from basic research to solve educational problems.
It is therefore, interested in testing the workability or usefulness of ideas or theories in practical
situation. The findings help educators to make rational practical decisions about specific problems.
Applied research can be divided into three sub-categories:
i. Action Research: This is research directed towards solving specific educational
problems at the classroom level. It is aimed at developing new skills, finding new
answers or approaches to solve problems of current concern. It is practical and directly
relevant to an actual life situation. It is empirical and relies on actual observation and
realistic information, or data rather than subjective opinions or past experiences.
ii. Evaluation Research: This is concerned with making decision relating to the value or
worth of educational programmes in order to improve their efficiency by making
necessary revisions or modifications.
iii. Research and Development: It aims at developing and testing more efficacious
educational products. According to Nkpa (1997), this is concerned with developing and
 testing curricula, methods and materials to ensure maximum efficiency of educational
products and practices.

On the Basis of Method of Investigation Typology

Classifying educational research based on the methods of investigation are:
(a) Experimental Research
In this type of research, independent variables are manipulated to observe the effects on the
dependent variables. It serves to determine possible outcomes given certain conditions. There are
two groups – experimental or treatment group and the control group. The experimental group
receives the treatment while the control group may not receive any treatment. The difference is
noted and used.
(b) Ex - Post Facto Research
This type of study is similar to experimental study in the sense that it also seeks to establish cause
– effect relationship but differs from it in that the researcher usually has no control over that
variables of interest and therefore cannot manipulate them. In fact, randomisation is not possible.
The subjects may be grouped on the basis of some naturally occurring characteristics. Such
variables like sex, race, intelligence, aptitude, creativity, personality, socio-economic status, etc.
cannot be directly manipulated. Sometimes, this type of research is referred to as causal
comparative studies.
(c) Descriptive Research
This is concerned with either description and interpretation of existing relationships, attitudes,
practices, processes, trends, etc. or the comparison of variables. It does not make attempts to
manipulate variables. It may be divided into many categories. Prominent among them are:
i. Surveys: In this type of research, a number of data-gathering psychometric tools and
procedures are used. These include questionnaires, tests, checklists, rating scales, score
cards, inventories, interviews, etc. The study can be used to ascertain the nature of a
phenomenon from a relatively large number of cases. When you study the entire target
population, the survey is called census. Most of the times, the entire population is too
large to be handled. In this case, you have to use appropriate sampling technique to
obtain a truly representative sample.
 ii. Documentary Analysis: In this type, documents and records are examined for relevant
information. Official gazettes, minutes of meetings, reports of panels and blueprints
can be examined. Content analysis of curriculum materials and classroom lessons can
be included as documentary analysis.
iii. Case Studies: The case study is an intensive study geared towards a thorough
understanding of a given unit. The social unit may be an individual, a group of
individuals, a community or an institution. In other words, you may decide to study a
phenomenon in one school, association, organisation, agency, one student, teacher,
administrator, etc. in order to solve specific problems through in-depth study.
(d) Historical Research
This is a past oriented research; which involves the location, documentation, evaluation and
interpretation of available evidence in order to understand past events. Understanding past events
may lead you to greater understanding of present and future events. It may also prevent future
pitfalls, or even suggest hypotheses which should be used for the solution of existing problems. Its
focus may be on social concerns, educational practices, educational institutions or the educators
themselves. In historical research, evidence from relics, artifacts, documents, records, oral
accounts etc. are usually relied on. If you are undertaking a historical research in education, then
your evidence may be sought from attendance registers, certificates, report cards, inventories,
manuscripts, equipment, records of news talk, etc.
There are two main sources of research information in a historical study. These are primary source
and secondary source. When evidence comes from direct source such as original documents,
photographs, eye-witness accounts, it is called primary source. But when a non-observer mediates
between the original evidence and the investigator as in books, research reviews, newspapers or
stories by non-participants, it is called secondary source. As a researcher, you should always
determine the authenticity of the evidence you use.

On the Basis of the Kind of Evidence and Analysis used Typology

Using this typology, research can be classified as quantitative, qualitative or multiple perspective.
(a) Quantitative Research
This type uses information or data expressed in numerical values. Most experimental studies fall
under this type or category. Data collection methods include tests of various types, experiments,
questionnaire, rating scales etc. Quantitative data are analysed using either descriptive or
inferential statistics.

(b) Qualitative Research

This type uses information which is verbal or non-numerical. It makes use of qualitative data
yielded through interviews, observations, artifacts, and documentary sources, audio and visual
materials among others. Information could be analysed using transcription, coding, historical and
philosophical analysis. This can introduce elements of subjectivity in explaining, describing,
collecting and even analyzing information.
(c) Multiple Perspective Research
This type uses both quantitative and qualitative approach otherwise known as eclectic, is usually
more comprehensive, yielding more generalisable and holistic findings which are more rigorous
than any one approach.

Exercise
1. List the steps in the research process.
2. What are the two types of research classified according to the kind of evidence and
analysis? What is the major difference between the two types?
Unit 3: THE RESEARCH PROBLEM

Introduction
In any educational research there must be a problem which must be identified to justify the
research. This is the goal you have to accomplish; this is because it is the problem that requires
attention. At this time, the questions that bother you should be hinged on what constitute a
researchable problem. In this unit you will read about the problem selection, sources of the
problems, etc.

Identification of Research Problems

Educational research is concern with finding solution to educational problem. As a beginner
researcher, identifying a problem which is worthy investigating is usually a difficult task. You may
spend considerable amount of time examining many research problems without being able to make
up your mind on which to select. This is so because you do not take time to first acquaint yourself
with relevant research literature. Acquainting oneself with relevant research literature helps one to
know what problem is trivial as well as those problems that have been solved by others.
Before you arrive at a researchable topic, there are three systematic stages of reductive deductions
which are very necessary for you to take from the general problem area. These include:
a. locating a general problem area in your particular field of interest;
b. pining this problem down to a manageable size, and
c. state it concisely in a form that can be empirically investigated.
These steps are not as easy to take as they look. But you have to note that a good research work
is a difficult thing to undertake. No easy research topic which can be effectively carried out without
careful and logical planning. But if you allow your interest to be your guide and you choose a topic
that personally appeals to you as a result of your previous activities and readings, you will be
motivated to spend the long and difficult hours and the painstaking efforts that are necessary for
the effective execution of the study. It is of equal important to note that a good research study is
the one that make a definite contribution to knowledge.
Sources of Research Problems
Problems in education could be located from the following sources:
a. Literature: this is a very useful sources of problem identification in education. When you
have an extensive review of literature, it provides you with some researchable problems.
Textbooks, theses and dissertations, research reports in journals and educational
periodicals, conference papers, encyclopedia among others would provide sufficient
insight on researchable problems or topics.
b. Theory: This is a set of propositions which can account for or explain certain phenomena,
events or situations. They propound general principles for which applicability or
educational situations require research. According to Nkpa, in Osuji, et al, theories are
fertile sources of research problem. This is because from theories, relationships among
variables can be predicted, tested and established.
c. Personal experience: one may encounter some problems requiring investigation in the
course of his/her professional engagement. Such encounter could come during field work,
lectures, seminars, teaching practice etc. This may lead to discovery of areas where
knowledge is lacking while answers have to be found.
d. Internet Sources: Through the internet one can avail oneself the opportunity of getting
current research findings or write-ups on any area of interest. You can see different
methods of solving the same problem you have and you can apply such methods into your
situation.

Qualities of a good Educational Problem

For any problem to be judged as a good educational problem, it must satisfy certain criterial. Such
a problem must be:
a. Researchable: This is an aspect that any research problem need to meet. Being
rechargeable means that the data for solving the problem are available and accessible.
Many philosophical and ethical problems cannot be studied empirically; otherwise the
findings can only yield useful information that can be used to find answers to those ethical
and philosophical questions.
b. Significant: A good educational problem should be the one whose solution can make
valuable contribution to either the theory or practice of education.
 c. Novel: A problem worthy of investigation should be such that no known solution to it exits
as at the time of the study. A problem should be new in some way to the extent that it has
not been solved before.
d. Educational in Nature: A problem is educational if it`s finding will have implications in
the field of Education.
e. Viability: A research problem should be viable. In this case, it can be expanded or
followed up in further researches. It should not be a dead end. As you answer the research
questions, further questions which require investigations should be generated.

Background to the Study

This is the introduction to the study which usually contains a brief explanation of the concepts in
the title as they relate to the study. It will put the research problem into perspective. It has to be
sharp and precise without an outburst of personal bias. At this stage, you have to highlight all the
variables of interest in the study as well as their interplay with the research problem. You can even
cite few works already done in the area which prompted your investigation in that area.

Statement of the Problem

This is a statement which pinpoints what is wrong with or about the subject matter of interest. It
needs to be very specific and direct. It is presented in a logical sequence. It starts with such
information required for the understanding of the problem, some justifications including citations
and a declarative statement or an amplification in the form of question. The statement of the
problem is concise, clear and persuasive information on the subject matter of investigation as well
as the variables planned for investigation.

Purpose of the Study

The purpose of the study provides you with an overview of the intention of the study as contained
in the research title and its breakdown in specific terms. What do you intend to do? This is
presented in a clearly and unambiguous statement which must be expressed in a language couched
in proposed undertaken action that is tenable, sensible and practical.
Significance of the Study
This is the rationale for the study. It explains the use or the importance of the findings as well as
who and how such information will be useful. It also explains how the findings would contribute
in terms of extending the frontiers of knowledge, raising new questions or suggesting variations in
the existing practice, or to reveal a gap which the study will close.

Scope of the Study

This gives you the extent of the content coverage which you could tackle with the available
resources. It involves the delimitation of the study which specifies the boundaries to be covered in
the study.

The Hypothesis
The word ‘hypothesis’ is a combination of two words – hypo and thesis. Hypo means “less than”.
Hypothesis means therefore “less than a thesis”. While a thesis is a proposition to be proved or
defended. Hypothesis is therefore a conjectural proposition, an informed, intelligent guess about
the solution to a problem. It is an assumption or proposition whose veracity and validity is to be
established. In other words, the hypothesis is in itself an explanation for certain observed or
observable events, behaviours, phenomena or predictions with regards to how they occur, why
they occur or when they occur. These expectations and explanations must be clear, succinct,
testable and verifiable. The aim of a hypothesis is not to prove anything, but to test whether it
should be accepted or rejected. This is done with data rigorously and painstakingly collected
through empirical research.

The Research Questions

These are questions posed by researcher, answer to which would lead to the solution of the
problem. In constructing research questions, it is important to note that simplicity and clarity of
language are very important.
Exercise
1. What is educational research problem?
2. What are the various sources of research problems?
3. Explain the criteria for selecting a research problem.
UNIT 4: THE REVIEW OF LITERATURE

Introduction
Literature review is the most important aspect of planning in carrying out research in education.
It provides much of the theoretical reference point or basis for undertaking a proposed study.
In this unit, you will be looking at the concept of literature review, sources, steps, and citations in
the review of literature.

The concepts of Literature Review

The review of literature 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.
According to Nkpa (1997); & Osujiet. all (2012), literature review is the systematic study of all
existing work that are relevant to the research work. It is concerned with locating, reading,
evaluating and citing reports of related research. Ali (1996); Osuji et al. (2012) described literature
review as the careful and systematic identification and or location, documentation, analysis and
reporting of information which are relevant, related and useful to the present study. In other words,
it means making an extensive searching, reading and assessment of all available materials written
and related to the problem of investigation. Though this exercise can be tedious but if this vital
aspect is ignored, there is risk of duplicating previous therefore not contributing to the
advancement of human knowledge.

Purpose and Value of Literature Review

Literature review according to Osuji et al. (2012) serves the following purposes in research:
1. It can reveal to you sources of data that you may not have known about their existence;
2. It can describe methods of dealing with problem situations that may be similar to your own;
3. It can reveal to you how other researchers have handled methodological and design issues
similar to yours;
4. It can introduce you to important research personalities whose works and findings you may
not have come across;
5. It can help you to evaluate your own research efforts comparing your work with similar
efforts of others;
 6. It can provide you with new ideas and approaches that may not have occurred to you;
7. It can increase your confidence in selected topic if you find that others have interest in the
topic or have found value in investing time, effort and resources into its study;
8. It can help you determine the sampling strategies that should be used in order to avoid
sampling problems encountered by other researchers.

Sources of Literature Review

Literature review sources can be divided into three categories as illustrated in table below:
Sources of literature Characteristics Examples
Primary sources High level of detail Reports
Little time needed to publish Theses
Emails
Conference proceedings
Company reports
Unpublished manuscript sources
Some government publications

Secondary sources Medium level of detail Journals

Medium time needed to publish Books
Newspapers
Some government publications
Tertiary sources Low level of detail Indexes
Considerable amount of time needed Databases
to publish Catalogues
Encyclopedias
Dictionaries
Bibliographies
Citation indexes

Books: Textbooks remain as the most important source to find models and theories related to the
research area. These provide knowledge in the area where the research intends to cover. In most
cases, textbook present the summary and the author1s interpretation of the conclusion of a series
of a studies in a particular field. You can research the most respected authorities in your selected
research area and find the latest editions of books authored by them.
Journal and periodicals: These contain the original works of others researchers. They contain
authors` own reports of their own studies. It is a very important source of literature review.
Magazine and newspaper: magazines are usually rich in scholarly articles and they can be
effective sources to learn about the latest trends and developments in the research area. Reading
magazines can be the most enjoyable part of the literature review. From this, the researcher could
obtain some useful ideas that could help him design and execute better research.
Students` projects, thesis or dissertations: This is another useful source of information and the
most commonly consulted by the students. In using this source are must be taken in order not to
duplicate the errors in the design of such work.
Encyclopedias and dictionary: These are very useful for adequate definitions and for clearer
understanding of the specialized meanings of key terms and concepts in the study.
The Catalogue: This provides information leading to the location and retrieval of books in the
library. This makes the searching of materials easier because the materials are arranged according
to subject area and provide details about the book.
The Index: This provides information leading to retrieval of article published in a wide range of
journals. Details of such articles are usually arranged according to subjecting area also according
to authors.

Steps in Reviewing Related Literature

Osuji et al. (2012) identify the following steps in conducting literature review:
1. Identify, select and list the key words or concepts which are associated with the topic of
investigation.
2. Use the selected key words or concepts to access relevant references from such preliminary
sources of information as the catalogue, the index and dissertation abstracts or
computerized referencing services.
3. When you have obtained the list of useful references you can now consult libraries and
other information depositories in order to access references materials. You can also make
use of the internet to access the materials.
4. Before you start reading, you should be able to get index cards on which to record the
information, prepare a note card or index card for each reference material consulted, such
that at the end you arrange them sequentially according to subheadings in the literature
review.
5. Read the reference materials and as you read make brief notes and pay attention to the
problem, procedure, design, result or just the summary for theoretical opinions.
 6. For quick identification, each entry on an index card should be clearly coded at the top
using the key words for the broad topic. After this, the author’s names and date of
publication come next, followed by the title of the work and the full citation of the work
and the ideas which you find useful.
7. In writing out the references in the index cards, you have to choose a referencing style and
be consistent with it. You will learn more on this in the section on citation.
8. Ensure that you review or dwell more on the primary sources of information than the
secondary sources. This will enhance the authenticity of your work, and provide you with
comprehensive, unadulterated and un-mutilated first - hand information.
9. Consult the most recent references first. This will help you to save time, get the most recent
materials and avoid redundant and unnecessary materials.
10. You will save a lot of time by first reading through the abstract and/or summary of any
referenced material. This will enable you to quickly ascertain the relevance of the
materials.
11. Lastly, organize and write-up all the insights you received while reading. This should be
a critical appraisal of the state of the art in the area of investigation. You have to ensure
that you understand all the related issues. The ideas must flow in accordance with the trend
of thought.

Citations
In carrying out a research project, we make use of ideas borrowed from other sources or works.
Whenever you lift the idea or ideas from any referenced materials, you must duly acknowledge
the source. This includes verbatim quotations, paraphrased statements, diagrams, tables etc. In
educational research reports some of the styles that are commonly used include:
1. American Psychological Association (A.P.A)
2. Modern Language Association (M.L.A)
3. National Educational Association of the United State (N.E.A)
4. Turabian, Kate L
5. Hubbel, George S
6. Ballou, Stephen V
But most institutions in Nigeria prefer and use the American Psychological Association (A.P.A)
style.
Works by a single author
a. When the name of the author forms part of the narrative only the date appears in
parentheses. For example,
Aminu (2020) demonstrated that students` teachers… or
According to Aminu (2020) students` teachers……
b. When the name of the author does not form part of the narrative, both the name of the
author and date are placed in parentheses with a coma separating them. E.g.
It has been demonstrated that (Aminu, 2020) students` teachers …

Works by two or more authors

a. When a work has two authors, always cite both names every time the reference occurs in
the text. In parenthetical material join the names with an ampersand (&). For example,
(Aminu & Yusuf, 2020).
In the narrative text, join the names with the word "and." For example,
Aminu and Yusuf, (2020) demonstrated that …..

b. When a work has three, four, or five authors, cite only the surname of the first author
followed by "et al." (Latin for "and others") and the year of publication. For example,
Adesanya et al. (2020) found out that…….
Corporate authors
These are associations, government agencies and other corporate bodies whose publications cannot
be credited to specific individuals as author(s). Instead, the name of such corporate body is cited
as the author.
a. The name of the corporate body is usually written in full each time they are cited. E.g. (Tai
Solarin College of Education, 2020).
b. If the name of the corporate author is long and cumbersome and the abbreviation is familiar,
the name may be abbreviated in subsequent citations
(Federal Ministry of Education (FME), 2019).
Subsequent citation can be (FME, 2019).
 (West African Examination Council (WAEC), 2020).
Subsequent citation can be (WAEC, 2020).

