Upload
phungkhuong
View
220
Download
5
Embed Size (px)
Citation preview
i
COURSE MANUAL
Research Methodology GCE 209
University of Ibadan Distance Learning Centre
Open and Distance Learning Course Series Development
ii
Copyright © 2014 by Distance Learning Centre, University of Ibadan, Ibadan.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, without the prior permission of the copyright owner.
ISBN 978-021-923-4
General Editor: Prof. Bayo Okunade
University of Ibadan,
Nigeria
Telex: 31128NG
Tel: +234 (80775935727)
E-mail: [email protected]
Website: www.dlc.ui.edu.ng
iii
Vice-Chancellor’s Message
The Distance Learning Centre is building on a solid tradition of over two decades of
service in the provision of External Studies Programme and now Distance Learning
Education in Nigeria and beyond. The Distance Learning mode to which we are
committed is providing access to many deserving Nigerians in having access to higher
education especially those who by the nature of their engagement do not have the
luxury of full time education. Recently, it is contributing in no small measure to
providing places for teeming Nigerian youths who for one reason or the other could
not get admission into the conventional universities.
These course materials have been written by writers specially trained in ODL course
delivery. The writers have made great efforts to provide up to date information,
knowledge and skills in the different disciplines and ensure that the materials are user-
friendly.
In addition to provision of course materials in print and e-format, a lot of Information
Technology input has also gone into the deployment of course materials. Most of them
can be downloaded from the DLC website and are available in audio format which
you can also download into your mobile phones, IPod, MP3 among other devices to
allow you listen to the audio study sessions. Some of the study session materials have
been scripted and are being broadcast on the university’s Diamond Radio FM 101.1,
while others have been delivered and captured in audio-visual format in a classroom
environment for use by our students. Detailed information on availability and access is
available on the website. We will continue in our efforts to provide and review course
materials for our courses.
However, for you to take advantage of these formats, you will need to improve on
your I.T. skills and develop requisite distance learning Culture. It is well known that,
for efficient and effective provision of Distance learning education, availability of
appropriate and relevant course materials is a sine qua non. So also, is the availability
of multiple plat form for the convenience of our students. It is in fulfilment of this,
that series of course materials are being written to enable our students study at their
own pace and convenience.
It is our hope that you will put these course materials to the best use.
Prof. Abel Idowu Olayinka
Vice-Chancellor
iv
Foreword
As part of its vision of providing education for “Liberty and Development” for
Nigerians and the International Community, the University of Ibadan, Distance
Learning Centre has recently embarked on a vigorous repositioning agenda which
aimed at embracing a holistic and all encompassing approach to the delivery of its
Open Distance Learning (ODL) programmes. Thus we are committed to global best
practices in distance learning provision. Apart from providing an efficient
administrative and academic support for our students, we are committed to providing
educational resource materials for the use of our students. We are convinced that,
without an up-to-date, learner-friendly and distance learning compliant course
materials, there cannot be any basis to lay claim to being a provider of distance
learning education. Indeed, availability of appropriate course materials in multiple
formats is the hub of any distance learning provision worldwide.
In view of the above, we are vigorously pursuing as a matter of priority, the provision
of credible, learner-friendly and interactive course materials for all our courses. We
commissioned the authoring of, and review of course materials to teams of experts and
their outputs were subjected to rigorous peer review to ensure standard. The approach
not only emphasizes cognitive knowledge, but also skills and humane values which
are at the core of education, even in an ICT age.
The development of the materials which is on-going also had input from experienced
editors and illustrators who have ensured that they are accurate, current and learner-
friendly. They are specially written with distance learners in mind. This is very
important because, distance learning involves non-residential students who can often
feel isolated from the community of learners.
It is important to note that, for a distance learner to excel there is the need to source
and read relevant materials apart from this course material. Therefore, adequate
supplementary reading materials as well as other information sources are suggested in
the course materials.
Apart from the responsibility for you to read this course material with others, you are
also advised to seek assistance from your course facilitators especially academic
advisors during your study even before the interactive session which is by design for
revision. Your academic advisors will assist you using convenient technology
including Google Hang Out, You Tube, Talk Fusion, etc. but you have to take
advantage of these. It is also going to be of immense advantage if you complete
assignments as at when due so as to have necessary feedbacks as a guide.
The implication of the above is that, a distance learner has a responsibility to develop
requisite distance learning culture which includes diligent and disciplined self-study,
seeking available administrative and academic support and acquisition of basic
information technology skills. This is why you are encouraged to develop your
computer skills by availing yourself the opportunity of training that the Centre’s
provide and put these into use.
v
In conclusion, it is envisaged that the course materials would also be useful for the
regular students of tertiary institutions in Nigeria who are faced with a dearth of high
quality textbooks. We are therefore, delighted to present these titles to both our
distance learning students and the university’s regular students. We are confident that
the materials will be an invaluable resource to all.
We would like to thank all our authors, reviewers and production staff for the high
quality of work.
Best wishes.
Professor Bayo Okunade
Director
vi
Course Development Team
Content Authoring Afusat Olanike Busari
Content Editor
Production Editor
Learning Design/Assessment Authoring
Managing Editor
General Editor
Prof. Remi Raji-Oyelade
Ogundele Olumuyiwa Caleb
Dara Abimbade
Ogunmefun Oladele Abiodun
Prof. Bayo Okunade
vii
Table of Contents
Study Session 1 Meaning of Research 1
Introduction 1
Learning Outcomes for Study Session 1 1
1.1 Definition of Research 1
1.2 Characteristics of Research 2
1.3. Steps in Research Process 5
Summary of Study Session 1 15
In Study Session 1, you have learned that: 15
Self-Assessment Questions (SAQs) for Study Session 1 15
Notes on the Self-Assessment Questions (SAQs) for Study Session 1 16
Study Session 2:Theories, Hypotheses Generation and Testing 17
Introduction 17
Learning Outcomes for Study Session 2 17
2.1 Generating Hypotheses 17
2.2 Characteristics of Hypothesis 18
2.3 Functions of Hypothesis 20
4. Logic: a hypothesis can be sourced from logic or the art of reasoning – that is, by
means of deductive and inductive reasoning. 22
2.5 Types of Hypothesis 22
Summary of Study Session 2 24
Self-Assessment Questions (SAQs) for Study Session 2 24
Notes on the Self-Assessment Questions (SAQs) for Study Session 2 25
Study Session 3: The Research Problems and Literature Review 27
Introduction 27
Learning Outcomes for Study Session 3 27
3.1 The Research Problem 27
3.2 Components of Setting a Problem 28
3.3 Time Constraints 32
3.4 Purpose of Literature Review 33
Summary of Study Session 3 40
Self-Assessment Questions (SAQs) for Study Session 3 40
Notes on the Self-Assessment Questions (SAQs) for Study Session 3 41
Study Session 4: Concepts, Constructs and Variables 43
Introduction 43
viii
Learning Outcomes for Study Session 4 43
4.1 Defining Concepts 43
4.2 Constructs 44
Summary of Study Session 4 52
Self-Assessment Questions (SAQs) for Study Session 4 52
Notes on the Self-Assessment Questions (SAQs) for Study Session 4 52
References 53
Study Session 5: Types of Research Approaches –Qualitative and Quantitative
Approaches 54
Learning Outcomes for Study Session 5 54
5.1 Qualitative Approaches 54
5.3 Longitudinal Research 57
Summary of Study Session 5 59
Notes on the Self-Assessment Questions (SAQs) for Study Session 5 60
Study Session 6: Survey Research and Research Tool 62
Introduction 62
Learning Outcomes for Study Session 6 62
6.1 Meaning of Survey Research 62
6.2 Questionnaire 65
6.3 Interview and Observations 69
6.9 Case Study 77
Summary of Study Session 6 78
Self-Assessment Questions (SAQs) for Study Session 6 78
Notes on the Self-Assessment Questions (SAQs) for Study Session 6 79
Study Session 7: Experimental and Quasi Experimental Research 80
Introduction 80
Learning Outcomes for Study Session 7 80
7.1 Experimental Research 80
7.2 Characteristics of the Experimental Method 80
7.3 Classification of Experimental Design 83
7.4 Quasi-Experimental Designs 87
Summary of Study Session 7 89
In Study Session 7, you have learned that: 89
Self-Assessment Questions (SAQs) for Study Session 7 89
Notes on the Self-Assessment Questions (SAQs) for Study Session 7 89
ix
Study Session 8: Research Design 91
Introduction 91
Learning Outcomes for Study Session 8 91
8.1 Definition 91
8.2 Experimental Vs Ex-post facto research 91
8.3 Method of controlling independent variables 93
Summary of Study Session 8 100
Self-Assessment Questions (SAQs) for Study Session 8 100
Notes on the Self-Assessment Questions (SAQs) for Study Session 8 100
Study Session 9: Population and Sample 102
Learning Outcomes for Study Session 9 102
9.1 Population and Samples 102
9.2 The Purpose of Sampling 103
9.3 Types, Methods and Procedures for Sampling 104
Summary of Study Session 9 114
Self-Assessment Questions (SAQs) for Study Session 9 115
Notes on the Self-Assessment Questions (SAQs) for Study Session 9 115
Study Session 10: Level of Measurement Validity and Reliability 117
Learning Outcomes for Study Session 10 117
10.1 Levels of Measurement 117
10.2 Validity and Reliability 120
10.2.1 Validity 120
10.2.2 Reliability 122
Summary of Study Session 10 124
In Study Session 10, you have learned that: 124
Self-Assessment Questions (SAQs) for Study Session 10 124
Notes on the Self-Assessment Questions (SAQs) for Study Session 10 125
Study Session 11: Test Administration and Interpretation 127
Learning Outcomes for Study Session 11 127
11.1 Process of Test Administration 127
11.2 Test venue 128
11.3 The actual testing session 130
Summary of Study Session 11 131
Self-Assessment Questions (SAQs) for Study Session 11 131
Notes on the Self-Assessment Questions (SAQs) for Study Session 11 132
x
Study Session 12: Writing Research Report 133
Introduction 133
Learning Outcomes for Study Session 12 133
12.1 Planning a Research Report 133
12.2 Format for the Research Report 134
12.2.1 Title Page 134
12.2.2 Acknowledgement 134
12.2.3 Abstract 134
12.2.4 Introduction and Problem Statement 135
12.2.5 Review of Literature 135
12.2.6 Methodology 136
12.2.7 The Statement of Results 136
12.2.8 Analysis and Discussion 137
12.2.9 Specification of Statistical Analysis 137
12.2.10 Statement of How to use Results 138
12.2.11 Results 138
12.2.12 Discussion 140
12.2.13 Summary and Conclusion 141
12.2.14 References 141
Summary of Study Session 12 145
In Study Session 12, you have learned that: 145
Self-Assessment Questions (SAQs) for Study Session 12 145
Notes on the Self-Assessment Questions (SAQs) for Study Session 12 146
1
Study Session 1 Meaning of Research
Expected duration: 1 week or 2 contact hours
Introduction
In this session, you will be exposed to the basic concepts in research. This course aims
at providing learners with tools to do research, to help them avoid some of the pitfalls
and common mistakes in research and to establish good research habits. Historical
research has been known for a long time; possibly longer than most other types of
research. Historical research differs from other studies because it cannot be classified
as either quantitative or qualitative. This chapter will discuss the meaning of research,
characteristics of research, goals of research and meaning of historical research.
Learning Outcomes for Study Session 1
When you have studied this session, you should be able to:
1.1 Define the term ‘research’. (SAQ 1.1)
1.2 Enumerate at least five characteristics of research. (SAQ 1.2,)
1.3 Explain the deductive approach to research (SAQ 1.3)
1.4 Describe the concept of historical research (SAQ 1.4).
1.1 Definition of Research
The word ‘research’ comprises two syllables, i.e. re and search. re is a prefix meaning
again, new or over again search is a verb meaning to examine closely and carefully, to
test and try, or to probe. Together they form a noun describing a careful, systematic,
patient study and investigation in some field of knowledge, undertaken to establish
facts or principles. Research is a structured investigation that utilizes acceptable
scientific methodology to solve problems and create new knowledge that is generally
applicable. Scientific methods consist of systematic observation, classification and
interpretation of data. Leedy (1997) defines research as the systematic process of
collecting and analysing information (data) in order to increase our understanding of
the phenomenon with which we are concerned or interested. Leedy goes on to state
what research is not. According to Leedy, research is not mere: gathering;
2
transformation of facts from one location to the other and searching for information.
Research goes beyond this. It could be define as the process of specifying problem
and collecting data to solve a problem, analyze the data and see whether the problem
is solved. It could be define as a procedure of specifying a problem, and find ways of
analyzing data collected on such problem. According to Leedy, research is an activity
or process through which we attempt to systematically and with the support of data,
give an answer to a question, the resolution of a problem or a greater understanding of
a phenomenon. This process is called research methodology.
Figure 1.1: The Concept of Research (Photo from www.cep-probation.org)
1.2 Characteristics of Research
According to Leedy, the research process has the following five characteristics:
i. Research originates with a question or problem: The world is filled with many
unanswered questions and unresolved problems. In every situation there are things
that make us wonder or ask questions. And by asking questions, we strike the first
spark igniting a chain reaction that terminates in the research process to observe in
thinking about one’s principal goal in research. Principal problems are broken down
much more frequently than is realised. Most researchers do not take time or the
trouble to isolate the lesser problems within the major problem and consequently their
research projects become cumbersome and unwieldy.
3
ii. Researchers are guided by specific research problems; question or hypothesis:
A hypothesis is a logical assumption, a reasonable guess, an educated inference. It
may direct thinking to the possible source of information that will aid in resolving the
research problem through the resolution of attendant sub-problems. Hypotheses are
constant recurring features of everyday life. They present the natural working of the
human mind. Each of these hypotheses provides a direction for exploration to locate
the information that may resolve the problem in the society. It is important to note that
hypotheses are never proved; they are either rejected or accepted. After the hypotheses
come data. The data either support or fail to support the hypotheses. Many of the
greatest discoveries in science have begun as hypotheses.
iii. Research accepts certain critical assumption: In research, assumptions are
equivalent to self-evident truths. The assumptions must be valid or else the research
cannot proceed. For this reason, careful researchers – certainly in academic research –
set forth a statement of the assumption as the bedrock upon which their study must
rest. In a research, therefore, it is important that others know what the individual
assume with respect to the study. An assumption is a condition that is taken for
granted, without which the result situation would be impossible. For the beginning
researcher, you need to state your assumptions, for it is better to be over explicit than
to take too much for granted.
iv. Research requires the collection and interpretation of data in attempting to
resolve the problem that initiated the research:
Having isolated the problem, divided it into appropriate sub-problems, posited
reasonable questions or hypotheses and recognised the assumptions that are basic to
the entire effort. The next step is to collect whatever data that seem appropriate and to
organise them in meaningful ways so that they can be interpreted. The significance of
the data depends on the way the human mind interprets the information from those
data. In research, unprocessed data are worthless.
v. Research is by its nature, cyclical: The research process follows a cycle and
begins from simple to complex. It follows logical developmental steps. What,
however, seems like a neat circle may never be conclusive. In exploring an area, one
comes across additional problems that need to be resolved. Research leads to another
research. To view research this way is to invest it with a dynamic quality i.e. its true
4
nature which is different from the conventional view, which sees research as static,
self- contained and an end in itself. Every researcher eventually learns that genuine
research creates more problems than it resolves. Such is the nature of the discovery of
knowledge. Best and Khan (1995) define research as the systematic and objective
analysis and recording of controlled observations that may lead to the development of
generalisations, principles or theories, resulting in prediction and possibly ultimate
control of events. Due to the abstract nature of this definition, they stated a summary
of characteristics of research which according to Best and Khan (1995) may help to
clarify its meaning. The characteristics are summarised below.
Research:
- emphasises the development of generalisations, principles or theories that will
be helpful in predicting future occurrences;
- is based upon observable experience or empirical evidence;
- involves gathering of new data from primary or first-hand sources or using
existing data for a new purpose;
- is often characterised by carefully designed procedures that apply rigorous
analysis;
- requires expertise;
- strives to be objective and logical, applying every possible test to validate the
procedures employed, the data collected and the conclusions reached;
- involves the quest for answers to unresolved problems;
- is characterised by patient and unhurried activity;
- is carefully recorded and reported; and sometimes requires courage.
Best and Khan believe that a researcher should be a scholarly imaginative person, with
a high integrity, who is willing to spend long hours painstakingly seeking truth. They
noted that researchers are human beings, and ideals that have been listed are never
completely realised.
Activity 1.1
It is important at this point to take a moment to reflect on what you have read so far.
Based on your learning experience, write down the characteristics of a research. Make
sure you do this, it will help you in monitoring rate at which you learn.
5
Activity 1.1 Feedback:
Basically, there are eight characteristics of a research and these are:
Research:
- emphasises the development of generalisations, principles or theories that will
be helpful in predicting future occurrences;
- is based upon observable experience or empirical evidence;
- involves gathering of new data from primary or first-hand sources or using
existing data for a new purpose;
- is often characterised by carefully designed procedures that apply rigorous
analysis;
- requires expertise;
- strives to be objective and logical, applying every possible test to validate the
procedures employed, the data collected and the conclusions reached;
- involves the quest for answers to unresolved problems;
- is characterised by patient and unhurried activity;
- is carefully recorded and reported; and sometimes requires courage.
-
1.3. Steps in Research Process
1. Formulating the Research Problem
2. Extensive Literature Review
3. Developing the objectives
4. Preparing the Research Design including Sample Design
5. Collecting the Data
6. Analysis of Data
7. Generalisation and Interpretation
8. Preparation of the Report or Presentation of Results
6
Figure 1.2: Steps in Research (Photo from www.slideshare.net)
Step1. Formulating the research problem:
It is the first and most crucial step in the research process
- Main function is to decide what a researcher wants to find out about.
- The way a researcher formulate a problem determines almost every step that
follows.
Sources of research problems
Research in behavioural science revolves around four Ps:
• People- a group of individuals
• Problems- examine the existence of certain issues or problems relating to their
lives; to ascertain attitude of a group of people towards an issue
• Programmes- to evaluate the effectiveness of an intervention
• Phenomena- to establish the existence of a regularity.
In practice most research studies are based upon at least a combination of two.
Every research study has two aspects:
1. Study population
• People: individuals, organizations, groups, communities(they provide
researcher with the information or the researcher collects information about
them)
7
2. Subject area
• Problems: issues, situations, associations, needs, profiles
• Programme: content, structure, outcomes, attributes, satisfactions, consumers,
Service providers, etc.
• Phenomenon: cause-and-effect relationships, the study of a phenomenon Itself.
Considerations in selecting a research problem include:
1. Interest: a research exercise is usually time consuming, and involves hard
work and possibly unforeseen problems. One should select topic of great
interest to sustain the required motivation.
2. Magnitude: It is important to select a topic that the researcher can manage
within the time and resources at his/her disposal. Narrow the topic down to
something manageable, specific and clear.
3. Measurement of concepts: The researcher should make sure that he /she is
clear about the indicators and measurement of concepts (if used) in the study.
4. Level of expertise: The researcher should make sure that he/ she have adequate
level of expertise for the task he /she is proposing since he /she needs to do the
work yourself.
5. Relevance: Should ensure that his /her study adds to the existing body of
knowledge, bridges current gaps and is useful in policy formulation. This will
help the researcher to sustain interest in the study.
6. Availability of data: Before finalizing the topic, make sure that data are
available.
7. Ethical issues: How ethical issues can affect the study population and how
ethical problems can be overcome should be thoroughly examined at the
problem formulating stage.
Steps in formulation of a research problem
Working through these steps presumes a reasonable level of knowledge in the broad
subject area within which the study is to be undertaken. Without such knowledge, it is
difficult to clearly and adequately explore a subject area.
Step 1 Identify a broad field or subject area of interest to you.
Step 2 Dissect the broad area into sub areas.
Step 3 Select what is of most interest to you.
Step 4 Raise research questions.
8
Step 5 Formulate objectives.
Step 6 Assess your objectives.
Step 7 Double check.
Apart from the research problem, every study in behavioural science has a second
element, the study population from whom the required information to find answers to
the research questions is obtained. As you narrow the research problem, similarly you
need to decide very specifically who constitutes your study population, in order to
select the appropriate respondents.
Step 2. Reviewing the literature: This involves the following facts;
- Essential preliminary task in order to acquaint yourself with the available body
of knowledge in your area of interest.
- Literature review is integral part of entire research process and makes valuable
contribution to every operational step.
- Reviewing literature can be time-consuming, daunting and frustrating, but is
also rewarding. Its functions are to:
a. Bring clarity and focus to the research problem;
b. Improve your methodology;
c. Broaden your knowledge;
d. Contextualise your findings.
a. Bring clarity and focus to your research problem;
The process of reviewing the literature helps you to understand the subject area better
and thus helps you to conceptualise your research problem clearly and precisely. It
also helps you to understand the relationship between your research problem and the
body of knowledge in the area.
b. Improve your methodology:
A literature review tells you if others have used procedures and methods similar to the
ones that you are proposing, which procedures and methods have worked well for
them, and what problems they have faced with them. Thus you will be better informed
to select a methodology that is capable of providing valid answer to your research
questions.
9
c. Broaden your knowledge base in your research area:
It ensures that the researcher reads widely around the subject area in which he/she
intends to conduct a research study. As you are expected to be an expert in your area
of study, it helps fulfill this expectation. It also helps you to understand how the
findings of your study fit into the existing body of knowledge.
d. Contextualise your findings:
How do answers to your research questions compare with what others have found?
What contribution have you been able to make in to the existing body of knowledge?
How are your findings different from those of others? For you to be able to answer
these questions, you need to go back to your literature review. It is important to place
your findings in the context of what is already known in your field of enquiry.
Procedure for reviewing the literature:
i. search for existing literature in your area of study;
ii. review the literature selected;
iii. develop a theoretical framework;
iv. develop a conceptual framework.
Search for existing literature: This involves the following;
- To effectively search for literature in your field of enquiry, it is important that
you have in mind at least some idea of broad subject area and of the problem
you wish to investigate, in order to set parameters for your search.
- Next compile a bibliography for this broad area. Sources are:
1. Books
2. journals
Step 3 The formulation of objectives:
- Objectives are the goals you set out to attain in your study.
- They inform a reader what you want to attain through the study.
- It is extremely important to word them clearly and specifically.
Objectives should be listed under two headings:
a. main objectives ( aims);
b. sub-objectives.
10
• The main objective is an overall statement of the thrust of your study. It is also
a statement of the main associations and relationships that you seek to discover
or establish.
• The sub-objectives are the specific aspects of the topic that you want to
investigate within the main framework of your study.
- They should be numerically listed.
- Wording should clearly, completely and specifically communicate to your
readers your intention.
- Each objective should contain only one aspect of the Study.
- Use action oriented words or verbs when writing objectives.
