NATURE AND FUNCTION OF RESEARCH

Imagine what the whole world would have been without research. Life will not only be
boring but short lived, static, uninspiring and perhaps difficult to live. The whole world
would have been divided of the spices, innovation and developments that awakening us each
day.

The contemporary world, it is unthinkable that any organization, institution or individual

would presumed or assumed that vital issues or decision can be made by rule of thumb or
virtually through guesswork hunches. Human mind has a quest for knowledge and a thirst to
explore and besides, an inordinate passion and desire to move his standard of living compels
him to undergo researches.

Man quest for knowledge is not without a price, his interaction with his environment makes
the world he lives to be intrinsically complex. The complexity of the world system and the
magnitude of the problems confronting mankind ranges from globalization’ economic
recession, crisis and adjustments, wars and conflicts, global warming, environmental
degradation, crime, food shortage, flood, urban slums, HIV/AIDS pandemics etc. These are
so enormous that systematic procedure is required in order to ensure that decisions are made
as much as possible with all the facts and statistics available. Hence, scientific research is
needed as a vital tool in advancing knowledge, for promoting human progress to explore
effectively and efficiently with the resources he is endowed with and to interact more
successfully with his environment in order to accomplish his desires, purposes, satisfactions
and to resolve his conflicts.

MEANING OF RESEARCH

 Research is an investigation undertaken in order to discover new facts or get

information about an existing fact that is related to livelihood for mankind and
learning (Mogbo, 2003).
 Research connotes a systematic and objectives search for and analysis of information
relevant to the identification and solution of any problem in any field of human
endeavour (Green et al, 1992).

Research Methodology Page1

  Research is an organized inquiry that aims at providing information for solving
identified problems. Here, research is seen as organized process of enquiry aimed at
obtaining information that is relevant to the solution of an identified problem or
problems.
 Research is a process of arriving at dependable solutions to problems through a
planned and systematic collection, presentation analysis and interpretation of data.
(Osuala, 1993).
 Research is the systematic and objective recording and analysis of controlled
observation that may lead to the development of generalization, principles or theories,
prediction and possibly ultimately control of events.
 Research is a systematic investigation into and study of materials and sources in order
to establish facts and reach conclusions.
 Research is a methodological search for new knowledge and practical solution in
form of answers.
 Research is a systematic inquiry to describe explain, predict and control the observed
phenomenon.
 Research is a general term which covers all kinds of studies designed to find
responses to worthwhile questions by means of systematic and scientific approach.

From these definitions above, the following are obvious as the characteristics of research

1. Research is a process and not just once and for all event or one-time event
2. It is an organized inquiry or investigation
3. It involves a search for information
4. It is the aim of research to solve the identified problem
5. It is a voyage that leads to discovery
6. Research is based upon accurate, observable experience and descriptions
7. It is conducted in a systematic and objective way
8. Research emphasizes the development of generalizations, principles or theories that
will be helpful in predicting future occurrenc

Research Methodology Page2

THE CONCEPT OF SCIENTIFIC RESEARCH

Research has to do with enquiry into the unknown; it is the search of or for solutions to
problems or answers to questions. When research employs a systematic approach i.e
scientific method, it is said to be scientific research. However, not every research qualifies to
be scientific, it must possess certain characteristics. All scientific investigations are;

1. Empirical: conclusions are not based on what the author feels or thinks, but on
concrete evidence from data collected.
2. Theoretical: This aims at producing a set of interrelated prepositions which can
explain certain phenomena among variables.
3. Cumulative: This builds upon existing facts and theories.
4. Non-ethical: It does not seek answers to questions such as whether an action is right
or wrong, rather, attempts to find the explanation for any action.
5. Verifiable: The process adopted in scientific investigation is such that it can be
replicated by other researchers with approximately the same results.

SCIENTIFIC RESEARCH AND COMMON SENSE

According to whitehead “in creative thought, common sense is a bad master, it sole criterion
for judgement is that the new idea will look like old ones” common sense and science differ
in the following ways;

1. The use of conceptual schemes and theoretical structures are strikingly different.
While the man in the street uses “theories and concepts’’ he ordinarily does so in a
loose fashion. He often blindly accepts fanciful explanations of natural and human
phenomena. An illness, for example may be thought to be a punishment for an evil
deed. An economic slump may be attributed to the gods. The scientists, on the other
hand; systematically builds his theoretical structures, test them for internal
consistency, and subject aspect of them to empirical test. He realizes that concepts he
is using are man-made terms that may or may not exhibit a close relationship to
reality.

Research Methodology Page3

 2. The scientist systematically and empirically tests the theories and hypothesis. The
man in the street tests his “hypothesis” too, but he tests them in what might be
regarded as a selective fashion. He invariably selects evidence simply because it is
consistence with his hypotheses. For example “the place of the woman is, in the
kitchen”. If an individual believes that, he can easily verify his belief by observing
that many women prepare meals for their husbands. Exceptions to this belief in which
some women are career women is not perceived. The sophisticated social scientists,
knowing this “selection” to be a common psychological phenomena, carefully guards
his research against his own pre conceptions and hypothesis.
3. In scientific research, control means several things. The scientists tries systematically
to rule out variables that are possible “cause” of the facts he is studying other than the
variables that he hypothesized to be “causes”. The layman seldom bothers to control
his explanations of observed phenomena in a systematic manner. He ordinarily makes
little effort to accept those explanations that are in agreement with his preconceptions
and biases.
4. The scientist consciously and systematically pursues relationships. The layman’s pre
occupation with relationship is loose, unsystematic and uncontrolled. He often sizes,
for instance on the fortuitous occurrence of two phenomena and immediately links
them indissolubly as causes and effect.

METHODS OF GAINING KNOWLEDGE

The means by which man seeks the answers to his problems can be classified under four
broad categories: Tenacity, Authority, A priori and science. These four categories are, of
course complimentary and overlapping.

Research Methodology Page4

 1. The method of Tenacity: Here, men hold firmly to the truth that, they know to be
true because they hold firmly to it, because they have always known it to be true.
Frequent repetition of such “truth” seems to enhance their validity.
2. The method of Authority: This is the method of establishing beliefs. If the Bible or
Quran say so, it is so. If the mother says it, it is true. This method is superior to the
method of tenacity because human progress, although slow, can achieved using this
method could not in fact go on without the method of authority. Thus, it should not be
concluded that the method of authority is "unsound" it is only unsound under certain
circumstances.
3. The "A” priori method: This method rests its case for superiority on the assumption
that, the propositions accepted by the a prior are self-evidence. A prior propositions
usually “agree with reason” and not necessarily with experience. The idea seems to be
that men, by free communication and intercourse, can reach the truth because their
natural inclinations tend toward truth, the difficulty with this rationalistic position,
however lies in the expression "agree with reason". The question is whose reason?
4. The method of science: This method had self-correction aspects which no other
methods of attaining knowledge has. There are built-in- checks all along the way of
this method; These checks are so conceived and used that they control and verify the
scientist’s activities; and conclusions to the end of attaining dependable knowledge
outside himself. A scientist does not accept a statement as true even though the
evidence at first looks like promising, he insists upon testing it.
The checks used in Scientific research anchored as much as possible in reality outside the
scientist and personal belief, biases, values, attitudes and emotions. The best single word to
express this is objectivity.

Research Methodology Page5

CLASSIFICATION OF RESEARCH BY PURPOSE

BASIC RESEARCH

Basic research which is also known as pure or fundamental research is aimed at discovering
basic truth or basic principles. It is under taken mainly for the expansion of knowledge and
not direct practical utilization. It is the type of research in which we see testing and
development of theory, and it may or may not be immediately concerned with direct field
application of the developed theory.

APPLIED RESEARCH

Applied or field research is concerned primarily with establishing relationships and testing
theories in the field setting e.g research that seeks to find out the relationship between
poverty and family size.

ACTION RESEARCH

Action research focuses on solution to problems in a particular setting. Put differently action
research places it emphasis on a problem here and now. In an ideal setting its funding are to
be evaluated in terms of local applicability not universal validity e.g problem of
environmental degradation in Niger area in Nigeria.