Author with the same Surname

a. When two or more authors have the same surname, the initials of the authors are included
even when the year of the publications are difference. For example,
L. O. Aminu (2020) and F. A. Aminu (2020) reported that …….

Citing two or more works within the same parentheses

a. Two or more works by different authors cited within the same parentheses are arranged in
alphabetical order by the first author`s surname and citation separated by semicolons. For
example,
Recent studies (Adesanya, 2019; Aminu, 2018; Yusuf, 2019) …

b. Two or more work by the same author(s) with different years of publication are arranged
by the year of publication. For example,
Previous finding (Aminu, 2016, 2018, 2020) ….

c. Two or more works by the same author(s) in the same years of publication are identified
by the suffixes a, b, c, after the years. The years are separated. For example,
Similar findings (Yusuf, 2018a, 2018b, 2018c,) …

Ethical Considerations in Research

At this moment, you have to adhere strictly to the rules of the game in order to minimize ethical
issues which are often encountered in the conduct of the research studies. According to Anaekwe
(2002) in Osuji et.al (2012), such ethical issues include:
1. Plagiarism: This refers to copying someone’s work without acknowledging him as the
source of such information. In the academic circle, it is a very serious offence which can
lead to litigation, denial of the award of degree or certificate, termination of appointment
among others.
 2. Arm-chair Researching: This refers to people staying on their tables and manipulating
figures and building up a theoretical framework, and later publishing the concoction as an
empirical study. According to Anaekwe (2002), it negates the purpose of research and kills
initiatives and intellectual growth in research business.
3. Faking: The ability to Fake things involves the documentation of false information or
sources of information. As a researcher, you should endeavour to present authentic sources
of information used in the study. This will enhance the content validity of your work as
well as the replication of the work and the retrieval of such source materials by subsequent
researchers.
4. Over-citation of a particular Author: Some students often over-cite the works of their
supervisors and/or lecturers. This is not very good as it limits the research to a microcosm
of available information. Expand the scope of your research literature.

Exercise
1. What is literature review?
2. Justify the need of literature review in educational research
3. Outline the scientific attitudes applicable to research and the ethical issues in research.
4. how does secondary data differ from primary data?
e. Take any textbook or journal of your choice and list ten (10) references cited.
Unit 5: POPULATION AND SAMPLES

1.0 INTRODUCTION
Population and sample are of fundamental importance to research. The research investigation is
based on studying of small representational and proportionate group that are generalised to all
others who were not investigated. Generalisability of research findings is dependent, as much as
possible, on the extent to which the population of the study is defined and on the adequacy of the
sampling procedure used in the in the study. Since this concept are important it shall be discussed
in this unit.

Population
This concept population ordinarily refers to describe the total number of people living in a defined
geographical entity or area. This is when we talk of population of a country. But in research,
population is not limited to human beings alone, it includes objects, events, people that have at
least a common characteristic which is common to all of them. Kerlinger (1981) defined population
as all members of any well-defined class of people, events or objects. It means therefore that any
entity, group or set which constitutes a population must have at least one attribute or characteristic
which is common to all of them. The factors that determine the choice of population is the problem
under investigation.
A population can be classified into target and accessible. The target population is all the members
of the specific group to which the investigation relates, while accessible population is defined in
terms of those elements in the group within the reach of the researcher.

Sample
In some studies, the group of items to which the study relates (i.e. the population) may be small
enough to warrant the inclusion of all the members in the study. When the geographical spread or
area are too wide to cover with the limited time, cost, time and other constraints, it may be very
difficult to study the entire population. It then becomes necessary, reasonable and only feasible to
study a portion of the population which is described as sample. It means therefore that, the portions
of populations that are studied on the bases of which conclusions are made on the entire
populations are called samples. These samples represent populations, so we can have samples of
people, objects, institutions and things. For the purpose of your research projects, you will be
constrained to limit your investigations to smaller controllable samples. To this effect, you may
be faced with a very crucial decision to make. That is, how to select samples that will truly
represent the population to which the findings would be generalised.

Sampling Techniques
A sampling technique is a plan specification how elements will be drawn from the population.
These may be classified into major two groups – the probability sampling techniques and non -
probability sampling techniques.
Sampling Techniques

Probability Sampling Non-probability Sampling

i. Simple random i. Quota sampling
ii. Stratified random ii. Snowball sampling
iii. Cluster sampling iii. Purposive/Judgment sampling
iv. Systematic sampling iv. Convenience sampling
Figure showing the various types of sampling techniques

Probability Sampling
Probability sampling means that every item in the population has an equal chance of being included
in sample. In ideal probability sampling plan, the size of both the population and the desired sample
must be known. The probability sampling techniques include: simple random sampling, systematic
sampling, stratified sampling, cluster sampling, etc.

Simple random sampling

The simple random sample means that every case of the population has an equal probability of
inclusion in sample. The main purpose of using random sampling techniques is to select a sample
which is representative of the population and which can yield data that can be used for
generalisation to a larger population. The simple random sampling is by far the easiest and simplest
probability sampling techniques, in terms of conceptualization and application. The limitation in
the use of simple random sampling is that, it can only be used with small population, since you
have to enumerate or number all the subjects in the population, the costs of obtaining the sample
can be high if the units are geographically widely scattered, and The standard errors of estimators
can be high. The simple random sampling techniques can be carry out through any of the following
means:
a. Balloting or Hat and Draw method: This is otherwise called the lottery method in which
all the subjects in the population are listed or numbered in cards or papers. These papers
are shuffled and re-shuffled. These cards are then drawn from the container one by one,
with or without replacement. Any number picked up forms part of the sample.
b. Table of Random Numbers: This is a more systematic, refined and scientifically
generated random numbers. It is a continuous sequence of numbers which do not appear
in any particular order and which is generated by computers. In recent times, many tables
of random numbers have been generated by computers. These include Snedecor and
Cochran’s (1969), Ten Thousand Random Numbers; The Million Random Digits with
100,000 Normal Deviates by Rand Corporation (1965), Table of 105,000 Random Decimal
Digits by the Interstate Commerce Commission (1979) etc. Some statistical textbooks
contain table of random numbers in their appendices. To use the table of random numbers,
you have to number the population serially from one to the last person, event, object, case
of element. You can then, at random, select a number from any page or point, row or
column and draw your sample using the first two, three or four digits or you can use the
last two, three or four digits. During the process of drawing your samples, if a number
appears twice, or a number is larger than the population size, you have to ignore such
numbers and continue until your sample size is composed.

Stratified random sampling

Stratified sampling is where the population is divided into strata (or subgroups) and a random
sample is taken from each subgroup. A subgroup is a natural set of items. Subgroups might be
based on company size, gender or occupation (to name but a few). Stratified sampling is often used
where there is a great deal of variation within a population. Its purpose is to ensure that every
stratum is adequately represented (Ackoff, 1953).
Systematic sampling
Systematic sampling is where every nth case after a random start is selected. This involves, first
listing in a serial order, all the events, persons, objects or things in the whole population. After
this, the population (N) is divided by the sample size (n) to get the Kth interval. Once the Kth case
is decided, all others are automatically selected. For instance, assuming you have a population of
1,000 people and your sample size is 100. Then Kth position will be given by N/n = 1000/100 =
10. It means that every 10th position or interval is automatically selected as part of the sample.
Thus, numbers 10, 20, 30, 40, etc. are selected. You can even select any number: 1, 2, 3, …. 10
as the Kth number. For example, if the Kth case is 5, then 5, 15, 25, 35, 35 etc. become members
of the sample.
You would have noticed that independence is not ensured in systematic sampling. This is because,
once the first member – Kth is selected, every other member of the sample is automatically
determined. This is a limitation in the use of this method.

Cluster Sampling
In cluster sampling, the population is divided into units or sections with distinct boundaries. Using
simple random sampling is taken from these clusters, all of which are used in the final sample
(Wilson, 2010). Cluster sampling is advantageous for those researchers whose subjects are
fragmented over large geographical areas as it saves time and money (Davis, 2005).
The stages to cluster sampling can be summarized as follows:
i. Choose cluster grouping for sampling frame, such as type of company or geographical
region
ii. Number each of the clusters
iii. Select sample using random sampling
Non probability Sampling
Non probability sampling is often associated with case study research design and qualitative
research. With regards to the latter, case studies tend to focus on small samples and are intended
to examine a real-life phenomenon, not to make statistical inferences in relation to the wider
population (Yin, 2003). A sample of participants or cases does not need to be representative, or
random, but a clear rationale is needed for the inclusion of some cases or individuals rather than
others.
Quota sampling
Quota sampling is a non - random sampling technique in which participants are chosen on the basis
of predetermined characteristics so that the total sample will have the same distribution of
characteristics as the wider population (Davis, 2005).

Snowball sampling
Snowball sampling is a non - random sampling method that
It uses a few cases to help encourage other cases to take part in the study, thereby increasing sample
size. This approach is most applicable in small populations that are difficult to access due to their
closed nature, e.g. secret societies and inaccessible professions (Breweton and Millward, 2001).

Convenience sampling
Convenience sampling is when we select participants because they are often readily and easily
available. Typically, convenience sampling tends to be a favored sampling technique among
students as it is inexpensive and an easy option compared to other sampling techniques (Ackoff,
1953). Convenience sampling often helps to overcome many of the limitations associated with
research. For example, using friends or family as part of sample is easier than targeting unknown
individuals.

Purposive or judgmental sampling

Purposive or judgmental sampling is a strategy in which particular settings of persons or events
are selected deliberately in order to provide important information that cannot be obtained from
other choices (Maxwell, 1996). It is where the researcher includes cases or participants in the
sample because they believe that they warrant inclusion.

Exercise
1. Differentiate between population and sample.
2. Differentiate between probability sampling and non-probability sampling.
3. List the probability sampling methods.
4. Enumerate the non-probability sampling methods.
MODULE 2: RESEARCH DESIGN
Unit 1: Research Design

Introduction
The framework of how researches are carried out must be clearly defined and universally accepted.
It is this framework that will provides the modus operandi for research in that particular field or
discipline. In education, this framework which is followed in conducting research is called research
design, while the activities carried out within the design specifications constitute the research
methodology. In other words, the research methodology is the knits-gritty work aspect of the
research. In this unit, function, types of research design and variables will be discussed.

The Concept of Research design

Basically, research design is the plan, structure and strategy of investigation intended to provide
answers to research questions or hypotheses and to control the variables. It provides the procedural
outline for the conduct of any given investigation. It has plan because it involves what the
researcher will do from the writing of the proposal, conceptual framework and hypothesis and to
the data collection and analysis. While the structure of research states the method to used, how to
gather, and analyse the data. It also involves how research objectives will be achieved and the
problem encountered in the research will be tackled.
Function of research design
Some of the major functions accomplished by research design include:
a. Providing the researcher with the necessary framework or blueprint for tackling a particular
educational problem.
b. Helping in the delineation of the research purpose and boundary
c. Familiarizing the researcher with the potential problems in the execution of the study.

Types of Research Designs

There are basically two types of research designs. All studies in education are either descriptive
research design or experimental design. Some of the times, we have a combination of both.
Included in these two types of designs are historical research, developmental research, case study
research, correlational research, survey research, experimental research, designs etc.
You need to note that in general, the type of design for a particular study could be correctly inferred
or derived from the title or topic of study. For instance, looking at the table below you will notice
that some operational words in your topic can determine your design for that particular topic.

The table showing some operational words and types of research design
S/N Nature of Research Topic Type of Designs

1. (a) Relationship between ………… Correlational Research Design

(b) A correlational study of……….
(c) X and Y as a covariant of…..….
(d) A comparative study of……….
2. (a) Influence of …………… Survey Research Design
(b) Incidence of ……………
(c) Perception of …………..
(d) Impact of ……………….
(e) Evaluation of …………..
(f) Attitude of ……………..
3. Effect of …………. Experimental Research Design

Historical research design

This is a systematized and object enquiry into events, development, and experiences of the past.
Although, this design is used more in the field of history, it has in recent time, come to find useful
applications in the investigation of certain types of educational problems, such as tracing the
history of certain programmes in education or bringing to limelight the contribution of notable
figures and/or establishments to the development of education.

Sources of Historical Data

There are two main sources of data in historical research. These are primary sources and secondary
sources. Primary sources are those sources which contain account of an event or phenomenon
given by someone who actually observed the event or phenomenon. It could be an eye-witness
account or testimonies, autobiography, oral evidence, remains of manuscripts, certificate,
equipment, attendance register, inventories, report cards etc. Primary sources include original
documents, photographs or records from eye-witnesses. Have you witnessed an accident before?
If you are called to say exactly what had happened as you saw it happened, the evidence you give
is a primary source of historical research.
Secondary sources are those materials which contain account of an event or phenomenon by
someone who did not actually witness the event or phenomenon. These include textbook,
newspaper reports, review of researches etc. In this case, a non-observer mediates between the
original evidence and the investigator i.e. from a third party.

Limitations of Historical Research

Historical research works are said to be error and bias bound. It most of the time reflects the lack
of precision in its enquiry. The reason is obvious and clear. Most secondary sources of data are
found to be subjective. This brings in a lot of errors as things get distorted in the telling of stories,
events and things as they had been explained. Data collected from oral traditions or archives lack
authenticity, accuracy and control which a reliable data can give. There is partial or no control
and randomization achieved.

Developmental Research Design

This is a type of descriptive study which involves an investigation of patterns and sequences of
growth or changes that take place with time. You can study the development of education in a
particular district, or the development of interest patterns in children. Developmental research
seeks to ascertain how some dimensions, variables or characteristics of given population change
with time. Its thrust generally lies in finding out how these characteristics of the target population
change over time, at what rate, in which direction and the factors which possibly contribute to
these changes. Developmental research can be longitudinal or cross-sectional.
a. Longitudinal Research
In this type, the same groups of subjects are studied for a period of time. Observations are carried
out on these subjects from time to time within the period of research to note any changes in the
particular characteristics under study. The data are collected and analysed to see if there are
patterns and sequences which underlie the development or unfolding of these characteristics. It
provides a more valid approach for studying developmental trends.
But it has its own limitations. You will recall that it involves observing the subjects over a period
of time. Don’t you think that some may dropout on account of transfer, ill-health, or death? What
if something happens to the researcher to prevent him from continuing the study? It takes a long
time to complete. That means that there must be enough funds to sustain the study. If this is not
the case, what happens? Again, by the very nature of longitudinal research, modifications are not
possible once the study has started.
b. Cross-Sectional Research
This involves sampling a cross-section of the subjects of varying ages for the study. It means that
instead of following the same group of subjects for a very long time, a cross-section is sampled to
be observed at the same time. The characteristics of the subjects at different age levels are
examined and analysed to reveal possible trends in development. This approach makes it possible
for many subjects to be studied at the same time. It is cheaper and quicker.
But it is less accurate than longitudinal research. Because the subjects differ in other relevant
respects apart from age, the differences in the developmental patterns and sequences observed at
different age levels may not solely be attributable to age alone.

Survey research design

A survey research design is the one in which a group of people or items is studies by collecting
and analyzing data from only a few people or items considered to be representative of the entire
group. Survey research can be classified into two namely; procedure based and purpose based.
a. Procedure – based:
In this scheme, the basis of classification is the procedure, technique or instrument employed in
data collection. (Borg & Gall, 1979. &Kerlinger, 1979). Using this scheme, we have:
i. Questionnaire
ii. Interview
iii. Observational, and
iv. Panel surveys.
i. Questionnaire Survey: Every survey studies which employ the use of questionnaires as
the major data collection technique or instrument are called questionnaire survey.
ii. Interview Survey: Here interview constitutes the major technique for gathering relevant
information. Interview involves getting out information through verbal interaction
between the respondents and the researcher.
iii. Observational Survey: This is the type of survey involving the use of observation to
collect data. It involves gathering information through observation for the purpose of
measuring variables.
 iv. Panel Survey: This employs a definite procedure or technique of data collection. Here,
data are collected from a given sample at two or more different time periods. The data are
then analysed to discover trends or changes in the opinions of the subjects over the period
of time under study. It is suitable for studying trends or fluctuations or changes in the
subjects, opinions, attitudes or behaviours. It can be useful in studying how stable a group
of people’s attitude towards an issue or object is over time or how an identifiable
intervening variable can influence such attitudes.

b. Purpose – Based Classification:

Another way of classifying surveys is by the purpose to which the particular survey intends
accomplishing. Here, we have:
i. developmental,
ii. descriptive,
iii. correlational and
iv. public opinion surveys.
i. Developmental Survey: This helps to ascertain how some variables or characteristics of a
given population can change with time. It can be longitudinal or cross sectional studies.
ii. Descriptive Survey: All those studies which aim at collecting data, analyzing them and
describing in a systematic manner the characteristics, features or facts about a given
population is called descriptive survey. The studies are only interested in describing certain
variables in relation to the target population. They are concerned with a description of
events as they are. For instance, let look at the topic The performance of students in
Mathematics in WASC examination in Ogun State”. This study is only interested in
describing the performance of the students in WASC examinations in Mathematics in
WASC examination in Ogun State only. It may not be interested in going into details about
the causes of the performances.
iii. Correlational Survey: This type of study seeks to establish what relationship exists
between two more variables. Usually the studies indicate the direction and magnitude.
iv. Public Opinion Survey: This is usually designed to find out the opinion of people in a
given area toward an issue or event that is of interest to the general public in the area. This
area can be a large one such as a country like Nigeria, a small town like Ijebu - ode, a
 College campus or even your study centre. Any results obtained from the sample may be
generalized to the entire population. For instance, Public Opinion survey may be used to
predict the outcome of elections in a country.