The objectives should start with words such as
‘to determine’,
‘to find out’,
‘to ascertain’,
‘to measure’,
‘to explore’ etc.
The wording of objectives determines the type of research (descriptive, correlational
and experimental) and the type of research design you need to adopt to achieve them.
o Dolapo, a final year student in the department of Teacher Education, is
preparing to begin his project and he is confused with different formats his
classmates recommend for him to follow in other to write a good project. Thus,
he intends to consult his course adviser to enlighten him on the steps to take in
carrying out this research. If you were the course adviser, what steps would you
advise him to follow to write a fantastic project?
• There are different steps recommended by scholars to carry out a research.
However, there are some basic things that should be present in the steps.
Generally, these steps could be followed to carry out the research:
1. Formulating the Research Problem
2. Extensive Literature Review
3. Developing the objectives
4. Preparing the Research Design including Sample Design
5. Collecting the Data
6. Analysis of Data
7. Generalisation and Interpretation
8. Preparation of the Report or Presentation of Results
11
1.4 Types of Research
Research can be classified from three perspectives:
1. application of research study
2. objectives in undertaking the research
3. inquiry mode employed
Application:
From the point of view of application, there are two broad categories of research:
- pure research and
- applied research.
Pure research involves developing and testing theories and hypotheses that are
intellectually challenging to the researcher but may or may not have practical
application at the present time or in the future. The knowledge produced through pure
research is sought in order to add to the existing body of research methods.
Applied research is done to solve specific, practical questions; for policy formulation,
administration and understanding of a phenomenon. It can be exploratory, but is
usually descriptive. It is almost always done on the basis of basic research. Applied
research can be carried out by academic or industrial institutions. Three systems are
described here, the first based on the goal or purpose of the research; second is the
qualitative – quantitative dichotomy that represents two distinct orientations to
phenomenon being studied. The third system is a classification of general methods
used in behavioural research. However, below are listed the different types of
research based on the three systems discussed above:
I. Basic and applied research;
II. Qualitative and quantitative research;
III. Experimental research;
IV. Quasi-experimental research;
V. Non-experimental quantitative research;
VI. Historical research; and
1.5 Meaning of Historical Research
Historical research is a systematic process of describing, analysing and interpreting
the past based on information from selected sources as they relate to the topic under
study. The main issue of the historical method, like other types of research, is not the
12
accumulation of facts, but the interpretation of the facts. The interpretation of
information is the main issue in any form of research. History is always dynamic
therefore it defines the role of the historical researcher. Historical researcher makes in-
depth investigation. To establish the meaning and relationship of events, researchers
should strive to get close to the source of the events. This researcher does this by
relying on documentary sources; also they at times study artefacts either alone or with
documentary evidence. Where possible, the researcher relies on only the primary
source of information to get to the root of the matter. Primary data on the other hand
consists of such original information about people who made history and influenced
human events as events they influence and that influenced them. The contemporary
scene of their days must be reconstructed as nearly as possible. The primary data
therefore consists of portraits, a record of the words they wrote and are reported to
have been spoken, the testimony of their friends and acquaintances, the personal
records they left behind, the objects they used, the houses and towns where they
worked and affected events important in their times and significant in their lives.
According to Leedy (1997), the use of primary data tends to ensure the integrity of the
study and strength its reliability.
1.5.1 Sources of Data
The sources of historical information are commonly classified as primary and
secondary. A primary source is an original or first-hand account of the event or
experience. They are eye-witness accounts. They are reported by an actual observer or
participant in an event. Secondary sources are accounts of an event that were not
actually witnessed by the reporter. Secondary sources may sometimes be used, but
because of the distortion in passing on information the historian uses them only when
primary data are not available.
Primary Sources
Generally, primary sources of data are of such materials as documents, relics or
remains and oral testimony. Some examples of these primary sources as they relate to
educational studies are stated below:
a. Official Records and Other Documentary Materials
Examples of these are records and reports of legislative bodies and state departments
of public instruction; principals, vice chancellors, deans, departmental heads,
educational committee, minutes of school boards, professionals and periodicals,
13
bulletins, courses of study, curriculum guides, textbooks, examination report cards,
pictures of teacher and pupils, personal files, samples of students’ work and
recordings among others.
b. Oral Testimony
Included in this category are – interviews with administrators, teachers and other
school employees, students and relatives, lay citizens and member of governing
bodies.
c. Relics
Included are buildings, furniture, teaching materials, equipment, decorative pictures,
textbooks, examination and samples of students work.
Secondary Sources of Data
Secondary sources are the reports of a person who relates the testimony of an actual
witness of participant in an event. The writer of the secondary source was not at the
scene of the event, but merely reports what the person who was there said or wrote.
Secondary source of data are usually of limited worth for research purposes because
of the errors that may result when information is passed from one person to another.
Most history textbooks and encyclopedia are examples of secondary sources, for they
are often several times removed from the original first- hand accounts of events.
Activity 1.2
Before you will continue studying the secondary sources of data, it is important that
you carry out this task. Based on what you have read so far, explain the primary
sources of data in a research.
Activity 1.2 Feedback:
Generally, primary sources of data are of such materials as documents, relics or
remains and oral testimony. Some examples of these primary sources as they relate to
educational studies are stated below:
a. Official Records and Other Documentary Materials
Examples of these are records and reports of legislative bodies and state departments
of public instruction; principals, vice chancellors, deans, departmental heads,
14
educational committee, minutes of school boards, professionals and periodicals,
bulletins, courses of study, curriculum guides, textbooks, among others.
b. Oral Testimony
Included in this category are – interviews with administrators, teachers and other
school employees, students and relatives, lay citizens and member of governing
bodies.
c. Relics
Included are buildings, furniture, teaching materials, equipment, decorative pictures,
textbooks, examination and samples of students work.
1. Inductive Approach: A researcher makes individual observation and then makes
generalization. Looking at the trend of individual observation one generalizes. For
example if a person is climbing a hill with thermometer in his/her hand, on top of hill
it is 80c and went up 300 feet 79, 78, 77 at every 300 feet it drops by 10c . These are
individual observations, generalization can be made that “the higher you go the cooler
it becomes” if a researcher observed the first Ghanaian, to dark be in complexion, ,
2nd Ghanaian dark, 3rd Ghanaian dark one can make generalization that Ghanaians are
always dark in complexion.
2. Deductive Approach: start from generalization and from it we make a number of
deduction e. g. All birds fly is a generalization, if one now see an object flying in the
sky, one then says it is bird, one starts with generalization and make deduction. The
approach can be right or wrong. The scientific method is often conceived as
inductive/deductive approach/method because we are always making observation and
reach conclusion and another person make use of conclusion to start making deduction
from it. When one reach the end of 1st step, one start the 2nd step. For example what is
the best method of teaching self-control in young children? One takes three methods
and find out which of the three methods is effective. If coping self-statement is
observed to be the best out of the three methods. Talking to oneself covertly is the
coping self-statement. One can now start under conclusion that if coping self-
statement is effective for teaching self-control in young children, it should be for
adolescent, adults, aged people, one now design another new experiment to see
15
whether the method will be effective for the ages mention above. Which is the starting
of the 2nd experiment. In this sense science is deductive inductive approach.
Another method of teaching self-control includes (1) cognitive restricting or value (2)
clarification. In value clarification one modifies a value support an unwholesome
behaviour, one tries to modify the values in individual.
Summary of Study Session 1
In Study Session 1, you have learned that:
• There are different definitions of research in education.
• There are some basic characteristics of research, goals of research.
• There are different types of research and one important form of research is the
historical research.
• Students have also learnt that historical research deals with meaning of past
events.
Self-Assessment Questions (SAQs) for Study Session 1
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 1.1 (tests learning outcome 1.1)
How can we correctly define ‘research’?
SAQ 1.2 (tests learning outcome 1.2)
There are many characteristics of a good research, enumerate at least five of these
features.
Box 1.1: Secondary Sources of Data
• Secondary sources are the reports of a person who relates the testimony
of an actual witness of participant in an event. The writer of the
secondary source was not at the scene of the event, but merely reports
what the person who was there said or wrote.
• Secondary source of data are usually of limited worth for research
purposes because of the errors that may result when information is
passed from one person to another.
16
SAQ 1.3 (tests learning outcome 1.3)
Can you list at least four goals of a research?
SAQ 1.4 (tests learning outcome 1.4)
Can you briefly describe the concept of historical research?
Notes on the Self-Assessment Questions (SAQs) for Study Session 1
SAQ 1.1: Research is a structured investigation that utilizes acceptable scientific
methodology to solve problems and create new knowledge that is generally
applicable.
SAQ 1.2: Research has the following characteristics:
- emphasises the development of generalisations, principles or theories that will
be helpful in predicting future occurrences;
- is based upon observable experience or empirical evidence;
- involves gathering of new data from primary or first-hand sources or using
existing data for a new purpose;
- is often characterised by carefully designed procedures that apply rigorous
analysis;
- requires expertise;
SAQ 1.3: Deductive Approach: start from generalization and from it we make a
number of deduction e. g. All birds fly is a generalization, if one now see
an object flying in the sky, one then says it is bird, one starts with
generalization and make deduction. The approach can be right or wrong.
SAQ 1.4: Historical research is a systematic process of describing, analysing and
interpreting the past based on information from selected sources as they
relate to the topic under study. The main issue of the historical method,
like other types of research, is not the accumulation of facts, but the
interpretation of the facts. The interpretation of information is the main
issue in any form of research. History is always dynamic therefore it
defines the role of the historical researcher.
17
Study Session 2:Theories, Hypotheses Generation and Testing
Expected duration: 1 week or 2 contact hours
Introduction
This chapter will explain how to formulate hypotheses and theories which are all
components of scientific knowledge. A good research follows the scientific approach
in its implementation. A scientific approach is a special organised form of all
reflective thinking and inquiry.
Learning Outcomes for Study Session 2
When you have studied this session, you should be able to:
2.1 Explain the term ‘hypothesis. (SAQ 2.1)
2.2 (SAQ 2.2)
2.3 Describe the characteristics of hypothesis. (SAQ 2.3)
2.4 Enumerate the functions of hypothesis. (SAQ 2.4)
2.1 Generating Hypotheses
To start with, it is important to tell you the meaning of hypotheses and the
characteristics of hypotheses. Hypotheses may be developed to provide more
specificity and direction. Hypotheses may be derived directly from the statement of
the problem. They may be based on the research literature, or in some cases, they may
be generated from data collection and analysis. A hypothesis is an assumption or a
guess at the solution to a problem or the status of the situation. In one word,
hypothesis take on some of the characteristics of a theory, which is usually considered
a larger set of generalisations about a certain phenomenon. Thus, a theory might
include several hypotheses. Logically, the approach is to proceed so that a decision
can be made about whether or not the hypotheses are tenable. This is called testing the
hypothesis; the results of such a test either support or refute the hypothesis.
18
2.2 Characteristics of Hypothesis
There should be a definite reason for the hypothesis either from a theory or from some
evidence that it is a useful and valuable hypothesis. Typically, a hypothesis states a
relationship or effect between variables and this should be done in a straightforward
and clear manner. Finally, a hypothesis should be testable; indeed much of research is
directed towards testing hypothesis. A weakness of many hypotheses is that they are
too broad to pinpoint the specific problem under study.
2.2.1 Formulating Hypothesis
1. Hypothesis should be clearly and unambiguously stated: this means that a
hypothesis should not contain terms with double or doubtful meaning.
Moreover, it should be stated in simple terms – that is, only a few concepts and
their relationships should be incorporated into one hypothesis at a time. This
will reduce the complex nature of the hypothesis thereby making it easy to
understand and easy to obtain information for testing it. A hypothesis should
clearly specify outcomes which can be tested directly.
2. Thus, if any word implies many concepts in a problem understudy, then more
than one hypothesis should be stated to ensure that more concepts than
necessary are not built into only one hypothesis.
3. Hypothesis should state the relationship between two or more variables: this
third characteristics implies that main features of a problem should be brought
into a cause and- effect or other forms of relationships. This means the
hypothesis would be to specify a guessed relationship between variables.
4. It should be testable: This refers to the verifiability of a hypothesis. This means
that a hypothesis should be stated in an operational form by showing that the
main concepts exist such that the claims of the hypothesis could be confirmed
by empirically collecting information. This also implies that the variables
should be measurable and accessible within a reasonable time. Stating a
hypothesis that would require waiting longer than necessary for certain
variables in the hypothesis to mature should be avoided. Variables that cannot
be measured using known and available tests or techniques are not encouraged.
19
Activity 2.1
At this moment, it is important that you write down some points in your note in form
of classroom activities. In own words, write down the characteristics of an hypothesis.
Make sure you do this, it will help you monitor how you progress with your learning.
Activity 2.1 Feedback:
Basically, the following are the characteristics of hypothesis:
1. Hypothesis should be clearly and unambiguously stated
2. If any word implies many concepts in a problem understudy, then more than
one hypothesis should be stated to ensure that more concepts than necessary are
not built into only one hypothesis.
3. Hypothesis should state the relationship between two or more variables
4. It should be testable: This refers to the verifiability of a hypothesis.
2.2.2 Sources of Hypotheses
Some of the sources of hypothesis are:
1. Personal Experience: the personal experience of the researcher affects the variety
and quality of guesses he can generate for solving the problem understudy.
2. Literature Review: A review of literature will expose the research workers to the
experiences of others and especially to the findings of previous studies which will
likely raise new questions as well as providing the research needed information for
reducing tentative answers to the current question under study. A review of literature
also gives him new ideas for improving the quality of his hypothesis.
3. Theory: the vast array of theory developed in social sciences, humanities and
behavioural science could be the basis for generating hypothesis by the researcher. A
theory based hypothesis can be derived from theory-based problem.
4. Logic: a hypothesis can be sourced from logic or the art of reasoning – that is, by
means of deductive and inductive reasoning. If the problem being studied is to
determine the relationship between X and Y, a hypothesis can be generated by either
the deductive or the inductive process of reasoning. Deductive reasoning was first
introduced by Aristotle (Cohen and Manion, 1989) and it is made up of a major
premise based on a self-evident proposition; a major premise providing a particular
20
instance and a conclusion. It assumes that new knowledge can be acquired or
problems solved through a sequence of normal steps from general to particular.
Tuckman explains it as follows: When general expectation about events based on
presumed relationships between variables are used to arrive at more specific
expectations (or anticipated observations), that process is called deduction. On the
other hand Mouly (1978) describes Francies Bacorts inductive reasoning by stating
that if one collected enough data without preconceived notion about their significance
or orientation thus maintaining a complete objectivity, inherent relationships
pertaining to the general case would emerge to be seen by the alert observation. That
general statement of relationship emerging from combined specific observation is the
hypothesis from inductive process.
Figure 2.1: The Sources of Hypothesis (Photo from www.slideshare.net)
2.3 Functions of Hypothesis
A hypothesis is a very important tool of research in the humanities, educational, social
and behavioural sciences researches in the following ways:
1. It directs attention to the problem. That is, hypothesis helps bring problem to
sharp focus by making clearer the important issues in the problem. The
researcher will now be in a better position to understand more, the problem he
is about to solve.
2. It guides the research by delimiting and directing the search for evidence
pertinent to the solution of the problem. In this regard the research worker will
collect only information that is relevant instead of dissipating energy on
21
aimless search which leads to the collection of unnecessary data that may have
little or no bearing on the solution of the problem.
3. It provides the frame work for deciding on appropriate research design
especially the tools and techniques for data collection and analysis.
4. It is a powerful and invaluable tool for the advancement of knowledge. This is
because hypothesis is testable.
2.3.1 Characteristics of hypotheses
Good hypotheses must fulfill the following conditions:
- They must have direct bearing on the problems stated.
- They should state the relationship or difference between two or more variables.
- They should be clearly and unambiguously stated.
- They should be testable such that their implications can be deduced.
- The guessed solution in a hypothesis should be reasonable such that it does not
form an open conflict with studies that have been confirmed, validated and
established Hypothesis when formulated lies at different level of specificity
which is related to conceptual and operational definition. Some hypothesis can
be conceptually stated some can be operationally stated and so be concrete. We
have different levels of specificity i.e low and high specificity.
Activity 2.2
Attempt this task and know whether you are getting the content right or not. Write
down the sources of hypotheses.
Activity 2.2 Feedback:
Some of the sources of hypothesis are:
1. Personal Experience: the personal experience of the researcher affects the
variety and quality of guesses he can generate for solving the problem
understudy.
2. Literature Review: A review of literature will expose the researcher to the
ideas and opinions of other scholars
3. Theory: the vast array of theory developed in social sciences, humanities and
behavioural science could be the basis for generating hypothesis by the
researcher. A theory based hypothesis can be derived from theory-based
problem.
22
4. Logic: a hypothesis can be sourced from logic or the art of reasoning – that is,
by means of deductive and inductive reasoning.
2.5 Types of Hypothesis
Hypotheses can be stated in two different forms either in a directional manner or in a
‘null’ form. Examples of directional hypotheses are:
i. Students attending private secondary schools Speaks good English better than
those students attending public secondary schools.
ii. Students from low socio –economic background do not perform academically
better than the students from high socio- economic background.
Null Form:
a. There is no significant difference in the spoken English of students in the
private and those in the public secondary schools.
b. There is no significant difference in the academic performance of students from
low socio- economic background and students from high socio –economic
background.
Hypotheses could also be stated in a correlational manner, such as: “There is no
relationship between students’ birth order and academic performance. The concise
title of the problem, the research objectives, the research questions and the hypotheses
constitute the very important part of the research known as the research problem. If
there is no problem, there will be no research because research is all about solving the
identified problem. Hypotheses are classified according to how they are stated. These
are: research hypotheses (which are stated in declarative form) and statistical
hypotheses (which are stated in null or no effect or negative form and alternate form).
Example of methods of formulating hypothesis could be seen in the three types of
hypothesis.
1. Athletic who engage in constant practice will show greater/better competition
performance than those not engaging in constant practice.
2. There will be no difference between the competition performances of Athletics
engaging in constant practice and those not engage in constant practice.
3. Athletics engaging in constant practice will show better performance in
competition than those not engage in constant practice.
a. The 1st one is called positive directional hypothesis.
23
b. The 2nd one is null hypothesis.
c. The 3rd one is negative directional hypothesis.
The null hypothesis comes from German word null which means zero. It is called
Zero hypothesis and is also hypothesis of no difference.
Beginning researchers are advised to stay clear of directional hypothesis to avoid bias.
They are suppose to use the null hypothesis so as to be neutral to avoid bias.
Figure 2.2: Null VS Alternative Hypotheses (Photo from www.skiripsi-fkip-
inqqris.blogspot.com)
i. The Research Hypothesis:
This states the relationship or difference expected between two variables. It can be
written in two forms, non-directional and directional. The non –directional shows that
a relationship or difference exists. For instance, the hypothesis can be stated as:
There is a significant difference in the level of performance of students after reduction
of test anxiety after exposure to cognitive restricting techniques .The level of
performance of students improved after the use of instructional games. The non-
directional hypotheses are tested with two-tailed test of significance while the
directional – hypotheses are tested with one-tailed test of significance.
ii. The Statistical Hypothesis
This is so called because it is the hypothesis tested in a research. It states a procedure
of ‘no effect’ in influencing the outcome. A null-effect means that there is no
relationship or difference between the variables under study. It implies that any
occurring relationship is a chance – relationship and not a true relationship. For
instance:
There is no significant difference in the level of HIV awareness of the students before
and after receiving HIV education. One major disadvantage of null hypothesis is that
24
it does not express the researcher’s true expectation with regard to the result of the
study. In order to solve this major crisis, an alternate hypothesis is stated: this
hypothesis is the non-directional research hypothesis and it is stated alongside the null
hypothesis.
Summary of Study Session 2
In Study Session 2, you have learned that:
Hypotheses link theories to particular concrete settings; the heart of the research
project is the problem.
Hypotheses can be stated in the directional and null forms.
A good hypothesis should have some basic characteristics to help researchers do a
smooth research exercise.
Self-Assessment Questions (SAQs) for Study Session 2
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 2.1 (tests learning outcome 2.1)
How will you define theory?
SAQ 2.2 (tests learning outcome 2.2)
Can you explain the meaning of hypothesis in a research?
Box 2.1: Types of Hypotheses
• Hypotheses can be stated in two different forms either in a directional
manner or in a ‘null’ form.
Example of directional hypotheses is:
Students attending private secondary schools Speaks good English better
than those students attending public secondary schools.
• Null Form: Example of null hypothesis is:
There is no significant difference in the spoken English of students in the
private and those in the public secondary schools.
25
SAQ 2.3 (tests learning outcome 2.3)
A good hypothesis should possess some characteristics, can you describe at least four
of these features?
SAQ 2.4 (tests learning outcome 2.4)
Briefly enumerate the functions of hypothesis in research.
Notes on the Self-Assessment Questions (SAQs) for Study Session 2
SAQ 2.1: A hypothesis is an assumption or a guess at the solution to a problem or the
status of the situation. In one word, hypothesis take on some of the
characteristics of a theory, which is usually considered a larger set of
generalisations about a certain phenomenon.
SAQ 2.2: The sources of hypothesis include;
Personal Experience
Literature Review
Theory
Logic
SAQ 2.3: There should be a definite reason for the hypothesis either from a theory or
from some evidence that it is a useful and valuable hypothesis. Typically, a
hypothesis states a relationship or effect between variables and this should
be done in a straightforward and clear manner. Finally, a hypothesis should
be testable; indeed much of research is directed towards testing hypothesis.
SAQ 2.4: It directs attention to the problem.
It guides the research by delimiting and directing the search for evidence
pertinent to the solution of the problem.
It provides the frame work for deciding on appropriate research design
especially the tools and techniques for data collection and analysis.
It is a powerful and invaluable tool for the advancement of knowledge. This is
because hypothesis is testable.
26
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi: Prentice
Hall. PP 20 – 23.
William, W. & Stephen, G.S. (2005). Research Methods in Education (8th Edition). Boston:
Allyn & Bacon, PP 1 – 18.
27
Study Session 3: The Research Problems and Literature Review
Expected duration: 1 week or 2 contact hours
Introduction
In this session, you will learn how to write research problems and review literature. In
the review of literature, researcher attempts to gather information relevant to the
research problem at hand. The process centres on three questions:
1. Where is the information found?
2. What should be done with information after it has been found?
3. What is made of the information?
The first question deals with the specific sources, both electronic and hard copy. For
most students, these sources can be found in or obtained through the library and the
internet. Finding the information often involves using reference works such as indexes
of periodical, literature.