RESEARCH AND DEVELOPMENT

This is a relatively new type of research. Research and development program aim at
developing and testing more efficacious products. These products are developed and trial
tested in the field to ensure their effectiveness.

EVALUATION RESEARCH

This is concerned making decisions relating to the value or worth of materials, methods and
programs 'based on empirical data. It is a systematic process of collecting pertinent data on
basis of which decisions could be made about an individual, a program, materials or
methods. Evaluation research often provides useful information on the effectiveness or
operation of a system.

Research Methodology Page6

CLASSIFICATION OF RESEARCH BY METHOD DESCRIPTIVE OR NON
EXPERIMENTAL RESEARCH

Descriptive research describes and it interprets what is. It seeks to find out the conditions or
relationships that exists, opinions that are held, processes that are going on, effects that are
evident or trends that are developing.

Descriptive research can further be classified as follows:

Survey Research: This studies both large and small populations by selecting and studying
samples chosen from the population to discover the relative incidence, distributions and
interrelations of variables. The survey researchers are interested in the accurate assessment of
the characteristics of whole populations, from those samples the researcher infer the
characteristics of the target population.

Survey researches can be conveniently classified lay the following methods of obtaining
information' [1] interview methods [2] observation methods [3] Telephone [4]. Questionnaire
methods: [5] focus group discussion.

Correlational studies or study: It seeks to establish what relationship exists between two or
more variables. The degree of relationship is expressed as a co relational study only indicates
if a relationship exist or not but does not indicate causation.

Case studies: -This singles out individuals, group of individuals, institutions, communities
etc. for studies. Usually case studies are motivated by problems emanating from the case or
cases being studied.

Ex-Post-Facto Research: This research seeks to find out the factors that are associated with
certain occurrence, outcomes conditions or types of behavior by analysis of past events or of
already existing conditions. In this type of research, the researcher has no control over
certain factors or variables. He cannot manipulate the variables because they already exist
and cannot be changed, for; It is unethical to do so.

Research Methodology Page7

Developmental Research: Development research involves investigations of pattern and
sequences of growth or changes that take place as a function of time. It seeks to find out how
some dimensions, variables or characteristics of a given population change with time e.g.
development of mental ability in children.

A developmental research could either be longitudinal or cross-sectional. In longitudinal

studies, the same group or subjects are studies over a given period of time. Within this period
the subjects are observed periodically for changes in characteristics. In cross- sectional
studies, large group of people or subjects are assessed on a particular variable at a time rather
than looking at how they change on the selected variable over a long period of time.

Historical Research: This interprets past trends attitude, events and facts. More
emphatically, historical research may be considered as embracing the whole field of human
past abroad as life itself. Although the data must be viewed with historical perspective as part
of the process of social development rather than as isolated attitude, events or facts, e.g.
“Resource control in Nigeria during military government or regimes”

EXPERIMENTAL RESEARCH

Experimentation is the most scientifically sophisticate research method, it is usually

confused with the scientific method. However, experimentation is only one aspect of the
scientific method which itself involves a great number of activities. The purposes of
experimentation are to derive verified functional relationships among phenomena under
control.

IMPORTANCE OF RESEARCH

1. Research is very important in all forms of human endeavor. It is applicable to almost

all the fields of learning. Accordingly, we have research in the fields of medicines,
business, education, marketing, advertising, pharmacy, engineering, technology,
agriculture, industrial, environmental, social, economic etc.
2. It helps in discovering the causes of problems in organizations, communities
environments etc. Discovering causes of problems is the first and important step to
problem solving.

Research Methodology Page8

 3. Research provides information or data that can serve as input for decision making and
also enhances the formulation of policy
4. Through research we reduce area of ignorance by discovering new truths which, leads
to better ways of doing better-products better environment, better ways of preventing
and treating .diseases, better understanding human behaviors and better understanding
of the world we living in.
In a nutshell, serves the following purposes research

1. It leads to the advancement knowledge

2. It increases understanding of phenomena
3. It provides solutions to problems
4. It brings about overall development and progress

PROBLEMS OF RESEARCH IN NIGERIA

1. Inadequate funding of research projects

2. Lack of necessary equipment, facilities and materials
3. Unattractive working conditions for research problems
4. The "publish or perish" syndrome
5. Lack of record keeping culture
6. Attitude of the government and society
7. The business factor
8. Illiteracy
9. Security/ inability of researcher to gain access to vital document
THE RESEARCH PROCESS

Research steps or process step to follow carrying out a research. It is the various stages the
researchers must undergo during the investigation before any solution can be proffered to the
identified research problems. Research process provides a systematic plan or procedure for
the researcher.

Research Methodology Page9

 The research process can best be conceptualized in a diagrammatic form as shown below

RESEARCH PROCESS

.....Identifying the Problem

formulate hypothesis
Review Literature
or research question

Design the research i.e sample

size method of data collection,
sampling technique

Data collection
and analysis

Discussion and
generalization

In relation to In relation to
hypothesis or research literature review
question

Suggestions

Research Methodology Page10

As shown above, the first step in the research process is the identification of the problem.
The problem could emanate from; various sources like textbook, research reports, personal
experience, etc. once the problem has been identified, it is described to make it clearer. The
description is usually declarative statement but also be in question form.

The next step is the review of relevance literature. Related literature is the writing of
recognized authorities or previous research relating to the problem of the study. The
statement of hypothesis which is the third step further clarifies the nature of the problem and
the logic underlying the investigation and gives direction to the data gathering process.

The research design is the next step and it usually consists of the description of the subjects,
the procedures for data collection and the method for data analysis. Data are collected
organized, analyzed and presented. Once this is done, the results are interpreted and
generalizations are done in terms of research questions, hypothesis and review of the related
literatures

If the research is conducted properly as planned, the research question should be answered,
the problem solved or at least a first-step taken toward achieving this end. It should however,
be emphasized that the research process described above should not be considered the only
satisfactory sequence or procedure since other; researchers may suggest other acceptable
research process.

Research Methodology Page11

IDENTIFICATION AND SELECTION OF RESEARCH PROBLEMS

The most daunting challenge that student may encounter is the selection of a research topic.
After many effort of exploration, some students may be forced to abandon their topic and
have to begin afresh, while others go on despite the unsuitability of their research problem
and end up having nothing to show up for their work. As a result, the research project/thesis
which ought to be most rewarding experience of the student’s academic life becomes a mere
exercise in futility.

Guidelines for selecting research topic

Although there are no general rules for selecting suitable research topic, however the below
suggestion may be of great assistance in selecting a research topic.

1. The topic must be of interest to the student while interest sometimes develops with
familiarity it does not seem likely that the student can do his best work on a topic that
has no personal meaning to him.
2. The topic should be sufficiently original that it does not involve objectionable
duplication.
3. The topic must be researchable; many problems are of a philosophical nature. They
can be discussed but not to the point where objective evidence can provide a solution.
Thus, the problem “should secondary school girl work? Having no referent is, as
stated a philosophical issue not subjected to scientific determination. Before it could
be investigated, it should have to be suffer academically.
4. The topic must be significant specifically the topic should be capable of adding new
information to the present state of knowledge. There are so many problems that need
to be investigated that it does not make sense to have a student devote himself to the
study of a trivial or nonsensical problem.
5. The research into the problem must be feasible. The student must make sure that data
are available in the situation in which he finds himself. Many significant problems
must be by passed perhaps because they would not get the necessary clearance or

Research Methodology Page12

 because they are not possible from the standpoint of the competence of the
investigator.

TYPES OF RESEARCH FORMAT

The various types of reports commonly written by professionals and business organizations
includes dissertation, articles for publication in journals or in proceedings of conferences,
seminars or workshops, project proposals, feasibility reports, design reports, tender reports, -
operations and maintenance manuals, progress report, environmental impact assessment
reports, etc

DISSERTATIONS

A dissertation is a written document embodying results of original research with a view to

advancing a position or proposition or substantiating a specific view. Such a document is
also often referred to as thesis may be written by a candidate for an academic degrees e.g.
Msc. Thesis.