Case Study Research

This is an in-depth intensive study or investigation of one individual, a small unit or a phenomenon.
When we say a small unit here, we are referring to a family, a school, a church, a classroom, an
association, a teacher, an administrator or a group of these. Case studies are used for solving
specific problems through in-depth study for documenting social realities, life cycle, change or
growth. A phenomenon can be taken as a case or an issue. For instance, the impact of cultism on
the peaceful coexistence of students in tertiary institutions in Ogun State. A case studies employ a
variety of data collection techniques such as questionnaire, observation, interview, test etc.

Correlational Studies
This is a type of research study which seeks to establish the type of relationship existing between
two or more variables. The investigation will indicate to you the direction and magnitude of the
relationship between the variables. Correlational research aims at investigating the extent to which
variations in one factor correspond with variations in one or more other factors; based on
correctional coefficients. This research method is concerned with discovering or clarifying
relationships among variables through the use of correlation coefficients

Ex-Post Facto Research

This is another type of research which is non-experimental or field study that can be recognized
by the process of studying, analysing and making recommendations about some activities in their
natural settings. According to Kerlinger (1974), ex post facto research is any research in which it
is not possible to manipulate variables, to assign subjects or conditions at random. From this
description of ex post facto research, you can see that most of the studies which do not fall into
experimental designs and do not involve randomization can be regarded as ex-post facto research.
Experimental Design
The main trust in experimental design is to establish cause and effect relationship. It is the most
powerful and valid design which can be used to identify confidently the cause of any given effect.
In experimental study, one or more independent variable(s) is/are manipulated by the researcher
under control conditions and its/their effect(s) observed. This involved exposing one or more
experimental groups to one or more treatment conditions and observing the difference(s) between
this/these group(s) and one or more control group(s) to which the treatment was not administered.
An experimental must therefore have experimental or treatment group(s) and control group(s). the
treatment group(s) is the group to which the treatment was administered. The experimental or
treatment group(s) is the variable(s) being manipulated and whose effect is under investigation.
Variables
A variable is a qualitative or quantitative entity which can take on different values or levels. It can
also be described as the conditions or characteristics which the researcher or the experimenter
manipulates, controls or observes to obtain results. For instance, gender is a variable. It can be
male or female. Age is another variable. It can be of several continuous values. In experimental
designs, there are so many and varied variables with different names. Let us now look at some of
them.

Independent and Dependent Variables

The conditions or characteristics which you manipulate in order to find out their relationships with
the observed phenomena are called independent variables. These are also called experimental or
treatment variables. In educational research, such variables like methods of teaching, reward,
instructional materials, age, sex, race, location, qualification or experience of teachers etc. are
examples of independent variables.
The conditions or characteristics which appear disappear or change as the researcher or
experimenter manipulates the independent variables is called dependent variables. Note that by
manipulation, we mean such actions like apply, remove, or change etc. When you measure a
variable to determine the effects of the experimental treatment on it, it is referred to as the
dependent or criterion variable. You should note that the manipulated variable is the independent
variable. Its effects are observed on the dependent variable.
Continuous and Discrete Variables
All the variables which yield continuous data are called continuous variables. Continuous data are
those which can be in both whole and fractional numbers e.g. time can be in hours, minutes and
seconds, length can be in meters, centimeters and millimeters, age can be in years, months and
days, including the fractions of these. For example, 5.5 seconds, 5.3 seconds etc. or scores –
20.5%, 45%, 52.6% etc.
All variables which yield only whole numbers and no fractions are called discrete variables e.g.
sex – man and woman group = group 1, group 2 or group A and group B. There is no 2½ man or
group 1½. It can only be whole numbers.

Intervening and Extraneous Variables

The intervening variables are the unwanted variables that intervene between the cause and the
effect. They may not be controlled or measured directly, but may have some important effect on
the outcome of the experiment. They tend to modify the effect of the independent variable on the
dependent variable and therefore, must be identified and removed or prevented. Take for instance,
in a laboratory experiment, some factors like fatigue, anxiety, motivation, patience; carelessness
etc. may influence the result, even though they cannot be observed directly. These are intervening
variables. You have to take care of them by holding them constant or equalizing their presence in
the experimental and control groups.
Extraneous variables are those variables not manipulated by the experimenter but which may have
significant influence on the dependent variable. They have to be controlled; otherwise they distort
and invalidate conclusions the result from the findings.

Exercise
1. Examine the term research design as it applies to educational research.
2. compare the following
a. the case study design and the survey design
b. the ex – post facto design and the experimental design
MODULE 3: TECHNIQUES AND INSTRUMENTS FOR DATA COLLECTION
Unit 1: Techniques for Data Collection
Unit 2: Validity and Reliability of Instruments

Introduction
Every scientific research involves the collection of pertinent data. Such data are necessary for
arriving at the solution(s) of the problem on hand. Data collection can therefore be referred to as
the way or process through which information are gathered, compiled, received by a researcher for
the purpose of writing a research work or project. It can also be defined as the various ways through
which information are got for the purpose of getting ideas about a particular events or thing. Data
are usually collected for a group of objects/subjects such as registration of birth, deaths, migration,
population distribution, etc. The collection of data is a very crucial step in the research process.
Therefore, in this unit, the various techniques for collection of relevant data will be discussed

3.1 Observation Technique

Observation can be regarded as a process of looking out and recording the presence or absence of
a particular trait or behaviour of a person or group of persons. According to Nworgu (1991),
observation, as a method of data collection, involves watching people, events, situations or
phenomena in order to obtain first-hand information about a particular aspect of such person,
event, situation or a phenomenon.

Types of Observation
In general, there are two types of observation. These are: participant and non-participant
observation.
i. Participant Observation in which the observer is either a member of the setting or
group being observed or has joined the setting or group just for the purpose of carrying
out the observation. In this case, the observer is always with the observed e.g. your
head of department, the principal and the teachers, the instructor and the apprentice etc.
ii. Non-Participant Observation in which the observer is not a member of the setting or
the group being observed. The observer visits the observed at different intervals e.g.
 Supervisors from the Ministry of Education to the schools, or Inspectors from the
Inspectorate, Teaching Practice etc.
Problems of Observation
Observation, as a method of data collection, has some problems. These are:
i. Observer Effects: These are virtually inevitable, but can be minimized by unobtrusive
methods.
ii. Observation Requires Enterprise: In order to have proper observation, you need to
be trained if you are not an expert. Assistant observers also need to be trained. This
will help to save time or maintain objectivity.
iii. The Number of Observations: The number of observations needed to obtain a
representative sample of events is most of the times prohibitive. This is why many
researchers resort to studying the target phenomena poorly.
iv. Interpretation Bias: If a researcher decides to report a coloured version rather than
the objective findings, it can distort the observed event. This can be minimized by the
use of blind observation whereby trained observers are used without them knowing the
objectives of the observation; the use of trained recorders for objective recording of
events; the use of multiple independent observers or the use of taped recorders which
can allow analysis and re-analysis.
v. Halo-Effects: it means later records of observations being affected by earlier
impressions. This can reduce the reliability of information collected through
observation.
vi. Rating Errors: This can occur when the rating system is used. It can be as a result of
ambiguities in the meaning of the scale points. Again, rating systems such as the
tendency to rate subjects towards the middle, rather than at either of the two extremes.

Advantages and Disadvantages

Advantages
i. It provides unique insights not attained by other methods.
ii. It yields direct first-hand information which is more valid than reported information.
iii. It is peculiarly suitable for the study of young children, handicapped persons and
illiterates.
Disadvantages
i. It requires enormous amounts of time, energy and resources to be properly executed.
ii. Samples are usually small and this tends to reduce the internal validity as well as the
external validity or generalisability of the findings. This is due to lack of representation
of the samples.
iii. Some behaviour is not explicit and observation may not apply in such situations.
iv. The faking of some participants when they know that they are being observed.

The Concept of Questionnaire

This is instrument used in eliciting written responses from the subjects of the research through a
series of questions or statements put together with some specific objectives in mind. It can be used
to ascertain facts, opinions, beliefs, attitudes, practices, etc. According to Obodo (1997); &Osuji,
et al. (2012) it as a series of questions or statements presented in a written form to a subject or
group of subjects and to which they are expected to answer in writing.

Types of Questionnaires
Based on the format for the statements/questions and the responses, questionnaires can be
classified into two major types. These are structured/closed or fixed response type and the
unstructured/open end type.
a. The Structure or Fixed Response Questionnaire
In this type the respondents are restricted to some response options. A question is asked or a
statement is made and a respondent has to choose from the available alternatives. You can see that
the respondents do not have the freedom and opportunity to express their views. This type of
questionnaire is usually preferred because it facilitates data analysis and the estimation of validity
and reliability indices for the instrument. Again, it is easier and demands less time to complete.
Look at some sample items of the fixed response type below:
How old are you?
10 – 14 years ( )
15 – 19 years ( )
20 and above ( )
What is your parent highest qualification?
 T.C. II ( )
N.C.E./ND ( )
B.Sc./B. Ed./HND( )
M.Sc./ M. Ed. ( )
Ph. D ( )
b. Unstructured or Open-Ended Questionnaire
The response options are not provided for the respondents. All you need do is to provide questions
pertinent to the problem and the respondents are free to supply their responses in their own words
and in any manner they deem fit. When you are not sure or cannot predict what the subjects’
responses are likely to be, this type of questionnaire is the most appropriate. The open-end
questionnaire can provide unanticipated and insightful information that could lead to a better
understanding of the problem. But they are difficult to complete and time-consuming. There may
be misinterpretation arising from the fact that some people may not be able to express themselves
very well; while others may use styles which are at variance with that of the researcher.
For sample:
How old are you --------------------------?
What is your parent highest educational qualification--------------?

Construction of questionnaire
Questionnaire should be constructed in such a way as to elicit the necessary information that would
be needed for the research from the respondents. In designing a questionnaire, the researcher must:
i. Have a knowledge of the respondent`s language and ability
ii. Make his questions to be simple and straight forward so that they can be easily
understood by the respondents
iii. See that the questions are capable of being answered by Yes or No, number, a date
or something precise.
iv. Tactfully frame the questions so as to break any barrier of suspicion or reservation
on the parts of the respondent.
v. Avoid ambiguity in questions at all costs.
vi. Not include any irrelevant question.
vii. Not include questions that require calculation.
Characteristics of a Good Questionnaire
To serve its purposes very effectively, the questionnaire you construct must be characterized by
the following features:
a. Relevance: The instrument should be relevant to the purpose of the research. It should
elicit all the information necessary for answering the research questions and testing the
hypotheses. It should also consider the background and experience of the respondents.
b. Consistency: The instrument should be able to yield responses that are consistent. The
responses of a group of people to the instrument on two different occasions should be as
close as possible on these occasions.
c. Usability: The instrument should as much as possible be usable. It should not be too long
or so bulky. The conditions for the administration and the method for interpreting the data
elicited should be fairly simple and easy.
d. Clarity: Both the instructions and the items should be clear enough to avoid possible
misinterpretations. You should note once again that a good questionnaire should not
contain ambiguous items or instructions.
e. Quantifiability: The responses from a good questionnaire must be easily quantifiable. It
should be easy to assign numerical values or figures to such responses in a manner that is
systematic.
f. Legibility: All the items should be legible without tiny characters. The words should be
properly spaced with appropriate side margins. Always use the computer to type your
questionnaire so as to produce very neat and legible instrument.

Advantages and Disadvantages

Advantages
i. It is economical in terms of time and cost;
ii. It can be administered to variety of people;
iii. It is easy to administer and even to distant respondents;
iv. It can be used to get information on non-cognitive constructs such as kindness, stress,
anxiety etc;
v. Great percentage of students or respondents can be reached at a time.
Disadvantages
 i. There may be low percentage return which may lead to the distortion of the findings;
ii. There may be misunderstanding or wrong responses if the item s are not clear or the
instructions are misleading;
iii. Wrong or negative or faked answers may be given if the instrument is too lengthy or if
it is intruding o the respondents’ private life;
iv. It is not suitable for the illiterates, semi-illiterates, and children.
Interviews
The interview as a method for data collection involves eliciting information through some verbal
interaction between the respondents and the researcher. In other words, an interview can be
regarded as a face-to-face interaction in which oral questions are posed by an interviewer to draw
out responses from the interviewee. There are basically two types of interview. These are the face-
to-face interview and the telephone interview. You have to note that whether it is face-to-face of
telephone interview you are employing; they have the same characteristics. Interaction between
the interviewer and the interviewee is purely verbal.

Phases of Interview
There are four major phases of interview. These phases overlap and interact. They are preparation,
establishment of rapport, the question, answer and the recording phases.
a. Preparation Phase: The degree of success in an interview is dependent on how well you
have prepared for it. The preparation stage is when you take decisions on the mode of
recording the responses. You need to check the recording instruments for validity and
reliability. You need to trial-test the instruments to ensure that they are in good working
conditions.
b. Rapport Phase: For collection valid data, there is need to establish a cordial atmosphere
for the interview. There is no rule for doing this, but as a mature investigator, you have to
survey the situation and evolve appropriate strategies. You may follow any of these
suggestions:
i. Courteously seek permission from the appropriate authority. In doing this, you
should provide information about the objectives of the study and the nature of the
interview.
 ii. Give notice to the interviewee and book appointment for date, time and venue.
The venue should be comfortable in terms of sitting, ventilation, lighting and
decoration. The venue should be noise-proof.
iii. Take note of your appearance. It should be appropriate, neat, and inoffensive.
Consider your dental and body deodorization and any other aspect likely to irritate
your subject.
iv. Your first contact with the interviewee must be friendly, pleasant and courteous.
Use appropriate salutation and address him with the correct title. Pay some
compliments, but not flatter. This will make him more relaxed.
v. You need to be relaxed too. Introduce yourself briefly and modestly and also
introduce the problems which are the focus of the interview. You may need to
start by asking the first few minutes for a short conversation to enable you and
your subject to relax. You also need to assure him that his responses will be treated
confidentially and only for the research purposes. Otherwise, he might be
suspicious and unwilling to respond to the questions freely.
c. Question-Answer Phase: This is where you have to make use of your skills and expertise
to make the session more permissive, flexible and interactive. You have to keep the
interviewee, interested and responsive till the end of the interview. In doing this, you have
to devise appropriate strategies, but you have to be pointed and business-like and not to
wonder aimlessly. When you ask a starter question, follow it up with prodding and probing
questions to get comprehensive information. Where you need to use pictures or related
objects, drafts, drawings etc. use them for probing to elicit further explanations and reasons
for earlier statements. But you should try to avoid interviewer bias by being as non-
directive as possible. In all, you have to be relaxed, not tensed up or nervous throughout
the duration of the interview.
d. Recording Phase: Recording is a very important aspect of interview. You must try to
comprehensively record information from an interview as unobtrusively as possible.
Recording may be done in any of the three methods.
i. Mental Note: If you want to commit the responses into memory, you can be sure;
it removes apprehension on the part of the respondent and therefore increases the
rapport. But you have to appropriately assess yourself in terms of memory
 retention before using mental notes. This is because any information forgotten is
as good as missing and may not be easily reconstructed.
ii. Written Note: It is advisable to use written records when there are too many
questions and responses. But extensive writing is likely to excite or offend
respondents. If this happens the rapport and validity of the responses will be
reduced. If you are versed in shorthand, it can be an advantage in written notes.
You can also use structure interview schedule where alternative responses are
provided. Here, there is minimum writing, but you have to make allowance for
unanticipated responses.
iii. Taped Records: In order to solve all the problems of memory loss and that of
extensive writing, tape recording of interview is the answer. It removes strain
from the interviewer. It can be replayed and transcribed at leisure. But you have
to bear at the back of your mind that audio and video recording instruments can
go faulty in the process. You must take care of this at the planning stage. Other
problems like instrument reactivity should also be taken care of. Recording can
also frighten or excite the subject. This may channel away his attention and bring
about distortion in his behaviour. As far as possible, conceal your recorders. You
can use micro recorders, remote censors or pick up buttons when available.