Learning Outcomes for Study Session 3
When you have studied this session, you should be able to:
3.1 Explain the term ‘research problem’. (SAQ 3.1)
3.2 Differentiate between assumption and significance of the study. (SAQ 3.2)
3.3 Enumerate the factors related to time constraints in a study. (SAQ 3.3)
3.4 Discuss three sources in review of the literature (SAQ 3.4)
3.1 The Research Problem
The heart of the research project is the problem. The first important issue in a research
is to articulate an acceptable problem. Whatever the student does should have one
purpose: to formulate a problem that is carefully phrased and represents the single
goal of the research efforts. The question that comes to the mind of every researchers
at every step in the investigation is to ask “what am I doing, and for what purpose am
I doing it?” Selection of a research problem does not necessarily mean it is adequately
stated. A problem may be stated broadly and then systematically restricted through a
28
review of the literature in the initial stages of the research effort. It is better to work in
this direction than to begin with a problem that is too narrow and then attach pieces to
expand it. Research problems may be stated in a declarative or descriptive manner or
in question form. Majority of the researcher prefer the question form, but either form
is acceptable. The question form may aid in focusing the problem, and it is especially
effective when sub -problems, are included within the larger research problem. The
most important characteristic of the problem statement is that it must provide adequate
focus and direction for the research .Example of adequate statement might be: “A
study of the effects of birth order on the academic performance of Senior secondary
school students in Oyo state’’ Or in form of question such as “What are the effects of
birth order on the academic performance of Senior secondary school students in Oyo
state’’ ”A good statement of the problem should provide the researcher with direction
in pursuing the research. The basic directive for the statement of the problem is:
Problem should be stated in a complete grammatical sentence in few words as much
as possible. A problem could be: Having stated the purpose, the next will be the
research questions. Once the research objectives are stated, both the research
questions and hypotheses will be easier to state.
Figure 3.1: Identification of Research Problem (Photo from www.slideshare.net)
3.2 Components of Setting a Problem
There are components of the setting of the problem, which help in shaping and
refining the problem and provide a focus and manageable scope for it. Each
component is discussed below.
29
3.2.1 Delimitation
These are the boundaries of the study. For example, a study on “Signs and Symptoms
of Juvenile delinquency among Junior Secondary Students in Akinyele Local
Government Area of Oyo State students with a defined location such as state, local
government or a school or schools in a community. Often, conclusions are not to be
extended beyond this population sampled unless the researcher can justify beyond
reasonable doubt that the additional groups have the same characteristics. Also, this
aspect of the study should provide full information of what the researcher intends to
do and not to do. For example, in choosing the subjects for the study, the researcher
may decide not to include all students who transferred from other schools and those
who are not resident of Akinyele LGA. The problem of delimitation often arises
because of large area that might be covered by the statement of the problem. Most
beginners tend to write statement of the problem that they could hardly manage. Best
and Khan (1995) provided some distinction between limitations and delimitations.
Limitations are those conditions beyond the control of a researcher that may place
restrictions on the conclusions of the study and their applications to other situations,
such as administrative policies that do not allow the use of more than one class in an
experiment, an instrument that has not been validated or due to some reasons, inability
to randomly select and assign subjects to experimental and control groups.
Delimitations on the other hand are the boundaries of the study.
o For more than two months now, Henry has been struggling to find a way to
properly begin his research work for his final year project. He taught the first
step should be to look for relevant literature on different issues in his area of
specialisation. But, he was confused when one of lecturers told him that was
not the right way to begin a research. So, he decided to consult his uncle, who
happened to be a lecturer in the faculty of education in one of these standard
universities. What first step do you think his uncle should tell him to take to
begin the research?
• For a student to carry out an acceptable research, the starting point should be
identification of a research problem. The first important issue in a research is to
articulate an acceptable problem. Whatever the student does should have one
purpose: to formulate a problem that is carefully phrased and represents the
single goal of the research efforts. The question that comes to the mind of
every researchers at every step in the investigation is to ask “what am I doing,
30
and for what purpose am I doing it?” Selection of a research problem does not
necessarily mean it is adequately stated. A problem may be stated broadly and
then systematically restricted through a review of the literature in the initial
stages of the research effort.
3.2.2 Assumptions
Assumptions are statements of what the researcher believes to be facts but cannot
verify. A basic assumption in studies that involve the completion of questionnaires is
that the respondents will complete the questionnaires to the best of their abilities and
provide honest and reliable information without any biases. Suppose the researcher
wants to determine by means of experimental method whether the treatment produced
the result hypothesised, the researcher must assume that the test is valid, and if the
treatment is a class test or examination, that it is within the capability of the
examinees. Assumptions are what the researcher takes for granted. If the assumptions
of a researcher are known, that makes it easier to evaluate the conclusions of that
result from such assumptions. You should note that in research, you do not leave
anything to chance; therefore, all assumptions that have bearing on a problem must be
clearly and unreservedly spelt out. According to Leedy (1997), to discover
assumptions, you have to ask yourself “what am I taking for granted.
3.2.3 Significance
In an attempt to set the problem, the researcher needs to state clearly the reasons for
undertaking the study. To address this aspect of the study, you must ask yourself such
questions like; of what use is the study? Who are the beneficiaries? For educational
studies, the beneficiaries of the findings are mostly the students, teachers, educators,
ministry of education and its various organs, parents, the public, examination bodies
etc. It is therefore necessary to state how the study will be useful to some, if not all the
groups. For example, a study on “causes of poor performance of students in secondary
school mathematics examinations” will be useful to:
i. teachers to assist them perhaps in their approaches to the teaching and learning
of mathematics.
ii. examination bodies – to improve their techniques of setting questions and
ascertain the students views as to the difficulty level of the questions and such
other related issues as the time allocated to the questions;
31
iii. parents, in terms of the type of support they should provide for their
children/wards to ensure better performance and curriculum developers, to
assist in curriculum review of programme etc.
iv. In writing the significance of the problem, attention should be concentrated on
each of the hypothesis/research questions.
3.2.4 Definition of Terms
It is important to define all unusual terms that could be misinterpreted. This definition
helps to establish the frame of reference with which the researcher approaches the
problem. The variables to be defined should be defined in operational terms. Such
expressions/terms as academic performance, intelligence, delinquency when used in
special contexts within a study, must be operationally defined. Academic grades
assigned by teachers or scores on standardised achievements are operational
definition. Without knowing explicitly what a term means, you cannot evaluate the
research or determine whether the researcher has carried out what, in the problem, was
mentioned as the major objectives of the study. The definitions must interpret the term
as it is employed in relation to the study. The researcher should bear in mind that in
defining a term in their study, they determine whatever they wish the term to mean
within the context of the problem or its sub-problems.
3.2.5 Feasibility of a Research Problem
By the time researchers have selected, shaped and refined a research problem, the
problem should be sufficiently clear that a consideration of practical issues involving
its feasibility is in order. Practical considerations of what can reasonably be
accomplished given the time and resources available can force researchers sometimes
painfully to reduce the scale of a project. A careful and honest appraisal of the time
and money required to accomplish a study will be useful in determining the feasibility
and reveal if a change in aims and objectives is necessary. The appraisal aspects of a
project’s feasibility centre primarily on two related issues – time and money. Also,
other issues of importance include special aptitude of the researcher to the problem,
availability of data and special equipment.
Activity 3.1
This session has so far exposed you to the basic issues in research problem. So, in
your own words, write down define the term ‘assumption’
32
Activity 3.1 Feedback:
Assumptions are statements of what the researcher believes to be facts but cannot
verify. A basic assumption in studies that involve the completion of questionnaires is
that the respondents will complete the questionnaires providing to the best of their
abilities, honest and reliable information without any biases.
3.3 Time Constraints
In determining a research, one of the major considerations is whether there will be
sufficient time to complete adequately what the researcher intends to do. This is very
important particularly the researcher is working within a specified time such as
completing a study or dissertation for a degree.
Some of the major factors related to time in a study are:
i. Population: If the population has characteristics that are fairly widespread, then a
sufficient number of people will be readily available from which to collect data. If
however a study focuses on people with special characteristics (e.g. childless couple)
that are somewhat rare, then data collection might be problematic and time
consuming.
ii. Proper Development of Measuring Instrument: All techniques for gathering data
should be tested before the actual study is conducted. This surely is time consuming.
If a study involves a host of pre-tests, then it means that a lot of time will be
consumed in modification and administration of the test as well as analysis of data
collected.
iii. Time Required for Data Collection: The time required may range from a short
period of less than one hour to years depending on the type of study. Due to the
variability of the time required for data collection, time should be properly looked into
when considering the feasibility of a study.
iv. Time for Analysis of Data: Generally, the less structured a questionnaire is, the
more time the analysis consumes. Just like the time for data collection, the time for
data analysis should also be considered by the researcher.
Financial Constraints
The financial expenditures associated with a study is another constraint on feasibility.
The major areas that involve finance are:
33
i. Production of questionnaire;
ii. Allowance(s) to research Assistant(s);
iii. Transportation cost;
iv. Cost of Analysis of Data
v. Office supplies and equipment.
vi. Use of computer.
Anticipating and Avoiding Problems
Problems related to time and financial considerations arise virtually in all research
studies, but their impact on the outcome of the research can be minimised if they can
be anticipated as much as possible, especially during the planning stage, when the
details of the study are easier to change. An experiment researcher may be able to
identify trouble spots in his/her proposal and suggest modifications to avoid them. It is
also necessary to obtain whatever permissions or consents that may be needed early in
the planning stage of the study. Also, it is necessary to conduct a pilot study, which is
a preliminary run through of all the procedures that surface during the main study.
Any problems that surface during the pilot study can then be dealt with before the
main study. If inadequate time or money is the problem, perhaps the project can be
scaled down. It is possible to reduce the sample size, or the number of hypotheses,
mail questionnaires instead of personal distribution of questionnaires.
3.4 Purpose of Literature Review
According to Leedy (1997), review of literature has several purposes. Primarily, it is
to assist the researcher in confronting the research problem. In any study, your own
problem is central. Everything you embark upon is because it helps you to resolve
your own problem or answer your research questions. When you know what others
have done, you will be in a position to investigate your chosen problem with deeper
insight and complete knowledge. Other benefits of literature review are:
i. Informing the researcher of what has already been done in the area.
ii. More specifically, limiting and identifying the research problem and possible
hypotheses.
iii. Providing possible research design and methodological
iv. Procedures that may be used in the research study.
v. Providing suggestions for possible modifications in the research to avoid
unanticipated difficulties.
34
vi. Identifying possible gaps in the research.
vii. Providing a backdrop for interpreting the results of the research study.
3.4.1 The Activities of the Review of the Literature
The existing body of knowledge related to the research problem provides information
for identifying the problem. The review of the literature itself consists of several
specific activities that, to a large extent, also take place in a sequence. Like most
activities or steps in a process, there are efficient and inefficient ways to review the
literature. Rather than going to the library or the internet and haphazardly beginning to
take notes, the researcher should follow a systematic process. Another important
procedural point in conducting activities is, for each activity, to do as complete and
accurate an initial job as possible. If a report is relevant enough to include in the
review, sufficient information should be obtained from it so that there is no confusion
later about what was done (conditions, procedures, individuals involved etc.) or about
the results. Doing the review of literature in this suggested manner, will not only
reduces frustration but also saves time.
Activity 3.2
It is important to reflect on all you have read so far. Write down the basic of literature
review in a research.
Activity 3.2 Feedback:
Primarily, it is to assist the researcher in confronting the research problem. In any
study, your own problem is central. Everything you embark upon is because it helps
you to resolve your own problem or answer your research questions. When you know
what others have done, you will be in a position to investigate your chosen problem
with deeper insight and complete knowledge.
3.5 Sources of Information
Studies are published in books, periodicals, technical reports, conference proceedings
and academic theses. Most of the recent reports are available in electronic form and
can be accessed at various institution’s library.
3.5.1 The Library
The library has some basic functions as a repository of writings, books and
manuscripts; a kind of literary museum where manuscripts and books are kept and
35
added as the slow advance in knowledge become available. In the 20th Century and
especially in the latter half of the century – the role of the library changed. An
explosion of information, and knowledge occurred. The availability of new knowledge
caught many libraries unprepared. Most are struggling to revamp their original
purpose of providing a repository of written thought and factual information. Libraries
have come to grip with two important facts. First, they can no longer hold all the
information written within their walls. With the vast amount of information being
produced daily, it is unfeasible to hold a majority of it in any one location. The space
cannot be afforded, nor can the ever increasing purchasing cost. Secondly, library
patrons are becoming more sophisticated in their wants and desires. Ease and speed of
access are two important priorities. New ways of storing vast amount of information
(e.g. CD-ROM) are replacing and augmenting the shelves of books and periodicals
that line up the walls of the library. The library must continue to evolve. With
advances in telecommunications, libraries may exist literally without limits. As a
powerful tool for the researcher, the library continues to evolve. In well organised
libraries, in place of card catalogue, you have computer terminals and keyboards.
These are standard equipment on most institution’s libraries. With this, the researcher
can access vast amount of information, search countless online databases and
manipulate factual information with a facility that saves time, increase accuracy and
boggles the mind at its efficiency. Research has become less disciplinary and more
global in both problems and it’s methodology. This change has created demands that
libraries had never faced before. In recent times, most college libraries have replaced
the card catalogue with an electronic database containing the total resources in the
library. Now, you can sit before a keyboard and type in the area of knowledge or the
title of the book. These days, the principal publishers of indexes are encoding their
information electronically. The use of the computer facilities for information
gathering is just at its gestation stage in most libraries in Nigeria.
36
Figure 3.2: People Reading in the Library (Photo from www.glenbrook225.org)
3.5.2 Online Computer Searches: Database Access to Literature
CD-ROM journal indexes is a valuable tool in identifying and locating different
references for the researcher. In most cases, the CD-ROM searches are carried out by
the end-user and they pertain to one specific database. Just like CD-ROM database
searches, online searches involve a computer looking for a specific keyword by
reading the text of a journal article title, abstract, as well as all other words in the
record. “Reviewing all the abstracts over many years would be utterly impossible for a
human being but it is easy for a computer” (Krathwol, 1993). Whereas the CD-ROM
search usually focuses on a single database, the online search can have access to
approximately 4,000 databases. The database provides access to literally billions of
records. Advantages of online computer search are:
- access to large number of record of the world’s published literature;
- reduced time required in comparison with a manual search;
- frequent update that narrows the gap between publication of documents and
user access.
- search requests tailored to use needs immense breadth of search
- topic that could not be performed manually; and fast turn -around from request
to results.
Limitations of online computer search are;
- Possibility of unavailability of cited documents, some charges for access to
various databases; and
- No guarantee that human input of indexed materials and
- keywords will provide complete retrieval relevant records;
- Databases documentation that varies in quality and coverage;
37
- Some training required by user;.
- Turn- around time dependent on system availability and competency of the
searcher.
- However, most libraries in Nigeria are yet to embark on computerised service
to the public.
3.6 Analytical and Theoretical Frameworks
Reading as much as time permits about your topic may give you ideas about approach
and methods which had not occurred to you, and may also give you ideas about how
you might classify and present your own data. It may help you to devise a theoretical
framework as a basis for the analysis and interpretation of data. It is not enough to
collect facts and to describe what it is. All researchers collect facts but must organise
and classify them into a coherent pattern. As the researcher is reading previous studies
he/she, gets into the habit of examining how authors classify their findings, how they
explore relationship between facts and how facts and relationship are explained.
Methods used by others may be unsuitable for their purposes, but they may give them
ideas about how they might categorise their own data, and ways in which they may be
able to draw on the work of other researchers to support or refute their own argument
and conclusions.
Activity 3.3
It is important to reflect on all you have read so far. Write down the sources of
information in literature review.
Activity 3.3 Feedback:
Basically, there are two sources of information that students could use in reviewing
relevant literature. These are:
The Library
The library has some basic functions as a repository of writings, books and
manuscripts; a kind of literary museum where manuscripts and books are kept and
added as the slow advance in knowledge become available. In the 20th Century and
especially in the latter half of the century – the role of the library changed. An
explosion of information, and knowledge has occurred. The library must continue to
evolve. With advances in telecommunications, libraries may exist literally without
38
limits. As a powerful tool for the researcher, the library continues to evolve. In well
organised libraries, in place of card catalog, you have computer terminals and
keyboards.
Online Computer Searches: Database Access to Literature
CD-ROM journal indexes is a valuable tool in identifying and locating different
references for the researcher. In most cases, the CD-ROM searches are carried out by
the end-user and they pertain to one specific database. Just like CD-ROM database
searches, online searches involve a computer looking for a specific keyword by
reading the text of a journal article title, abstract, as well as all other words in the
record. “Reviewing all the abstracts over many years would be utterly impossible for a
human being but it is easy for a computer” (Krathwol, 1993).
3.7 How to Write on the Related Literature
As individual begin to collect information, articles, etc. about their study, they will
need some type of structure to organise their information. One easy way to organise is
to set up a database that includes records of each article and book you possess. These
records can include fields containing the title, author, date, journal reference, special
keyboard, an identification number, as well as notes you have taken while studying
the item. These items can be filled or neatly packed together. When any is needed, a
quick search of the database should reveal where they are located.
Computerise as much as possible. If you have a personal computer, open a file and
computerise your data as much as possible. You can always work out the format of
presentation e.g in an alphabetical order.
Be systematic and thorough: In research, you should “make haste slowly” be careful
not to make careless, half complete notes\ that, when consulted later – are lacking in
essential information that they are practically useless.
The original time spent seeking out the item can be wasted, it would be much better to
take care and do the job right in the first place. Little is gained by rushing so that you
fail to get adequate or correct information the first time around and cannot read it
when referring to it later. After you have amassed an impressive literature, you will
need to arrange your information in some order.
39
Get the proper psychological orientation: Be clear in your mind what you are trying
to do. The review of literature is a discussion of previous publications that bear
directly on the topic you are investigating. Consider the review of related literature to
be a review with someone about what others have written in relation to what you plan
to do. Viewing the literature section in this way, will help you develop the proper
psychological perspectives and will help you see your own effort in relation to other
researchers.
Have a Plan: Writing a review of literature takes planning and organisation. This
effort requires structure, unity and coherence; perhaps a careful consideration of the
problem should suggest relevant areas for discussion and indicate the direction that the
discussion of the related literature should take.
Emphasis relatedness: Keep your reader constantly aware of how the literature you
are discussing is related to your problems. Point out precisely what that relationship is.
Remember that you are writing a review of the related literature. Literature review
should not create a chain of pointless, isolated summaries of the writing of others This
is not a discussion of related literature as no attempt is made to demonstrate the
relatedness of the literature to the present problem. Whenever you cite a study,
account for it in terms of the problem you are investigating. Specifically explain
precisely what the relationship is. Unless you can establish such a relationship, you
may consider whether there is need to include the study at all.
Review the literature, do not reproduce it: A sound discussion shows the
relationship of the problem to the broad environment of similar studies done by others.
This is the sole purpose of the literature discussion. To show the connection between
the discussion of the literatures that others have produced and the investigation you
are carrying out, Leedy (1997) has made the following recommendations: Write out
your problems at the top of the page where you cannot lose sight of it. By this, it will
continue to remind you of the central axis on which the entire study revolves. Divide
the problem by numbering its various parts.
Cite each specific study separately. Gather together all citations that refer to a
particular subdivision of the problem and the rationale for inclusion in the review of
literature.
40
Gather together all the citations that refer to a particular subdivision of the problem
that you have as many group as you have subdivisions of your main problem. Study
these groups in relation to each other. Write the review, label each section with a
heading whose wording contains the identical words found in the statement of the
problem. Summarise what you have said. A summary is necessary, in which you
gather up all that has been said and set forth in importance in terms of the research
problem.
Summary of Study Session 3
In Study Session 3, you have learned that:
In setting up a research, the following terms need to be addressed: limitation,
delimitation and significance of the study.
Also, assumptions and operational definitions of some terms and expressions need
attention. Time and money are also two major constraints to a study.
How to conduct a feasibility study was also discussed.
You found that time and money are two major constraints in a study.
You also discovered that identification of possible problems helps to strengthen the
study by reviewing it.
Self-Assessment Questions (SAQs) for Study Session 3
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 3.1 (tests learning outcome 3.1)
Can you define the term ‘research problem’?
Box 3.1: How to Write on the Related Literature
• As individual begin to collect information, articles, etc. about their study,
they will need some type of structure to organise their information.
• One easy way to organise is to set up a database that includes records of
each article and book you possess.
• These records can include fields containing the title, author, date, journal
reference, special keyboard, an identification number, as well as notes you
have taken while studying the item.
41
SAQ 3.2 (tests learning outcome 3.2)
Assumption and significance are importance aspects of a research, how will you
distinguish between the two?
SAQ 3.3 (tests learning outcome 3.3)
Can you enumerate the factors related to time constraint in a study?
SAQ 3.4 (tests learning outcome 3.4)
Discuss three sources in review of the literature.
Notes on the Self-Assessment Questions (SAQs) for Study Session 3
SAQ 3.1: The heart of the research project is the problem. The first important issue in
a research is to articulate an acceptable problem. Whatever the student does
should have one purpose: to formulate a problem that is carefully phrased
and represents the single goal of the research efforts.
SAQ 3.2: Assumptions are statements of what the researcher believes to be facts but
cannot verify. A basic assumption in studies that involve the completion of
questionnaires is that the respondents will complete the questionnaires to the
best of their abilities and provide honest and reliable information without
any biases. On the other hand, significance of the study tells us about the
essence of the research. In an attempt to set the problem, the researcher
needs to state clearly the reasons for undertaking the study. To address this
aspect of the study, you must ask yourself such questions like; of what use is
the study? Who are the beneficiaries?
SAQ 3.3: The factors include:
Population
Proper Development of Measuring Instrument
Time Required for Data Collection
Time for Analysis of Data
42
SAQ 3.4: The Library
The library has some basic functions as a repository of writings, books and
manuscripts; a kind of literary museum where manuscripts and books are
kept and added as the slow advance in knowledge become available.
Online Computer Searches: Database Access to Literature
CD-ROM journal indexes is a valuable tool in identifying and locating
different references for the researcher. In most cases, the CD-ROM searches
are carried out by the end-user and they pertain to one specific database.
43
Study Session 4: Concepts, Constructs and Variables
Expected duration: 1 week or 2 contact hours
Introduction
In this session, you will be exposed to the meaning and characteristics of concepts,
constructs and variables. Concepts are defined as ideas, persons, things, or events.
Concepts are often given both nominal definitions which explain their meaning and
operational definitions that indicate how they are measured. Constructs and variables
are to be discussed in this chapter. Accurate operational definitions are essential to
any research work. Such traits as intelligence, anxiety or motivation are not directly
observable and are generally referred to as constructs; implying that they are mere
constructions of the scientist imagination. On the other hand, variables are conditions
or characteristics that the researcher manipulates, controls or observes. Examples of
some important variables in behavioural science are sex, education, age class,
occupational mobility, verbal aptitude, anxiety, strength, intelligence, achievement
etc.