ARTICLES FOR PUBLICATION

The author of a dissertation may decide to publish the work. Similarly, a person who has
conducted an original research may decide to publish the results. The works are written out
as articles for publication in a journal. If the article is presented at a conference, seminar or
workshop, it may be published along with others as part of the proceedings of the
conference.

PROJECT PROPOSALS/RESEARCH PROPOSAL

Project proposal are written to put forward comprehensive suggestions for consideration, e.g.
a proposal, prepared by a candidate for a; higher academic degree, such as HND; BSc, MSc,
PhD and so on. It may be a proposal prepared by a consulting engineering firm for
Electrification of building constructed in a specific institution or community or a proposal,
by an Estate Management firm on a construction of Estate Building in newly located city or
town.

Research Methodology Page13

A research proposal is an account of what a researcher intends to do and how he intends to
do this. It is necessary, for students to presents a proposal to their supervisors, or departments
before they can be allowed to go ahead with detailed proposals of their research work.

FEASIBILITY REPORTS

Following acceptance of a report proposal; the client (an institute, a community, a company
or an Individual) may order a feasibility study. Key elements to the preparation of a
feasibility report are an understanding and clear expression of the f problem, consideration of
alternatives and making a decision based on correct analysis.

ESSENTIAL PARTS OF RESEARCH REPORTS Preliminaries:

[Link] page [Link] page [Link] or approval page [Link]

[Link] vi. Abstract vii. Table of contents.

CHAPTER ONE – INTRODUCTION

1.1 background of the study or general introduction

1.2 statement of the problem
1.3 aim and objective of the study
1.4 importance of the study
1.5 statement of hypothesis
1.6 delimitations and limitation
1.7 definition of terms
CHAPTER TWO – LITERATURE REVIEW

CHAPTER THREE – RESEARCH METHODOLOGY

3.1 Research variables

3.2 Population and sampling procedure

3.3 Data collection

3.4 method of data analysis

Research Methodology Page14

CHAPTER FOUR – Data Analysis

CHAPTER FIVE –Summary And Conclusion

5.1 Discussion and Implication of the study

5.2 Summary and conclusion

5.3 Recommendations

5.4 Suggestion for further research

References

Appendices

ABSTRACT

An abstract is a summary of the research report consisting of not more than 250 - 400 words.
It should-contain in summary form the objective of the study, characteristics of the subject,
the method used, the findings of the study and the conclusions reaches by the researcher.

Research Methodology Page15

CHAPTER ONE

BACKGROUND OF THE STUDY

In the background of the study, the researcher explain clearly how he became interested in
the problem, the circumstances surrounding the problem and why he felt the study is worth
pursuing. Here, the problem is articulated and brought to focus in such a way that whoever
reads through would have had an idea of what the research problem is all about. This section
thus provide the necessary information needed for a clearer understanding of the research
problem. After giving the general background of the study, the investigator carefully
establishes the need for the study, telling why the information resulting from the study is
needed.

STATEMENT OF PROBLEM

The statement of the problem serves to elaborates upon the information provided in the title
of the study. It is simply sentences which announces the existence of a problems that
necessitate the need for the investigation. By this, the researcher is able to convince the
reader that the area chosen merits research attention. In stating this section, the researcher
should be able to isolate, clarify and establish specific focus that clearly shows the direction
of the study. Statement of the problem may be in the form of clear and precise problem
statements or may sometimes be blended with some bothering questions called research
questions which reveals assumed relationship between two more variables.

AIM AND OBJECTIVES

This is sometimes referred to as the purpose of the study. This section has to do with the
researcher's intention. A problem may arise here if the statement of the problem is not well-
stated. Every study needs to be objective-oriented. Such objectives must be clearly stated in
such a way that no doubt is left about the goals of the study. These goals are usually designed
to show the researchers intention to give answers to the research problem under study. The
goals are either expressed in a paragraph of statements of objectives, or by itemizing specific
objective which the investigator intends achieve. The objective or purpose, of the study must

Research Methodology Page16

have relevance to the questions raised by the problem of the study, which also provide
direction for the study.

IMPORTANCE OF THE STUDY

Importance of the study otherwise known as significance of the study attempts to show the
potential relevance of the study to the public or end users. In other word, it tells for whom
the study will be meaningful and give some indication of its value for others. In doing this, a
researcher is expected to state clearly the importance of the study and why he thinks his
research problems worth investigation.

STATEMENT OF HYPOTHESIS

A statement of hypothesis .is a tentative statement, a proposed explanation of a phenomenon

which still has to be rigorous tested. It is a formal affirmative statement predicting a single
research outcome. It is logical step taking toward solving a research problem. A researcher
after conceiving a problem, makes guesses by stating his ideas about the causes and solution
to the problem under declarative statements of assumptions or calculated guesses held by the
researcher, which serves as a tentative answer to the problem under investigation. They give
suggested causes of the problem, which imply suggested solutions as well. Hypothesis have
therefore become a tentative proposal for the explanation of phenomenon or situation that
has some degree of empirical substantiation or probability which is subject to proof by the
investigator. A hypothesis is also viewed as an explanation of the relationship between two
or more variables.

The importance of the role of the hypothesis in research cannot be overemphasized. The
hypothesis is in a very real sense the core of the study. It guides the researcher in planning
the course of the inquiry in choosing the kind of data needed, in deciding the proper
statistical treatment and in examining the results of the study. However, not all research work
required hypothesis. They should be incorporated if statistical testing is to be used.

Research Methodology Page17

TYPES OF HYPOTHESES

There are basically two major types of hypothesis namely the null hypothesis and the
alternative hypothesis. This alterative hypothesis are the direct positive assumptions with
respect to the cause and causes and solutions to the problem under investigation while the.
null hypothesis refute the assumptions asserted in the alternative hypothesis.

The null hypothesis is the one being tested and is a useful tool in testing the significant
difference. The null hypothesis is akin to the legal principle that a man is innocent until he Is
proven guilty. It constitutes a challenge, and the function of an experiment is to give facts a
chance to refute or fail this challenge. As earlier said, the hypothesis that is being tested is
the null hypothesis and is denoted by Ho: while the hypothesis we are willing to accept if we
reject the null hypothesis is called the alternative hypothesis denoted by H 1.

DELIMITIATIONS AND LIMITATIONS

Delimitations are those restrictions imposed on the study by investigator, for example the
variables of problems that could hinder a student from putting in maximum effort to the
cause of his/her research work. Therefore, evidences such as financial constraints, time
factor, facilities and other academic pursuits including socio-economic problems form
aspects of the delimitation of study.

Limitation refers to the scope of study. The target audience age group, sample size, types of
instrument, to use, etc. it is believed that those variables may impose limitations which could
guide the making of generalization.

In nutshell, delimitation are those restrictions deliberately imposed by the researcher himself
while limitation refers to those limitations to those limiting conditions or the constraints
suffered during the course of the study. Limitations are not the researchers deliberated
actions but may be inherent in the research work. Examples are limitation on questionnaires
no response or truthfulness in response.

Research Methodology Page18

DEFINITION OF TERMS

This simply the explanation of those words or terms that may be ambiguous in meaning to
other readers. The aim of this section is to allow the researcher to define some keywords,
concepts or construct as well as technical terms used in the work. The purpose of definition
of terms is to achieve a good understanding of the expressions contained in the report in
order to avoid misinterpretation. The definition of terms is a most important facet of the
research report. Since without common understanding of essential terms, the value of the
research may be loss through misinterpretation.

CHAPTER TWO

LITERATURE REVIEW

Literature review is a summary of the writing of recognized authorities and of previous

research problem one intends to investigate. This exercise provides a theoretical backing of/
to the study.

The review of related literature should not be composed merely of bibliographical summaries
of quotations, nor should be merely a series of quotations connected by a few words.
Quotations should be used when the individual whose work is being reviewed has uniquely
expressed an idea that no other wording would convey. Paraphrasing is much preferred to
paragraph after paragraph of quotation.