Advantages and Disadvantages

Advantages
i. It provides opportunity for face-to-face interaction.
ii. The subjects can respond to questions the way they like.
iii. Indepth information which the respondents may not wish to write down can be
obtained.
iv. Information recorded is reliable because it is recorded by the researcher himself.
v. It is very useful for collecting data from children and illiterates or those who cannot fill
the questionnaire.
Disadvantages
i. It can be very expensive to conduct.
ii. It consumes a lot of time.
iii. Subjective information derived from unstructured interview may prove difficult to
analyse.
iv. The validity of verbal responses collected may be questionable.
v. Conducting interviews demands a lot of skills.
vi. Intra and inter-interviewer variabilities can affect the results. For instance, the mood
of the interviewer-fatigue, hunger or other types of pressure on the interviewer (intra-
interviewer) or two different interviewers getting different results in a single study
(inter-interviewer).

Documentary Sources: Information are collected, particularly during the planning stage of
collecting data, from various sources that keep records and documents that pertain to the study.
Such sources could be from personal, organizational or public libraries, diaries, files, record etc.

Measuring Scales
The Summated or Likert – Type Rating Scale:
This is a type of rating scale used to measure attitudes or opinions. The scale was named after
Rensis Likert the person who devised it. It involves a list of questions or statements about the
phenomenon to be measured, with a set of graduated response options. With this scale, respondents
are asked to rate items on a level of agreement or disagreement with the statements or questions
raised. The responses to the statements are then summed up and a total score or the average score
is obtained. This will help to determine the people’s position on the phenomenon which is
measured. A positive statement attracts higher value while negative statement attracts lower value.
For example,
Positive Statement “I like Mathematics”.
Strongly Agree (4) Agree (3) Disagree (2) Strongly Disagree (1)
Negative Statement “I do not like Mathematics”.
Strongly Agree (1) Agree (2) Disagree (3) Strongly Disagree (4)

The scale does not have to state “agree” or “disagree”; dozens of variations are possible on themes
like frequency, quality and importance. For example:
• Frequency: Often to never.
• Quality: Very good to very bad.
• Likelihood: Definitely to never.
• Importance: Very important to unimportant.

The Thurstone Equal-Appearing Intervals Scale

Developing the scale is time consuming and relatively complex compared to other scales (like the
Likert Scale).This is because it has a laid down principle. To use the scale, you will first of all
construct a set of more than 100 items which will measure the attribute you want. These items
will then be presented to a panel of about 25 judges. These judges will be required to rate each
item on the set, according to the degree of intensity, on a seven-point scale or more.
Using the averages of the ratings given to the items by the judges, a numerical value is computed
for each item. The items are selected in such a way that they cover the entire scale with equal
intervals between any two consecutive items.
For example, 25 judges were asked to sort a number of statements into 7 categories of varying
intensity as follows:
Very High (7), High (6), Slightly Above Average (5), Average (4), Slightly Below Average (3),
Low (2), Very Low (1)
For a hypothetical item X, the number of judges who placed it in each of the 7 categories is given
below;
Category 7 6 5 4 3 2 1 Total
Number 3 1 2 3 7 4 5 25
of Judge
21 6 7 12 21 8 5 80
∑𝑥 80
Mean Value, (𝑥̅ ) = 𝑁
= 25 = 3.2
In the same way, the numerical values or weights of the other items are determined. You can see
why it is not very popular in educational researches.

The Guttmann Scale or Cumulative Scale

Unlike the Likert and Thurnstone scales, the Guttmann scale determines the unidimensionality of
the items making up a given scale. Unidimensionality refers to whether all the items measure all
aspect of the particular variable in question; or just one aspect of it. When the items on a scale are
one dimensional, we say they form a perfect scale. When the items of a perfect scale vary in
intensity, the scale is said to be perfectly reproducible. But, when there are errors or inconsistencies
in the responses or when we have errors in the scale, the scale is said not to be reproducible. You
must have to note that it is difficult to come across a perfect reproducible scale. Guttmann produced
a formula which can be used to estimate the extent of reproducibility of a scale. It is called
coefficient of reproducibility and it is given by the equation:
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑟𝑟𝑜𝑟𝑠
Coefficient of reproducibility = 1 – 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒𝑠

If the coefficient of reproducibility resulting from the use of this formula is 0.90 and above, we
say the scale is reproducible. If otherwise, the scale is not reproducible.

Exercise
1. distinguish between participant and non – participant observation
2. what is difference between structure and unstructured questionnaire?
3. state three each of advantages and disadvantages of interview techniques
4. explain the following measuring scales
a. Likert – type scale
b. Thurstone scale, and
c. Guttman scale.
UNIT 2: VALIDITY AND RELIABILITY OF INSTRUMENTS

Introduction
An instrument is a measuring device, it could be a questionnaire, observation or test measuring
intelligence or achievement. You should know that if a study is very well designed but uses faulty
instruments, the findings would be completely invalidated. It is therefore very important that when
you design a very good study, it has to be valid and reliable to serve the purpose of the study. In
this unit, you will be studying the validity and the reliability of the instruments constructed.

Measurement Error
Errors in measurement could arise from faulty instruments incorrect interpretations of the values
obtained or instability in the behaviour of the respondents or testees. These errors could be
systematic or random. A systematic error can occur when the errors are very frequent and are made
in one direction away from the true score. Random errors can occur when measurement values
deviate from the true score and as frequently in one direction as another.

Validity
Validity refers to the extent to which an instrument measures exactly what it purports to measure
and nothing else. Validity is always specific to some particular use. An instrument may be valid
for one purpose, but not for another. For instance, an instrument that has a high validity in
reasoning ability may have very low validity for measuring arithmetic ability. An instrument may
be valid for one culture or geographical setting, but not valid for another.
Validity can be classified into three major types. These are content, construct and criterion-related
validities.

Content Validity
This refers to the degree with which the items of an instrument are representative of the content
and behaviours specified by the theoretical concept being measured. It is estimated using the
sample of items and comparing them with the content and behaviours which they should represent.
A high degree of content validity is achieved if the sample of items covers all aspects of the content
and behaviours.
A way of estimating the content validity of a test is by constructing a test-blue point otherwise
called table of specification. This systematically specifies the content, objectives and evaluation
techniques in the process of generating valid test instruments.
Face-to-face validity – This is a type of content validity. Most people, some of the times,
erroneously equate it with content validity. But it refers to the subjective judgement of assessors
about what an instrument appears to be measuring, that is, on the face value. There is no systematic
procedure adopted for this purpose.

Construct Validity
This refers to the extent to which a particular instrument reflects hypothetical constructs presumed
to underlie the performance and also the extent to which the instrument reflects the theories
underlying the constructs. Some psychological concepts such as: intelligence, creativity, anxiety,
attitude, reasoning etc. which cannot be seen with the eyes but their existence can only be inferred
from manifested characteristics or behaviour are called constructs. For you to design a test, you
have to ask: To what extent do certain explanatory concepts or qualities account for students’
performance on a test?
The process of construct validity is easy to determine only when the construct is specified.
Therefore, the construct should be precisely defined before you undertake to develop the
instrument. Let us take students’ interest in your subject, for instance. In the definition of interest,
you specify such behaviour as prompt attendance, alertness, carrying out assignments, smiles on
their faces when they understand, curiosity to learn, asking relevant questions etc. So when you
administer a testing the subject and notice that students who exhibit the abovementioned
behaviours perform better, you can say that the test has good construct validity.

Criterion Related Validity

This refers to the extent to which an instrument yields the same results as a more widely accepted
measure. If you want to verify the degree of criterion-related validity of your test, you can compare
the result of your test and that from a known test like the Standford – Binet Intelligence Test, using
correlation coefficients. If your test correlates highly with the known test, you say it possesses a
high degree of criterion-related validity. This is possible if the two tests are on a related area. For
instance, Eyesenk Personality Inventory can be used as a criterion for a new personality measure.
Criterion-related validity is of two types. They are concurrent and predictive validities.
a. Concurrent Validity: This is applicable when a new instrument is administered at the same
point in time as well as a known instrument. It can be very useful if the equivalent form of an
instrument is required concurrent validity is attained when the correlation between the results
of the newly developed instrument and those of suitable equipment is sufficiently high.
b. Predictive Validity: This is concerned with the prediction of future performance. It is the
degree to which predictions made by an instrument are confirmed by the later behaviour of
respondents. For instance, you can use the results from an intelligence test or aptitude tests to
predict success at school. You can also use the results from JAMB or UME to predict the
performance of students in undergraduate courses. A second instrument can be administered
after the behaviour which the first instrument attempts to predict has occurred. The results
yielded by the two instruments are correlated. A sufficiently high correlation index shows
predictive validity.

Reliability
This is a common word in everyday usage. But in educational research, it refers to the extent of
consistency with which an instrument measures what it is measuring. An instrument is regarded
as reliable if it yields the same results time after time, or if similar results come out consistently
under the same or slightly different test conditions. It is very important that the test you are going
to use for your research project is reliable, especially if your research study is going to involve
pretest and posttest, experimental and control groups. If your instrument has low level of
reliability, it may not be able to discriminate between pretest scores and posttest scores. It may not
discriminate effectively between the performance of the experimental and control group.
Correlation coefficient is an important statistical procedure for determining the degree of
reliability.
Test-Retest Method
This involves the repeated administrations of the instrument to the same people on two occasions.
It is usually recommended that the time interval be between two weeks and one month. The scores
resulting from the two administrations of the test are correlated to determine the coefficient of
stability. You will have to note that if the time interval is too short, the memory of the responses
to the first test will affect the second. A long-time interval may create opportunity for candidates
to learn more or to forget what they had known.

Alternate-Form Method or Equivalent Form Method

In this method, two parallel forms of an instrument are administered to the same respondents at a
single sitting or with a short time interval between the two. The scores from the two tests are
correlated to determine the coefficient of equivalence. You should note that parallel forms of an
instrument are expected to have the same specification of content and objectives and measures the
same behaviours.

Split-Halves Method
This is a measure of internal consistency. It requires the administration of a single test instrument
to the students once then the items of the instrument are split into two parts. In other words, the
total set of items is divided into halves. The scores on the halves are correlated to obtain the
estimate of reliability. You can split the items using odd and even numbers, or randomly dividing
the items into two groups etc. You can see that the result you get from it for a half test. Therefore,
it is corrected using the Spearman-Brown formula:
1
2 𝑥 𝑅𝑒𝑙𝑖𝑎𝑏𝑖𝑡𝑦 𝑜𝑓 𝑡𝑒𝑠𝑡
2
Reliability = 1
1+ 𝑅𝑒𝑙𝑖𝑎𝑏𝑖𝑡𝑦 𝑜𝑓 𝑡𝑒𝑠𝑡
2

For Example, Suppose the correlation coefficient between the two halves of a test is 0.90 the
reliability of the full test is
2 𝑥 0.90 1.80
rxx = 1+ 0.90 = 1.90 = .95

Internal Consistency Method

The estimate here is obtained through an analysis of the individual items following a single
administration of the measuring instrument.
a. The Rational Equivalence Method:
This uses the formation of equivalent halves by considering all possible splits and computing the
reliability coefficient by employing either Kuder Richardson formula (for dichotomous items) or
Cronbach’s alpha (for Likert scales). Kuder and Richardson (1937) developed coefficients for
estimating the reliability of instruments composed of dichotomously scored items. You may note
that dichotomous items are scored one or zero for presence or absence and for positive or negative
responses to characteristics under investigation.
The most widely used formulae are numbers 20 and 21 otherwise called KR20 and KR21. KR21
is simpler and can be used for instruments developed by individual researchers, while KR20 which
is more technical is used for determining the degree of reliability of standardized tests.
𝑁 (1− 𝛴𝑝𝑖 𝑞𝑗
KR20 is given by; KR20 = 𝑁−1 𝑄𝑡 2

where N = the number of dichotomous items

Pi = the proportion of positive responses to the ith item
qi = 1 – Pi
∑ = summation
Qt2 = the variance of the total composite
𝑁 (1− 𝛴pq
KR21 is given by; KR21 = 𝑁−1 𝑄𝑡 2

where N = the total number of items

P = the proportion of positive responses to each item
q = the proportion of negative responses to each item
∑ = summation
Qt2 = variance of the total composite
Cronbach’s alpha is a unique estimate of the expected correlation of one instrument with an
alternative form that can be used for polychotomous items while Cronbach alpha is a
generalization of it.
𝑁 (1− 𝛴𝑄 2 (p𝑝𝑖 ))
Cronbach’s alpha is given by = 𝑁−1 𝑄𝑡 2

Where N = the number of items

∑Q2(Yi) = the sum of item variances
Qt2 = the variance of the total composite

(b) Standard Error of Measurement

This is an estimate of test reliability obtained from the reliability coefficient and the standard
deviation of test scores. It is inversely related to the reliability coefficient.
(c) Factor Analysis
This is used to obtain estimates of reliability which approximate the true reliability better than all
other coefficients. It is represented by coefficient theta (Ø) which is derived from principal
components factor analysis. It can be used to estimate reliability. Another coefficient called omega
(Ω) also derived from factor analysis can also be used to estimate reliability.

Exercise
1. What are the two types of measurement error?
2. What is validity?
3. Describe any two types of validity.
4. What are the methods of estimating reliability?
MODULE 4: STATISTICAL METHODS
Unit 1: Introduction to Statistics
Unit 2: Organization and Representations of data
Unit 3: Measure of Central Tendency
Unit 4: Measures of Variability
Unit 5: Measure of Association and Correlation
Unit 6: The Testing of Hypothesis

UNIT 1: Introduction to Statistics

Introduction
In the previous modules/units, you worked through the different methods of collecting data in
research. The question is: what do you do with this seemingly unmanageable bulk of data? This
question will take us to ‘Data Analysis’, which we shall describe “as the process of organizing and
summarizing data in order to provide answers to the research questions or test hypotheses stated
in the study”. This process, most of the times, involves the use of statistical procedures to
summarise and describe the characteristics of samples and populations of the study. In this unit,
we shall first look at the meaning of statistics, the types of statistics, as well as benefits of statistics

Meaning of Statistics
Statistics as a branch of mathematics focus on the organization, analysis, and interpretation of a
group of numbers. According to Spiegel (1972), statistics is concerned with scientific methods for
collecting, organizing, summarizing, presenting and analysing data as well as drawing valid
conclusions and making reasonable decisions on the basis of such analysis.
According to Osuji et. al, (2012) statistics can be viewed in the following ways
1. As a state arithmetic: In this case, it involves observing, recording and computing the
number of resources, financial, human and material, available to a government for the
purpose of governance or war. Every government needs accurate statistics to make
governance easier.
2. As pieces of information: Statistics imply data or pieces of information e.g. the age of
Adedayo, the height of Yusuf, the weight of Adu, the number of students in classes SS. 1.
 Others are: number of accidents on road A for a year, number of candidates employed by
Ogun State Government 1999 etc.
3. As summaries of information: In this case, it can be used as summaries of information
about a small group of individuals selected from large group for the purpose of
investigating the large group. This is called sample statistics. This can be in the form of
sample size, mean, median, variance, standard deviation, mode, etc. Each of these is
regarded as a statistic.
4. As Mathematical function or models: In this case, it is used for comparison of two or
more samples. In other words, it can be used for pair wise differences, ratios of 2- test, 2-
score, t-score, t-test, f-test etc. are examples.
5. As academic discipline: In this case, it is regarded as a subject or field of study, in which
case, it is an aspect of applied mathematics.

Types of Statistics
These are Descriptive Statistics and Inferential Statistics
Descriptive Statistics
This can be described as a type of statistical application which is concerned with the organisation
and presentation of data in a convenient, usable and communicable form. Spiegel (1972) described
it “as the set of methods serving the functions of organizing, summarizing and communicating
data. You can use descriptive statistical methods when you are interested in merely describing the
characteristics of the group or the sample of study. It means that the descriptive analysis which
you make will not generalize beyond the particular group or sample observed. In the same way,
conclusions drawn from the study are limited and apply only to that group of study.

Inferential Statistics
These are statistical methods used for arriving at conclusions extending beyond immediate data.
They are the phases of statistics which can be used to deal with conditions under which conclusions
are drawn about a larger group based on data collected from some smaller group or groups chosen
from and related to the larger group. Inferential statistics can be described as a statistical procedure
which makes use of sample statistics to make inferences about the population parameters. It
involves the process of sampling that is representative of the population. It makes use of the aspect
of inferential statistics called parametric statistics which are powerful tests that make use of the
normal probability model, or making comparison involving the setting up of confidence limit,
setting up of the degree of freedom etc.

Benefits of the Study of Statistics

In studying statistics, you stand to derive some general benefits. These benefits focus on the useful
knowledge, skills, capabilities or dispositions which you will acquire from the study of, or training
in statistics. They vary, according to the extent and level of study, or training in the subject. Osuji
et al. (2012) identified some of these benefits to include:
1. Acquire knowledge and skills in observation, collection, organisation, communication,
analysis of data, drawing inferences from the analysis of data and making sound decisions;
2. Make meaningful contributions to local, national or international debates on topical issues;
3. Read, understand and interpret communicated data, follow inferences drawn therefrom
and appreciate decisions made consequent upon the inferences drawn;
4. Successfully execute empirical research. No reasonable or worthwhile empirical research
can be carried out or reported without statistics for answering research questions, testing
hypotheses or taking decisions and making predictions;
5. Read, interpret and make use of research reports or articles;
6. Follow and critique contributions to debates presented with facts and figures;
7. Acquire the skills and techniques for estimating, predicting and projecting into the future
based on the previous and present data;
8. Draw sound conclusions based on some pieces of information that are probable or not
quite certain.