Learning Outcomes for Study Session 4
When you have studied this session, you should be able to:
4.1 Define the term ‘concept’. (SAQ 4.1)
4.2 Differentiate between concepts and constructs. (SAQ 4.2)
4.3 Describe the term ‘variable’ (SAQ 4.3)
4.4 Discuss the two major types of variables. (SAQs 4.4).
4.1 Defining Concepts
An important part of theories are concepts. These are mental constructs or images
developed to symbolise ideas, persons, things or events. Concepts are the building
blocks that are interrelated in propositions that form the explanatory statements of a
theory. Monnete et al (1994) state that scientific analysis involves two types of
definitions of concepts – each functioning at a different level of analysis and serving
different purposes. Concepts are groups of behaviour, objects or situation that have
44
common characteristics. It is a collection of behaviour that have something in
common. Concept could be concrete or abstract. e. g. if one is talking about type of
furniture use in the house, this group is called concrete concept because one can touch
it. They could be abstract; e. g. honesty is an abstract concept in research refers to a
group of behaviour that has something in common. Intelligence is also abstract
concept which is similar to construct but they are different. Concepts are groups of
behaviour, objects or situation that have common characteristics. It is a collection of
behaviour that have something in common.
Abstract concepts are also known as construct. A construct is an abstract concept. In
research we use a lot of construct or abstract. Before using them we must obtain their
indicators. Constructs have indicators which is a concrete behaviour which is
equivalent to that construct or abstract concept e.g. intelligence indicators include;
i. High mark in achievement test,
ii. Good memory
iii. Ability to solve problem
Motivation: indicators (i) long hour of reading
Without indicators, constructs cannot be measured.
Before obtaining indicators a construct must be defined.
Concept and construct are similar
An important step in moving from the abstract level of theory to the concrete level of
research is to give concepts operational definition: definitions that indicate the precise
procedures or operations, to be followed in measuring a concept. The process of
moving from nominal to operational definitions can be complex because concepts are
more general than abstract, and controversy often arises over exactly what they refer
to. So it should be evident that operationalising concepts can be difficult, complex and
sometimes, controversial.
4.2 Constructs
Construct cannot be seen, heard or felt. They can only be inferred by phenomena such
as test scores or by observed hostile or aggressive acts, skin responses, pulse rates, or
persistence at a task. A construct is a concept. But it has the added meaning of having
been deliberately and consciously invented or adopted for a specific purpose. For
example, concept, an abstraction from the observation of presumable behaviours. But
45
as a scientific construct, intelligence means both more or less than it may mean as a
concept. This means that scientists use it in different ways i.e. school achievement is
in part a function of intelligence and motivation. Also, constructs can be observed and
measured. The children can be observed by administering an intelligence test to them
while teachers can estimate the relative degrees of intelligence of their pupils.
o Busayo just finished a class on research methods where the lecturer told them
that concepts are strategic part of theories. The lecturer also many a distinction
between concepts and constructs, but because the class was too large, the
young man could not properly understand the basic issues raised in the
classroom. Fortunately for him, he had an uncle who happened to be a lecturer
in the faculty of education and he decided to consult him for more explanations
on the topic. If you were Busayo’s uncle, how would you enlighten him on this
topic?
• An important part of theories are concepts. These are mental constructs or
images developed to symbolise ideas, persons, things or events. Concepts are
the building blocks that are interrelated in propositions that form the
explanatory statements of a theory. Concepts are groups of behaviour, objects
or situation that have common characteristics. It is a collection of behaviour
that have something in common. Concept and construct are similar. Construct
cannot be seen, heard or felt. They can only be inferred by phenomena such as
test scores or by observed hostile or aggressive acts, skin responses, pulse rates,
or persistence at a task. A construct is a concept. But it has the added meaning
of having been deliberately and consciously invented or adopted for a specific
purpose.
4.3 Variables
Kerlinger (1977) defines a variable as a symbol to which we assign numerical values.
For example, if a researcher is interested in the effects of two teaching methods on the
science achievement of JSS three students. After the different teaching methods have
been implemented, the JSS three students involved would be measured with a science
achievement test. It is very unlikely that all of the JSS three students would receive the
same score on this test, so the score on the science achievement test becomes a
variable, because different individuals will have different scores. Therefore, the score
46
on the science achievement test is a variable. There is another variable in the example
above – the teaching method. In contrast to the science achievement test score, which
undoubtedly would be a scale with many possible values, teaching method is a
categorical variable consisting of only two categories, the two methods. So we have
different kinds of variables and different names or classifications for them. There are
many classification systems given in the literature; so many that there is a
considerable overlap and opportunity for confusion. The names we use are
descriptive, they play a role in research study. The variables described below by no
means exhaust the different systems and names that exist, but they are the most useful
for communicating about behavioural research.
4.4.1 Independent and Dependent Variables
Variables can be categorised as independent and dependent variables. This
categorisation is very important in research. An independent variable is the presumed
cause of the dependent variable i.e. the presumed effect. According to Kerlinger
(1977), the independent variable is the antecedent and the dependent is the
consequent. The independent variable is the variable manipulated by the research. If a
researcher decides to study the effects of teaching methods on students’ performance
in mathematics, you will then manipulate the method, the independent variable, by
using different methods. The dependent variables (Y) are the presumed effect which
varies concomitantly with changes or variations in the independent variable (X). It is
the variable that is not manipulated. Rather, it is observed for variation as a presumed
result of variation in the independent variable. You could notice from the discussion
that there are two types of independent variables, viz: Treatment and Organismic or
Attribute variables. Treatment variables are those factors that the researcher
manipulates and to which he or she assigns subjects. Attribute variables are those
characteristics that cannot be altered by the researcher. Examples are: sex, age, race or
tribe and already determined intelligence level. However, the researcher can decide to
include or remove them as variables to be studied. Variables are the conditions or
characteristics that the experimenter manipulates, controls or observes. They can be
divided into two, namely: independent and dependent variables.
The independent variables are the conditions or characteristics that the experimenter
manipulates or controls in his or her attempt to ascertain their relationship to observed
phenomena. It is the factor manipulated by the experimenter in his attempt to
47
establish, or to ascertain relationship to an observed phenomenal e. g. given knock to a
child on the head, give to another child, then watch their reaction. It is likened to a
stimulus which is the factor presented to a subject. The independent variable is usually
shown on the horizontal axis of the graph. It is always compared to a stimulus which
experimenter presented to the variables. In behavioural research, an independent
variable may be a particular treatment programme, a type of teaching material, a
reward or a period of exposure to a particular condition or an attribute such as: sex or
level of intelligence.
The dependent variables are the conditions or characteristics that appears, disappears
or changes as the experimenter introduces, removes or changes the independent
variables. It is the factor which changes as a result of the manipulation of the
independent variables that is, it is the response aspect of the result or outcome.
Dependent variables results from independent variables. It is the effect of independent
variables. It is always shown on vertical or Y axis e. g. cry is the result of knock given
to a child making noise. Cry is the dependent variables e. g. teaching through two
methods and then score the result of test the score is the dependent variable. The
dependent variable may be a test score, the number of errors in a comprehensive
exercise etc. The dependent variables are the measured changes in pupils’
performance attributable to the influence of the independent variable.
There are two types of independent variables: treatment and orgasmic or attribute
variables. Treatment variables are those factors that the researcher manipulates and to
which he or she assigns subjects. Attribute variables are those characteristics that
cannot be altered by the experimenter such as: age, sex, race or intelligence quotient
(IQ). The experimenter can decide to include them or remove them as variable to be
studied.
Activity 4.1
At this moment, it is important to take note of the difference between dependent and
independent variables in research. In your own words, distinguish between dependent
and independent variables.
Activity 4.1 Feedback:
An independent variable is the presumed cause of the dependent variable i.e. the
presumed effect. According to Kerlinger (1977), the independent variable is the
48
antecedent and the dependent is the consequent. The independent variable is the
variable manipulated by the research. The dependent variables (Y) are the presumed
effect which varies concomitantly with changes or variations in the independent
variable (X). It is the variable that is not manipulated. Rather, it is observed for
variation as a presumed result of variation in the independent variable.
4.4.2 Discrete and Continuous Variables
Discrete variables are variables with a finite number of distinct and separate values
e.g. sex, race, family size, number of days present. All these are discrete variables
because they can be measured only in a discrete set of units such as 2, 3, etc. No
meaningful measurement values lies between these distinct and separate values.
Continuous variables are those that at least theoretically can take an infinite array of
values. Age is normally measured in years, but it can also be measured in months,
days, hours, minutes and seconds. There is no theoretical limit to how precise the
measurement of age might be. For most behavioural research purposes, the
measurement of age in terms of years is quite appropriate, but age is nonetheless a
continuous variable. Nominal variables are by definition, discrete in that they consist
of mutually exclusive or discrete categories. Ordinal variables are also discrete. The
mutually exclusive categories of an ordinal variable may be ranked from low to high,
but there cannot be a partial rank. For example, in a study of the cadre of lecturers,
rank might be ordered 1 = 2 etc, but you cannot talk of 1.3 or 2.6 etc. Family size,
number of counts, etc. can only be whole numbers or discrete intervals. A continuous
variable can take an ordered set of values within a certain range. It is important to
recognise that a variable is continuous or discrete by its very nature and the researcher
cannot change that. It is however possible to measure a continuous variable by
specifying a number of discreet categories as it was explained for age, but this does
not change the nature of the variable itself. The nature of variable, discrete or
continuous, determines how it is used in data analysis. Sometimes discrete data are
treated as continuous in order to use statistical models, but care must be taken to
assure that the results will be meaningful.
49
Figure 4.1: The Variables (Photo from www.socratic.org)
4.4.3 Confounding Variables
Confounding variables are those aspects of a study or sample that might influence the
dependent variables (outcome measure) and whose effects may be confused with the
effect of the independent variable. Confounding variables are of two types: extraneous
variables and intervening variables.
i. Intervening Variables
In experimental research, certain variable that cannot be controlled or measured
directly may have an important effect upon the outcome. They intervene between the
cause and the effect. In an achievement test, the researcher may suspect that certain
variable may be influencing the student’s performance though they may not be
observed directly, such factors as anxiety, fatigue, motivation and English
comprehension ability. These are difficult to define operationally, but you cannot
ignore them, therefore, they must be controlled through the use of appropriate designs.
ii. Extraneous Variables
Extraneous variables (i.e. variables not manipulated by the researcher) that may have a
significant influence upon the results of a study. Many research results are
questionable because of the influence of these extraneous variables. In a study of
teacher’s competence and pupils’ performance, different teachers and whole classes
will be used. Such variables as academic ability, teacher personality, age and
readiness, constitute extraneous variables that could affect the result of the study. It
should be noted that for an extraneous variable to confound the result of a study, it
must be correlated strongly enough with both the independent and dependent variables
50
that its influence can be mistaken for that of the independent variables. It is almost
impossible to eliminate the influence of the extraneous variable totally, but a good
design could minimise its influence.
4.5 Constructs, Observables and Intervening Variables
The explanations provided earlier in this unit shows that there is a clear difference
between constructs and observable variables. Constructs are non-observable variables,
when operationally defined are observable. This is important because if you are not
aware of the level of discussion when talking about variable, you can hardly be clear
about what you are doing. Constructs are called intervening variables. Intervening
variable is a term invented to account for internal and directly unobservable
psychological processes that in turn account for behaviour. Intervening variable is an
“in the head” variable. It cannot be seen, heard or felt. It is inferred from behaviour
e.g. learning is inferred from among other things – improved performance. Anxiety is
inferred from test scores, skin responses and from heartbeat, etc. The researcher using
such terms should be aware that he or she is using invented ‘constructs’ the reality of
which has been inferred from behaviour. For example, motivation can only be judged
by observing behaviour. But it cannot be measured directly because it is on “in-the
head” variable, an intervening variable, an unobservable entity. This means that, to
judge ‘motivation’, you have to measure the presumed indicators of motivation and
not motivation itself. You must in other words always measure some kind of
behaviour, be it marks on paper, spoken words etc. and then make inferences about
presumed characteristics. This explains why it is always important to have operational
definition of these constructs to be sure that the investigation you are embarking upon
is quite clear to you and the readers will be able to follow and understand what the
research is all about.
Activity 4.2
At this moment, it is important to take note of the differences between discrete and
continuous variables. In your own words, distinguish between the two types of
variables.
Activity 4.2 Feedback:
Discrete variables are variables with a finite number of distinct and separate values
e.g. sex, race, family size, number of days present. All these are discrete variables
because they can be measured only in a discrete set of units such as 2, 3, etc. No
51
meaningful measurement values lies between these distinct and separate values.
Continuous variables are those that at least theoretically can take an infinite array of
values. Age is normally measured in years, but it can also be measured in months,
days, hours, minutes and seconds.
4.6 Controlling Variables
Confounding Variables that are of interest to the researcher can be controlled by
building them into the study as independent variable. For example, a researcher
comparing two teaching methods may wish to control teacher capabilities by making
it an independent variable. Several variables that are not of interest to the research
may be removed or their influence minimised as follows:
Removing the Variable: Variables may be controlled by removing them completely.
For example, teacher concerns for the good performance of his/her pupil could be
eliminated by using a neutral teacher to supervise the examination.
Randomisation: Randomisation involves pure chance selection and assignment of
subjects to experimental or control group. This is a method whereby everyone selected
for the study has equal chance of being assigned to either the experimental or control
group e.g. use of tossing the coin.
Matching Cases: Where randomisation is not feasible, selecting pairs of individuals
with identical or nearly identical characteristics and assigning one to experimental or
control reduces the influence. Where the variables are more than one, matching may
not effectively solve the problem.
Balancing Cases or Group Matching: Balancing cases consists of assigning subjects
to experimental and control groups in such a way that the means and variances of the
groups are as nearly equal as possible.
Analysis of Variances: This method permits the research to eliminate initial
differences on several variables between the experimental and control groups by
statistical methods. The use of per-test mean scores as covariance is more effective
than the matching process.
Box 4.1: Controlling Variables
It is important to take note of the methods in controlling variables in a research. These are:
• Removing the variable
• Randomisation
• Matching Cases
• Group Matching
• Analysis of Variances
52
Summary of Study Session 4
In Study Session 4, you have learned that:
A variable is a symbol to which individual can assign numerical values.
Independent variables are variables the researcher could manipulate and dependent
variables vary concomitantly with changes in the independent variable.
Discrete variables are variables that can take finite numbers while continuous
variables can take an infinite number of values.
While the knowledge of data is important for meaningful analysis and interpretation of
data, constructs, observable and intervening variables are important in that they
explain why researcher should define higher terms clearly and operationally for
meaningful evaluation and understanding of the research.
Self-Assessment Questions (SAQs) for Study Session 4
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 4.1 (tests learning outcome 4.1)
If you were told to define concept, how would you define it?
SAQ 4.2 (tests learning outcome 4.2)
Can you differentiate between concepts and constructs?
SAQ 4.3 (tests learning outcome 4.3)
How will you describe the term ‘variable’?
SAQ 4.4 (tests learning outcome 4.4)
Discuss the two major types of variables.
Notes on the Self-Assessment Questions (SAQs) for Study Session 4
SAQ 4.1: An important part of theories are concepts. These are mental constructs or
images developed to symbolise ideas, persons, things or events. Concepts
are the building blocks that are interrelated in propositions that form the
explanatory statements of a theory.
53
SAQ 4.2: Concepts are groups of behaviour, objects or situation that have common
characteristics. Construct cannot be seen, heard or felt. They can only be
inferred by phenomena such as test scores or by observed hostile or
aggressive acts, skin responses, pulse rates, or persistence at a task.
SAQ 4.3: Kerlinger (1977) defines a variable as a symbol to which we assign
numerical values. For example, if a researcher is interested in the effects of
two teaching methods on the science achievement of JSS three students.
After the different teaching methods have been implemented, the JSS three
students involved would be measured with a science achievement test.
SAQ 4.4: An independent variable is the presumed cause of the dependent variable
i.e. the presumed effect. According to Kerlinger (1977), the independent
variable is the antecedent and the dependent is the consequent. The
dependent variables (Y) are the presumed effect which varies concomitantly
with changes or variations in the independent variable (X). It is the variable
that is not manipulated.
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi: Prentice
Hall. PP 20 – 23.
Cresswell, J.W. (1994). “Research Design: Quantitative and Qualitative Approaches”
Thousand Oaks, C.A.: Sagem PP 105 – 107.
Leedy, P.D. (1997). Practical Research: Planning and Design (6th Edition). New Jersey:
Morill, PP 103 – 110.
Monette, D.R., Sullivan, T.S. and Dejong, R.D. (1994). Applied Social Research: Tool for the
Human Services (3rd Edition) Chicago: Harcourt Brace College Publishers, 3 – 8.
54
Study Session 5: Types of Research Approaches –Qualitative
and Quantitative Approaches
Expected duration: 1 week or 2 contact hours
Introduction
The terms qualitative and quantitative are often used to identify different approaches
to answering research question. You have to note that in any study, different questions
provide different types of information. The entire research methodologies can be
classified into two major approaches namely: qualitative and quantitative approach.
Learning Outcomes for Study Session 5
When you have studied this session, you should be able to:
5.1 Explain the meaning of qualitative approach (SAQ 5.1)
5.2 Differentiate between qualitative and quantitative approaches (SAQ 5.2)
5.3 Describe longitudinal research (SAQ 5.3)
5.4 State the defining characteristics of cross-sectional research (SAQ 5.4).
5.1 Qualitative Approaches
Qualitative approach is research that describes phenomena in numbers and measures
(Krathwol, 1993). Qualitative approach has its origin in descriptive analysis, and is
essentially an inductive process, reasoning from the specific situation to a general
conclusion. As Lancy (1993) points out, qualitative research is typically thought of as
a method with a set of procedures for conducting research. Qualitative research in its
purest sense follows the naturalist paradigm, i.e. that research should be conducted in
the natural setting and that the meanings derived from research are specific to that
setting and its conditions. The qualitative approach is sometimes referred to as
interpretative, the naturalistic, the constructivist or the post-positivist approach. The
common qualitative research designs are case study research design. Cresswell (1994)
defines qualitative study as “inquiry process of understanding a social or human
problem based on building a complex holistic picture formed with words, reporting
detailed views of informants and conducted in a natural setting. Most of the time,
55
qualitative researchers often start a study with general questions, collect an extensive
amount of verbal data from a smaller number of participants and present their findings
with words or descriptions that are intended to accurately reflect the situation under
study. A qualitative study may conclude with tentative answers on hypotheses about
what was observed. These tentative hypotheses may then form the basis of the future
quantitative studies designed to test the proposed hypotheses. Qualitative research
involves data in the form of words, pictures, descriptions or narratives. Qualitative
researchers tend to adopt an attitude of discovery or exploration that lead to
discovering, building or enhancing theory as opposed to testing it. For example, an
investigation of the teaching and learning of mathematics in Senior Secondary
School Two (SS.II) at Abadina College UI, Ibadan, where the researcher spends 8
weeks asking questions transpires within the period. Qualitative researchers remain
detached from their subjects in order to make unbiased, universal and context-free
generalisations. Qualitative research is very holistic and emergent with specific focus,
design, interview instruments and interpretations developing and changing along the
way.
Qualitative studies tend to use an inductive form of analysis where by no observations
of particular cases may be generalised to a class of cases. Inductive reasoning
emphasises after-the-fact explanation; theory emerges from a careful consideration of
the evidence (data).
Figure 5.1: Research Methods (Photo from www.slideshare.net)
5.2 Quantitative Approaches
The quantitative approach is typically used to answer questions about the relationship
among measured variables with the purpose of explaining, predicting and controlling
phenomena. The quantitative approach is sometimes referred to as the traditional, the
positivist, the experimental or the empiricist approach. Cresswell (1994) defines a
56
quantitative research as “an inquiry into a social or human\ problem, based on testing
a theory composed of variables, measured with numbers and analysed with statistical
procedure, in order to determine whether the predictive generalisations of the theory
hold true”. The quantitative researchers usually start with a preformed hypothesis to
be tested. The relevant variables are isolated, extraneous variables are controlled,
standardised data are collected from a sizeable number of participants and the data are
analysed in such a manner that the original hypotheses can be rejected or not rejected
and conditions that can be generalised are stated. Often quantitative researchers seek
explanations and predictions that can be generalised. The objective is to establish,
confirm or validate relationships and to develop generalisations that contribute to
theory. Careful guidelines exist for conducting quantitative studies.
1. Concepts, variables and hypotheses are often defined before the study begins
and remain fixed throughout.
2. Methods are chosen by the researchers so that they can objectively measure the
variables of interest.
3. Researchers remain detached from their ‘subjects’ in\ order to make unbiased,
universal, context-free generalisations.
4. Quantitative researchers often use experimental or correlational designs to
reduce error, bias and extraneous variables. They believe there is a relatively
stable reality ‘out there’ that can be measured by well-designed questionnaires
or instruments. In all cases, generalisations receive greater acceptance if the
instruments are valid and reliable. Data are normally collected from whole
population or sample that is representative of the population. Quantitative
researchers are always interested in describing the norm of the population
sampled. Quantitative studies tend to rely on deductive form of analysis.
Activity 5.1
To guide you in understanding this session, attempt this activity. Briefly distinguish
between qualitative and quantitative approaches in research.
Activity 5.1 Feedback:
Qualitative approach is research that describes phenomena in numbers and measures
(Krathwol, 1993). Qualitative approach has its origin in descriptive analysis, and is
essentially an inductive process, reasoning from the specific situation to a general
conclusion. As Lancy (1993) points out, qualitative research is typically thought of as
57
a method with a set of procedures for conducting research. Qualitative research in its
purest sense follows the naturalist paradigm, i.e. that research should be conducted in
the natural setting and that the meanings derived from research are specific to that
setting and its conditions. The quantitative approach is typically used to answer
questions about the relationship among measured variables with the purpose of
explaining, predicting and controlling phenomena. The quantitative approach is
sometimes referred to as the traditional, the positivist, the experimental or the
empiricist approach.
5.3 Longitudinal Research
According to Gall et. al. (1996), a longitudinal study is one that involves “collecting
data from a sample at different points in time in order to study changes or continuity
in the sample characteristics. Thus, “secondary school students’ attitude to
mathematics” might be started in their first year in JSS.1 and end in their final year, in
SSS.3, which means the study will last for a period of six years. In such a study, data
are collected at different times of the period of study. According to Leedy (1997),
though longitudinal studies are difficult to implement, they are however essential for
exploring developmental issues. It is only longitudinal studies that can determine the
different experiences of individuals over time. One disadvantage of longitudinal
studies is that they can be reactive. People responses or behaviour at one time may be
influenced by the fact that they have been observed earlier. Another disadvantage is
that people who participated earlier may not want to or be able to participate later.
People die, move away, become uninterested or in other ways, become unavailable.
This loss of participants can adversely affect the validity of the research findings. The
major disadvantage of all longitudinal studies is that they are difficult and expensive
to conduct, especially if they span a long period of time. Longitudinal studies may or
may not use survey methods to gather data. Its most distinguishing characteristics,
unlike most research designs, is not related to how data are collected and analysed, but
to the time period in which data are collected.