Paraphrasing means that the researcher reorders the original sentence by using his own words
to write what someone has said or written without losing the original meaning.

IMPORTANCE OF LITERATURE REVIEW

1. it provides a theoretical background to the study

2. It enables the researcher to familiarize himself with previous researches and works in
the area of investigation.
3. Literature review enables the researcher to know how others have approached similar
problem in the past.
4. It sharpens the researcher’s understanding the problem under study

Research Methodology Page19

 5. Literature review helps the researcher to establish the current situation and opinion
about the research problem.
6. Through literature review the researcher discovers the gaps or missing links that
needs to be filled and this enriches the body of knowledge.
7. By reviewing the work of others and knowing what others have done the researcher
does not all fall a victim of duplicating of what has been done already.
8. It gives the research insight about the appropriate research methods and design to use.
9. It helps avoid plagiarism

LITERATURE CITING

Literature citing (referencing) is the practice of directing a reader to the source of

information being used, quoted or paraphrased in a piece of writing. The source may be
journals, books or personal communication. These are cited as source(s)

There are two main purposes for citing references, namely;

THE NUMBERING SYSTEM

This system has been referred to as the Kate Turabian Style, also known as the Chicago
style. It is a system where numbers are used to indicate citations in the text. The oldest form
of this is the “footnotes only” where the researcher is expected to place the source cited at the
end (foot) of the page where the citations occurred, and nothing more is added. Yet an
improved form requires that after the citations in the text and the corresponding footnotes,
the writer goes further to prepare a list of references for all the sources cited and footnoted at
the end, of the work in tine order of citations. Example: (1) Ujah et al noted that while 25%
of Females of reproductive age lived in developed countries, they contributed only one
percent to maternal deaths worldwide, while the remaining 99% of the deaths occur in
developing countries (11) Krugman et al 2 logically examine the direct of currency exchange
rates on international trade.

It will however be noted that this method becoming obsolete and no longer in vogue

Research Methodology Page20

AUTHOR-AND-YEAR METHOD

In this method, citing is done with the surname of the author and the year of publication e.g
Agunloso (2006) or Agunloso & Akinwumi (2012). The citing could be located at the
beginning of a sentence, at the end or any suitable place in between. The number of authors
involved in the publication being cited is also a factor example

(a) Sheriki (2002) states that deregulation of the oil sector would enhance competition and
thereby makes petroleum products more available.

(b) According to Nda-Umar 91999) Statistics in Nigeria is tantamount to lies and

misinformation

(c) Cholera is in some ways similar to Typhoid, but is causative bacteria are more fragile and
the clinical course is extremely dramatic (Adeyemo et al 1980)

Sometimes, two or more articles are written by an author in one year. To distinguish between
the articles, small case alphabets a,b,c etc. are placed after the year. This is demonstrated in
the examples bellows:

Fundamental steps that Africa and its people must take, both nationally and regionally, to
ensure that remote sensing technologies become catalysts to the development and economic
growth of the continent have been adequately specified (Abiodun, 1993a, Abiodun 1993b)

LISTING OF REFERENCES

One of the aims of Indicating the source(s) of information used in a write-up as already said
is to afford the reader the opportunity to get to the original source of the Information so that
he, can make an independent check or get more information. It is therefore not sufficient to
cite references; the references cited should be listed. In listing, sufficient information should
be given so that the interested reader has a reasonable chance of findings the referenced
material the chooses to obtain it.

The following table provides a summary, of the type of information that should be given
about the different sources; journal books, edited books and published theses.

Research Methodology Page21

 Sources Information given in listing
Journal Surname(s) and initials of all authors of the
articles; year of publication, title of the
article; name of journal; volume; number
and pages.
Book Surname(s) and initials of all the authors of
the book, year of publication (i.e year of
the:
. latest edition as indicated on the copyright
Page); title of the book; publisher city.
Dissertation/thesis Surname and initials of the author, year
indicated: title. of these, followed by the
word; ‘unpublished PhD)Msc, or Bsc HND
etc thesis; Name of University; city
Edited book (containing articles Witten Surnames and initials of all the authors of
by different authors) the particular article; year of publication of
the book, title of the article; the word “in”
followed by the name of the book;
surname(s) and initials of editor followed
by “ed”; publisher city
Internet Surname of the author ot the article (if it is
credited to an individual); The title of the
article and the year with address i.e
[Link].

The citation formats generally agreed upon and are still in vogue is that of APA style known
as American Psychological Association.

Example of citation using APA style (Book)

1. Osuala E. C. (2005) Introduction to Research methodology Onitsha; Africana-first

publishers limited
2. Evborokhai, J. A.; et al (2006) Business policy and strategy. Ibadan; Oniboje press

Journal

Aluya, M. S. (2001) Marketing challenges in a depressed journal of marketing management,

2, 6, 8

Article in a Newspaper

Research Methodology Page22

Azubike O. (2008, May 20). Deregulation will bring hardship, people’s daily p4

(Assignment: students should look for books, Article, journal, etc and practice referencing.
Also how would you reference article down loaded or copied from the internet).

CHAPTER THREE

RESEARCH DESIGN AND METHODLOGY

According to green (1987), it is the specification of method sand procedures for acquiring the
information need to solve the problem under investigation. It is the overall plan of the work
that stipulates what information is to be collected, from which source and by what procedure.
Data collection must be planned; what data is needed i.e the type; what instrument to use;
from what source as well as how to obtain the needed data. A good design therefore will
ensure that the information obtained in relevant to the research problem and that it is
collected by an objectives procedure. Although research design may be classified by many
criteria what dictates which design is applicable or appreciate for any investigation is the
nature of the research problem.

RESEARCH VARIABLES

Research variables are the conditions or characteristics that the researcher manipulates,
controls or observes to obtain data. Variables may be classified into two major types;
independent and dependent variables.

An independent variable is a type of variable that the researcher manipulates or controls in

order to ascertain its relationship to observed phenomenon. On the other hand, a dependent
variable is a response variable of interest which response to the variability in the explanatory
variable x.

POPULATION AND SAMPLING PROCEDURE

The word ’’POPULATION” in research may include tangible and intangible objects.

It is different from the meaning given to it as it is viewed or used in the case of population
census where population simply means a group of people in a given geographical area. The

Research Methodology Page23

population under study may be group of persons, organization, events, animals, etc. In
conducting a research, the researcher has to decide whether to study the entire population or
to study only a certain portion of the population. If a researcher chooses to look into the
effect of strike action on Academic programme in Nigeria Polytechnic, the population of
study would be the entire staff of polytechnic. Every research report must describe with
precision the population that is involved in the study. The characteristic features of a
population must be properly described and understand by the researcher as it may enable,
him discover certain variables which might, be associated with the dependent variable or the
research problem.

Apart from this, understanding the characteristic features of the population guides the
investigator in the election of sample by ensuring that the features inherent in the population
are well represented in the chosen sample. This is important for the purpose of
generalization.

Sample is a part of the population that is being investigated. It is a representative portion of

the of the population. This population is the unit of study or analysis which every researcher,
must determine before, drawing the sample from it. The unit of study is known as a case
hence the term, case study.

DETERMINATION OF SAMPLE SIZE

This is simply the determination of the numerical value of the sample to be selected from the
study. The three factors that determine the size of an adequate sample are; the nature of the
population, the type of the investigation and the degree of precision desired. The
determination of sample size as well as the selection of the sample size as well as the
selection of the sample is easier from a homogenous population than from a heterogeneous
population. If the desired precision is high, then the sample should be highly adequate
numerically.

All things being equal, the higher the sample size, the higher the degree of precision. The
most common method of determining sample size is the use of percentages. Here, the
researcher decides on a particular percentage of the population he wishes to use as his

Research Methodology Page24

sample, ensuring that it is numerically adequate. This means that a researcher can from 40%
or 50% of the population for example. In research any sample size from 40% of the
population is acceptable. This is implied from Taname (1974) who provided a formula for
determining sample size from a heterogeneous population.