Exercise
1. What is statistics?
2. What are the two types of statistics?
3. explain the benefit that will be gained when learn statistics
UNIT 2: ORGANISATION AND REPRESENTATIONS OF DATA

Introduction
Data collected in education can be from various sources and can be in various forms, such as:
opinions, scores/marks, frequencies, verbal etc. The data can then be organized or arranged to
make them meaningful. In this unit, we shall look at sequencing, tables, and frequency distribution
tables. Sometimes data are organized in form of graphs. This is also known as graphical
representations. There are different types—bar chart, pictogram, histogram, frequency polygon,
pie chart and ogive. In this section we shall discuss these graphs or charts.

Organisation of Data
Sequencing
This involves arranging the data in order of magnitude – ascending or descending order.
Example:
Given that the test scores of 10 students in statistics are: 5, 6, 2, 9, 3, 7, 4, 2, 1, 8.
This could be arranged in ascending order thus: 1, 2, 2, 3, 4, 5, 6, 7, 8, 9. or descending order thus:
9, 8, 7, 6, 5, 4, 3, 2, 2, 1.
If the data consists of names, they can be arranged in alphabetical order. If they consist of objects,
events, animals, etc. they can be arranged according to kinds, species, groups etc.

Tables
A table can be regarded as a two-dimensional representative of statistical information or data.
Tables can be simple or complex as shown in the example on the enrolment of pupils in TASCE
Staff school Igbeba Ijebu Ode from 2015 to 2020.

Pupils` enrollment in TASCE staff school, 2015 -2020

S/N Year Boys Girls Total
1. 2015 45 50 95
2. 2016 57 34 91
3. 2017 50 52 102
4. 2018 45 55 100
5. 2019 54 40 94
6. 2020 68 54 122
Frequency Distribution Table
A frequency distribution table shows the number of times each score, value or item occurs in a
distribution. For easy counting tallies are made in bundles of 5. Then, under the last column, add
up the number of tallies against each score and write the value as frequency.
Example
The scores of some students in a PDE 5110 test are given below. Present the scores in a frequency
table.
9, 10, 0, 1, 5, 2, 4, 9, 8, 0, 1, 4, 6, 0, 3, 7, 6, 3, 5, 5, 7, 8, 5, 5, 7, 7, 8, 2, 4, 5
Score Tally Frequency
0 /// 3
1 // 2
2 // 2
3 // 2
4 /// 3
5 //// / 6
6 // 2
7 //// 4
8 /// 3
9 // 2
10 / 1

Grouped Frequency Distribution

When summarizing large masses of raw data, it is often to distribute the data into classes or
categories and to determine the number of individuals belonging to each class called the class
frequency.
In order to determine the interval or class size:
i. Determine the number of groups. Must not be too small or too large. This should be
between 10 and 20
ii. Find the range. This is given by the highest score minus the lowest score.
iii. Divide the range by the number
iv. Draw a table and tally the scores according to groups.
Example
Present the scores below in a grouped frequency table.
55, 62 60, 50, 52, 58, 55, 60, 51, 55, 68, 55, 47, 39, 58, 42, 47, 42, 48, 55, 48, 46, 55, 51, 58, 65,
52, 35, 54, 55, 52, 56, 46, 65, 53, 34, 48, 50, 39, 59, 53, 52, 33, 48, 65, 60, 36, 68, 45, 62, 59, 60,
33, 40, 61, 38.
Number of group = 10
Range = highest score – lowest score = 60 – 33 = 27
27
Range divided by the number = 10= 2.7 = 3 (approximate).

Class Interval Tally Frequency

33– 35 //// 4
36 – 38 // 2
39 – 41 /// 3
42 – 44 // 2
45 – 47 //// 5
48 – 50 //// / 6
51 – 53 //// /// 8
54 – 56 //// //// 9
57 – 59 //// 5
60 – 62 ///// // 7
63 – 65 /// 3
66 – 68 // 2
Total 56
Table: Grouped Frequency Distribution Table

3.2 Bar Chart

This consist of rectangular bars of equal width whose height are proportional to the quantities or
sizes of the items that are being represented. Each bar is however separated by equal gaps. To draw
a bar chart the class interval is plotted on the X-axis against the frequency on the Y-axis.
Example
The table below shows the number of items produced by TASCE venture over a five-year period.
Year 2015 2016 2017 2018 2019
Items Produced 2500 3000 1500 4000 1000
Prepare a bar chart for this information.
Solution
Items Produced
4500
4000
3500
3000
2500
2000
1500
1000
500
0
2015 2016 2017 2018 2019

Years
Pie Chart
This is used to represent both discrete and continuous data. It involves using a circle to represent
a set or groups of items or scores. Each group or item is represented by a sector of the circle. The
angle subtended at the centre by the sector is proportional to the frequency of the items or scores
represented. It implies that the total frequencies of the set are represented by 3600.
Example
In Ogun State, a census taken regards the various ethnic group, gives the following results, Egba
– 200, Ijebu – 100, Remo – 90, Yewa – 60, Awori – 50, Water side – 80, others – 140, Represent
the above information in a pie chart.
Solution
Group No of People Angle of Sector
Egba 200 200/ o
720 x 360 = 100
Ijebu 100 100/ o
720 x 360 = 50
Remo 90 90/ o
720 x 360 = 45
Yewa 60 60/ o
720 x 360 = 30
Awori 50 50/ o
720 x 360 = 25
Water Side 80 80/ o
720 x 360 = 40
Others 140 140/ o
720 x 360 = 70
Total 720 = 360o
 Ethnic Groups in Ogun State

Others, 140
Egba, 200

Water Side, 80

Awori, 50 Ijebu, 100

Yewa, 60
Remo, 90

Egba Ijebu Remo Yewa Awori Water Side Others

Pictogram
This is an aspect of statistics in which picture are drawn to interpret data or information.

Example
The table below shows the masses (in kilogram) of 175 students in a certain school.
Masses in Kg 43 56 67 82 95 108
No of Students 45 50 25 15 30 10
Represent the information in pictogram.
Solution
Let 10 students represent a picture, i.e. = 10 students
Let 5 students represent half picture, i.e. = 5 students
Numbers of students that weigh 43kg = = 45
Numbers of students that weigh 56kg = = 50
Numbers of students that weigh 67kg = = 25
Numbers of students that weigh 82kg = = 15
Numbers of students that weigh 95kg = = 30
Numbers of students that weigh 108kg = =10

Histogram
This is made up of rectangular bars of equal joined to one another, and it is used for continuous
data. At the vertical axis, we have the frequencies and at the horizontal, we have the corresponding
class intervals. The difference between the two is that, for bar chart the class intervals are used
while for histogram the exact class boundaries are used. There are two exact class boundaries—
upper and lower exact class boundaries. These are obtained by subtracting 0.5 from the upper
boundary and adding 0.5 to the lower. To construct a histogram:
i. Compose a composite table having the class interval, the exact class limits, and the
frequencies
ii. Choose suitable scales and draw the vertical and horizontal axes.
iii. Mark of the frequencies on the vertical axis and the exact limits or real limits on the
horizontal axis.
iv. Draw the rectangular bars on each boundary with the height corresponding to the
frequencies.
v. Draw arrows to show what is on the vertical and horizontal axes.
Example:
Represent the information in table below in histogram
Mark Frequency Class Boundary
1 – 10 4 0.5 – 10.5
11 – 20 12 10.5 – 20.5
21 – 30 20 20.5 – 30.5
31 – 40 25 30.5 – 40.5
41 - 50 38 40.5 – 50.5
51 – 60 43 50.5 – 60.5
61 – 70 30 60.5 – 70.5
71 – 80 16 70.5 – 80.5
81 – 90 8 80.5 – 90.5
91 – 100 4 90.5 – 100.5
Total 200

Solution
45
40
35
30
F 25
20
15
10
5
10.5 20.5 30.5 40.5 50.5 60.5 70.5 80.5 90.5 100.5
Class Boundary

Frequency Polygon
This is a line graph plotted using the frequencies against the mid-points of the class intervals.
To construct the polygon:
i. Draw up a composite table having the class interval, the mid-points and the frequencies.
ii. Choose suitable scales for the vertical and horizontal axes.
iii. Plot the graph using the frequencies against the mid-points of the class interval.
 iv. To complete the polygon, add an interval at the top and below. Let the two intervals
have zero frequencies.
Example
Represent the information in table below in frequency polygon.
Mark Frequency Class Mid – Points
6–8 0 7
9 – 11 3 10
12 – 14 5 13
15 - 17 7 16
18 – 20 10 19
21 - 23 8 22
24 - 26 4 25
27 - 29 3 28
30 – 32 2 31
33 – 35 5 34

10 •
9
8 •
7 •
F 6
5 •
4 •
3 • •
2 •
1
0 • •
7 10 13 16 19 22 25 28 31 34
Mid - Point
Ogive
This is a graph which involves the use of a smooth curve to join the Cartesian coordinate plots of
cumulative frequencies against the real class boundaries. In other words, instead of the
frequencies, it makes use of the cumulative frequencies. The graph gives shape like shallow ‘S’.
To draw the ogive:
 i. Compose a composite table having the class boundaries, the exact class limits,
frequencies and cumulative frequencies.
ii. Choose a suitable scale to accommodate the highest cumulative frequency on the
vertical axis and the class boundaries on the horizontal axis.
iii. Plot the points on the cumulative frequencies against the corresponding class
boundaries.
iv. Join with a smooth curve.

Example: Represent the information in table below in ogive

Mark Frequency Cumulative Class Boundary
frequency
1 – 10 1 1 0.5 – 10.5
11 – 20 3 4 10.5 – 20.5
21 – 30 5 9 20.5 – 30.5
31 – 40 7 16 30.5 – 40.5
41 - 50 10 26 40.5 – 50.5
51 – 60 8 34 50.5 – 60.5
61 – 70 8 42 60.5 – 70.5
71 – 80 4 46 70.5 – 80.5
81 – 90 3 49 80.5 – 90.5
91 – 100 1 50 90.5 – 100.5
Total 50

70
60
50
40
30
20
10
0.5 10.5 20.5 30.5 40.5 50.5 60.5 70.5 80.5 90.5 100.5
Class Boundaries.
Exercise
(1) The data below shows the expenditure of a housewife for a particular day.
Items Amount
Rice 120
Yams 100
Beans 140
Vegetable 60
Meat 140
Fish 60
Palm Oil 80
Display the information using (a) Bar Chart (b) Pie Chart (c) Pictogram

2. The table below shows the frequency distribution of marks of 30 students in a mathematics test.

Marks 1 – 10 11 – 20 21 – 30 31 – 40 41 – 50 51 – 60

Frequency 8 2 4 2 4 10

(a) Construct a histogram for the distribution.

(b) Construct a frequency polygon for the distribution.
(c) Construct an Ogive/curve for the distribution.

3. The achievement test scores of some students in PDE 5110 examination are given below:
80, 61, 57, 52, 48, 46, 35, 59, 78, 66, 59, 51, 37, 57, 70, 63, 57, 50, 47, 59, 28, 25, 73, 64, 58, 47,
40, 25, 65, 55, 72, 48, 41, 28, 69, 55, 50, 45, 40, 24, 60, 63, 35, 71, 40, 45, 33, 31, 38, 36.
Using a class interval of 5, construct a frequency distribution table for the data.
UNIT 3: MEASURE OF CENTRAL TENDENCY
Introduction
Measures of central tendency are typical and representative of a data. Every value in the
distribution cluster around the measure of location. The popular average which in statistics is called
the arithmetic mean is one of such measures. Others are the median, mode, harmonic mean and
geometric mean. Let us discuss them in that order.

The Mean
This is otherwise called the arithmetic average. It is the sum of the scores in a distribution divided
by the total number of scores. It is usually denoted by 𝑥̅ (pronounced “X bar”) for a sample and
the Greek letter  (pronounced “mu”) for population. The formula is
∑𝑥 𝑋1 +𝑋2 +𝑋3 ,+ . . . +𝑋𝑛
The mean (𝑥̅ ) = =
𝑁 𝑁

where ∑X is the sum of scores

N is the total number of scores X.

Arithmetic Mean of an Un-grouped frequency Distribution

Example 1
Find the mean of 2.3, 5.4.0, 6.2, 7.9, 8.1, 0, 3.4
∑𝑥 2.3 + 5.4 + 0 + 6.2 + 7.9 + 8.1 + 0 + 3.4 33.3
The mean (𝑥̅ ) = = = = 4.1625
𝑁 8 8

Example 2
The table below shows the number of children in the families of twenty teachers in a school.
Number of Children 1 2 3 4 5

Number of Teachers 4 2 6 5 3

Calculate the mean

Solution
We represent the number of children by the variable X and the number of teachers by f. So
𝑓1 𝑋1 + 𝑓2 𝑋2 + . . . + 𝑓𝑛 𝑋𝑛 ∑𝑓𝑥
𝑥̅ = =
𝑁 ∑𝑓

Where N = f (i.e. sum of all the frequencies) and so

(1 𝑥 4)+(2 𝑥 2)+(3 𝑥 6)+(4 𝑥 5)+(5 𝑥 3) 4 + 4 + 18 + 20 + 15 61
𝑥̅ = = = = 3.05
4+ 2+ 6+ 5+ 3 20 20
or in tabular form:
X 1 2 3 4 5 Total

F 4 2 6 5 3 20

FX 4 4 18 20 15 61
∑𝑓𝑥 61
Mean = 𝑥̅ = = 20 = 3.05.
∑𝑓

Arithmetic Mean of a Grouped Frequency Distribution

A grouped distribution usually has intervals e.g., 9-11, 12-14, etc. The same formula applies i.e.
∑𝑓𝑥
𝑥̅ =
∑𝑓

However, the X values are the mid-mark of each group. So the following steps are recommended:
a. Find the mid mark for each class. Call this X.
b. Find the product of each mid mark and its corresponding frequency i.e. Xi fi.
c. Find the sum of these products i.e., fX.
d. Add up the total frequency
∑𝑓𝑥
e. Divide the sum of the product by the total frequency i.e. find
∑𝑓

Example
The table below shows the distribution of weights of 30 female undergraduates.
Weight kg 48 – 50 51 – 53 54 – 56 57 – 59 60 – 62 63 – 65
No of students 3 6 9 5 5 2

Calculate the mean.

Solution
We set up the workings as follows:
Group X Mid mark X F FX
48 – 50 49 3 147
51 – 53 52 6 312
54 – 56 55 9 495
57 – 59 58 5 290
60 – 62 61 5 305
63 – 65 64 2 128
 Total 30 1677
∑𝑓𝑥 1677
Mean = (𝑥̅ ) = = = 55.9
∑𝑓 30

The Median
A measure of central tendency that can be used to overcome the weakness of being distorted by
extreme value of the mean or not being computed from incomplete data is the median. It does not
take the exact value of the extreme values into account and this dives it an edge over the mean as
a measure of central tendency. It is the value of the middle item (or the mean of the values of the
two middle items) when the items are arranged in an increasing order of magnitude.
Generally speaking, if N is “odd”, the median of N numbers is given by the value of the
𝑛+1
( ) 𝑡ℎ counting from the end, provided, of course, that the numbers are arranged according to
2

size. If N is “even”, the median of N number is given by the value of the (n/2)th + (n/2 + 1)th.
Median of an Ungrouped Frequency Table
Example 1: Find the median of 6, 4, 1, 9, 8, 3, 7, 10, 5.
Solution
First re-arrange and obtain 1, 3, 4, 5, 6, 7, 8, 9, 10.
The middle value is easily found as 6.

If the number of items is even the median is the average of the two-middle items.
Example 2: Find the median of 163, 149, 152, 160, 195 and 180.
Solution
We first rearrange and obtain 149, 152, 160, 163, 180, 195.
160 + 163 323
Now, 160 and 163 are in the middle. Thus the median = = = 161.5
2 2

Median of an Ungrouped Frequency Table (Using formulae)

𝑛+1
If N is “odd”, the median of N numbers is given by the value ( ) 𝑡ℎ.
2
𝑛 𝑛
( )𝑡ℎ + ( +1)𝑡ℎ
If N is “even”, the median of N number is given by the value ( 2 2
) 𝑡ℎ
2

Example 3:
The table below shows the distribution of marks scored by some students in a mathematics test.
 Marks 22 24 39 42 45 48 56 60
%
Freq. 11 2 7 13 10 3 9 5
Find the median

Solution

Marks 22 24 39 42 45 48 56 60
%
Freq. 11 2 7 13 10 3 9 5
Cum. 11 13 20 33 43 46 56 60
Freq.
From the table, there are 60 members as indicated by the terminal (last) cumulative frequency.
Since this 60 is even, the median mark will be
𝑛 𝑛 60 60
( )𝑡ℎ + ( +1)𝑡ℎ ( )𝑡ℎ + ( +1)𝑡ℎ 30𝑡ℎ + (30+1)𝑠𝑡
2 2 2 2
Then the median value = ( ) 𝑡ℎ = ( ) 𝑡ℎ = ( ) 𝑡ℎ
2 2 2

42 + 42 84
The 30th member is 42%, and 31st member is 42% = = = 42%
2 2

Example 4:
Calculate the median age from the following data.
Age 9 12 13 14 16 17 18 20
(years)
No of 6 15 11 7 12 9 4 7
Students
Find the median.
Solution
Age (years) 9 12 13 14 16 17 18 20
No of 6 15 11 7 12 9 4 7
Students
Cumulative 6 21 32 39 51 60 64 71
No. of
Students

𝑛+1 71+1 72
Since the members are odd, the median age will be ( ) 𝑡ℎ = ( ) 𝑡ℎ =( 2 ) 𝑡ℎ = 36𝑡ℎ
2 2

36th member falls within the cumulative of up to 40 and this is under 14 years of age
Median of a Grouped Frequency Distribution
 𝑁
( −𝐶𝑚 )
2
The median here is found using the formula = 𝐿1 + ( )𝑖
𝐹𝑚

Where
L1 = the lower class boundary of the median class,
N = the total frequency,
i = the class interval,
fm = the frequency of the median class.
Cm = the cumulative frequency of the class just before the median class.