58
Figure 5.2: Longitudinal Research (Photo from www.lynn-library.libguides.com)
5.4 Cross Sectional Research
Cross-sectional research is a research method often used in developmental
psychology, but also utilized in many other areas including social science and
education. This type of study utilizes different groups of people who differ in the
variable of interest, but share other characteristics such as socioeconomic status,
educational background, and ethnicity. For example, researchers studying developmental
psychology might select groups of people who are remarkably similar in most areas,
but differ only in age. By doing this, any differences between groups can presumably
be attributed to age differences rather than to other variables. Cross-sectional studies
are observational in nature and are known as descriptive research, not causal or
relational. Researchers record the information that is present in a population, but they
do not manipulate variables. This type of research can be used to describe
characteristics that exist in a population, but not to determine cause-and-effect
relationships between different variables. These methods are often used to make
inferences about possible relationships or to gather preliminary data to support further
research and experimentation.
Defining Characteristics
• Takes place at a single point in time
• Does not involve manipulating variables
• Allows researchers to look at numerous things at once (age, income, gender)
• Often used to look at the prevalence of something in a given population
Activity 5.2
Briefly explain the meaning of longitudinal research.
59
Activity 5.2 Feedback:
According to Gall et. al. (1996), a longitudinal study is one that involves “collecting
data from a sample at different points in time in order to study changes or continuity
in the sample characteristics. Thus, “secondary school students’ attitude to
mathematics” might be started in their first year in JSS.1 and end in their final year, in
SSS.3, which means the study will last for a period of six years.
Potential Challenges
While the design sounds relatively simple, finding participants who are very similar
except in one specific variable can be difficult. Also, groups can be affected by cohort
differences that arise from the particular experiences of a unique group of people.
Individuals born in the same time period may share important historical experiences,
while people born in a specific geographic region may share experiences limited
solely to their physical location.
Summary of Study Session 5
In Study Session 5, you have learned that:
Quantitative could be called traditional, positivist, etc. Quantitative is used to confirm,
validate and test a theory, it is focused, has established guidelines, static design,
context-free and detached view, adopts a deductive approach to analysis, and report is
communicated through numbers, statistics aggregate data and in scientific style.
Qualitative is to describe, explain, explore, interpret and build theory. It is process-
oriented and holistic, context-bound and personal view.
Data are collected using observations, interviews and inductive analysis of data is
adopted.
Self-Assessment Questions (SAQs) for Study Session 5
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
Box 5.1: Cross-Sectional Research
The defining characteristics of cross-sectional research are:
• Takes place at a single point in time
• Does not involve manipulating variables
• Allows researchers to look at numerous things at once (age, income, gender)
• Often used to look at the prevalence of something in a given population
60
SAQ 5.1 (tests learning outcome 5.1)
How will you explain the meaning of qualitative approach?
SAQ 5.2 (tests learning outcome 5.2)
Differentiate between qualitative and quantitative approaches
SAQ 5.3 (tests learning outcome 5.3)
How will you describe longitudinal research?
SAQ 5.4 (tests learning outcome 5.4)
State the general characteristics of qualitative studies.
Notes on the Self-Assessment Questions (SAQs) for Study Session 5
SAQ 5.1: Qualitative approach is research that describes phenomena in numbers and
measures (Krathwol, 1993). Qualitative approach has its origin in descriptive
analysis, and is essentially an inductive process, reasoning from the specific
situation to a general conclusion.
SAQ 5.2: Qualitative research in its purest sense follows the naturalist paradigm, i.e.
that research should be conducted in the natural setting and that the
meanings derived from research are specific to that setting and its
conditions. The qualitative approach is sometimes referred to as
interpretative, the naturalistic, the constructivist or the post-positivist
approach. The quantitative approach is typically used to answer questions
about the relationship among measured variables with the purpose of
explaining, predicting and controlling phenomena.
SAQ 5.3: According to Gall et. al. (1996), a longitudinal study is one that involves
“collecting data from a sample at different points in time in order to study
changes or continuity in the sample characteristics.
SAQ 5.4: Defining Characteristics of a Cross-sectional Research
• Takes place at a single point in time
• Does not involve manipulating variables
• Allows researchers to look at numerous things at once (age, income, gender)
• Often used to look at the prevalence of something in a given population
61
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi: Prentice
Hall. PP 20 – 23.
Cresswell, J.W. (1994). “Research Design: Quantitative and Qualitative Approaches”
Thousand Oaks, C.A.: Sagem PP 105 – 107.
Gall, M.D.; Borg, W.R. & Gall, G.R. (1996). Educational Research: An Introduction (6th
Edition). White Plains, NY: Longmans, PP 376 – 380.
Leedy, P.D. (1997). Practical Research: Planning and Design (6th Edition). New Jersey:
Morill, PP 103 – 110.
62
Study Session 6: Survey Research and Research Tool
Expected duration: 1 week or 2 contact hours
Introduction
In this session, you will learn the basic concepts in survey research and the research
tools used in carrying out this type of research. The aim of any survey is to obtain
information which can be analysed and patterns extracted and comparison made. In
surveys, all respondents are asked the same question while the same circumstances are
assumed to prevail. Information, most of the time, can be gathered by means of self –
completed questionnaire, or by means of questionnaire, schedules or checklists
administered by an interviewer. Often, information is provided by observing,
interviewing, questioning and other methods. Questionnaire needs to be well-designed
so that they will provide the information a researcher needs, which will be acceptable
to the respondents and pose no problem at the analysis and interpretation stage. In
designing questionnaire, the researcher must ensure that care has to be taken to ensure
researcher is well guided in question writing, in the design, piloting, distribution and
return of questionnaires. In this chapter, you will study the design of questionnaire
under them following subheadings: what a questionnaire is, steps in constructing
questionnaires, choosing a response format, types of questionnaire, evaluating the
questions and piloting the questionnaire. Observation and case study will also be
discussed.
Learning Outcomes for Study Session 6
When you have studied this session, you should be able to:
6.1 Explain the meaning of survey research (SAQ 6.1)
6.2 Describe the characteristics of survey.(SAQ 6.2,)
6.3 Explain the term ‘questionnaire’ (SAQ 6.3)
6.4 Discuss at least five steps in constructing questionnaire (SAQ 6.4).
6.1 Meaning of Survey Research
A survey is a data collection technique in which information is gathered from
individuals called respondents by having them respond to questions. This is probably
63
the most widely used research methodology. The survey method gathers data from a
relatively large number of cases at a particular time. It is not concerned with
characteristics of individuals but as generalised statistics that result when data are
abstracted from a number of individual cases. It is essentially cross sectional. Survey
provides such information that each statement pictures a prevailing condition at a
particular time. The survey requires an imaginative planning, careful analysis and
interpretation of the data gathered and logical and skillful reporting of the findings.
Generally, a survey as a descriptive research methodology, describes and interprets
data. It is concerned with conditions or relationships that exist, opinions that are held,
processes that are ongoing, effects that are evident or trends that are developing.
Surveys typically involve collecting data representative of populations too large to be
dealt with by other methods. Indeed, the generalisability of survey findings is another
major attraction of the method. All surveys, involve presenting the respondents with a
series of questions to be answered. These questions may tap matters of facts, attitudes
and opinions or future expectations. The questions may be simple single-item
measures or complex multiple-item scales. In whatever form, however, survey data
are basically what people say to the investigator in response to a question and data can
be collected in survey research in two basic ways through questionnaire or interviews.
For instance an individual who is interested in which variable will be relevant in the
major research conducts a preliminary research to know which variable is relevant. It
is thus a pilot, preliminary research e. g. If a researcher is interested in teacher
effectiveness, main concern is to train teachers to become effective, there are many
variables which can make teachers effective, one has to decide which one an
individual is going to use one then conduct preliminary research as to know which
important variables to use on teacher which is a survey research. It is primarily
concerned with the present, although it often considers past events and influences as
they relate to current conditions. The survey is included in the quantitative tradition
rather than the qualitative because descriptive researchers tend to convert their data
into numerical indices and to employ statistical analysis technique to generalise their
findings from a sample of the respondents to a population. The method of research
that looks with intense accuracy at the phenomena of the moment and then describes
what the researcher sees is called descriptive survey.
64
Figure 6.1: People carrying out Survey Research (Photo from www.news.virginia.edu)
6.1.1 Characteristics of Descriptive Survey
The characteristics of the descriptive survey are as indicated below:
The principal means of collecting data is through observation technique to ensure
discreteness of the population by carefully choosing: clearly defining problems and
specifically delimiting the study area. Data in descriptive survey research are
particularly susceptible to distortion through the introduction of bias into that research
design. Therefore, particular attention should be given to safeguarding the data from
the influence of bias. Although the descriptive survey method relies on observation for
the acquisition of the data, those data must then be organised and presented
systematically so that valid and accurate conclusions can be drawn from them.
6.1.2 Advantages and Disadvantages of Survey Research
Survey research has the advantage of wide scope: a lot of information can be obtained
from a large population. Compared with other research methodologies, survey
research is relatively economical. Also, because it uses existing educational facilities
and personnel to reduce cost, survey research data are accurately within sampling
error. One major disadvantage is that survey information does not penetrate very
deeply below the surface. The scope of the information sought is usually emphasised
at the expenses of the depth. Survey research takes a lot of time and money.
Interviews require skill, time and money. Also, survey research requires a good deal
of research knowledge and sophistication. The survey investigator must be familiar
with sampling question and schedule construction, interviewing analysis of data and
other technical aspects of survey.
65
o Assuming you are a lecturer and you instructed one of your final year students
to find out teachers level of use of technology for classroom instruction. Before
the student would embark on this research, you wanted to be sure that he
understood the type of research he was to carry out. Then, you asked your
student to briefly describe survey research for you. What kind of response
would you expect from him?
• A survey is a data collection technique in which information is gathered from
individuals called respondents by having them respond to questions. The
survey method gathers data from a relatively large number of cases at a
particular time. It is not concerned with characteristics of individuals but as
generalised statistics that result when data are abstracted from a number of
individual cases. Generally, a survey as a descriptive research methodology,
describes and interprets data. It is concerned with conditions or relationships
that exist, opinions that are held, processes that are ongoing, effects that are
evident or trends that are developing.
6.2 Questionnaire
A questionnaire contains written questions that people respond to directly on the
questionnaire form itself without the aid of an interviewer. Best and Khan (1995) state
that a questionnaire is used when factual information is desired. When opinions rather
than facts are desired, an opinionaire or attitude scale is used. Also, Okpala, Onuoha
and Oyedeji (1993) define a questionnaire as “a self-reporting instrument that has
received good use in educational research, psychological research, programme
evaluation etc. It is sometimes described as the most common type of research
instrument. It is important for teachers, students and research officers to master how to
construct a questionnaire. This would enable such people to develop appropriate
questionnaires when they cannot get already existing ones that is valid and reliable.
Like tests, questionnaires are constructed for specific purposes. It is therefore,
necessary for the researcher to think of a specific study and design before determining
whether it is appropriate for him to use a questionnaire. Characteristics of a good
questionnaire are:
i. it deals with a significant topic, one which the respondent will recognise as
important enough to warrant spending his/her time on. The significance should
be clearly and carefully stated on the questionnaire or in the letter that
accompanies it;
66
ii it seeks only the information which cannot be obtained from other source such
as: school reports or census data;
iii. it is as short as possible, and only long enough to get the essential data. Keep
the writing required of the respondent to a minimum and make the response
system clear and easy to complete;
iv. it is attractive in appearance, neatly arranged and clearly duplicated or printed;
and directions for a good questionnaire are clear and complete. Important terms
are defined. Each question deals with a single idea and is worked as simply and
clearly as possible.
6.2.1 Steps in Constructing a Questionnaire
Okpala, et. al. (1993) state that usually a researcher is faced with respondents who
have great amount of information that would go untapped unless the questionnaire
items are valid and reliable enough to elicit the information. Process of developing
and using a questionnaire could be divided into eight steps as suggested by Henerson
et. al. (1978). The steps are listed below:
i. Identifying the programme objectives and specific information to be obtained;
ii. Selecting a response format;
iii. Identifying the frame of reference of the respondents;
iv. Writing the item/questions;
v. Preparing a data summary sheet;
vi. Critiquing the questions, trying them out, and revising them;
vii. Assembling the questionnaires; and
viii. Administering the questionnaires.
Some of these steps will be followed in an effort to explain to you how to design a
questionnaire.
Programme Objectives and Response Format
The researcher needs good direction which can only be provided by clearly stating the
objectives of the programme. Clearly stated objectives help to clear issues concerning
the type of information the researcher requires. Objectives of a programme could be:
i. to determine the view of students about teaching of mathematics
ii. to find out the attitude of students toward statistics.
67
Choosing a Respondent Format
Having clearly stated the objectives of the questionnaire, the researcher need to
determine the response format. The two major types of questions used in
questionnaires are close-ended and open-ended. The researcher could use either a
combination of both of them in constructing his/her question.
Close-Ended and Open-Ended Questionnaires
Two basic questions can be used in questionnaires and interviews. Close-ended
questions are those questions that provide respondents with a fixed set of alternatives
from which they are to choose. The response formats of multiple items scales, for
example, are all close-ended.
Open-ended questions are questions to which the respondents write their own
response, much as you do for an essay type examination questions.
Close-Ended Questionnaire
Questionnaires that call for short response are known as the restricted or closed form
type. In this type of question, you mark a yes or no, write a short response, or check an
item from a list of suggested responses.
Examples are:
i Yes or No Type:
Statistics is a difficult subject.
ii. Writing a Short Response:
Yes/No
The science subject that deals with human behaviour is ………..
Activity 6.1
Before you will continue with the steps in creating questionnaire, attempt this activity.
Briefly explain the characteristics of a good questionnaire.
Activity 6.1 Feedback:
Characteristics of a good questionnaire are:
i. it deals with a significant topic, one which the respondent will recognise as
important enough to warrant spending his/her time on. The significance should
be clearly and carefully stated on the questionnaire or in the letter that
accompanies it;
68
ii it seeks only the information which cannot be obtained from other source such
as: school reports or census data;
iii. it is as short as possible, and only long enough to get the essential data. Keep
the writing required of the respondent to a minimum and make the response
system clear and easy to complete;
iv. it is attractive in appearance, neatly arranged
Improving Questionnaire Items
The questions that make up a survey are the basic data gathering devices, therefore,
the researcher should choose word with care, especially with questionnaires that allow
no opportunity to clarify questions for the respondents. Therefore, the following steps
should be taken to improve questionnaire items:
Questions should be simple and direct, expressing only one idea. Complex statements
containing more than one idea should be avoided e.g. the 6-3-3-4 system of education
programme is good and all parents should pay levy to support it. This is not a good
question because it is possible respondents would agree with the first part, but not
with the second part .For the questions designed for research, slang terminology
should be avoided.
Before administering the questionnaire, some issues have to be ascertained. They are
being discussed under the following sub-topics:
Frame of Reference of Students
It is important for the researcher to consider the frame of reference of the respondents
before administering the questions. In fact, it is much better before phrasing the
question. This is necessary in order to ensure that the questions are within the
capability of the respondents.
Pilot the Questionnaire
All data gathering instruments should be pilot tested how long it takes to complete
them, to check that all questions and instructions are clear, and to enable the
researcher to remove any items which do not yield useable data. There is a temptation
in a small study to go straight to the distribution stage, but however pressed for time
you are, give your best to give the questionnaire a trial run. Note that, it should be
tried out on a group similar to the one that will form the population of your study.
69
This will enable you to come out with a good questionnaire which your respondents
will respond to, also you can carry out a preliminary analysis of the data collected.
Figure 6.2: The Role of Questionnaire in Research (Photo from www.123rf.com)
Administering the Questionnaire
There is a need to make early decision about how to distribute your questionnaire and
what to do about no response. There are advantages in being able to give
questionnaires to respondents personally. You can explain the purpose of the study
and in some cases; the questionnaire can be completed on-the-spot. Colleagues and
friends may be persuaded to lend a hand. If all fails, you may have to mail your
copies, but postal service is expensive and response rate is generally low. Therefore,
mailing should be the last resort. Whether the researcher is distributing the
questionnaire by direct distribution or by mailing, the instrument must be
accompanied with a covering letter. The letter should inform the respondents about
the objectives of the questionnaire, what will be done with the information, the
deadline for the return of the questionnaire. It might be necessary to assure the
respondent about the anonymity and confidentiality of the information supplied.
6.3 Interview and Observations
In an interview, there is a face-to-face question and answer session between the
researcher and the respondents. Observation, however, is no ‘natural gift’, but a highly
skilled activity for which an extensive background knowledge and understanding is
required, and also a capacity for genuine thinking and the ability to spot significant
events. It is certainly not an easy option. In this chapter, interviews and observations
will be discussed under the following headings: interview, what interview is, purpose
of interview, types of interview, conducting an interview, validity and reliability of
interview. Observation: what it is, characteristics, uses, observation plans and tool for
recording observation.
70
Questionnaires are relatively inexpensive for reaching a substantial number of people,
but they have some disadvantages associated with non-response and occasionally with
careless response. The interview is a more effective method of conducting a survey. In
interview, the investigator or his/her assistant reads the questions directly to the
respondents and records their answers. Interviews offer the investigator a degree of
flexibility that is not available in questionnaire. Interview items, like questionnaire
items, can be open-end or close-ended.
The purpose of an interview is to find out what is in someone’s mind. The purpose of
open-ended interviewing is not to put things in someone’s mind, but to assess the
perspective of the persons being interviewed. Too often, the interviewees provide
information based upon what they think the interviewer wants to hear. Therefore, it is
critical for the interviewer to make sure the person being interviewed understands that
the researcher does not hold any\ preoccupation notions regarding the outcome of the
study. Interviews are used to gather information regarding an individual’s experience
and knowledge, his or her opinions, beliefs and feelings and demographics data.
Interview questions can be asked so as to determine past or current information as
well as predictions for the future. The preferred method for data collection is to tape
record the interview if the respondent agrees. Otherwise, the interviewer must keep
brief notes and expand on them as necessary immediately after the interview while the
information is still fresh in the interviewer’s mind. A structured interview includes a
series of closed form questions similar to those used in a questionnaire. Semi-
structured interviews go one step further by following these closed form questions
with probes designed to obtain additional, clarifying information. Regardless of which
form is used, interviewing involve much more than just asking questions. The
questions for the interview should be carefully planned and the words should be
accurate as that of the items in the questionnaire. Furthermore, it should be pilot tested
just like the questionnaire, to ensure that the questions are clear, precise and free from
bias. Planning is necessary for interview to ensure that the set objectives will be
achieved.
Activity 6.2
To guide you in understanding this session, attempt this activity. Briefly distinguish
between interview and observation
71
Activity 6.2 Feedback:
In an interview, there is a face-to-face question and answer session between the
researcher and the respondents. Observation, however, is no ‘natural gift’, but a highly
skilled activity for which an extensive background knowledge and understanding is
required, and also a capacity for genuine thinking and the ability to spot significant
events. It is certainly not an easy option.
6.4 Focus Group Interviewing
It is rapidly becoming one of the major research tool to understand people’s thoughts
and feelings. It is usually conducted by inviting six to ten people to gather for a few
hours with a trained moderator to talk about a product, service, organization an issue
and a problem. The meeting is held in a pleasant place, and refreshments are served to
create a relaxed environment. The moderator needs objectivity, knowledge and nature
of problem to be discussed and some understanding of group and participants’
behaviour. The moderator starts with a broad question before moving to more specific
issues, encouraging open and easy discussion to bring out true feelings and thoughts.
At the same time, the interviewer focuses the discussion, hence the name focus group
interviewing. It is often held to help determine the subject areas on which questions
should be asked in a later, large-scale, structured-direct interview
Comments are recorded through note taking or videotaped and studied later to
Understand. This method is especially suited for participants feelings dealing with
HIV/AID. STIS and related areas. Focus group interview provides opportunity for
participants to make clarifications and freely pass their comments.
Figure 6.3: People Participating in Focus Group Discussion (Photo from
www.mummytales.com)
72
How to Arrange an Interview
Below are the steps for successfully handling the interview as a technique for
gathering data for research purpose.
i. Set up the interview well in advance;
ii. Send the agenda of questions you will ask the interviewee;
iii. Ask for permission to tape the interview;
iv. Confirm the date immediately in writing;
v. Send a reminder together with another agenda for questions, 10days before you
expect to arrive;
vi. Be prompt, follow the agenda, have a copy of your question for your
interviewee in case he/she has misplaced his/her copy;
vii. Following the interview, submit a typescript of the interview and get either a
written acknowledgement of its accuracy or a correct copy from the
interviewee;
6.4.1 Strengths of Interview
The interview is often superior to other data-gathering devices. One reason is that
people prefer to talk than to write. After the interviewer gains rapport or establish a
friendly relationship with the interviewee, certain types of confidential information
may be obtained that an individual might be reluctant to put in writing. The
interviewer can explain more explicitly the investigations purpose and what
information he/she wants. If the subject misinterprets the questions, the interviewer
may follow with a clarifying question;
It is possible to seek the same information in several ways during � the interview. It is
possible to stimulate the interviewee and � possibly explore significant area not
anticipated on the original plan of investigation. �� Interview is particularly
appropriate when dealing with children.
6.4.2 Conducting an Interview
The interview is a social avenue designed to exchange information between the
respondent and the interviewer. The quantity of information exchanged depends on
how astute and creative the interviewer is at understanding and managing that
relationship. Note that the goal is to collect data and not to make friends. Appearance
must be neat, clean and business like but friendly. The first contact of the interviewer
with the interviewee must be appealing. At times, characteristics such as: socio-
73
economic status, age, race and ethnicity might influence the interview. Also, begin
with simple non-threatening questions. If an interview schedule is used, the interview
will progress according to the schedule. As needed, the interviewer will use probes or
follow up questions that are intended to elicit clearer and more complete
Activity 6.3
To guide you in understanding this session, attempt this activity. Briefly discuss the
strengths of an interview in a research.
Activity 6.3 Feedback:
The interview is often superior to other data-gathering devices. One reason is that
people prefer to talk than to write. After the interviewer gains rapport or establish a
friendly relationship with the interviewee, certain types of confidential information
may be obtained that an individual might be reluctant to put in writing. The
interviewer can explain more explicitly the investigations purpose and what
information he/she wants.
6.4.3 Recording Responses
An integral aspect of interview, is to record responses of respondents. The four most
common ways of recording responses are: classifying responses into predetermined
categories, summarising the “high point” of what is said, taking verbatim notes or
recording the interview with a tape recorder or video machine. Recording is easiest
when an interview schedule is used.
6.4.4 Controlling Interview
Once the interviewers go into the field, the quality of the resulting data is dependent
on the proper supervision during interviewer training. Interviewers should be
informed that their work will be thoroughly checked and failure to follow procedure
will not be accepted.