The formula is stated below

N = N
1+Nee
Where n = sample size sought

N = population size

I = is a constant

With a chosen level of significance population of one thousand (1000) €, of 50% for
example, and from a heterogeneous population an adequate sample size can be obtained by
using this formula.

SAMPLING AND SAMPLING METHODS

Sampling is the process of selecting elements to be included in a sample for use in a sample
survey.

It is one thing to determine sample size. It is another to select the sample from the population
using appropriate method. The act of drawing (selecting) samples from a population is called
sampling method. The researcher must specify clearly the method used in drawing the
sample form the population. The applicable sampling method depends largely on the nature
of the population. For instance, the method employed to obtain a sample from a
homogeneous population cannot be entirely applied to a heterogeneous population. The
method of selecting sampling are of two basic group, namely; random and non-random
sampling.

Research Methodology Page25

A RANDOM SAMPLING

It is a sampling technique which gives every element in the population an equal chance of
being included in the sample, it includes the following

(1) Simple Random Sampling

This refers to' random sampling. techniques in which each element in the population has. an
equal chance of being included in the sample. The sample elements are selected by balloting,
tossing coins or by use of a table of random number.

It is generally accepted that the best method of drawing a sample random sample is by means
of random sampling numbers for example if an auditor wishes to sample 29 movies put a
total of 583, received in a financial year. The procedure that could be followed is Each;
movie would be numbered, from 001 through 58.3. Then select a starting row or. column
from a table of random sampling numbers and begin reading groups of three digits
sequentially, (from random sampling table we have 34819, 80011,: 17751, 032.75 etc then
the movie number 348, 198, 001,117, 032 etc while number 751 would-be rejected as too
large.)

The advantages of this method of sampling include the selection of sample members being
unbiased and the general acceptance by the' layman-that the method id fair.

(2) Stratified Random Sampling

This extends the idea of sample random sampling to ensure that a heterogeneous population
has its definite strata levels taken account of in an independent simple random sample of the
assigned sample elements are drawn and these are pooled together to obtain the total sample.

The general procedure for taking a stratified sample is

a. Stratify the population defining a number of separate partitions

b. Calculate the proportion of the population lying in each partition
c. Split the total sample size up into the above proportion
d. Trace a separate sample [normally simple random) from each partition, using the
sample sizes as defined-in (c)
Research Methodology Page26
 e. Combine the results to obtain the required stratified sample
Advantages of this method of sampling is the fact that the sample itself (as well as the,
method, of selection) is free from bias, since it takes into account significant strata levels
(attributes of a population under considered important to the investigation.

(3) Systematic Sampling

This is a method of sampling that can be used where the population is listed (such as the total
number of students Federal Polytechnic Bida) or some of it is physically in evidence (such as
a row of hoses, items coming off a production line or customers leaving a supermarket). The
technique is to choose a random starting place and then systematically sample every nth item
in the population. The nth item chosen is based on the size of sample required. Systematic
sampling is particularly useful for populations that (with respect to the investigation to hand)
are of the same kind or are uniform. These are referred to as homogeneous populations.

(4) Cluster and Multi-stage sampling.

Where a population is spread over a relatively wide geographical area, random sampling will
almost certainly entail traveling to all parts of the area and thus could be prohibitively
expensive. Multi-stage sampling which is intended to over-come this particular problem,
involves the following

(a) Splitting the area up into a number of regimes

(b) Randomly selecting a small number of the region

(c) Confining sub-samples to these regions alone, with the size of each sub sample
proportional to the size of the area. Once the final sub samples have been selected, the final
sampling technique could be simple or stratified random or systematic depending in the
existence or otherwise of a sampling frame.

NON-RANDOM SAMPLING

This is sampling methods that do not rely on scientific method but based on the use of the
investigators judgment of specific criteria., as against chance in determining the elements to
be included in the sample. This method includes:
Research Methodology Page27
Judgement Sampling

This technique involves investigator’s judgement or specified criteria as against chance in

determining the elements to be included in a sample.

Quota Sampling

This is a method in which a quota is assigned to each group of the population but the
selection of each group’s quota may, pot be at random. The investigator is at liberty to
merely handpick any element to be included based on his judgment.

Convenience Sampling

In this method, the investigator selects elements that are readily available or that are
convenient to use. This is most common is televised market survey.

METHOD OF COLLECTING DATA

Data collection can be thought of as the means by which information is obtained from the
selected subject of an investigation'. There are various data collection methods which can be
employed. Sometimes a sampling technique will dictate which method is used and in other
cases there will be a choice depending on how time and man power (and inevitably money)
is available. The following list gives the most common methods.

Experimental method: refer to data collection through tests and direct measurement of
variables. Experimental dominates most of the data collection exercise in basic and medical
sciences. It tends to generate accurate and reliable, information, but may be time consuming
expensive and may be difficult be use when the population is large.

Observation Method: This is the method used when the investigator visit the location of the
event or individual being selected, and directly observe and record data on the required
variables. Observation method is commonly used in studies of consumer, movement in
stores, banks and other service stations for studying traffics on a road and for studying
certain pattern of behaviour. It may produce accurate and reliable, information eliminates
subjective bias and reflects current happenings, However, it. is. Usually expensive while
information provided is limited.
Research Methodology Page28
Interview Method: An interview is a data collection exercise involving direct oral
communication between, the. investigator and respondents either face - to - face or even a
telephone system. The investigator must call an Interviewer who will record the response
with pen on paper, or with a tape recorder. Interview conducted face to face are called
personal interview while those conducted over a telephone system are called telephone
interview.

Telephone Interview: is flexible, faster and cheaper than any other method.

The response is high and access can sometimes be gained to respondent who would neither
be interviewed personally nor 'answer ‘questionnaire. It is the most useful for radio and
television survey. However little home is given to respondents to consider answer and fit is
restricted to respondents who have telephone. Also a number of question that can be asked
from respondent are. usually limited.

Personal interview is also flexible and provide opportunity to restructure questions.

Interviewer can physically cross check response and note reactions to question. Personal data
regarding respondent can be more easily obtained- However, it is expenses, time consuming
and may not be suitable for large sample scattered geographically, and certain types of
people such as busy executives and people in high income group may be difficult to
interview.

Questionnaire Method: This refers to the process of collecting data by sending a set of
question printed in paper called "questionnaire" to the respondent either by hand or by post.
The respondents fill in his response into space provided or the questionnaire for that purpose,
and returned it to the investigator thereafter.

Questionnaire method is by far the most popular and widely used survey involving
questionnaire that are administered through the postal system are caller postal inquiry.
Questionnaire method is. relatively cheap, free from interview bias, suitable for busy-
executives that are difficult to reach through interview and given enough time for the
respondents to answer while the respondent’s identity remains anonymous. It is suitable

Research Methodology Page29

where the distance between investigator and respondent is long and could be used to each a
large sample dispersed over wide geographical area in good time.

However, it not flexible and not suitable when dealing with illiterates and the lateof response
may be low. Problem of non - response may respondent do not understand the question.

FACTOR TO BE CONSIDERED IN QUESTIONNAIRE DESIGN

i. Number of questions: Questionnaires must not be unnecessary lengthy. The fewer

the number, the better. Because respondents tends to show apathy toward lengthy
questioning. Its result only relevant questions should be Included in the questionnaire.
ii. Types of question: Factual questions are most preferred if-we are to include memory
questions. We must take into consideration the time lag between the event occurred
and when the question is. asked because people tends to forget events with time. To
quicken their response. We may provide options as in objective test
iii. Order of presentation: Questions must be asked in a logical sequence. Simple
questions should however, come before more difficult ones, while questions that are
likely to be refused should be reserved for later part of the questionnaire.
iv. Wording: Technical terms vogue generalizations and expressions capable of different
interpretation should be avoided. Words that are used in an emotional or abusive way
as well as leading question should be avoided.
v. Simplicity and Ease: of Analysis: The overall design of a questionnaire should be
simple and easy to analyse Question may be structured to facilitate machine
processing.
vi. Needs for cross checking the validity of response: some dummy questions may be
introduced for the purpose of checking the validity of response. Response, to a
question on income may be cross checked with another on expenditure.
vii. The target population: The population of interest must be borne in mind when
designing a questionnaire. Their language, literacy level, religion believes, etc and
how these-may influence peoples response to questions.
viii. Time and cost consideration especially as they affect resources and time available to
the investigator must also be put in mind when designing questionnaires.