So, the steps are:

i. Compute half of the total frequency
ii. Compute the cumulative frequency
iii. Use the cumulative frequency to identify the group where the median belongs (i.e. the
median class).
iv. Compute the lower-class boundary of the median class.
v. Apply the formula to obtain the median.

Example 5:
The age distributions of sixty members of staff of an establishment are as follows.
Age in years 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49
No of 9 11 16 12 8 4
workers
Calculate the median
Solution
The working is set up as follows.
Group Class Boundaries Frequency Cumulative frequency

20 – 24 19.5 – 24.5 9 9
25 – 29 24.5 – 29.5 11 20
30 – 34 29.5 – 34.5 16 36 median class
35 – 39 34.5 – 39.5 12 48
40 – 44 39.5 – 44.5 8 56
 45 – 49 44.5 – 49.5 4 60
The total number of workers = 60.
60𝑡ℎ
The median class is where the = 30th member is located.
2

The cumulative frequency column shows that the 30th value is located in group 30 – 34 with 29.5
as the lower class boundary.
 L1 = 29.5, i = 5, Cm = 20, fm = 16
𝑁
( –𝐶𝑚 ) (30 −20) 10
Median = 𝐿1 + ( 2
)𝑖 = 29.5 + ( )5 = 29.5 + (16) 𝑥 5
𝐹𝑚 16

50
= 29.5 + = 29.5 + 3.125 = 32.625
16

The Mode
The mode is the highest occurring item in a set of observation.

Finding Mode of a Set of Observation

The mode of a set of observation is easily found by identifying the value that occurs most.
Example 1
Find the mode of the data set 6, 2, 5, 3, 2, 8, 1, 9, 6, 2, 7.
Solution
The highest occurring number is 2 since it occurs three times. So the mode is 2.

Example 2
Find the mode of 14, 36, 28, 14, 10, 16, 10, 19, 2.
Solution
In this example, two numbers occur most. They are 14 and 10 which both occur twice. The modes
are 14 and 10. This is a bimodal distribution i.e. one with two modes. Tri-modal has three modes
while multi-modal distribution has many modes. So the mode, unlike the median and mean, may
not be unique.
Example 3
Find the mode of 100, 152, 165, 190, 300.
Solution
This data set has no mode since every item occurs equally (once). This confirms that the mode
may not exist in some situations.

Finding the Mode of an Ungrouped Frequency Distribution

For an ungrouped frequency distribution, the mode is the value corresponding to the highest
frequency of the distribution.
Example 1
Find the mode of the distribution below:
X 10 20 30 40 50
F 85 76 92 102 98
In this data, the highest frequency is 102 and this corresponds to 40. So the mode is 40.
Please note that the mode is not 102 but the value whose frequency corresponds to 102, which
happens to be 40 in this case.

Finding the Mode of a grouped Frequency Distribution

The mode of this type of distribution is usually computed using a standard formula as follows
𝛥1
Mode = 𝐿1 + (𝛥 )𝑖
1 + 𝛥2

where L1 = lower class boundary of the modal class.

𝛥1 = freq. of modal class – freq. of next lower class.
𝛥2 = freq. of modal class – freq. of next upper class.
Example 1
The data below shows the distribution of suspended particulate matter (micrograms per cubic
metre) in samples of air taken from 60 large cities.
Concentration 10 – 19 20 – 29 30 – 39 40 – 49 50 – 59 60 – 69 70 – 79

Number of cities 5 19 10 13 6 4 3

Calculate the mode

Solution
First identify the modal class. The highest frequency is 19 and this belongs to group 20 – 29. So
the modal class is 20 – 29. The modal class boundaries are 19.5 – 29.5. Also, we can easily find
the class interval to be 10 (i.e. 29.5 – 19.5). So,
L1 = 19.5, i = 10, 𝛥1 = 19 – 5 = 14, 𝛥2 = 19 – 10 = 9
𝛥1 14 140
 Mode = 𝐿1 + (𝛥 )𝑖 = 19.5 + (14+ 9) 𝑥 10 = 19.5 + ( 23 )
1 + 𝛥2

= 19.5 + 6.0869 = 25.59

Exercise
1. Given the following two sets of data:

i. 1, 7, 3, 4, 3, 9, 8, 6. ii. 12, 6, 7, 3, 15, 10, 18, 5, 9, 3, 8, 9, 18.

Calculate the (a) the mode, (b) the median, (c) the mean,
2.

Group 50 – 54 55 – 59 60 – 64 65 – 69 70 – 74

F 3 5 2 3 1

Using the data to calculate (a) the mode, (b) the median, (c) Mean
UNIT 4: MEASURES OF VARIABILITY
Introduction
The measures of variability indicate the degree to which a set of scores differs from each other in
the distribution. These measures present a measure of homogeneity within the group of scores. In
this unit, we shall look at the range, the quartiles, the deciles, the percentiles, the mean deviation,
the variance and the standard deviation.

The Range
The range is simply defined as the difference between highest value and lowest value in a
distribution. So,
Range = Highest value – Lowest value
Coefficient of range is defined as
Hihest Value − Lowest Value
Coefficient of range = Highest Value + Lowest Value

Example 1: Given the values 6, 5, 11, 3, 19, 14, 30, 7 find the range
Solution
The highest value is 30 while the lowest is 3.
The range is thus 30 – 3 = 27.
Also,
30 − 3 27
Coefficient of range = = = 0.82
30 + 3 33

Example 2:
Given the following frequency table, find the range and coefficient of range
X 14 22 12 18 26 24
F 3 6 2 4 5 1
The same formula holds whether for a set of data or a frequency distribution. Inspecting the x
values we identify the highest value as 26 and the lowest as 12.
The range = highest value – lowest value = 26 – 12 = 14.
26 − 12 14
Coefficient of range = = = 0.37
26 + 12 38
Measures of Partition
These are measures that divide a distribution into specified number of parts. They are also called
fractiles. Examples are quartiles, deciles, and percentiles.

Quartiles
This deals with division into four parts. The first quartile (Q1) is known as the lower quartile and
is ¼ of a distribution. The second quartile (Q2) is 2/4 of the distribution. Of course 2/4 is ½ of the
distribution which is the same as the median. The third quartile is ¾ of the distribution.
The lower quartile for a set of n numbers is [¼ (n + 1)]th item.
The upper quartile is also [¾ (n + 1)]th items.
a. Lower and Upper Quartiles of a Ungrouped Frequency Distribution
Example 1
Given the data 3, 4, 1, 2, 7, 12, 5. Find
(i) the Lower Quartile
(ii) the upper Quartile

Solution
The first step is to rearrange the data in ascending or descending order and obtain 1, 2, 3, 4, 5, 7
and 12.
Now the total number of figures in the data is 7. So n = 7.
7+1 8
i. Lower Quartile = [¼ (n + 1)]th= = = 2 = 2nd item.
4 4

Checking through the rearranged data for the second item we obtain Q1 = 2.
3(7+1) 3𝑥8 24
ii. The upper quartile = [¾ (n + 1)]th = = = = 6th item.
4 4 4

Running through the data you obtain 7 as the upper quartile.

Example 2
Find Q1 and Q3 of 2, 14, 6, 10, 2, 6, 14, 18, 18.
We 1st rearrange and obtain 2, 2, 6, 6, 10, 14, 14, 18, 18.
There are 9 numbers, so n = 9
9+1 10
i. Q1[¼ (n + 1)]th= = = 2.5th item.
4 4

i.e. it is between the 2nd and 3rd item. The 2nd item is 2, the 3rd is 6, difference = 4.
So 2.5th item = value of 2nd item + 0.5 of 4 = 2 + 2 = 4
ii. For Q3, ¾ (n + 1) = ¾ (9 + 1) = 7.5th item.
The seventh item is 14 and the 8th is 18.
So, Q3 = 7th item + 0.5 of (18 – 14) = 7th item + 0.5 x 4. Q3 = 14 + 2 = 16
b. Lower and Upper Quartiles of a Grouped Frequency Distribution
The procedure for this is similar to that used for median. So Q1 is located at N/4th position while Q3
is at ¾ Nth position.
𝑁
( –𝐶𝑄1 )
4
𝑄1 = 𝐿1 + ( )𝑖
𝐹 𝑄1

3𝑁
( –𝐶𝑄3 )
4
And 𝑄3 = 𝐿1 + ( 𝐹𝑄3
)𝑖

Where
L1 =is the lower class boundary of the class where the quartile is located;
fQ1 = frequency of the class to which the lower quartile belongs;
fQ3 = frequency of the class to which upper quartile belongs.
CQ1 = cumulative frequency of class preceding the class of the lower quartile class
CQ3 = cumulative frequency of class preceding the upper quartile class;
i = class interval.
The same formula goes for other fractiles like deciles and percentiles. You only need to make the
necessary adjustment.

Example
The data below shows the age distribution of dwellers of a village.
Age Frequency
20 -29 12
30 – 39 13
40 – 49 39
50 -59 28
60 -69 11
70 -79 7
80 -89 10
Calculate the first and third quartile for the distribution.
Solution
Age Frequency Cum. Freq. Class boundary
20 -29 12 12 19.5 – 29.5
30 – 39 13 25 29.5 – 39.5
40 – 49 39 64 39.5 – 49.5 Q1Class
50 -59 28 92 49.5 – 59.5
60 -69 11 103 59.5 -69.5
70 -79 7 110 69.5 -79.5
80 -89 10 120 79.5 – 89.5 Q3 Class

(i) From the table, Q1 = N/4 = 120/4 = 30th cumulative frequency

Hence,
L1 = 39.5, N = 120, CQ1 = 25, FQ1 = 39, i = 10
𝑁
( –𝐶𝑄1 ) (30 –25) (5) 50
4
𝑄1 = 𝐿1 + ( )𝑖 = 39.5 + ( ) 10 = 39.5 + ( 39 ) 10 = 39.5 + (39)
𝐹 𝑄1 39

= 39.5 + 1.28 = 40.78 ≈ 40.8

(ii) From the table Q3 = 3N/4 = 3 x 120/4 = 90th cumulative frequency
Hence,
L1 = 49.5, N = 120, 𝐶𝑄3 = 64, 𝐹𝑄3 = 28, i = 10
3𝑁
( –𝐶𝑄3 ) (90 –64) 26 260
4
𝑄3 = 𝐿1 + ( ) 𝑖 = 49.5 + ( ) 10 = 49.5 + (28) 10 = 49.5 + ( 28 )
𝐹𝑄3 28

= 49.5 + 9.29 = 58.79 ≈ 58.8

Deciles
These are values that divide a distribution (or, better, the total area of the triangles of its histogram)
into 10 equal parts. We shall write D1 for the first decile which exceeds 10 per cent and is exceeded
by 90 per cent of the data, D2 for the second decile which exceeds 20 per cent and is exceeded by
80 per cent of the data, and, in general, Di for the ith decile which exceeds i. 10 per cent and is
exceeded by 100 – 𝑖. 10 per cent of the data. In order to reach any one of the Di (𝑖 can be either 1,
2, 3, . . . , or 9), we will have to count n. i/10 cases starting at the bottom of the distribution or n
minus that many starting at the other end. The formula for calculating decile is
𝑁
(10 – 𝐶𝐷𝑖 )
𝐷𝑖 = 𝐿1 + ( )𝑖
𝐹𝐷𝑖
 Where,
L1 = The lower-class boundary of the ith decile.
N = Total Frequency.
𝐶𝐷𝑖 = Cumulative frequency before the ith decile.
𝐹𝐷𝑖 = Frequency of the ith decile.
i = Size (width) of the ith decile.
Di = the required decile.

Percentile
These are values that divide a distribution (or, better, the total area of the triangles of its histogram)
into 100 equal parts.
The formula for calculating Percentile
is
𝑁𝑖
(100 – 𝐶𝑃𝑖 )
𝑃𝑖 = 𝐿1 + ( )𝑖
𝐹𝑃𝑖

Where,
L1 = The lower-class boundary of the ith Percentile.
N = Total Frequency.
𝐶𝐷𝑖 = Cumulative frequency before the ith Percentile.
𝐹𝐷𝑖 = Frequency of the ith Percentile.
𝑖 = Size (width) of the ith Percentile.
Pi = the required Percentile.

Example
The table shows the marks distribution of 50 students.
Marks Frequency
16 – 20 2
21 – 25 4
26 – 30 15
31 – 35 9
36 – 40 7
41 – 45 13
For the distribution, calculate the following:
(i) 15th percentile
(ii) 6thdecile

Solution
Marks Frequency Cum. Freq. Class Boundary
16 – 20 2 2 15.5 – 20.5
21 – 25 4 6 20.5 – 25.5
26 – 30 15 21 25.5 – 30.5
31 – 35 9 30 30.5 – 35.5
36 – 40 7 37 35.5 – 40.5
41 – 45 13 50 40.5 – 45.5

𝑁𝑖
(100 – 𝐶𝑃𝑖 )
𝑃𝑖 = 𝐿1 + ( )𝑖
𝐹𝑃𝑖

15 𝑥 50
( –6) (7.5 –6)
100
(i) 15th percentile = 𝑃15 = 25.5 + ( ) 𝑥 5 =25.5 + ( )𝑥 5
15 15

1.5 7.5
= 25.5 + ( 15 ) 𝑥 5 = 25.5 + ( 15 ) = 25.5 + 0.5 = 26

(ii)
𝑁
(10𝑖 – 𝐶𝐷𝑖 )
𝐷𝑖 = 𝐿1 + ( )𝑖
𝐹𝐷𝑖

6 𝑥 50
( –21) (30 –21) 9
10
6thDecile = 𝐷6 = 30.5 + ( ) 𝑥 5 =30.5 + ( ) 𝑥 5 = 30.5 + (9) 𝑥 5
9 9

= 30.5 + 5 = 35.5

Mean Deviation
Mean Absolute Deviation
∑|𝑋−𝑋̅ |
Mean absolute deviation = for a set of numbers and
𝑁
̅̅̅
∑𝑓|𝑋−𝑋|
Mean absolute deviation = for a frequency distribution.
∑𝑓

The above formula is known as the mean absolute deviation or popularly known as Mean
Deviation.
Example 1: Find the mean absolute deviation of 5, 6, 7, 10, 12,
Solution
∑𝑋 5+ 6 + 7 + 10 + 12 40
The mean (𝑋̅) = 𝑁 = =8
5 5
∑|𝑋−𝑋̅ | |5 − 8|+ |6 − 8|+ |7 − 8|+ |10 − 8| + |12 − 8| |−3|+ |−2|+ |−1|+|+2| +|+4|
M.D = = =
𝑁 5 5
3+ 2+ 1 +2 + 4 12
= = = 2.4
5 5

Example 2:
Find the mean absolute deviation of the data

Group 1- 5 6 – 10 11 – 15 16 - 20
Frequency 6 14 8 2

Solution
Set up the working as follow:
Group Mid-mark (x) F fx |x – 𝑥̅ | f|x - 𝑥̅ |
1-5 3 6 18 6 36
6-10 8 14 112 1 14
11-15 13 8 104` 4 32
16- 20 18 2 36 9 18
30 270 100
∑𝑓𝑥 270
Mean (𝑋̅) = = = 9
∑𝑓 30
̅̅̅
∑𝑓|𝑋−𝑋| 100
Mean Absolute deviation = = = 3.33
∑𝑓 30

Variance and Standard Deviation

The population variance is often denoted by the Greek letter 2 (pronounced sigma square). Thus,
∑|𝑋−𝑋̅ |2
2 = for a set of data and
𝑁
∑𝑓|𝑋−𝑋̅ |2
2 = for a frequency distribution
∑𝑓

The standard deviation is the positive square root of the variance. Thus, Population Standard
deviation  (pronounced sigma) is
∑|𝑋−𝑋̅|2
 =√ for a set of data
𝑁
and
∑𝑓|𝑋−𝑋̅ |2
√ for a frequency distribution.
∑𝑓

The sample variance is denoted by S2 and is given by

∑|𝑋−𝑋̅ |2
S2 = for a set of data
𝑁−1

and
∑𝑓|𝑋−𝑋̅ |2
for a grouped data.
𝑁−1

However, as the sample size N increases (usually n > 30) the sample variance approximates the
population variance.
Example 1:
A set of 5 experimental animals was fed a special diet for one week and produced the following
gains in weight: 6, 4, 8, 10, 12.
Calculate the variance and standard deviation for this sample.
Solution
∑|𝑋−𝑋̅ |2
Variance = 𝑁−1

Since it is a sample
∑𝑓𝑥 6+4+8+10+12 40
Mean, (𝑋̅) = = = =8
∑𝑓 5 5

|6−8|2 + |4−8|2 + |8−8|2 +|10−8|2 + |12−8|2 |−2|2 + |−4|2 + |0|2 +|2|2 + |4|2
 Variance = S2 = =
5−1 4
4+16+ 0+4+ 16 40
= = = 10
4 4

Standard deviation = √variance = √10 = 3.16.