6.5 Observation: Definition and Characteristics
Observation continues to characterise all research: experimental, descriptive and
qualitative. Observation is the act of looking out for and recording the presence or
absence of verbal and non-verbal behaviour of a person or group of persons. The use
of specially designed evaluation instruments to collect observational data is referred to
74
as observational techniques. All methods of data collection involve some forms of
observation. Observational techniques, according to Monette et. al. (1994), is the
collection of data through direct visual or auditing experience of behaviour. With
observational techniques which include video or audio recording of behaviour, the
researcher actually sees or hears the behaviour or words that are the data of the
research.
Observational method varies in terms of the extent to which the researcher participates
in the activities of the people being observed. The two general possibilities are: that of
participant observer, and that of a non-participant observer. Participant observer is a
method in which the researcher is a part, and participates in the activities of the
people, group or situation that is being studied. Participant observation is highly
qualitative research.
Observation methods vary in terms of the extent to which the investigator participates
in the activities of the people being observed. The two general possibilities are that of
participant observer and that of non-participant observer. Each technique has its
unique difficulties, advantages and disadvantages. The principal ways of getting
information are by either experiencing something directly or by having someone tell
us what happened. Participant observation is a method in which the researcher is a
part of, and participates in the activities of the class, group, people or situation that is
being studied. In some cases, the researcher may have belonged to the group prior to
the start of the research and can use this position as a group member to collect data.
For example, a teacher in a science vacation course wishes to observe the reactions of
his colleagues to the different lectures, demonstrations and other activities during the
course. In many types of research, the relationship between the researcher and those
participating in the research is fairly clear-cut. In surveys, for example, participants
know who the researchers are and that they as respondents are providing data to the
researchers. In this participant observation – the researcher – participant relationship
becomes more problematic in that it can take a number of different forms: two major
issues arise. The extent to which the observer will change the setting that is being
observed and the extent to which people should be informed that they are being used
for research purposes. The way in which a researcher resolves these issues determines
the nature of the observer – participant relationship for a given project.
75
Figure 6.4: Observation of a Process (Photo from www.ndi.org)
6.5.1 Steps in Participant Observer Research
The first step is to establish the specific goals of the research and decide that the
participant observation is the most appropriate strategy. The second step is to decide
which specific group to study. The third step is very challenging, gaining entry into
the group to be studied. The fourth step is to develop rapport and trust with the people
being studied so that they will serve as useful and accurate sources of information.
The fifth step is to observe and record.
6.6 Non-Participant Forms of Observation
Some research activities require the investigator to refrain from participation in the
group being investigated. The concern is that the intrusive impact of an outsider might
change the behaviour of group members in a way detrimental to the research question.
The researcher under such conditions adopts a complete observer’s role; the observer
has no direct contact with or no substantial influence in those being observed. One
way of doing such non-participant observation is to use an observation technique that
has been called unobtrusive or nonreactive observation and physical traces.
Recording Observation: The manner in which observations are carried out may be
primarily quantitative or qualitative in nature. Quantitative observation typically calls
for more structured recording of data on\ coding sheets whereas qualitative
observation may use less structured field notes.
6.7 Characteristics of a Good Observation
Best and Khan (1995) noted that: observation, as a research data gathering process,
demands rigorous adherence to the spirit of scientific inquiry.
The following standards characterise observers and their observations:
76
i. Observation is carefully planned. Systematic and perceptive observers know
what they are looking for and what is relevant in a situation. They are not
distracted either by the dramatic or the spectacular.
ii. Observers are aware of the wholeness of what is observed. Although they are
alert to significant details, they know that the whole is often greater than sum
of its parts.
iii. Observers are objective. They recognise their likely biases, and they strive to
eliminate their influence upon what they see and report.
iv. Observers separate the facts from the interpretation of facts. They observe the
facts and make their interpretations at a later time.
v. Observation are checked and verified, whenever possible by repetition or by
comparison with those of other competent observers.
vi. Observations are carefully and expertly recorded. Observers use appropriate
instruments to quantify and preserve the results of their observations.
Observations are collected in such a way as to make sure that they are valid and
reliable.
What should be Observed?
Monette et. al. (1994) stated that it is possible to organise one’s thought around some
general categories of things to be observed and recorded:
i. The setting: field notes should contain some description of the general
physical and social setting being observed.
ii. The people: field notes should include a physical and social description of the
main characters who are the focus of your observations.
iii. Individual behaviour – the central observation in most studies are the
behaviour of the people in the settings.
iv. Group behaviour: In some cases, the behaviour of group may be an important
bit of information.
Box 6.1: Characteristics of a Good Observation
• Observation is carefully planned.
• Observers are aware of the wholeness of what is observed.
• Observers are objective.
• Observers separate the facts from the interpretation of facts.
77
6.8 Measurement of Evaluation
If it does not create a barrier between the observer and those observed, simultaneous
recording of observations is recommended. This practice minimise the errors that
result from faulty memory. There are other occasions when recording would more
appropriately be done after observation. The recording of observation should be done
as soon as possible, while the details are still fresh in the mind of the observer.
However, many authorities agree that objectivity is more likely when the
interpretation of the meaning of the behaviour described is deferred until a later time,
for simultaneous recording and interpretation often interfere with objectivity.
Systematising Data Collection
To aid the recording of information gained through observation, a number of devices
have been extensively used. Checklists, rating scales, scorecards and scaled specimens
provide systematic means of summarising or quantifying data collected by observation
or examination. For example, the teaching practice assessment sheet is an observation
scale except the part that deals with quality of the written lesson note.
6.9 Case Study
The case study is a way of organising social data for the purpose of viewing social
reality. It examines a social unit as a whole. The unit may be a person, a family, a
social group, a social institution or a community. The purpose is to understand the life
cycle as an important part of the life of that unit. The case study probes deeply and
analyses interactions between the factors that explains present status or that influence
change or growth. It is a longitudinal approach, showing development over a period of
time. A case study is conducted to shed light on a phenomenon, be it a process,
event, person or an object of interest to the researcher. A case study constitutes a
single instance of the phenomena. Data may be gathered by a wide variety of methods
including:
- observation by the researcher or his/her assistants on physical characteristics,
social qualities or behaviour;
- interviews with the subject(s) relatives, friends, teachers, counsellors and others;
- questionnaire, seeking opinion about a situation, psychological tests and
inventories; And
- recorded data from newspapers, school, courts, clinics, government agencies,
classroom, laboratory and other sources (Best and Khan, 1995).
78
A single case study emphasises analysis in depth. But if the objective analysis leads
researchers to consistent observations of significant variable relationships, hypotheses
may be confirmed, leading to valid generalisations. Often a substantial amount of data
is gathered from a wide variety of sources to present a description of the phenomena
or expensive from the perspective of the participants. Although the case study is
useful method or organised research observation, certain precautions should be
considered:
- The method may look deceptively simple. To use it effectively, the researcher
must be familiar with existing theoretical knowledge of the field or inquiry, and
skillful in isolating the significant variables from many that are irrelevant.
- Subjective bias is a constant threat to objective data-gathering and analysis.
- Effects may be wrongly attributed to factors that are merely associated and
rather than cause and effect related.
A case study final report takes the form of a rich descriptive narrative that attempts to
reconstruct the participants reality.
Case study researchers who have used interpretational or structural analysis methods
tend to use an objective writing style and\ make effective use of tables, figures and
matrices to display their findings.
Summary of Study Session 6
In Study Session 6, you have learned that:
Survey is a data collection technique in which information is gathered from
individuals called respondents.
Survey has specific characteristics that distinguish it from other research
methodologies e.g. the population for the survey must be carefully chosen, clearly
defined and specifically delimited to set precise parameter for ensuring discreteness to
the population.
Information could be generated for the research through interviews and observations
Self-Assessment Questions (SAQs) for Study Session 6
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
79
SAQ 6.1 (tests learning outcome 6.1)
How will you define survey research?
SAQ 6.2 (tests learning outcome 6.2)
Survey research has many characteristics, describe at least three of these
characteristics.
SAQ 6.3 (tests learning outcome 6.3)
Can you explain the term ‘questionnaire’?
SAQ 6.4 (tests learning outcome 6.4)
Discuss at least five steps in constructing questionnaire
Notes on the Self-Assessment Questions (SAQs) for Study Session 6
SAQ 6.1: A survey is a data collection technique in which information is gathered
from individuals called respondents by having them respond to questions.
This is probably the most widely used research methodology. The survey
method gathers data from a relatively large number of cases at a particular
time.
SAQ 6.2: The principal means of collecting data is through observation technique to
ensure discreteness of the population by carefully choosing: clearly
defining problems and specifically delimiting the study area. Data in
descriptive survey research are particularly susceptible to distortion
through the introduction of bias into that research design.
SAQ 6.3: A questionnaire contains written questions that people respond to directly on
the questionnaire form itself without the aid of an interviewer.
SAQ 6.4: Henerson et. al. (1978). The steps are listed below:
i. Identifying the programme objectives and specific information to be obtained;
ii. Selecting a response format;
iii. Identifying the frame of reference of the respondents;
iv. Writing the item/questions;
v. Preparing a data summary sheet.
80
Study Session 7: Experimental and Quasi Experimental Research
Expected duration: 1 week or 2 contact hours
Introduction
In the previous sessions, you have been exposed to the basic concepts in research and
review of literature. In this session, you will learn other research designs and these are
experimental and quasi experimental research design. Experiment means that
something is trying to determine its effects. That something has one or more
independent variable (s) manipulated to determine the effects. An independent
variable manipulated in an experiment is called an experimental variable. In its
simplest form, the experimental study attempts to control the entire research situation
except for certain input variables that then become suspect as the cause of whatever
change has taken place within the investigative design.
Learning Outcomes for Study Session 7
When you have studied this session, you should be able to:
7.1 Define experimental research (SAQ 7.1)
7.2 Describe at least five characteristics of experimental research (SAQ 7.2,)
7.3 Discuss the classifications of experimental research (SAQ 7.3)
7.4 Explain the concept of quasi experimental designs (SAQ 7.4).
7.1 Experimental Research
Experimental research method is a systematic and scientific approach to research in
which the researcher manipulates one or more variables, and controls and measures
any change in other variables. It is used to establish cause and effect. There is
consistency in a causal relationship (a cause will always lead to the same effect.
Experiments are conducted to be able to predict phenomenon and causes of events.
7.2 Characteristics of the Experimental Method
Before specific experimental designs are introduced, some general criteria for a well-
designed experiment will be considered. Essentially, the characteristics that make for
81
a good research design also apply to the design of an experiment. The criteria are
briefly discussed as follow:
Adequate Experimental Control – This means that there are enough constraints on
the conditions of the experiment so that the researcher can interpret the results. The
experimental design is so structured that if the experimental variable has an effect, it
can be detected. This may also mean controlling other variables through
randomisation or by building them into the design as independent variables.
Lack of Artificiality – This criterion is especially important in educational research if
the results of the experiment are to be generalised to a non-experimental setting – for
example, a classroom. It means that the experiment is conducted in such a way that
the results will apply to the real educational world.
Basis for Comparison – There must be some way to make a comparison to determine
whether or not there is an experimental effect. In some experiments, a control group is
used which is the group that does not receive an experimental treatment. The control
group in an instructional experiment usually consists of a group of students taught by
a traditional method. Certainly not all experiments require control groups.
Comparisons can be made between two or more experimental treatments and on
occasion with some external criterion.
Adequate Information from the Data – The data must be adequate in order to test
the hypotheses of an experiment. The data must be such that the necessary statistics
can be generated with enough precision to make decisions about hypotheses.
Uncontaminated Data – the data should adequately reflect the experimental effects.
They should not be affected by poor measurement errors in the experimental
procedure. The individuals from the various groups should not interact in such a way
as to cancel the experimental effects or to cause misrepresentation of the experimental
effects.
No Confounding of Relevant Variables – This criterion is closely related to
adequate experimental control group. There may be other variables operating that
have an effect on the dependent variable. If so, these effects must not be
82
misinterpreted as experimental effects. Their effects must be separated or controlled,
usually through the experimental design.
Representativeness – Representativeness usually means to generalise the
experimental results to some individuals, conditions, methods and so forth. To obtain
representativeness, experimenters commonly include some form of random selection
of subjects from the population to which they intend to generalise.
Figure 7.1: Experimental Research (Photo from www.ro.uwe.ac.uk)
Parsimony – The criterion of parsimony means that, with all other characteristics equal, a
simpler design is preferred to a more complex one. Of course, a design must be
complex enough for the purposes of the experiment, but complexity is not encouraged
for its own sake. The simpler design is usually easier to implement and possibly easier
to interpret. Experiments, like any other types of educational research, are susceptible
to technical and procedural errors. The development of an appropriate experimental
design and its adequate implementation require a considerable and careful planning,
but they provide the best safeguard against errors. Experimental designs require
simultaneous attention to a variety of details. This planning is done prior to
conducting the experiment.
Activity 7.1
At this point, it is important to reflect on what you have learnt so far. Based on what
you have read, describe the characteristics of the experimental method.
83
Activity 7.1 Feedback:
Adequate Experimental Control – This means that there are enough constraints on
the conditions of the experiment so that the researcher can interpret the results. The
experimental design is so structured that if the experimental variable has an effect, it
can be detected.
Lack of Artificiality – This criterion is especially important in educational research if
the results of the experiment are to be generalised to a non-experimental setting – for
example, a classroom. It means that the experiment is conducted in such a way that
the results will apply to the real educational world.
Basis for Comparison – There must be some way to make a comparison to determine
whether or not there is an experimental effect.
7.3 Classification of Experimental Design
There are three categories of experimental designs, namely: pre experimental designs,
the true experimental designs and the quasi experimental designs. The non/quasi-
experimental designs lack both the random assignment to conditions and the control
group that are such a central part of the good experimental designs. While they are
sometimes useful, they illustrate some inherent weakness in terms of establishing
internal validity. The better designs are called true experimental designs. The strength
of experiments as research techniques is that they are designed to enable us make
inference about the causality. The element that makes this possible is the control.
In experiments, the researcher has considerable control over determining who
participates in a study, what happens to them and under what condition it happens. At
the core of experimental research is the fact that the investigator exposes the people in
an experiment, commonly referred to as experimental participants to some conditions
or variables called the experimental group. The experimental group or experimental
treatment is an independent variable directly manipulated by the experimenter in order
to assess its effect on behaviour. Independent variables are those variables in a study
that are hypothesised to produce changes in another variable. The variable affected by
the independent variable is the dependent variable – so called because its value is
dependent on the value of the independent variable. An experimental group is a group
of subjects who are exposed to the experimental stimulus. Experimental condition is
the term used to describe the group of people who receive the experimental stimulus.
84
The third is the quasi-experimental design. The logic underlying experimentation is
illustrated by means of a series of symbols. The following symbols are commonly
used to describe experimental designs:
O = an observation or measurement of the dependent variable;
X = exposure of the people to the experimental stimulus or independent variable;
R = random assignment to conditions.
In addition, the symbols constituting a particular experimental design are presented in
time sequence, with those to the left occurring earlier in the sequence than those
further to the right. If OXO is a simple experiment, it can be described thus:
In this experiment, the researcher measures the dependent variable (this is called
pretest), exposes the subjects to the independent variable and then re-measures the
dependent variable. So one major way for assessing whether the independent variable
in an experiment has had an effect or not is to make a comparison of the pre -test
scores or measures with the Post-test. The symbols O, X, R will be used in discussing
the different types of design.
Pre-experimental Designs
This will be discussed under the following sub-topics.
One Group Pre-test – Post-test Design 0 x 0
This is a type of experiment in which a single group has a pre –experimental test, it
administers the experimental treatment and is given the same test. A science teacher
wanted to find out the effect of using a particular method in teaching his/her pupils.
He/She gave them a test before using the method. After using this method consistently
for at least six weeks; he/she varies the test items either by rearranging the numbering
or by rephrasing some of the items. The test is re administered. Any effect observed is
attributed to the treatment which is the new teaching method. It is necessary to
mention that there are series of weakness in this design. With the exception of attrition
and selection which are irrelevant owing to the lack of a control group, the design is
subject to the other five threats to internal validity. Maturation pre-test and the Post-
test scores, if paper and pencil tests are used, a shift of scores from pre-test to Post-test
could occur owing to testing effects. Regardless of the measurement used,
instrumentation changes could produce variations in the pre-test – post-test as a result
of regression towards the mean. In all, these variations on the dependent variable
85
produced by one or more of these validity treats could be mistaken for variation due to
the independent variable.
Static Group Comparison
The static group comparison is used to determine the influence of variable on one
group and not on another.
Group I x O
Group II – O
In this case, there are two groups; the experimental group is exposed to treatment, X,
the control group is not. There is no provision for establishing the equivalence of the
experimental and control group and this is a very serious limitation. A researcher gave
homework on the topic he/she is teaching for a period of one month to a class and
gave no homework at all to another class. At the end of one month, he administered
the same achievement test to both classes. The major validity threat to this design is
selection.
True Experimental Design
In true experiment, the equivalence of the experimental and control group is provided
by random assignment of subjects to experimental and control groups. Although it is
difficult to arrange a true experimental design, particularly in school-classroom
research, it is the most effective type of design and should be used whenever possible.
Three experimental designs are discussed under this section.
The Pre-test – Post-test Control Group
Experimental Group I RO1 XO2
Control Group II RO1 – O4
X Gain = O2 – O1
– Gain = O1 – 4
O1 x O3 Pre tests
O2 x O4 Post tests
This design utilises a true control group, including random assignment to equalise the
comparison group, which eliminates all the threats to external validity except perhaps
experimental attrition. Because of this, we can have considerable confidence that any
difference between experimental and control group on the dependent variable is due to
86
the effect of the independent variable. Most of the threats are minimised in this design.
History is removed because both groups experience the same events except for the
experimental stimulus. Because the same amount of time passes for both groups,
maturation effects can be assumed to be equal, both groups are pretested therefore any
testing influences on the post-test should be the same. Instrumentation effects are
controlled because any unreliability in the measurement process that could cause a
shift in scores from pre test to post-test should be the same for the groups.
Randomisation controls threat of selection by making sure that the comparison groups
are equivalent. Attrition is the only threat that is not controlled due to its nature
because people die, move away, get disinterested etc. For analysis, gain scores may be
compared and subjected to a test of the significance of the difference between means.
Pre test scores can also be used in analysis of covariance .
Activity 7.2
At this point, it is important to reflect on what you have learnt so far. Based on what
you have read, describe true experimental research.
Activity 7.2 Feedback:
In true experiment, the equivalence of the experimental and control group is provided
by random assignment of subjects to experimental and control groups. Although it is
difficult to arrange a true experimental design, particularly in school-classroom
research, it is the most effective type of design and should be used whenever possible.
The Pre-test – Only Control Group Design
R X O
R – O
This design is one of the most effective ways in minimising the threats to
experimental validity. It differs from the static group comparison design in that
experimental and control groups are equated by randomisation. At the conclusion of
the experiment, the difference between the mean test scores are subjected to test of
statistical significance of variance analysis with a chosen topic to provide a paradigm
for a true experimental design.
87
7.4 Quasi-Experimental Designs
These designs provide control of when and to whom the measurement is applied, but
because random assignment to experimental and control treatment has not been
applied, the equivalence of the group is not assumed. A quasi-experiment is an
empirical study used to estimate the causal impact of an intervention on its target
population. Quasi-experimental research shares similarities with the traditional
experimental design or randomized controlled trial, but they specifically lack the
element of random assignment to treatment or control. Instead, quasi-experimental
designs typically allow the researcher to control the assignment to the treatment
condition, but using some criterion other than random assignment (e.g., an eligibility
cutoff mark). In some cases, the researcher may have control over assignment to
treatment condition.
Figure 7.2: Quasi Experimental Research (Photo from www.ehow.com)
Quasi-experiments are subject to concerns regarding internal validity, because the
treatment and control groups may not be comparable at baseline. With random
assignment, study participants have the same chance of being assigned to the
intervention group or the comparison group. As a result, differences between groups
on both observed and unobserved characteristics would be due to chance, rather than
to a systematic factor related to treatment (e.g., illness severity). Randomization itself
does not guarantee that groups will be equivalent at baseline. Any change in
88
characteristics post-intervention is likely attributable to the intervention. With quasi-
experimental studies, it may not be possible to convincingly demonstrate a causal link
between the treatment condition and observed outcomes. This is
particularly true if there are confounding variables that cannot be controlled or
accounted for.
7.5 Empirical research
Empirical research design refers to research based on actual experience as distinguish
from historical or philosophical. The three above are example of empirical research
because they are based on actual experience distinguish from historical and
philosophical research which are based on speculations e. g. what is purpose of man?
Is an example of speculation which is philosophical research.
Different between philosophical, historical and empirical is basically in the
methodology and not in the content of what is being research on.
In the empirical research, one collects the data and not just form opinion
A. e. g. we want to find out the best method of teaching 5 years old children we
can conduct experiment and see which method is better, this is empirical
research.
B. Finding out effect of colonialism in the emergence of Nigeria nationalism. This
one is historical research.
Recently historical and philosophical are gradually moving towards the use of
empirical research.
Relationship between Empirical and Experimental Research
Although all experimental research is empirical research but not all empirical is
experimental researcher. Experimental research is only one type of empirical research;
there are all other types of research that are not experimental e. g. Survey and
Correlational research. Our emphasis is on experimental research.
Box 7.1: Quasi Experimental Design
• These designs provide control of when and to whom the measurement is
applied, but because random assignment to experimental and control
treatment has not been applied, the equivalence of the group is not
assumed.
• A quasi-experiment is an empirical study used to estimate the causal
impact of an intervention on its target population.
89
Summary of Study Session 7
In Study Session 7, you have learned that:
In this chapter, students have learnt that experimental design can be classified into
three major groups mainly: pre-experimental designs, true experimental designs and
quasi experimental designs. Examples of the various designs were discussed:
The major difference between true experimental design and others is the randomness
of the sample.
Self-Assessment Questions (SAQs) for Study Session 7
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 7.1 (tests learning outcome 7.1)
How will you define experimental research?
SAQ 7.2 (tests learning outcome 7.2)
Describe at least five characteristics of experimental research
SAQ 7.3 (tests learning outcome 7.3)
Discuss the classifications of experimental research.
SAQ 7.4 (tests learning outcome 7.4)
How will you explain the concept of quasi experimental designs
Notes on the Self-Assessment Questions (SAQs) for Study Session 7
SAQ 7.1: Experimental research method is a systematic and scientific approach to
research in which the researcher manipulates one or more variables, and
controls and measures any change in other variables.
SAQ 7.2:
Adequate Experimental Control
Lack of Artificiality
Basis for Comparison
Adequate Information from the Data
90
SAQ 7.3: There are three categories of experimental designs, namely: pre
experimental designs, the true experimental designs and the quasi
experimental designs.