Research Methodology Page30

METHODS OF DATA ANALYSIS

Analysis means the categorizing, ordering, manipulating and summarizing of data to obtain
answers to research questions. The purpose of analysis is to reduce data to intelligible and
interpretable form so that the relations of research problem can be studied and tested.

Many statistical tools and technique are designed for analysis of data. This range from
Arithmetic Mean, Median, Mode Range, Quartile, Standard Deviation, Variance,
Correlation, Regression, Chi Square, Factor Analysis, Analysis of Variance and so on. We
shall examine few of these statistical tools

REGRESSION ANALYSIS AND CORRELATION

Regression analysis is used to estimate the relationship between two or more variables. It is
also referring to the methods by which estimates are made of the values of variable.
Illustration 1

Example 1

The moisture of the wet mix of a product is considered to have an effect on the finished
product density. The moisture of the wet mix was controlled and finished! product density
measured as shown in the following table

Mix moisture (X) 4.7 5.0 5.2 5.2 5.9 4.7 5.9 5.2 5.3 5.9 5.5 5.0

Density (Y) 3 3 4 5 10 2 9 3 7 6 6 4

(i) fit the model Y = a + bx to the data (ii) find the regression equation X and Y using
appropriate model (iii) predict the value of the density given the moisture of the wet mix.
When moisture is 5.4, 5.8, 4.5 and 4.8; what is the density?

Research Methodology Page31

Solution

No X Y XY X2 Y2

1 4.7 3 14.1 22.09 9.0

2 5.0 3 150 25.00 9.0

3- 5.2 4 20.8 27.04 16.0

4 5.2 .5 26.0 27.04 25.0

5 5.9 10 59.0 34.81 100.0

6 4.7 2 9.4 22.09 4.0

7 5.9 9 53.1 34.81 81.0

8 5.2 3 15.6 27.04 9.0

9 5.3 7 37.1 28.09 49.0

10 5.9 6 35.4 34.81 36.0

11 5.6 6 33.6 31.36 36.0

12 5.0 4 20.0 25.00 16.0

Y^ =a+bx

n ∑ XY −∑ X ∑ Y
b=
n ∑ X −¿ ¿ ¿
2

12 ( 339.1 ) −63.6(62)
b=
12 ( 339.18 )−¿ ¿

b = 126/25.2

b = 5.0

y=
∑Y
n

= 62/12

y = 5.16

a = y - bx

a = 5.16 – 5.0(5.3) = 21.34

Research Methodology Page32

Y^ = 21.34 + 5 x

When moisture; 5.4 Y^ = -21.34 + 5(5.4) = 5.66 ≈ 5.7

5.8 Y^ = -21.34 + 5(5.8) = 7.66 ≈ 7.7

4.5 Y^ = -21.34 + 5(4.5) = 1.16 ≈ 1.2

4.8 Y^ = -21.34 + 5(4.8) = 2.66 ≈ 2.7

Research Methodology Page33

MULTIPLE LINEAR REGRESSION

In the first part of this course, we discussed simple linear regression i.e. the regression of the
dependent variable Y on just one independent variables X. If there are two or more
independent variables that exist significant influence on a dependent variable, the regression
of the dependent variable on just one of them is inadequate. A statistical techniques that
takes into consideration the influence of more than one independent variable on a single
dependent variable is the MULTIPLE LINEAR REGRESSION. This technique take into
account simultaneously the relationship between all variables when two Or more
independent variables are being used to predict the value of one dependent variable.

The General Equation

The general linear regression model with K independent variables is of the form.

^y =a+b 1 x 1+ b2 x 2 +b 3 x 3 +bk x k or

Where y = a + b 1 x 1+ b2 x 2 +b 3 x 3 +b k x k +e

y is the variable which is to be predicted x1 x2 x3 ... xk are the k known variables on which
predictions are to the based, and a, b1, b2, b3 ... bk are the numerical constants which must be
determined from the observed data. The simplest multiple regression model with two
independent variables x1 and x2 is given by the model equation: y = a+ b1x1 + b2 x2 + e
with the following parameters
a = y - b1 x 1 - b2 x 2
Sx 1 ySx 2−Sx 2 ySx 1 x 2
b^ 1 =
Sx 1 Sx 2−¿ ¿ ¿

Sx 2 ySx 1−Sx 1 ySx 1 x 2

b^ 2 =
Sx 1 Sx 2−¿ ¿ ¿

Where

Sx1 =∑ x 21−¿ ¿

Sx2 = ∑ x 22−¿ ¿

Research Methodology Page34

Sx1x2 = ∑ x 1 x2 −¿ ¿

Sx1y = ∑ x 1 y−¿ ¿

Sx2y = ∑ x 2 y−¿ ¿

Interpretation of the parameters (constants) is quite simple. The intercept is the average value
of r when all independent variables equal zero. Each regression slope bk ≥ 1 is called a partial
regression coefficient and it indicates the average change in y associated with a limit change
in xk when the other independent variables are held constant. Thus:
b1= The change in the mean response per unit increase in x1 when x2is held constant.
b2 = The change in the mean response per unit increase in X1 when x2 is held constant.
The aim of adding to the simple two variable model is to improve the fit of the data. The
closeness of fit is measured by the COEFFICIENT OF MULTIPLE DETERMINATION, R 2
for which a useful computational formula is given below:

R2 =a ∑ y + b1 ∑ x 1 y+ b2 ∑ x 2 y−¿¿ ¿ ¿ ¿ ¿

Example 1

The following data shows the number of bedrooms the number of baths, and the prices at
which eight of one family houses sold recently in a certain community:
a. Fit the model y = a + b1x1 + b2x2 to the data and interpret the regression line.
b. Predict the average sale price of 3 bedroom houses with 2 baths i.e. x1=3, x2 = 2
c. Compute and interpret R2.

Research Methodology Page35

 No. of No. of Price N Solutions
bedrooms baths
x1 x2 y 2
x1 x2
2
x1 x2 x1y x2y y2
3 2 88800 9 4 6 266400 177600 7885.44
2 1 84300 4 1 2 168600 84300 7106.49
4 3 93800 16 9 12 375200 281400 8798.44
2 1 84200 4 1 2 168400 84200 7089.64
3 2 89700 9 4 6 269100 179400 8046.09
2 2 84900 4 4 4 169800 169800 7208.01
5 3 98400 25 9 15 492000 295200 9682.56
4 2 92900 16 4 8 371600 185800 8630.41
25 16 717000 87 36 55 2281100 145770 6447.08
0
64447080

Solution

a. ^y = a + b1x1 + b2X2
625
Sx1 = ∑ x 21−¿ ¿ = 87 - ¿ ¿ = 87 - = 87 – 78.125 = 8.875
8

256
Sx2 = ∑ x 22−¿ ¿ = 36 - ¿ ¿ = 36 - = 36 – 32 = 4
8

(25)(16) 400
Sx1x2 = ∑ x 1 x2 −¿ ¿ = 55 - = 55 - = 55-50 = 5
8 8

(25)(717000)
Sx1y = ∑ x 1 y−¿ ¿ = 22881100 - = 22881100 – 2240625 = 40475
8

(16)(717000)
Sx2y = ∑ x 2 y−¿ ¿ = 1457700 - = 1457700 – 1434000 = 23700
8

Sx 1 ySx 2−Sx 2 ySx 1 x 2 ( 40475 ) ( 4 )−(23700)(5) 161900−118500 43400

b^ 1 = = = = = 4133.33
Sx 1 Sx 2−¿ ¿ ¿ ( 8.875 ) ( 4 )−¿ ¿ 35.5−25 10.5

Research Methodology Page36

 Sx 2 ySx 1−Sx 1 ySx 1 x 2 ( 23700 ) ( 8.875 )−(40475)(5) 210337.5−202375 7962.5
b^ 2 = = = = =
Sx 1 Sx 2−¿ ¿ ¿ ( 8.875 ) ( 4 )−¿ ¿ 35.5−25 10.5
758.33
a = y - b1 x 1 - b2 x 2
= 89625 – [(4133.33)(3.125)] – [(758.33(2)]