Example 2:
The age distribution of all members of staff of an organization is given below.
Age in years 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44
No of workers 4 6 15 8 7

Calculate the mean and variance. What is the standard deviation?

Solution
The population variance and standard deviation are required. We set up the working as follows:
Group Mid-mark, f Fx |x – x| |x – 𝑥̅ |2 f|x – 𝑥̅ |2
(x)
20 – 24 22 4 88 -11 121 484
25 – 29 27 6 162 -6 36 216
30 – 34 32 15 480 -1 1 15
35 – 39 37 8 296 4 16 128
40 – 44 42 7 294 9 81 567
40 1320 1410

∑𝑓𝑥 1320
Mean, 𝑋̅ = = = 33
∑𝑓 40

∑𝑓|𝑋−𝑋̅|2 1410
Variance = = = 35.25
∑𝑓 40

Standard deviation = √variance = √35.25 = 5.94.

Coefficient of Variation
standard deviation 100
This is defined as 𝑥
mean 1

Exercise
1. Given the following two sets of data:
(i) 1, 7, 3, 4, 3, 9, 8, 6.
(ii) 12, 6, 7, 3, 15, 10, 18, 5, 9, 3, 8, 9, 18.
Calculate the (a) the lower quartile, (b) the upper quartile of each of the data set.
2. Given the set of data 12, 6, 7, 3, 15, 10, 18, 5, 9, 8, 9, 18. Calculate
(a) the range (b) the coefficient of range (c) the mean deviation (d) the
variance
3. The following shows the scores of 80 students in a test
39, 29, 44, 31, 46, 37, 26, 34, 32, 42, 30, 35, 25, 18, 26, 37, 15, 22, 26, 28, 21, 15, 40, 25, 28, 33,
32, 21, 43, 32, 29, 30, 38, 22, 20, 29, 38, 49, 13, 38, 26, 30, 45, 21, 35, 26, 28, 39, 33, 32, 41, 15,
39, 33, 34, 26, 33, 30, 23, 31, 37, 47, 18, 30, 30, 37, 29, 43, 14, 22, 35, 28, 26, 31, 41, 32, 24, 20,
21, 32, 14
(a) Using interval of 12 – 14, construct a group frequency table
(b) Calculate the quartile deviation
(c) Calculate the 10 – 90 percentile range.
(d) Compute the mean absolute deviation
(e) Find the standard deviation.
UNIT 5: MEASURE OF ASSOCIATION AND CORRELATION
Introduction
In this unit, you will learn about the degree of relationship between the attributes or variables
expressed as a coefficient of correlation. The result of the unit will teach you the most common
types of correlation which are Pearson Product Moment and Spearman Rank Order.

The Concept of Correlations

Correlation is the statistical technique for establishing the extent of relationship or association
between two variables. Correlation values ranges from –1 to +1. It means that a correlation
coefficient of –1 indicates a perfect negative relationship, while +1 shows perfect positive
relationship and 0 correlation coefficient implies no relationship at all.
A Positive coefficient. This indicates a direct relationship i.e. it means having high scores in one
variable mean having high scores in the other variable. It also implies that having low scores in
one variable means having low scores in the other variables.
A Negative coefficient: This is opposite of positive coefficient. It means that having high score in
one variable score means low score on another variable. It also implies that those who score low
on one variable score high on the other variable.

Pearson Product Moment Correlation Coefficient (r)

This type of correlation coefficient, named after the man who developed it, is used when the two
sets of data are continuous or interval data. There are two major approaches of calculating the
Pearson Product Moment correlation coefficient (r).
a. The deviation Method
∑(𝑋−𝑋 ̅ ) (𝑌−𝑌̅ ) ∑𝑋𝑌
The formula is given by: or
√(𝑋−𝑋̅ )2 (𝑌−𝑌̅)2 √(∑𝑋 2 ) (∑𝑌 2 )

where x = 𝑋 − 𝑋̅ and y = 𝑌 − 𝑌̅
Example. Let us considered the two-score obtained by the same group of 10 students in PDE 5110
test (X) and PDE 5111 and (Y) respectively.
PDE 5110 (X) 10 13 16 18 19 25 7 8 14 20
PDE 5111 (Y) 5 9 8 7 10 18 4 11 13 15
Solution
S/N PDE PDE 𝑋 − 𝑋̅ 𝑌 − 𝑌̅ (𝑋 − 𝑋̅ ) (𝑌 − 𝑌̅ ) (𝑋 − 𝑋̅)2 (𝑌 − 𝑌̅)2
5110 5111 (x) (y) (xy) 𝑥2 𝑦2
score score
(X) (Y)
1 10 5 -5 -5 25 25 25
2 13 9 -2 -1 2 4 1
3 16 8 1 -2 -2 1 4
4 18 7 3 -3 -9 9 9
5 19 10 4 0 0 16 0
6 25 18 10 8 80 100 64
7 7 4 -8 -6 48 64 36
8 8 11 -7 1 -7 49 1
9 14 13 -1 3 -3 1 9
10 20 15 5 5 25 25 25
Total 150 100 159 294 174

150
Mean (𝑋̅ ) = = 15
10
100
Mean (𝑌̅ ) = = 10
10

∑(𝑋 − 𝑋̅ ) (𝑌 − 𝑌̅ ) = 159
(𝑋 − 𝑋̅)2 = 294
(𝑌 − 𝑌̅)2 = 174
Now, substitutes the above value in the formula
∑(𝑋−𝑋̅ ) (𝑌−𝑌̅ )
r=
√(𝑋−𝑋 )2 (𝑌−𝑌̅)2
̅

159 159 159

r= = = 226.18 = +0.70
√294 𝑥 174 √51156
 b. Calculating Pearson r using the Raw Score Method
𝑁∑𝑋𝑌− ∑𝑋 ∑𝑌
The formula is given by r =
√[𝑁∑𝑋 2 − (∑𝑋)2 ][𝑁∑𝑌 2 − (∑𝑌)2 ]

Let us use the same data in example above

PDE 5110 10 13 16 18 19 25 7 8 14 20
(X)
PDE 5111 5 9 8 7 10 18 4 11 13 15
(Y)

Solution
S/N PDE 5110 PDE 5111 X2 Y2 XY
Score (X) Score (Y)
1 10 5 100 25 50
2 13 9 169 81 117
3 16 8 256 64 128
4 18 7 324 49 126
5 19 10 361 100 190
6 25 18 625 324 450
7 7 4 49 16 28
8 8 11 64 121 88
9 14 13 196 169 182
10 20 15 400 225 300
Total 150 100 2544 1174 1659

N = 10
∑𝑋 = 150
∑𝑌 = 100
∑𝑋 2 = 2544
∑𝑌 2 = 1174
∑𝑋𝑌 = 1659

Now, substitute these values in the raw score formula, we obtain

𝑁∑𝑋𝑌− ∑𝑋 ∑𝑌
r=
√[𝑁∑𝑋 2 − (∑𝑋)2 ][𝑁∑𝑌 2 −(∑𝑌)2 ]
10 𝑥 1659−150 𝑥 100
r=
√[10 𝑥 2544 − (1502 )][10 𝑥 1174− 1002 ]
16590−15000
r=
√[25440− 22500][11740− 10000]
 1590
r=
√[2940][1740]
1590 1590
r= = 2261.77 = +0.70
√5115600

you can see that the two approaches give the same result.

Spearman Rank Order Correlation Coefficient – rho

This correlation coefficient was developed independently by Spearman and Brown. This is why
it is sometimes referred to as Spearman-Brown Rank Order Correlation Coefficient. It is used
when the scores in each variable are ranked in the same direction, with respect to magnitude. This
6𝛴𝐷 2
is defined by the formula 𝑝 = 1 - N(𝑁2−1)

Let illustrate solve using the same set of scores obtained by 10 students in PDE 5110 and PDE
5111 tests, respectively.
PDE 5110 10 13 16 18 19 25 7 8 14 20
(X)
PDE 5111 5 9 8 7 10 18 4 11 13 15
(Y)

Solution
S/N PDE 5110 PDE 5111 Rx RY (Rx – Ry) D2
Score (X) Score (Y) D
1 10 5 8 9 -1 1
2 13 9 7 6 1 1
3 16 8 5 7 -2 4
4 18 7 4 8 4 16
5 19 10 3 5 2 4
6 25 18 1 1 0 0
7 7 4 10 10 0 0
8 8 11 9 4 5 25
9 14 13 6 3 3 9
10 20 15 2 2 0 0
Total 60
 N = 10
𝛴𝐷2 = 60
Now, substitute in the formula
6𝛴𝐷 2
𝑝 = 1 - N(𝑁2−1)
6 𝑥 60
𝑝 = 1 - 10(102 −1)
360
𝑝 = 1 - 10(100 −1)
360
𝑝 = 1 - 10 x 99
360
𝑝 = 1 - 990

𝑝 = 1 – 0.36 = 0.64

Exercise
1. What is the interpretation of the correlation results?
2. The interest scores and achievement scores of a group of students in a particular subject
are presented below:
Interest 25 21 14 16 18 19 18 18 19 23 16 19 18 25 22 20 18 20
Score
Achievement 93 96 60 97 76 67 67 77 61 75 69 94 60 88 72 66 74 75
Score
Calculate the extent of the relationship between the students’ interest and achievement in the
subject using
i. Pearson Product Moment Correlation Coefficient
ii. Spearman Rank Order Correlation Coefficient.
UNIT 6: THE TESTING OF HYPOTHESIS
Introduction
The purpose of testing a hypothesis is to determine the probability that it is supported by facts. If
a hypothesis is successfully verified or tested and confirmed to be true, it is then used to support a
theory. In other words, theories are developed, tested and confirmed in research through the
process of hypothesis testing. This leads to the generation or advancement of knowledge. In this
unit, you are going to be exposed to the rudiments of the processes involved in testing hypotheses.

Hypothesis Testing
Drawing inferences about a population based on a random sample from that population involves
formulating and testing hypotheses. Hypothesis testing is really a systematic way to test claims or
ideas about a group or population. The hypothesis to be tested is usually stated in null hypothesis
represented with the symbol HO. The null hypothesis (H0), is a statement about a population
parameter, such as the population mean, that is assumed to be true. It is hypothesis of no difference
or no effect. For instance, in a study to compare the performance of male and female students in
mathematics, a simple null hypothesis could be: there is no significant difference between male
and female students in mathematics performance. i.e. 𝜇𝑏 = 𝜇𝑔
The null hypothesis is usually stated against the alternative hypothesis. alternative hypothesis is
usually represented by the symbol H1. An alternative hypothesis (H1) is a statement that directly
contradicts a null hypothesis by stating that the actual value of a population parameter is less than,
greater than, or not equal to the value stated in the null hypothesis. It is the hypothesis we accept
when the null hypothesis is rejected. An alternative hypothesis could be: there is significant
difference between male and female students in mathematics performance. i.e. 𝜇𝑏 ≠ 𝜇𝑔 or 𝜇𝑏 -
𝜇𝑔 = 0

level of significance for a test.

Level of significance, or significance level, refers to a criterion of judgment upon which a decision
is made regarding the value stated in a null hypothesis. The criterion is based on the probability of
obtaining a statistic measured in a sample if the value stated in the null hypothesis were true. In
most researches in education, two alpha levels are used. These are 5% (0.05) and 1% (0.01). If you
choose 5% in a particular study, the implication is that if your study is replicated 100 times, the
same outcome will occur 95 out of 100, and 5 out of 100 may vary due to chance. If it is 1% (0.01)
level of significance, it means that if your study is replicated 100 times, you are sure 99 out of 100
will be correct while 1 out of 100 may vary due to chance factors. This is a more rigorous
confidence level.

Degrees of Freedom
This is the number of observations which are free to vary when certain restrictions have been
placed on the data being considered. Take for instance, in your class, you ask your students to
provide any four numbers which would be added to 6 to add up to 22. In this case, it is fixed, other
numbers can vary. But no matter how they vary, 6 must be added to sum up to 22. Therefore, the
degree of freedom here is N – 1, where N is the total number of choices while 1 is the fixed
variable. As we go on, you will see the modal for getting the degrees of freedom for different
tests.

Making a Decision.
You use the value of the test statistic to make a decision about the null hypothesis. The decision is
based on the probability of obtaining a sample mean, given that the value stated in the null
hypothesis is true. If the probability of obtaining a sample mean is less than 5% when the null
hypothesis is true, then the decision is to reject the null hypothesis. If the probability of obtaining
a sample mean is greater than 5% when the null hypothesis is true, then the decision is to retain
the null hypothesis. Also, you can compare the calculated value of the test statistic with a critical
or table value of the test statistic. The critical or table value of the test statistic, therefore serves as
a criterion value. This serves as the bases for rejecting or not rejecting the null hypothesis. In sum,
there are two decisions a researcher can make:
1. Reject the null hypothesis. The sample mean is associated with a low probability of occurrence
when the null hypothesis is true.
2. Retain the null hypothesis. The sample mean is associated with a high probability of occurrence
when the null hypothesis is true.
Types of Error
You decide whether to retain or reject the null hypothesis. Because we are observing a sample and
not an entire population, it is possible that a conclusion may be wrong. Table below shows that
there are four decision alternatives regarding the truth and falsity of the decision we make about a
null hypothesis:
1. A true null hypothesis is not rejected.
2. A true null hypothesis is rejected.
3. A false null hypothesis is rejected.
4. A false null hypothesis is not rejected.
When a true null hypothesis is not rejected or a false null hypothesis is rejected, the correct decision
is taken and no error is involved as in 1 and 3 above. However, when a true null hypothesis is
rejected or a false null hypothesis is not rejected, an incorrect decision is made.
These represent the two types of errors encountered in hypothesis testing. These errors are called
type I and type II errors respectively.
Type I error is made when a false null hypothesis is rejected
Type II error is made when a false null hypothesis is not rejected.

Table showing types of errors in decision making

State of Nature DECISION
Reject Null Hypothesis Do not reject the null hypothesis

Null Hypothesis is true Type I error Correct decision

Null Hypothesis is false Correct decision Type II error

You have to note that as you try to minimize type I error by becoming too rigorous, maybe you
reduce the significance level from 5% to 1%, you stand the chance of making type II error by
increasing the level of significance from 1% to 5%.

One-Tailed and Two-Tailed Tests

One Tailed Tests: This is used to test hypotheses concerning the mean in a single population with
a known variance. It is stated to indicate the direction of difference. For example, people who live
in high altitude areas perform better in long distance races. People who have stout bodies do better
in short-put. Expensive cars are better in performance etc.
Two-Tailed Tests: Two-tailed tests or non-directional test, are hypotheses tests where the
alternative hypothesis is stated as not equal to (≠). The researcher is interested in any alternative
from the null hypothesis. It is stated in such a way that it does not indicate a direction of difference,
but agrees that a difference exists. They merely state that there is no significant difference between
A and B. For instance, there is no significance difference in academic performance between male
and female students in mathematics.

The T-Test
It was developed by William Gosset in 1908. You will have to note that there are conditions for
the use of t-tests. These are:
i. there must be two groups to be compared;
ii. the population from which the samples are drawn must be normally distributed;
iii. the population variances are homogenous;
iv. the samples are independent or randomly drawn from the population;
v. the variables have continuous values;
vi. suitable for both large and small samples (but not less than ten).
Note that any sample size less than 30 is regarded as small, but when the sample size is more than
30, it is regarded as large. The procedure for carrying out z-test is the same to that of t-test. While
z-test is specifically used for large samples, t-test can be used for both small and large samples.
When t-test is used for large samples, it approximates to z-test.

Difference between Population and Sample Means

When comparing a population and sample means, you will use this mode:
𝑥̅ − 𝜇
t= S
√𝑛−1

where 𝑥̅ = sample mean

𝜇 = population mean
S = standard deviation
n = number.
For instance, you are given that the mean achievement score of all SS I students in Odogbolu, in a
Physics standardized test is 50%. A teacher conducted a study to verify this claim. He used 25 SS
II students in that locality. He drilled them on the different aspects of Physics syllabus for SS I,
for about eight weeks. At the end, the teacher administered the Physics test on the 25 students. His
results are 55.75 as mean and 7.40 as standard deviation.
The first step is to propose a hypothesis (H0). You can say the sample mean of 55.75 is not
significantly greater than the population mean of 50, at an σ level of 0.05 or you can say that there
is no significant difference between the sample mean of 55.75 and population mean of 50.
𝑥̅ − 𝜇
t= S
√𝑛−1

where 𝑥̅ = 55.75 𝜇 = 50 S = 7.40 n = 25.