SAQ 7.4: These designs provide control of when and to whom the measurement is
applied, but because random assignment to experimental and control
treatment has not been applied, the equivalence of the group is not assumed.
91
Study Session 8: Research Design
Expected duration: 1 week or 2 contact hours
Introduction
In this chapter, effort will be made to present some further explanation of correlation
studies as a research design and link it up with Ex-Post Facto designs. Some authors
describe Ex-Post Facto design as an experimental design in a reverse direction.
Generally, Ex-Post Facto research is a sub-type of the non-experimental method.
Nature of empirical research and differences between empirical and experimental was
also discussed
Learning Outcomes for Study Session 8
When you have studied this session, you should be able to:
8.1 Explain correlation design. (SAQ 8.1)
8.2 Define ex-post facto research. (SAQ 8.2)
8.3 Describe method of controlling independent variables. (SAQ 8.3)
8.4 Discuss the limitations of ex-post facto research. (SAQ 8.4).
8.1 Definition
A design is the framework in which a research is conducted and against which results
are to be evaluated. Given a research problem and context, there may also be some
other satisfactory design. It is the purpose of this chapter to enable students match
their studies with particular deigns described as follow. First a distinction is drawn
between ex-post factor and experimental research for the purpose of design.
8.2 Experimental Vs Ex-post facto research
The fundamental principles of experimental research are as follows:
1. Purpose: The main purpose of an experiment is to describe the effect of the
independent variable on the dependent variable.
92
2. Independent variable: this term refers to any induced or selected variation in
the experimental methods or materials or subjects whose effect is to be
observed and evaluated.
3. Evaluation: the effects of the independence variables on the dependent is to be
evaluated. The outcome of the experiments will be described in terms of
variation in the depending variable. Such outcome is often in terms of
comparison of means or medians of various experimental groups or frequency
of occurrence of certain events.
4. Control: provision is made for controlling the effects of other variables whose
presence might influence the variation of the dependent variable and thus
confound the results.
5. Generalization: The investigator often wish to generalize the results of his
research to subjects who did not participate in the research. The techniques of
statistical inference (discussed later) give scientific objectivity to this process
of generalization. Experimental research differ from ex-post facto research in
three important respects:
i. Manipulation Vs selection: in experimental research, the investigator
manipulates the independent variable whose effects are to be studied, in ex-
post factor research, the investigator merely selects the independent variable
whose effect are to be studied.
ii. Control Vs, non-control: in general, in experimental research, the
investigator has a wider range of options with respect to the manner in
which he controls the effects of other independent variables. No controls are
necessary in ex-post factor research.
iii. Definitive Vs tentative conclusions: Experimental research ordinarily yields
more definitive conclusions especially where the investigator intends to
explore cause and effect relationships. Ex-post factor research yields
association at best. Results of ex-post factor research may be spurious
owing to factors outside the study but which affect its outcome. (e.g.
finding out that length of time spent in preparation for a test and test scores
are negatively correlated).
Activity 8.1
Based on what you have read so far, how would you distinguish between experimental
and ex-post facto research designs.
93
Activity 8.1 Feedback:
Experimental research differ from ex-post facto research in three important respects:
1. Manipulation Vs selection: in experimental research, the investigator
manipulates the independent variable whose effects are to be studied, in ex-post
factor research, the investigator merely selects the independent variable whose
effect are to be studied.
2. Control Vs, non-control: in general, in experimental research, the investigator
has a wider range of options with respect to the manner in which he controls
the effects of other independent variables. No controls are necessary in ex-post
factor research.
3. Definitive Vs tentative conclusions: Experimental research ordinarily yields
more definitive conclusions especially where the investigator intends to explore
cause and effect relationships. Ex-post factor research yields association at
best.
8.3 Method of controlling independent variables
Four methods are available to control independent variables. Not all of them may be
employed in a given study
1. Holding the variable constant
An independent variable is held constant if subjects who take a particular value or
range of values of the variable only are studied. Sex is held constant when only males
or females are studied. Similarly age or IQ is held constant when subjects belong to
the same age group or have comparable IQ as measured by adequate test. A researcher
may study subjects categorized into low, medium and high socio economic status
separately and pool his results.
2. Randomization of effects of the variable
The effects of a variable are randomized whenever subjects are selected by
randomization. That is subjects are selected after consultation of the table of random
numbers. Subjects are assigned to treatment groups at random also. The result is
dominant in any group. In fact with this process, the effects of other variables, not
known to the researcher, are also randomized.
94
3. Matching of subjects
Subjects may be matched with respect to scores on a variable critical to a particular
study, say achievement in mathematics. For a study which requires two samples,
subjects should first be separated into pairs as closely matched with respect to known
mathematics scores as possible. Then one subject from each pair should be randomly
assigned to each sample. The result is two samples closely matched with respect to
one or more variables including those unknown to the investigator.
4. Statistical control
Whenever two variables are correlated, one can be used to predict performance on the
other. Pre-tests and post tests are usually correlated. In particular IQ scores can be
used to predict performance on any educational achievement test sine IQ scores
account for a portion of the variance of the depended variable, the effects of IQ can be
systematically controlled by eliminating this portion of the variance from the final
analysis. This is the rationale for ANCOVA. This method is not usually adopted
manually.
The use of randomization is required if the techniques of statistical inference are to be
used. Usually a good researcher design will also make provision for holding one or
more variables constant. Often one or more additional methods of controlling
variables will be used in the same design.
5. The number of samples
Some statistical techniques are designed for use with one sample. Some for use with
two samples and some for use with two or more samples, this will become clearer
later.
6. The methods of experimental or statistical control
Randomization and holding variables constant pose no restriction on choice of method
of analysis. Techniques of statistical inference assume the use of these methods of
control. Matching restricts choice of techniques to those adequate for data organized
by matching. Also statistical controls restrict choice of analysis to particular methods.
In general, whenever the more sophisticated techniques of experimental or statistical
control are required, the parametric statistics will be preferred.
o Oludare is a final year student, preparing to write his project. He has selected
some variables to work on. However, to get more positive result, he needs to
95
control some independent variables in his study. If are to advise him as PhD
student, what basic methods would you recommend for him in controlling these
independent variables?
• Independent variables could be controlled by using any of these methods:
Holding the variable constant: An independent variable is held constant if
subjects who take a particular value or range of values of the variable only are
studied. Sex is held constant when only males or females are studied.
Randomization of effects of the variable: The effects of a variable are
randomized whenever subjects are selected by randomization. That is subjects
are selected after consultation of the table of random numbers. Subjects are
assigned to treatment groups at random also. The result is dominant in any
group. In fact with this process, the effects of other variables, not known to the
researcher, are also randomized.
Matching of subjects: Subjects may be matched with respect to scores on a
variable critical to a particular study, say achievement in mathematics.
8.4 Types of Research Design
Simple Randomized Design (SPD)
In the simple randomized design, subjects are selected and designed to groups in
consultation with a table of random numbers. Two cases of this design exist.
a. Ex-post Factor Research, the SRD involves the selection of samples,
preferably of equal sizes, from two/more populations with respective
parameters m1, m2, m3. These are population means. The object is to test
whether the population parameters are equal on the basis of sample means.
b. Experimental Research: in experimental research SRD involves the selection
of two or more samples, preferably of the same size, from the same population.
The subjects in the samples are treated. The objective is to evaluate these effect
of treatment on the sample with respect to a given criterion.
8.5 Factorial Design (FD)
A research in which the effects of two or more independent variables are studied
simultaneously is a factorial design independent variables are called factors. This
design has two advantages.
i. Two experiments are conducted simultaneously thus resulting in savings in
time, money and resources.
96
ii. The interaction between the effects of the factors could be investigated
The factorial design is applicable to both ex-post and experimental research.
8.5.1 FD in Ex-post Factor Research
The layout of a 2x 2 factorial design in ex-post factor research is as follows:
Factor 1 (social economic status)
Factor 2
In this example, the research concerns the effect of socio-economic status and sex on
attitude to HIV/AIDS. Hence the factors are SES and sex each held at two levels.
There are four populations with respective to parameters m11, m12, m21, m22. These are
to be tested for equality with respect to a criterion. The students should note that,
samples of about equal size, are selected from each of the four populations, (male
High), (male, Low), (female High and (female, Low).Low).
8.5.2 Factorial Design in Experimental Research
The layout for experimental research is the same except that the population is the
same one not 4 in number. The sample of a given size is selected before randomly
classifying the subjects into (.(male High), (male, Low), (female High and (female,
Low). There are now four experimental groups who may be treated for some attitude
change and later tested for equality on some criterion.
Figure 8.1: Factorial Design (Photo from www.slideshare.net)
Sex; High Low
M m11 m12
F m21 m22
97
The Meaning of Interaction:
For either ex-post factor or experimental research, we say there is an interaction
between the effects of the two factors if:
m11 – m12 m21 – m22
or m11 – m21 m12 – m22
If there is no interaction the differences between the attitudes of people in high and
low socio-economic status should be the same regardless of their sex. Similarly, the
differences between the attitudes of male and female should be the same regardless of
their socio-economic status.
8.5.3 Randomized Blocks Design (RBD)
This design is similar to the factorial design except that one of the factors is now an
independent variable to be controlled, e.g. IQ. This design is utilized as follows: test is
administered on subjects to assess some critical variables to the research. Subjects are
grossly matched with respect to this variable by randomly assigning medium and low
scorers to experimental groups. The controlled variable is now referred to as blocking
variable or covariate. The effect of instructional methods, independent variables may
now be assessed on a given criterion. A3 x 2 RBD is as follows:
Covariate
(Verbal ability)
Medium High
Method I
Method II
Control
Interaction is interpreted as before. The RBD has three advantages over SRD:
i. Some control over blocking variable is achieved through matching samples
grossly on this variable.
ii. The sensitivity of analysis to possible relationship between independent
variables whose effects are to be studied and the dependent variables may be
investigated.
iii. Data for the various levels of the blocking variable may be analyzed separately
Randomized blocks design is usually experimental.
Activity 8.2
Based on what you have read so far, how would you describe the randomised blocks
design.
98
Activity 8.2 Feedback:
This design is similar to the factorial design except that one of the factors is now an
independent variable to be controlled, e.g. IQ. This design is utilized as follows: test is
administered on subjects to assess some critical variables to the research. Subjects are
grossly matched with respect to this variable by randomly assigning medium and low
scorers to experimental groups. The controlled variable is now referred to as blocking
variable or covariate.
8.5.4 Matched – Pairs Design (MPD)
This is a special case of RBD. Subject are paired on the basis of scores on critical
variable to research. The subject from each pair is randomly assigned to each sample
this have 2 samples closely match with respect to the critical variable of interest. The
experiment is carried out as in SRD. Subjects are repaired and analysis is performed
on the differences in the pairs. An elementary analysis is the computation of the mean
gain/loss scores. The more sophisticated analysis is the t-test for related samples
8.5.5 Repeated Measurements Design (RMD)
In RMD, experimental treatments are administered to each subject sequentially. The
criterion variable is administered to each subject following experimental treatment.
The structure is as follows:
Limitations
This design is usable if effects of treatment, (expect last) are temporary and do not
influence the effect of following treatments. Similar limitation also applied to the
criterion i.e. effect of one administration should not influence the next. No learning
effect should occur in administration or treatment. The degree of control is unique
since only one sample is used.
Control is usually affected by not applying the first treatment, that administration of
criterion is to the sample as the control group; the second administration is to the
sample as the experimental group.
Population Treatment
i
Criterion
var. m1 Treatment
2
CR m2
99
8.5.6 Design using ANCOVA
These deigns are used subjects that are not quite equal at the beginning of
experimentation. Samples are selected from same population, independent variables
are controlled (covariates identified), treatments are administered and evaluated on
given criterion. The ANCOVA affects equalization of subjects by a combination of
regression and analysis of variance. Given the covariate, regression methods are used
to predict the criterion scores on the basis of the pretest scores. Analysis of variance is
performed to test equality of groups. Criterion scores are now adjusted to obtain
adjusted scores which are to be used in all further inference.
8.6 Correlation Design
Correlation designs are usually attempts to establish relationship between two sets of
data.
It refer to collection of data from a larger number of people so as to identify trends,
patterns in the opinion on data collected e. g. of this is the opinion Pole. In the survey
research one is not seeking relationship, one rather identify cluster. Often, researchers
believe that if they show that a positive relation exists between two factors, then it
means that one factor is the cause of the other. Correlation does not mean causality.
That one can demonstrate a statistical positive correlation between two sets of data
does not by anyway mean any causality between them.
Correlational research refers to when a researcher collect data and find relationship
between two or more variables e. g. what is the relationship between military and
civilian regime (1) Age and prefer (2) system of government e. g. find performance in
mathematics by urban and rural students that is place of residences versus
mathematics performance.
Experimental research: one manipulates variables so as to establish cause and effect
relationship that is we want to know what causes what. We have to do something so as
to find out cause and effect relationship that is to know whether one variable is the
cause of another variable.
Box 8.1: Correlation Research Design
• Correlation designs are usually attempts to establish relationship between
two sets of data.
• It refer to collection of data from a larger number of people so as to identify
trends, patterns in the opinion on data collected
100
Summary of Study Session 8
In Study Session 8, you have learned that:
Positive correlation does not mean causality.
Ex-post facto studies are not experimental studies because they lack control and
randomness
Lack of control or manipulation is the major difference between ex-post facto and
experimental study;
Two types of self-selection into groups and into samples were identified;
Self-Assessment Questions (SAQs) for Study Session 8
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 8.1 (tests learning outcome 8.1)
How can we correctly explain correlation design.?
SAQ 8.2 (tests learning outcome 8.2)
Define ex-post facto research..
SAQ 8.3 (tests learning outcome 8.3)
Describe method of controlling independent variables.
SAQ 8.4 (tests learning outcome 8.4)
Ex-post facto research has some limitations, discuss at least three of them
Notes on the Self-Assessment Questions (SAQs) for Study Session 8
SAQ 8.1: Correlation designs are usually attempts to establish relationship between
two sets of data. It refer to collection of data from a larger number of
people so as to identify trends, patterns in the opinion on data collected e.
g. of this is the opinion Pole.
SAQ 8.2: In ex-post factor research, the investigator merely selects the independent
variable whose effect are to be studied.
101
SAQ 8.3:
Holding the variable constant
Randomization of effects of the variable
Matching of subjects
Statistical control
SAQ 8.4: This design is usable if effects of treatment, (expect last) are temporary and
do not influence the effect of following treatments. Similar limitation also
applied to the criterion i.e. effect of one administration should not
influence the next. No learning effect should occur in administration or
treatment.
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi: Prentice
Hall. PP 20 – 23.
Cresswell, J.W. (1994). Research Design: Quantitative and Qualitative Approaches Thousand
Oaks C.A.: Sagem PP105 – 107.
Kaufman, R. and Thomas (1980). Evaluation with Fear. New York: New Viewpoint in Best
and Khan, Yorsten, H. (ed.) (1977).
International Study of Achievement in Mathematics. New York: John Wiley, PP10 – 15.
Monette, D.R.; Sullivan, T.S. Dejong, R.D. (1994). Applied Social Research: Tool for the
Human Services (3rd Edition) Chicago: Harcourt Brace College Publishers, PP3 – 8.
102
Study Session 9: Population and Sample
Expected duration: 1 week or 2 contact hours
Introduction
In this session, you will be exposed to the concepts of population and sample in a
research. The primary purpose of research is to discover principles that have universal
application, but to study population to arrive at generalisation is often times
impracticable, if not impossible. Some populations are so large that their
characteristics cannot be measured before the measurement could be completed, the
population itself would have changed. The process of sampling makes it possible to
draw valid inferences or generalisations on the basis of careful observation of
variables within a relatively small proportion of the population. Population and sample
under the following sub-topics: definition of population and characteristics of
population, sampling, sample size, kinds of sampling and sampling frames will be the
focus of this chapter.
Learning Outcomes for Study Session 9
When you have studied this session, you should be able to:
9.1 Define population (SAQ 9.1)
9.2 Explain the term ‘sample’ (SAQ 9.2)
9.3 Describe at least three different types of sampling techniques (SAQ 9.3)
9.4 State the various stages in sampling designs. (SAQ 9.4)
9.1 Population and Samples
A population is any group of individuals that have one or more characteristics in
common that are of interest to the researcher. The population may be all the
individuals of a particular type or a more restricted part of that group. All primary
school pupils, all female secondary school teachers, all university science educators
may be a population”. A sample is a small proportion of a population selected for
observation and analysis. By the characteristics of the sample, one can make certain
inferences about the characteristics of the population from which it is drawn. Contrary
103
to some popular opinion, samples are not selected haphazardly, they are chosen in a
systematically random way, so that chance or the operation of probability can be
utilised. A sample is drawn from a population which refers to all possible cases of
what one is interested in studying. In most behavioural studies, the target population is
often people who have some particular characteristic in common, such as: all students,
all adolescents, all women. A sample consists of one or more elements selected from a
population. The manner in which the elements are selected for the sample has great
implications for the scientific utility of the research based on that sample. To select a
good sample, the researcher need to define clearly the population from which the
sample is to be drawn. Failure to define the population clearly can make generalising
from the sample observations highly ambiguous and result in drawing inaccurate
conclusions. The definition of a population should specify four things – content, unit,
extent and time. The content of the population refers to the particular characteristics
that members of the population have in common. The unit indicates the units of
analysis. The extent of the population refers to its special or geographic coverage. The
time factor refers to the temporal period during which a unit would have to possess the
appropriate characteristics in order to qualify for the sample. An example of a
population could be all second year senior secondary school students in Bodija,
Ibadan North Local Government Area of Oyo State.
9.2 The Purpose of Sampling
A major reason for studying the sample rather than the entire population is that the
population often times very large and it is not feasible to study it for example, if a
study of adolescents consisted of all adolescents in Nigeria, it is not really possible to
study each adolescents in the entire country; the time and weight of report will be very
prohibitive. Sampling allows the researcher to study a workable number of cases from
the large group to derive findings that are relevant to all members of the groups. A
second reason for sampling is that information drawn from an entire group maybe too
bulky and tedious to realise. Also, it offers opportunity for better control. An
improperly drawn sample renders the data collected virtually useless. An important
consideration regarding samples is how representative they are of the population from
which they are drawn. A representative sample is one that accurately reflects the
distribution of relevant variables in the target population. A researcher should always
assess the representativeness of a sample in order to make accurate conclusion.
104
Figure 9.1: Population (Photo from www.thehindu.com)
9.3 Types, Methods and Procedures for Sampling
One basic rule holds whenever a researcher considers methodology in relation to data.
It does not matter whether the methodology concerns sampling, statistical procedure
or any other type of operation. According to Leedy (1977), the general rule is: Look
carefully at the nature of and the quality of the data, not all data subject themselves to
sampling. It is only appropriate whenever large populations that have outward
semblance of homogeneity are to be studied. Sampling can be divided into two major
categories, namely: probability and non-probability sampling.
9.3.1 Probability Sampling
In this type of sampling, the researcher can specify in advance that each segment of
the population will be represented in the sample. This is the distinguishing
characteristic feature that sets it apart from non-probability sampling. The
composition of the sample is derived by selecting units from those of a much larger
population. In survey studies, the way the sample units are selected is important.
Generally, the components of the sample are chosen from the larger population by a
process known as randomisation. Such a sample is known as a random sample. Other
types of probability sampling are: systematic sampling, stratified sampling,
proportionate and disproportionate sampling and area sampling.
Activity 9.1
Before you will continue with the types of samples, it is important to do this. Briefly
explain what population means.
105
Activity 9.1 Feedback:
A population is any group of individuals that have one or more characteristics in
common that are of interest to the researcher. The population may be all the
individuals of a particular type or a more restricted part of that group. All primary
school pupils, all female secondary school teachers, all university science educators
may be a population”.
9.3.2 Non-Probability Samples
Probability samples are not required or even appropriate for all studies. Some research
studies, by their nature, are suited for non-probability samples, samples in which the
researcher does not know the probability of each population being included in the
sample. Although nonprobability samples can be very useful, they do have some
important limitations. First, without the use of probability in the selection of elements
for the sample, no real claim of representative can be made. There is simply no way to
ascertain precisely what population, a non-probability sample represents. This greatly
limits the ability to generalise findings beyond the level of sample cases. A second
limitation is that the degree of sampling error remains unknown and unknowable.
With no clear population being represented by the samples, there is nothing with
which to compare it. It also means that the techniques for estimating sample size are
not applicable to nonprobability samples. The only factor impacting on sample size
for nonprobability samples is that sufficient case can be selected to allow the types of
data analysis that are planned. A third limitation of non-probability samples involves
statistical test of significance. These commonly used statistics indicate to the
researcher whether relationships found in sample data are sufficiently strong and
generalisable to the whole population. All these tests incidentally are based on laws of
probability and assume that a random process is utilised in selecting the sample
elements. Therefore, non-probability samples do not qualify for test of significance. It
should be noted that it is not the size of the sample that makes for accurate results,
rather the procedures used for selecting elements into the samples.
In non-probability sampling, the researcher has no way of forecasting, estimating or
guaranteeing that each element in the population will be represented in the sample.
Non-probability sampling can be divided into two types, namely: convenience or
accidental sampling and quota sampling. Some non-probability selection procedures
may produce samples that do not accurately reflect the characteristic of a population
106
of interest. Such samples may lead to unwarranted generalisations and should not be
used if random selection is possible.
In certain types of descriptive studies, the use of available samples may restrict
generalisations to similar populations. For example, using a Professor of counselling
psychology who uses 100 level guidance and counselling students for a study, the
likely thing is that the generalisation will be limited to that group. A sample made up
of those who volunteers to participate in a study may represent a biased sample.
Volunteers are not representative of a target population, for volunteering results in a
selection of individuals who are different and who really represent a population of
volunteers. In a sense, those who respond to a mailed questionnaire are volunteers and
may not reflect the characteristics of all who were on the mailing list (Best and Khan,
1995).
9.3.3 Convenience or Accidental Sampling
Convenience sampling makes no pretence of being representative of population. It
takes the units as they arrive on the scene or as they are presented to the researcher by
chance. No attempt is made to control bias. Depending on the purpose and the
researcher’s awareness of the crudity of this type of sampling, convenience sampling
may be appropriate tom some less demanding kinds of data collection.
9.3.4 Quota Sampling
It is a variant of convenience sampling. It selects respondents in the same ratio as they
are found in the general population. For example, supposing the researcher want to
interview 50 Psychology students each from two nearby universities on the use of
Cognitive Behaviour Therapy (CBT) on every subject. Quota sampling demands that
the researcher just select the first available 50 psychology students from each
universities . The only assumption is that the numbers of psychology students in both
universities are approximately equal. This is also an unregulated type of sampling. It
has no limitation except the size of each category. The researcher have no probability
of how the 50 will arrive at your interview site. Other types of non probability
sampling are: Availability Sampling, Snowball Sampling, Purposive Sampling and
Dimensional Sampling. The ultimate objective of any research is to collect data
(information) so as to provide some useful principles, theories and generalisations.