= 89625 - 12916.66 – 1516.66

= 75191.68

Hence ^y = a + b1x1+ b2X2

= 75191.68 + 4133.33X1 + 758.33X2

Interpretation
a = average sale price of houses when no. of bedrooms and bath (x1, x2) are zero.
b1 = each extra bedroom adds on the average N4133.33 to the sale price of a house when no.
of bath is held constant.
b2 = each extra bedroom adds on the average N758.33 to the sale price of a house when no.
of bath is held constant.
(b) ^y = 75191.68 + 4133.33x1 + 758.33x2 when x 1= 3, x 2= 2

= 75191.68 + 4133,33(3)+ 758.33(2)

= 75191.68 + 12399.99 + 1516.66 9
^y = #89108.33

R2 = a ∑ y + b1 ∑ x 1 y +b 2 ∑ x 2 y −¿ ¿ ¿ ¿ ¿ ¿

= 75191.68 ( 717000 ) +4133.33 ( 2281100 ) +758.33 ( 1457700 )−¿ ¿ ¿ = 99.9%

Interpretation: This indicates that 99.9% of changes in (y) sale price of a house can be
explained by number of bedrooms and baths.

Research Methodology Page37

Example 2

The Nigeria Institute of Management is investigating the relationship between performance

in Quantitative Techniques and hours studied per week and the general level of intelligence
of candidates. The institute has data on 10 students which are:

Student Hours I.Q Exam

studied
2 2
S/N x1 x2 y x1 x2 x1y x2y x1x2 y2
1 9 99 56 81 9801 504 5544 891 3136
2 6 100 45 36 10000 270 4500 600 2025
3 12 119 80 144 14161 960 9520 1428 6400
4 14 95 73 196 9025 1022 6935 1330 5329
5 11 110 71 121 12100 781 7810 1210 5041
6 6 117 55 36 13689 330 6435 702 3025
7 19 98 95 361 9604 1805 9310 1862 9025
8 16 101 86 256 10201 1376 8686 1616 7396
9 3 100 34 9 10000 102 3400 300 1156
10 9 115 66 81 13225 594 7590 1035 4356
105 105 661 1321 111806 7744 69730 10974 46899
4

a. Fit an equation of the form y = a + b1x1+ b2x2 to the data and interpret your results
b. Predict the expected score of a candidate who has worked the 13 hours per week and who
has IQ of 102
c. Compute and interpret R2.

Research Methodology Page38

Solution
Sx1 = ∑ x 21−¿ ¿ = 1321 - ¿ ¿ = 1321 – 1102.5 = 218.5

Sx2 = ∑ x 22−¿ ¿ = 111806 - ¿ ¿ = 111806 – 111091.6 = 714.4

(105)(1054)
Sx1x2 = ∑ x 1 x2 −¿ ¿ = 10974 - = 10974 – 11067 = 93
10

(105)(661)
Sx1y = ∑ x 1 y−¿ ¿ = 7744 - = 7744 – 6940.5 = 803.5
10

(1054)(661)
Sx2 = ∑ x 2 y−¿ ¿ = 69730 - = 69730v – 69669.4 = 60.6
10

Sx 1 ySx 2−Sx 2 ySx 1 x 2 ( 803.5 ) ( 714.4 )−(60.6)(93) 574020.4−5635.8 579656.2

b^ 1 = = = = =
Sx 1 Sx 2−¿ ¿ ¿ (218.5)(714.4)¿ ¿ 156096.4−8649 147447.4
3.93

Interpretation

Each extra hour studied adds on the average 3.93 marks to exam score when I.Q is held
constant

Sx 2 ySx 1−Sx 1 ySx 1 x 2 ( 60.6 ) ( 218.5 )−(803.5)(−93) 13241.1+ 7472.5 87966.6

b^ 2 = = = = =
Sx 1 Sx 2−¿ ¿ ¿ (218.5)(714.4)¿ ¿ 156096.4−8649 147447.4
0.60

Each extra I.Q adds on the average 0.60 mark to exam score when hour studied is held
constant.

a= y -b 1 x 1 - b 2 x 2
a=66.1−[(3.93)(10.5)]−[(0.60)(105.4)]

= 66.1 – 41.265 – 63.24

Research Methodology Page39

a = - 38.405

Interpretation
Average exam score when I.Q and hour studied are zero
Hence,
a . ^y = a + b1x1 + b2X2

= -38.405 + 3.93x1 + 0.60x2

b. ^y =−38.405+3.9 x1 + 0.60x2
w h en x 1=13. x 2=102.
¿−38.405+3.93 ( 13 ) +0.60(102)
^y =73.885

c. R2 = a ∑ y + b1 ∑ x 1 y +b 2 ∑ x 2 y −¿ ¿ ¿ ¿ ¿ ¿
= (−38.405) ( 616 ) +( 3.93) ( 7744 )+(0.6) ( 69730 )−¿ ¿¿ = 0.996 = 99.6%
This indicates that 99.6% of changes in (y) exam score can be explained by hours studied
and I.Q of candidates.
ANALYSIS OF VIARANCE (ANOVA)

BASIC CONCEPTS

Experiment: This is an operation designed to discover something unknown or to establish it

when discovered. Treatment: Is any kind of factor whose different states are under
investigation e.g. level of fertilizer or manure applied in agricultural investigation.

Replication: Simply means repetition, that is, performing the same experiment more than
once under almost the same experimental condition.

Research Methodology Page40

Experimental Plot (Unit): This is the small to generate an observation. I denotes the group
of materials to which a treatment is applied in a simple trial of experiment e.g. plot of land,
patient in a hospital etc.

ONE WAY CLASSIFICATION

The model is given as

y ij =μ+ τ i+ ε ij i = 1, 2, ….. p j = 1, 2, ….. n

Where

y ij = represent the jth observation taken under treatment

μ = overall due to ith treatment

ε ij = random error in observing y ij

STEPS INVOLVED IN TESTING ANOVA

1. Setup the hypothesis

2. State the significance level α
MST
3. Test statistic e.g F =
MSE
4. Decision Rule: Reject Ho if Fc > F 1−∝ or Fα, v1, v2 otherwise do not reject Ho.
5. Computation
2
Y ..
CF =
N
P N
Y ..2
SST = ∑ ∑ y 2ij −¿ ¿
I=1 J=1 N
N 2 2
yi . Y ..
SStr = ∑ −¿ ¿
J =1 n N
SSE = SST – SStr

Research Methodology Page41

 ANOVA TABLE

Source of D.F SS MSS F-ratio

variation
Treatment P-1 SStr SStr/P-1 MST/MSE
Error (SSE) N-P SSE SSE/N-P
Total N–1 SST

Example
In an attempt to determine whether the yield of a particular plant depends on the types
of fertilizer (A, B, C) applied yields of the plant from different plots are recorded
below:

A 6 9 6 5 7 5 6 6
B 5 9 9 8 8 7 7 7
C 3 4 3 6 8 5 5 6
At α = 0.05, Is there evidence of difference in the yield of the plant produced with the
3 different fertilizer?
Solution
Ho: T1 = T2 = T3 (No significance difference)
H1: T1≠ T2≠ T3 (There is significance difference)
Decision Rule: Reject Ho if Fc > F1-α , v1, v2 from table
Computation
P N
Y ..2
SST = ∑ ∑ y ij −¿
2
¿
I=1 J=1 N
= (62 + 52 + 32 + …. + 62 - ¿ ¿ = 1006 – 937.5
= 68.5
N 2 2
y Y ..
SStr = ∑ i . −¿ ¿
J =1 n N
2 2 2 2
50 + 60 +40 150
= − = 962.5 – 937.5
8 24
= 25

Research Methodology Page42

 SSE = SST – SStr
= 68.5 – 25.0
= 43.5
ANOVA TABLE

Source of D.F. SS MSS F-ratio

variation
Treatment 2 25 12.5 6.07
Error (SSE) 21 43.5 2.06
Total 23 68.5
F1-0.05, 2, 21 = F0.95, 2, 21 = 3.47

Conclusion

Since 6.07 ¿ 3.47, we reject Ho and conclude that there is a difference in the effect of
the three fertilizer on the plant yield.