55.75 − 50 5.75 7.4 5.75 x √24
t= 7.4 = ÷ = = 0.77 x 4.899 = 3.772
1 √24 7.4
√25−1

At this point, you have to take a decision. This will be based on the comparison of the calculated
value of t-test and the value of t-test on the table or the critical region.
Now that tcal = 3.772, df = 25 – 1 = 24, alpha level = 0.05
ttab at (25: 0.05) = 2.060.
From this result, tcal is greater than the ttab i.e., 3.772> 2.060. We therefore reject that there is no
significant difference between the population mean and the sample, mean.

Difference between Two Independent Samples’ Means

Most of the times, you will be confronted with a situation where two samples are randomly and
independently drawn from a normal population. If the variances of the samples as estimates of the
population variance do not differ significantly or are homogenous, we can then say that they have
a t-distribution. This is particularly when the samples’ sizes are not large. Remember that a large
sample size is from 30 and above. The t-statistics which you can use in this case is as follows:
𝑥̅ 1 − 𝑥̅ 2
t=
[(𝑛1 −1)𝑠2 2
1 +((𝑛2 −1)𝑠2 )] 1 1
√ ( + )
(𝑛1 + 𝑛2 −2) 𝑛1 𝑛2

where 𝑠1 = Standard deviation

For instance, a teacher wanted to compare the academic performance of male and female students
in mathematics with a view to finding out whether their mean performances are significantly
different. He collected samples of the two groups. His results are shown in the table below:

Gender Mean Performance Standard Deviation No. of Samples

Male 24.92 5.89 65
Female 16.69 5.53 60
Solution:
Propose a null hypothesis H0: There is no significant difference between the mean performances
of the students from the two sets.
𝑥̅1 = 24.92, 𝑥̅2 = 16.69, 𝑠1 = 5.89, 𝑠2 = 5.53, 𝑛1 = 65, 𝑛2 = 60
24.92−16.69
t=
[(65−1)5.891 +((60−1)5.532 )] 1 1
√ ( + )
(65+ 60−2) 65 60

8.23
t= [64 𝑥 34.69 +59 𝑥 30.58]
√ ( 0.015 +0.017)
123

8.23
t= [2220.16+1804.22]
√ ( 0.032)
123

8.23
t= 4024.38
√ ( 0.032)
123

8.23
t=
√32.72 ( 0.032)
8.23
t= = 7.84
√1.05

df = (𝑛1 + 𝑛2 − 2) = (65+60 – 2: 0.05) = ttab at 123: 0.05

tcal = 7.84. ttab = (123 : 0.05) = 1.96
Since tcal is greater than ttab, we reject H0. It means that there is a significant difference between
the mean performances of the two sets of students.
The Chi-Square (𝒙𝟐 ) Test
The chi-Square test is a non – parametric inferential statistics method used in the analysis of
frequencies or nominal data. The word chi is pronounced kai. The test allows us to determine
whether or not a significant difference exists between the observed frequencies of cases in each
category of variables studied versus the expected frequencies or data or number of cases in each
category of variables based on the null hypothesis. The observed frequency is data obtained from
the actual frequency count while the expected is the data that would be obtained if equal numbers
responded to the same variables equally. The larger the margin between the observed and the
expected frequency counts, the higher the chi-square value. You can compare the calculated chi-
square against a given critical value to determine whether it is significant. The formula for chi-
square is:
(𝑂−𝐸)2
𝑥2 = 𝛴 𝐸

Where 𝛴 = Sum of,

O is the observed frequency, and
E is the expected frequency in each cell.
Example
A researcher studying the interest of 300 students towards the learning of mathematics from
different socio – economy background on a 4 – point Likert – type scale, obtained the following
information
Socio Mathematics Interest
Economy
Strongly Agree Disagree Strongly Total
Background
Agree Disagree
High 25 30 20 30 105
Middle 35 40 15 10 100
Low 30 45 12 8 95
Total 90 115 47 48 300
The researcher is interested in knowing whether or not mathematics interest is dependent on the
socio – economy status of the students. The appropriate null hypothesis under the chi – square test
of independence is formulated as follows:
Ho: The interest of students towards mathematics learning is not significantly independent of
socio-economy status
H1: The interest of students towards mathematics learning is significantly dependent of socio-
economy status
Let 0.05 level of significance be for testing the hypothesis
𝐹𝑅 𝑥 𝐹𝐶
E(RC) = 𝑁

Where E(RC) = Expected frequency of the cell

𝐹𝑅 = Total row frequency
𝐹𝐶 = Total column frequency
N = Total frequency
Expected frequencies can now be computed as:
90 𝑥 100
Row 1, Cell 1: = = 30
300
90 𝑥 115
Row 1, Cell 2: = = 34.5
300
90 𝑥 47
Row 1, Cell 3: = = 14.1
300
90 𝑥 38
Row 1, Cell 4: = = 11.4
300

100 𝑥 100
Row 2, Cell 1: = = 33.3
300
100 𝑥 115
Row 2, Cell 2: = = 38.3
300
100 𝑥 47
Row 2, Cell 3: = = 15.7
300
100 𝑥 38
Row 2, Cell 4: = = 12.7
300

190 𝑥 100
Row 3, Cell 1: = = 63.3
300
190 𝑥 115
Row 3, Cell 2: = = 72.8
300
190 𝑥 47
Row 3, Cell 3: = = 29.8
300
190 𝑥 38
Row 3, Cell 4: = = 24.1
300
These expected frequencies are now presented along with the corresponding observed frequencies
as shown below.
Socio Interest in Mathematics
Economy
Strongly Agree Disagree Strongly Total
Background
Agree Disagree
O E O E O E O E
High 25 31.5 15 34.5 20 14.1 30 11.4 90
Middle 35 30.0 40 38.3 15 15.7 10 12.7 100
Low 30 28.5 45 72.8 12 29.8 15 24.1 190
Total 100 115 47 38 300

You now compute the chi – square as follows

(𝑂−𝐸)2
𝑥2 = 𝛴 𝐸
(25 − 31.5)2 ( − 34.5)2 (20 − 14.1)2 (30 − 11.4)2 (35 − 33.3)2 (40 − 38.3)2 (15 − 15.7)2
𝑥2 = + + + + + + +
31.5 34.5 14.1 11.4 33.3 38.3 15.7
(10 − 12.7)2 (30 − 63.3)2 (45 − 72.8)2 (12 − 29.8)2 (15 − 24.1)2
+ + + +
12.7 63.3 72.8 29.8 24.1
(−5)2 (−19.5)2 (5.9)2 (18.6)2 (1.7)2 (1.7)2 (−0..7)2 (−2.7)2 (−33.3)2 (−27.8)2 (−17.8)2
𝑥2 = + + + + + + + + + +
30 34.5 14.1 11.4 33.3 38.3 15.7 12.7 63.3 72.8 29.8
(−9.1)2
+
24.1
25 380.25 34.81 345.96 2.89 2.89 0.49 7.29 1108.89 772.84 316.84 82.81
𝑥2 = + + + + 33.3 + 38.3 + 15.7 + 12.7 + + + +
30 34.5 14.1 11.4 63.3 72.8 29.8 24.1

𝑥 2 = 0.83 + 11.02 + 2.47 + 30.35 + 0.09 + 0.08 + 0.03 + 0.57 + 17.52 + 10.62 + 10.63 + 3.44
𝑥 2 = 87.65
Degree of freedom (df) = (R – 1)(C – 1),
Where R = The number of rows,
C = The number of columns
Then df = (3 – 1) (4 – 1) = 2 x 3 = 6
The critical 𝑥 2 value for 6 df and 0.05 level of significant is 12.592 [i.e., 𝑥 2 (6, 0.05) = 12.592]
𝑥 2 – cal = 87.65
𝑥 2 – crit = 12.592
The calculated value exceeds the critical value; hence we reject the null hypothesis. This implies
that the interest of the students towards the learning of mathematics depends on socio – economy
status. in other words, students from different socio – economy backgrounds hold different interest
toward mathematics.
MODULE 5: WRITING RESEARCH REPORTS
Unit 1: Writing Research Reports

Introduction
Research report is the final stage of any research process. The main aim is to communicate or
disseminate the research findings to the literate audience. In writing the research report, the
impersonal mode is preferred. That is to say, instead of say “I did this”, you should say “the study
was carried out to do this”. You will have to note that in presenting a research report, you have to
use the required format. Most institutions have their own format. These formats or house-styles
do not vary significantly from the general format

Sample Format of a Research Report

The research project report is usually presented following a particular format. The format varies from
place to place and from discipline to discipline. The format considered in the unit is said to be widely
used in educational sector. Importantly, a research project report comprises three sections. They are:

1. The preliminaries
2. The Main Body;
3. The Supplementary page

1. Preliminary pages:
i. Title page
ii. Approval / Acceptance page
iii. Certification page
iv. Dedication
v. Acknowledgement page
vi. Abstract
vii. Table of Contents
viii. List of tables
ix. List of figures
x. List of appendices
2. The Main Body
Chapter 1: Introduction
i. Background to the Problem
ii. Statement of the Problem
iii. Purpose / Objectives of the Study
 iv. Significance of the Problem
v. Scope of the Study
vi. Research Questions and/or Hypotheses
vii. Definitions of Terms
Chapter 2: Literature Review
i. Review of Related Literature
Chapter 3: Research Methodology
i. Research design
ii. Population
iii. Samples and Sampling techniques
iv. Instrumentation – construction of instruments, validation, reliability of instruments,
administration and scoring
v. Methods of data analysis
Chapter 4: Presentation of Results
i. Data analysis and findings
ii. Summary of major findings
Chapter 5: Discussion
i. Interpretation of findings
ii. Discussion of findings
iii. Implication of the study
iv. Recommendations
v. Limitations
vi. Suggestions for further study

3. Supplementary page:
i. Bibliography
ii. Appendices
iii. Index
Steps in Research Report Format
You have already noted that a research report is a straight forward, clearly and precisely written
document in which you attempt to explain how you have resolved the problem before you. The
presentation, in this unit, is consistent with the most acceptable formats. So let us explain them.
Preliminary Pages
i. The title page: This is the first page of this section. It carries information on the project.
It contains the title of the study, the name of the author, the relationship of the research
to a course or degree requirement, the department and the institution where the report
is to be submitted, and the date of presentation.
ii. Approval/Acceptance page: This contains some of the following information: the
names, signatures of the head of department, the dean, the supervisor(s) and dates, the
names(s) of the student(s). The specifications vary from institution to institution.
iii. Certification page: This contains the attestation of originality of the research project.
It may also include the name and signature of the external examiner.
iv. Dedication: Here, tributes are paid to persons who are dear to the author or those who
contributed in one way or the other to the success of the project. Emotionally-laden
words may be permitted in order to pay the tribute.
v. Acknowledgement page: This is used to express gratitude to those who helped in the
process of conducting the research and preparing the report. It should be simple and
restraining.
vi. Abstract: This is a succinctly summarised form of the report containing the aim of the
investigation, the sample, methods of investigation, the instruments used for data
collection, the analysis and findings and as well recommendation(s).
vii. Table of Contents: This serves an important purpose of providing the outline of the
contents of the report. It lays out in a tabular form, the chapters, headings and sub-
headings of the report. It is sequentially arranged and numbered from the preliminary
to the supplementary pages. Page references for each topic are so indicated.
viii. List of tables and figure and appendices: If tables and/or figures are used in the
report, a separate page is included for each list. It should indicate the page numbers in
which the tables or figures presented in the report are located. The numbers and titles
are serially listed. Also contained is the list of appendices that are embodied in or
 annexed to the report. The pages of the preliminary section are numbered with lower-
case Roman numerals (i, ii, iii, iv, v, etc.).
The Main Body
This section represents the major report of the project. The materials of this section are usually
arranged in five chapters as follows:
Chapter 1: Introduction
This is arranged under sub – section as given below
i. Background to the Problem: In this section, the researcher traces such factors or
circumstances that informed the investigation of the study. Here, the researcher
presents all materials which will help his reader understand the development of the
problem under investigation. It is presented using reasoned statements to show that it
is worthwhile to dissipate resources to carry out the study. It shows the nature, scope,
and current status of the factors affecting the problem. It has to be presented in a way
as to be clear and convincing to the reader.
ii. Statement of the Problem: The problem as already highlighted is stated in simple,
clear and unambiguous terms. This is not required to be too long.
iii. Purpose of the Study/Objectives of the Study: These go interchangeably, but it states
the specific aspects of the study and the reasons for focusing on them. It includes
statements of what should be accomplished and all that would be investigated.
iv. Significance of the Problem: this can also be called the rationale or the importance of
the study. It has to do with major contributions the researcher hope the work will make
either to the theory or the practice of his field. The usefulness, the utility value of the
research or findings of the research should be articulated. The institutions, groups or
individuals who are expected to profit or benefit and the benefits expected to accrue to
them are to be stated in this section.
v. Scope of the Study: This is interchanged with the delimitation of the study. Here, you
will have to indicate the extent to which the study will be covered. It involves the
geographical area, time period, and variables to be covered.
vi. Research Questions and/or Hypotheses: These are formulated based on the type of
research and the variables under investigation. They should be formulated to provide
answers to the problems under study. Definitions of Terms: The essence of definition
 is to ensure that the reader understands the specific meanings ascribed to the terms by
the author. So you have to use this to educate the readers on the operational meaning
of any coined, technical words, phrases or expressions which cannot otherwise be
understood because of their unconventional usage.
vii. Definition of terms: In this section, terms or concepts whose meaning are not obvious
in the context of the study are defined or explained.

Chapter 2: Literature Review

i. Review of Related Literature: This is the second chapter of your project report. It is
meant to give the reader an understanding of some of the works or study already carried
out in the area of the study. It will also give the reader an overall picture of the problem
you are solving. These is usually organized under relevant sub - heading

Chapter 3: Research Methodology

i. Research design: This lays out the master-plan for the research project. what is require
here is to specify the type of the design followed in the study. You should therefore
describe any plan used clearly, even if it cannot be classified under a conventional label.
All lapses should be reported as a limitation.
ii. The Population: In this case, you need to specify the aggregate of items or persons
from which the data were actually collected. You should specify all the necessary
parameters to ensure that the constituents and characteristics of the target population
are unambiguous. The target population may be people, animals, objects or events.
iii. Samples and Sampling techniques: The sample (that is, that portion of the population
from which the data were actually collected) and sampling techniques should be so
described in such a way as not to leave the reader in doubt about what you have done.
Specify the method in which the simple random sampling was used.
iv. Instrumentation: This involves a description of the instrument(s) used in collecting
data. Such instruments like questionnaire, attitude scales, tests, etc. should be fully
described to show their characteristics. You will have to report the reliability indices
and validation procedures. Where you used a standard instrument, in your report, you
have to give the rationale for the appropriateness. Where a new instrument is
 developed, you have to outline the necessary procedures followed in both the
construction and validation.
v. Data Collection: This involves the reporting of the steps taken to collect the pertinent
data. What methods did you use in your data collection? Did you use research
assistants? If yes, did they undergo training? Did you collect the data personally, or by
post? What problems did you encounter in the process of data collection? All the steps
which you have taken to ensure the collection of valid that should be reported.
vi. Methods of data analysis: In this section, you will describe the techniques which you
applied in the data analysis and the reasons for the choice. The reasons may be in
relation to the type of design, nature of the samples on the type of data. Try to use the
simplest, well-known method of data analysis. But where you use a mode of analysis
not widely known details of such method should be reported.

Chapter 4: Result and Discussion

i. Presentation and Analysis of data: This chapter expresses the results of findings of
the study. Tables, figures, graphs and textual descriptions are used to clarify significant
relationships. They should be serially numbered and titled so as to be self - explanatory.
They should be simple and should be directly related to the hypotheses and/or the
research questions.
ii. Interpretation of the finding: The most important task which you have to undertake
in writing the results of your study is to identify and interpret the major findings. You
should be able to discuss possible reasons why the results occurred the way they did.
You should try to fit them into the findings of previous research, suggest the
applications to the field and make theoretical interpretations.

Chapter 5: Summary and Conclusions

i. The Summary: In this section, you should provide a summary of the entire work, that
is a brief account of the entire work. The summary must be very brief, but consistent
with a clear presentation of all important information about the problem, method and
findings.
 ii. The Conclusion: The major finding that emerges from the study should constitute the
conclusion.
iii. Implication of the study: In this section, you may include ideas on the relevance of
the findings to educational theory and practice. But these ideas should be directly
generated from the study.
iv. Suggestions for further study: it is believed that solving one problem gives rise to
other problems. It is therefore expected that you should state those problems areas
encountered during the cause of the study for further investigation.

Supplementary Pages
i. Bibliography/ References: This is considered part of the main body; it is a list of the
sources which were made use of in preparing the study. The difference between
bibliography and references is that a bibliography consists of a list of all sources that
cited in the report and those not cited, but consulted to shed light on the problem, while
a reference is the list of sources that were only cited in the problem. Most institutions
in Nigeria adopt the APA format. References are done serially and alphabetically.
ii. The Appendices: This contains extra information which is part of the report the reader
should know about, but not necessarily for inclusion in the main report. They include
instruments used, raw data, instruction aids, items analysis data, scoring protocols and
procedures, lengthy quotations etc.

Exercise
1. List and explain the elements that constitute the preliminaries in a research project
2. Go to institution`s library and select three different research projects. List the items on the
table of content and compare them.
3. What are the major sections of the main body of research project report?