Exercises.
1. Three groups of six guinea pigs were injected, respectively, with 0.5mg, 1.0mg
and 1.5mg of a new tranquilizer, and the following are the number of minutes it
took them to fall asleep.

0.5mg 21 23 19 24 25 23
1.0mg 19 21 20 18 22 20
1.5mg 15 10 13 14 11 15

Test at α = 0.05 where the null hypothesis that difference in dosage have no effect can
be rejected.

2. The data given below are the weight losses of certain machine parts (in milligram)
due to friction when 10 different instruments were used under controlled
condition.

1 2 3 4 5 6 7 8 9 10
12 21 5 30 6 11 12 3 24 30

Research Methodology Page43

 13 20 6 25 8 15 18 4 25 23
11 19 3 25 4 17 15 1 26 34
15 18 7 28 8 20 20 7 22 32
Analyze the data test for the treatment effect.

CHI-SQUARE TEST (2)

The sum of squares of independent standard normal values is called a chi-square virate and
its distribution is called the chi-square distribution. The chi-square test written as 2-test, is a
useful measure of comparing experimentally obtained results with those expected
theoretically and based on the hypothesis. It is used as a test statistic in testing a hypothesis
that provides a set of theoretical frequencies with which observed frequencies are compared.
In general chi-square test is applied to those problems in which we study whether the
frequency with which a given event has occurred is significantly different from one as
expected theoretically. The measure of chi-square enables us to find out the degree of
discrepancy between observed frequencies and theoretical frequencies and thus, to determine
whether the discrepancy so obtained between observed frequencies and theoretical
frequencies is due to error of sampling or due to change. Chi-square is not a parameter as its
value is not derived from the observations in a population. Hence, chi-square test is a non-
parametric test. It is not concerned with any population distribution and its observations.

The 2-test was first used in testing statistical hypothesis by Karl Pearson in the year 1900. It
is defined as;
n
 = ∑ ¿¿
2

i=1

Where 0i = observed frequency of ith event

e i = expected frequency of ith event

Research Methodology Page44

Properties of 2-test

1. Mean = n
2. Variance = 2n
3. Mode = n-2

4. Karl pearson’s coefficient of skewness SKp =

mean−mode
SD
=
√ 2
n
n
5. If 0i and ei (i= 1,2,3…n) are set of observed and expected frequencies the ∑ ¿¿ ¿  2
i=1

Use of 2-test

1. It is used to test whether the population variance is equal to a specified variance

2. It is used to test for goodness of fit
3. It is used to test the independence of attributes in contingency table.

Steps required in calculating 2-test

1. Calculate all the expected frequencies I i.e e1 for all values of I = 1, 2, ….n
2. Take the difference between each observed frequency o i and the corresponding
expected frequency ei for each value of I, i.e find (oi- ei) for all values of I = 1,2,3… n
3. Square the difference for each value of, i.e calculate (oi- ei) for all values of I=1,2,3..n
4. Divided each square difference by the corresponding expected frequency i.e calculate
¿ ¿ for all values of I, = 1, 2, 3, …, n
n
5. Add all these quotients obtained in step 4, then, 2= ∑ ¿¿ is the required value of chi-
i=1

square
Note:

a. The value of 2 is always positive as each pair is squared one.

b. 2 will be zero if each pair is zero and it may assume any value extending to infinity
when the difference between the observed frequency and expected frequency in each
pair is unequal. Thus 2 lies between 0 and 
c. The significance test on 2 is always based on one- tailed test of the right hand side of
standard normal curve as 2 is always non-negative.

Research Methodology Page45

 d. As 2 is a statistic and not a parameter, so it does not involve any assumption about
the form of original distribution from which the observation has come.

Degrees of Freedom

The number of data that are given in the form of a series of variables in a row or column or
the number of frequencies that are put in cells in a contingency table, which can be
calculated independently is called the degrees of freedom and is denoted V.

Case 1: If the data is given in the form of a series of variables in a row or column then the
degrees of freedom = (number of items in the series) -1, i.e V= n −1 , where n is the number
of variables in the series in a row or column.

Case II: When the number of frequencies are put in cells in a contingency table, the degrees
of freedom will be the product of (number of rows less one) and the (number of column less
one) i.e V=( r −1 ) ( c−1 ), where r is the number of rows and c is the number of columns.

Contingency Table

The table usually used for chi-square test is called contingency table, that is, table of
attributes used to test the independence or association between two or more criteria for
classification. It is generally of the form;

Attribute A

Sample 1 2 3 Total
N A B C A+B+C
M D E F D+E+F
TOTAL A+D B+E C+F GRAND TOTAL

Research Methodology Page46

Example 1: An investigator wishes to experiments peas colours and shapes, he observed the
result of peas and the result below was presented;

Round Winkled
Yellow Green Yellow Green
Observed counts 513 102 101 22
Proportions 9/23 5/23 7/23 2/23
Is there any evidence to doubt his theory at 1% and 5% levels of significance?

Solution

Hypothesis

Ho: There is no evidence to doubt his theory i.e (There is no different)

H1: There is evidence to doubt his theory i.e (There is different)

Level of significance at = 0.01 and 0.05

Test statistic
n
2 = ∑ ¿¿
i=1

Decision rule

Reject Ho if 2cal ≥ 2tab otherwise do no reject Ho i.e (accept Ho)

Calculations

Total proportions = 9+5+7+2 = 23

Total value = 738

Round and Yellow peas = 9/23 x 738 = 288.78

Round and Green peas = 5/23 x 738 = 160.43

Wrinkled and Yellow peas = 7/23 x 738 = 224.61
Wrinkled and Green peas = 2/23 x 738 = 64.17

Research Methodology Page47

 O E o–e (o – e)2 (o – e)2/e
513 288.78 224.22 50274.61 174.09
102 160.43 -58.49 3421.06 21.28
101 224.61 -123.61 15279.43 68.03
22 64.17 -42.17 1778.31 27.71
n

∑ ¿¿ = 291.11
i=1

Critical value

2tab 3, 0.01 = 11.34, 2tab 3,0.05 =7.81

Conclusion: Since 2cal = 291.11 >2tab = 11.34 at 1% we reject Ho and conclude that there is
evidence to doubt his theory. However, at 5% 2cal = 291.11 > 2tab = 7.81, we also reject Ho
and still conclude that there is significant difference in his theory.

Example 2

One thousand girls in a college were graded according to their I.Q and the economic
conditions of their homes. Use X2-test to find out whether there is any association between
economic conditions at home and I.Q of girls at 5% level of significance.

Economic condition High Low Total

Rich 100 300 400
Poor 350 250 600
Total 450 550 1,000

Solution:

Hypothesis:

Ho: There is no association between economic condition at home and I.Q

H1: There exist an association between economic condition at home and I.Q

Level of significance

Research Methodology Page48

at  = 0.05

Test statistic
n
2 = ∑ ¿¿
i=1

Decision rule: Reject Ho if 2cal ≥ 2tab otherwise do not reject Ho

Calculation

Research Methodology Page49

 row total x columntotal
Using the formulae = grand total
the expected frequencies are calculated

in the table below:

Observed value Expected value

o e o-e (o – e)2 (o – e)2/e

100 180 -80 6400 35.6

350 270 80 6400 23.7
300 220 80 6400 29.1
250 330 -80 6400 19.4
∑ ¿¿ = 107.8
Critical value

2tab = 2(r-1)(c-1), 0.05 = 2(2-1)(2-1) =21, 0.05 = 3.84

Conclusion

Since 2cal = 107.8   tab

2
= 3.84 we reject H o and conclude that there exist an
association between economic condition at home and I.Q

Research Methodology Page50