107
The importance of valid and reliable data in research work requires no emphasis
whatsoever.
9.4 Probability Samples
Every sample approach is expected to provide a representative sample. But the
concern of a researcher is to be as certain as possible that his/her is a representative.
Sampling approaches that make use of the probability theory can both greatly reduce
the chances of getting a non-Representative sample and what is more important,
permit the researcher to estimate precisely the likelihood that a sample differs from
the true population by a given amount.
Figure 9.2: Probability Sample (Photo from www.sesre.wsu.edu)
In probability samples, each element in the population has equal chance of being
selected by the researcher. The calculated sampling error is an estimate of the extent
to which the values of the sample differs from those of the population from which it is
drawn.
Activity 9.2
Make sure you do this activity, it will help you monitor your progress in learning.
Distinguish between accidental and quota sampling techniques.
Activity 9.2 Feedback:
Convenience sampling makes no pretence of being representative of population. It
takes the units as they arrive on the scene or as they are presented to the researcher by
chance. No attempt is made to control bias. On the other hand, quota sampling is a
108
variant of convenience sampling. It selects respondents in the same ratio as they are
found in the general population. For example, supposing the researcher want to
interview 50 Psychology students each from two nearby universities on the use of
Cognitive Behaviour Therapy (CBT) on every subject.
9.4.1 Probability Sampling Techniques
In this section of the unit, we shall study, in details, each of the probability sampling
techniques. These are: random sampling, systematic sampling and area sampling.
The nature of the data is expected to determine the methodology of the investigation.
This fact is also applicable to sampling. Careful consideration of the target population is
most important. The researcher should select from the general population a sample
that will be both logically and statistically defensible. The first step in selecting any
sampling design is to analyse carefully the integral characteristic of the target
population. In view of this, the researcher could then select the sampling technique
most appropriate for the population type. Here, there is need to understand the general
characteristics of a population. These are:
- the population may be generally homogenous, the separate units may be similar
in observable characteristics;
- population may contain definite strata but each stratum may differ from every
other stratum by a proportionate ratio of its separate units;
- the population may consist of cluster whose characteristic are similar, but
whose units characteristics are as homogenous as possible.
According to Leedy (1997), in all sampling, the following three processes are
indigenous to the selection of the sample:
- the population must be identified, an analysis must be made of its structure,
and an assessment made of its characteristic;
- the process of randomisation must be outlined, and the selection of the sample
from the total population must be made in accordance with a method of
randomisation;
- the data must be extracted from the sample population.
9.4.2 Random Sampling
This is based on the assumption that, while individual events cannot be predicted with
accuracy, aggregate events can. For example, although it may not predict with great
109
accuracy an individual’s academic achievement, it will predict the average academic
performance of a group. Randomisation has two important applications in research:
i. to select a group of individuals for observation who are representative of the
population about which the researcher wishes to generalise; and
ii. to equate experimental and control groups in an experiment. Assigning
individuals by random assignment is the best method of providing for their
equivalence. to equate experimental and control groups in an experiment.
Identical representation of the population. Characteristics of successive random
sample drawn from the same population may differ to some degree, but it is possible
to estimate their variation from the each other. The variation known as sampling
error, does not suggest that a mistake has been made in the sampling process. Rather,
sampling refers to the chance variations that occur in sampling with randomisation,
these variations are predictable and taken into account in data analysis techniques.
9.4.2 Simple Random Sample
The most basic method of selecting a sample from population is random sampling. It
is not haphazard sampling. It is a procedure of selection of a sample from population
where individuals are chosen in such a way that each has an equal chance of being
selected, and each choice is independent of the other. One way to ensure this equality
of chance of selection is to number all the members of the population and then make
the simple selection by using a table of random. Thus in taking a sample of 200
students in a particular faculty on whom to administer Student Academic Stress Scale
(SASS), one may number all the students in the faculty. Suppose the total number of
students is 1000. A table of random numbers may be used to select the first 200
numbers which fall in the range of 0001 and 1000. The 200 students designated by
those numbers constitute a sample size of 200. Other approximate method of selecting
a random sample is to put those student numbers 0001 to 1000 on slips of paper. If
you wish to select two hundred students from a list of one thousand students in a
college, all you need to do is to assign a number to each of the students, fold their
numbers and place them in a basket, after shaking them thoroughly, you ask one of the
students or you do it yourself and select any two hundred from the basket. You will
discover that each of the one thousand students had equal chance of being selected.
Simple random sample for example Drug Abuse among 1991/92 session students
University of Ibadan. The population cannot be identified what we do is to draw
110
simple random sample which is defined as a group in which all elements have equal
chance of appearance or shown. The language is derived from sampling e. g. all
faces of dice must have chances of appearance e. g. choosing five people from class of
50, they must have equal chance of being chosen. Simple random sample is selected
from infinite, unknown and un-accessible population.
Figure 9.3: Random Sampling in Research (Photo from www.mediablogs.kesha.com)
Random sampling is likely to be effective in selecting subjects for psychological
studies in the sense that the sample will ordinarily represent the population because
the relevant population itself is usually homogenous with regard to the attribute, that is
the behaviour being investigated.
Table of Random Numbers
The use of table of random numbers is a more convenient way of selecting a random
sample, or assigning individuals to experimental and control groups so that they are
equated. It is perhaps the most frequently used method of random selection of a
sample.
9.4.3 Systematic Sampling
This is a sampling technique in which the elements are listed and numbered. They are
known, accessible and can be numbered e. g. this class is listed population because
they have their names written down. This method may be used whenever the
population is much larger than the size of the sample required and the investigator
does not wish to take a simple random sample. For example, if a sample of five
hundred people is to be selected from a list of 500,000 people, the first name would be
selected randomly from the list. This can be done by tossing coin and decide to start
111
with an even number, or odd number then every thousandth number would be
selected until the desired number is selected. If the desired number has not been
selected, the count would continue from the beginning again until the desired number
is obtained. For example, if you have a school admission register of 500,000 people,
you could select the required 500 candidates using this procedure.
To take systematic sample a number of specific step must be taken.
i. The 1st step is to give them number. Total number is known here.
ii. Determine the size of sample an individual want e. g. sample size of five out of
40.
iii. Determine the sample interval by dividing the total population by the sample
size 40 ÷ 5 = 8.
iv. Determine the starting point for the sampling which is usually done randomly.
It is not research wise to use Alphabetic names which may introduce a bias into
research design. A starting point can be used by closing eyes and put biro on the paper
to show the starting point 1st person out of five. Count 8 from there and that is going
to be 2nd etc. until we get the five people.
Activity 9.3
Make sure you do this activity, it will help you monitor your progress in learning.
Briefly discuss the concept of random sampling.
Activity 9.3 Feedback:
This is based on the assumption that, while individual events cannot be predicted with
accuracy, aggregate events can. For example, although it may not predict with great
accuracy an individual’s academic achievement, it will predict the average academic
performance of a group. Randomisation has two important applications in research:
i. to select a group of individuals for observation who are representative of the
population about which the researcher wishes to generalise; and
ii. to equate experimental and control groups in an experiment.
9.4.4 Stratified Random Sampling
In stratified sampling, the unit is divided into smaller subgroups called strata before
drawing the sample, and then separate random samples are drawn from each of the
strata. In this sampling procedure, the researcher could have proportionate stratified
sampling and disproportionate sampling. Stratified population is divided into different
112
layers e. g. 500 people in the class divided into boys and girls later subdivided into
Boarder/Day students then sub-divided into Yoruba/Ibo. The best example is
University population e. g. Faculties, departments, levels, sex which is divided into
structural groups e. g if in population of 500 we want a sample of 100; the structural
division of 100 must follow that of 500. To find total sample of boys we divide the
total population by the sample population of boys multiply by 100
The rationale for stratification is to obtain greater precision by sampling from
homogeneous sub- populations of the same population. The reader may have rightly
guessed that stratification is at its best whenever a population is heterogeneous.
9.4.5 Proportionate Sampling
In this sampling procedure, the size of the samples taken from each stratum is
proportionate to the stratum’s presence in the population. For example, you want to
study the effect of religious belief on the performance of students in a course on
evaluation. If there were 600 Christians, 400 Muslims and 200 Traditional
worshippers in the class, it means you will select the sample of 180 students based on
the proportion of each of the strata that is in the ratio of 3:2:1 and that will be 90
Christians, 60 Muslims and 30 Traditional worshippers. You will subsequently use a
simple random sampling technique to select the required number from each group.
9.4.6 Disproportionate Sampling
Stratified samples could be used to enable a researcher to make comparisons among
various subgroups in the population when one or more of the subgroups are relatively
uncommon. For example, if you want to ascertain the quality of science techniques in
a local government area and you want to compare the views of students from the local
government area. You may find that students outside the local government area may
constitute less than 10% of the students in each school. Which means that, for every
100 students in a school, only 10 will come from outside the local government area.
This number would be far too small to make meaningful statistical comparison.
Stratifying, in this case, would allow us to draw a larger sample of the students from
outside the local government area to provide enough cases for reliable comparison to
be made. This is called disproportionate stratified sampling because the strata are not
sampled proportionately to their presence in the population. This type of sample is
different from most probability samples where representative is achieved by giving
every element in the population an equal chance of appearing in the sample, but the
113
element in some strata have a better chance of appearing in the overall sample than do
the elements of other strata. The characteristics of the entire population together with
the purposes of the study must be carefully considered before a stratified sample is
decided upon.
9.4.7 Multi - Stage or Cluster Sampling
This is simply sampling in stages. It is often referred to as cluster sampling. The
method is at its best when the population attribute of interest occur in clusters or
stages.
This is a different type of population sampling from a large population spread across a
large area. Most of the time, it is not feasible to make a list to select a sample for study
through normal randomisation procedures. Because of this, the area has to be
subdivided into subunits for effective handling. A state, for example, could be
subdivided into local governments and further to zones or wards in a series of
sampling clusters to smaller ones. For example, if you want to seek the opinions of
retired Professor of counselling psychology about the status of counselling practises in
Nigeria secondary schools, you could first use random sampling to select five states.
Then in each state, you use random sampling to select ten local governments: from
each local government, five wards could be selected, also by simple random sampling.
9.5 The Sample Size
It has become necessary to present two views on sample size here. It appears
educators have not agreed on sample size. Best and Khan (1995) explain that there is
usually a trade-off between the desirability of a large sample and the feasibility of a
small one. The ideal sample is large enough to serve as an adequate representation of
the population about which the researcher wishes to generalise, and small enough to
be selected economically – in terms of subject availability, expense in both time and
money and the complexity of data analysis. There is no fixed number or percentage of
subjects that determines the size of an adequate sample. It may depend upon the
nature of the population of interest or the data to be gathered and analysed. It is often
stated that samples of 30 or more are to be considered large samples and those fewer
than 30 are considered as small samples. It is approximately at this sample size of 30
that the magnitudes of students’ critical values for small samples approach the Z
critical value of the normal probability table for large samples. Leedy (1997), on the
114
other hand, provided a statistical formula for determining the size of the sample
statistically thus;
N = (z/e)2 (p) (1 – p)
Where:
N =150
size of the sample
Z = the standard score corresponding to a given confidence level
e = the proportion of sampling error in a given situation
p = the estimated proportion or incidence of cases in the population.
the probability sampling strategies that lend themselves to random sampling are
preferred for research than the non probability sampling techniques;
Leedy recommends a maximum of 400 as an adequate sample size;
most authors emphasise the careful and painstaking selection of sample to a large
sample size.
Summary of Study Session 9
In Study Session 9, you have learned that:
- a population is any group of individual that have one or more characteristics in
common that are of interest to the researcher;�a sample is a small proportion
of a population selected for observation and analysis;
- four factors associated with population are: content, unit, extent and time;
sample is studied because population could be very large and near impossible to
study;
carefully selected representative sample tends itself to generalization of findings;
there are two major sampling techniques; probability sampling techniques comprise:
random sampling, systematic sampling, stratified sampling, proportionate and
Box 9.1: Cluster Sampling
• This is simply sampling in stages. It is often referred to as cluster
sampling.
• The method is at its best when the population attribute of interest occur
in clusters or stages.
• This is a different type of population sampling from a large population
spread across a large area.
115
disproportionate sampling; sampling is of two types: convenience or accidental and
quota.
Other major types of non-probability sampling are: availability sampling; snowball
sampling, purposive sampling and dimensional sampling; and non-probability
sampling has some limitations which include level of generalisability of finding and
acceptability of findings.
Self-Assessment Questions (SAQs) for Study Session 9
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 9.1 (tests learning outcome 9.1)
Define population.
SAQ 9.2 (tests learning outcome 9.2)
What do you understand by sample?
SAQ 9.3 (tests learning outcome 9.3)
Describe at least three different types of sampling techniques.
SAQ 9.4 (tests learning outcome 9.4)
State the various stages in sampling designs
Notes on the Self-Assessment Questions (SAQs) for Study Session 9
SAQ 9.1: A population is any group of individuals that have one or more
characteristics in common that are of interest to the researcher. The
population may be all the individuals of a particular type or a more
restricted part of that group.
SAQ 9.2: A sample consists of one or more elements selected from a population.
The manner in which the elements are selected for the sample has great
implications for the scientific utility of the research based on that sample.
SAQ 9.3:
Probability Sampling
Convenience sampling
Random Sampling
116
SAQ 9.4: To take systematic sample a number of specific step must be taken.
i. The 1st step is to give them number. Total number is known here.
ii. Determine the size of sample an individual want e. g. sample size of
five out of 40.
iii. Determine the sample interval by dividing the total population by
the sample size 40
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi:
Prentice Hall. PP 20 – 23.
Cresswell, J.W. (1994). Research Design: Quantitative and Qualitative Approaches
Thousand Oaks C.A.: Sagem PP105 – 107.
Kaufman, R. & Thomas (1980). Evaluation with Fear. New York: New Viewpoint in
Best and Khan, Yorsten, H. (ed. )(1977). International Study of Achievement in
Mathematics. New York: John Wiley, PP10 – 15.
Monette, D.R.; Sullivan, T.S. Dejong, R.D. (1994). Applied Social Research: Tool for
the Human Services (3rd Edition) Chicago: Harcourt Brace College Publishers,
PP3 – 8.
117
Study Session 10: Level of Measurement Validity and Reliability
Expected duration: 1 week or 2 contact hours
Introduction
In this chapter, we shall look into different forms of measurement as well as how they
are considered defendable by considering their validity and reliability. Our earlier
discussion shows that there are numerous ways of measuring phenomena such as: by
asking questions or noting observations. Measures also differ from one another in
terms of what is called the level of measurement: nominal, ordinal, interval and ratio.
It is important to note that variables are things that can take on different values,
measurement basically involves assessing the value, or category into which a
particular entity falls. For example, measuring age is the process of placing each
person into a particular age category. Whatever procedure for collecting data is
selected, it should always be examined critically to assess to what extent it is likely to
be reliable and valid
Learning Outcomes for Study Session 10
When you have studied this session, you should be able to:
10.1 Define the term ‘Measurement’ (SAQs 10.1)
10.2 Describe the different levels of measurement (SAQ 10.2)
10.3 List the appropriate reliability methods to be used on a set of data (SAQ 10.3)
10.4 Explain three different types of validity (SAQ 10.4)
10.1 Levels of Measurement
Measurement is “the assignment of numerals to objects or evens according to rules”.
Data may be characterised and classified in four ways: nominal data, ordinal data,
interval (continuous) data and ratio data. These four levels of measurement are the
recognised ones in measurement and are often referred to as scales of measurement.
Nominal Level (Scale)
Measurement, in this case, could be seen as some form of restriction or limitation.
Assign a specific name to anything and you have restricted that thing to the meaning
118
of its name. For example, you can measure a group of children by dividing it into two
groups: boys and girls, each group is thereby measured – “restricted” by virtue of
gender to a particular category. By assigning a name, you create a measurement.
Things can be measured nominally in an infinite number of ways. Variables which
assume only two integral values, often of the form, either/or true /false, yes /no, etc
.are nominal. Such variables are encountered in behavioural research. Examples
include, sex, socio–economic status (high /low),marital status (married/single).Even
when values such as 0 or 1 are forced on these variables such values are strictly not
subject to arithmetic operations. Hence the question of finding statistics such as mean
and variance does not arise. Consequently for statistical purposes these variables are
only nominal rather than being manipulated. For example, you can measure the
groups above, according to their religion, where they live, etc. The data can be
statistically treated by finding mode, percentage, chi-square etc.
Ordinal Level (scale)
Sometimes, measures can be ranked in ascending or descending order of magnitude
.Consider the measures 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. In most cases, all that may be valid
about this distribution is that the first position is preferable to the 5th for example. The
first cannot be said to be five times more preferable than the first. Nor the 4th twice
preferable than the 8th. Consequently the above measure are not subject to the
operations of arithmetic just as the nominal. They are only ordered, hence referred to
as ordinal. Neither mean nor variance of the two statistics on which statistical
inference depends can be calculated from such measures. This scale orders the scores
on some basis, such as: low to high or least to most. Although the scores are ordered,
equal intervals between scores are not established. In measuring on the ordinal scale,
the relationship is always asymmetrical. Something is always greater than, less than,
older than, younger than etc. Using ordinal scale, you can find the mode and the
median, determine percentage or the percentile rank and test by the chi-square.
119
Figure 10.1: Measurement Level (Photo from www.socialresearchmethods.net)
Interval Level (scale)
The interval scale is superior to the first two scales earlier discussed. Variables whose
values may be group as in 5-10, 11-15, 16-20, etc. for example are said to be in
interval scale. They are subject to arithmetic operations. So statistics such as the mean
and variance can be computed. But the interval scale may lack true zero. Scores on
constructs such as motivation, truancy, perception etc. lack true zero.
The interval level of measurement is characterised by two features, namely:
i. It has equal units of measurement;
ii. Its zero point has been established arbitrarily.
For example, the thermometric scale is equal to the others and the zero point has been
established arbitrarily. A common use of the interval measurement is in the rating
scale employed by many businesses, survey groups, and professional organisations.
Unlike the nominal and ordinal data, any data generated by using the interval scale
can be subjected to all statistical operations starting with the basic ones of counting,
ranking, adding, subtracting to the use of analysis of variance etc.
Ratio Level (scale)
The advantage of the ratio scale over the interval scale is its possession of a true zero.
This is what makes it superior to all the other scales. Parametric methods of statistical
inference presupposes data on this scale. Data in interval scale are also acceptable for
use in parametric statistics as long as the investigators conscious of the absence of a
true zero in drawing inferences. The highest level of measurement is ratio measures.
Ratio measures have all the characteristics of interval measure, but the zero points is
120
absolute and meaningful rather than arbitrary. As the name implies, you can make
statements to the effect that some score is a given ratio of another score.
Activity 10.1
Differentiate between nominal and ordinal scales.
Activity 10.1 Feedback:
In nominal scale, measurement, could be seen as some form of restriction or
limitation. Assign a specific name to anything and you have restricted that thing to the
meaning of its name. For example, you can measure a group of children by dividing it
into two groups: boys and girls, each group is thereby measured – “restricted” by
virtue of gender to a particular category. On the other hand, ordinal scale orders the
scores on some basis, such as: low to high or least to most. Although the scores are
ordered, equal intervals between scores are not established. In measuring on the
ordinal scale, the relationship is always asymmetrical. Something is always greater
than, less than, older than, younger than etc.
10.2 Validity and Reliability
Validity and reliability are two words that are used repeatedly in research
methodology. They are used primarily in connection with measurement instruments.
The integrity of your research may well stand or fall on the basis of how well you
understand their meaning and carefully obey their demands. They govern the
acquisition of data and the skillfulness with which you design the research structure
and create the instruments of measurement as an integral part of it
10.2.1 Validity
Validity is the extent to which a test result measures what it has been designed to
measure. In a standardised test, for example, validity would raise such questions as:
What does the test measure? Does it in fact measure what it is supposed to measure?
How well? How comparatively? How accurately does it measure it? In general, a test
is valid, if it measures what it claims to measure. Validity can also be thought of as
utility. There are several types of test, and uses of test need different types of validity.
The seven most common types with a brief discussion of each are listed below;
Face Validity: This type of validity relies basically on the subjective judgement of the
researcher. It asks two questions that the researcher must finally answer in accordance
121
with his or her best judgement is the instrument measuring what it is supposed to
measure? And is the sample being measured representative of the behaviour or trait
being measured?
Criterion Validity: This type of validity is determined by relating performance on
one measure (e.g. a test) to performance on another measure called a criterion. The
essential component in criterion validity is a reliable criterion – a standard against
which to measure the results of the instrument doing the measuring. The data of the
instrument should correlate highly with equivalence data or the criterion.
Content Validity: This type of validity is sometimes equated with face validity.
Content 88validity is the accuracy with which an instrument measures the factors or
situations under study – the accuracy with which an instrument measure the factors or
situations under study, that is, the content being studied. If, for example, you are
interested in the content validity of questions being asked to elicit familiarity with a
certain area of knowledge, content validity would be concerned with how accurately
the question asked tend to elicit information sought.
Construct Validity: A construct is any concept that cannot be directly observed or
isolated, e.g. attitude. Construct validity is concerned with the degree to which the
construct itself is actually measured. Different methods of measuring the same
construct should ‘converge’ or ‘focus’ in their result. Discriminability means that the
measuring instrument should be able to discriminate or differentiate the construct
being studied from other similar constructs.
Internal Validity: Internal is the freedom from bias in forming conclusions on
various data. It seeks to ascertain that the changes in the dependent variable are the
results of the influence of the independent variable rather than of the way the research
was designed.
External Validity: This type of validity is concerned with the generalisability of the
conclusions reached through observation of a sample to the universe, or more simply
stated, can the conclusions drawn from a sample be generalised to other cases.
122
Predictive Validity: This refers to the usefulness of a test in predicting some future
performance such as: the college performance. If a test is designed to pick out good
candidates for appointment on a job, and test score show a high positive correlation
with later actual success on the job, the test has a high degree of predictive validity,
whatever factor is actually measured. It predicts well. It serves a useful purpose. But
before determining a test’s predictive validity, success on the job must be accurately
described and measured. Predictive validity is not easy to assess. It is often difficult to
discover whether the faults of prediction lie in the test, in the criteria of success or
failure or both. Validity looks to the end result of measurement. The principal
question that validity asks is: “is the measure measuring what it is supposed to
measure?” Unfortunately, many students limit their assessment to content validity
with all its heavy reliance on the subjective judgment, but content validity should be
used with caution.
10.2.2 Reliability
Reliability is the consistency with which a test result measures what it has been
designed to measure (Oloyede, 2007). To ensure the integrity of any research, the
researcher should consider it mandatory to state clearly and definitively the
specifications of the measuring instrument. Here, you are dealing with the foundation
factors of the entire research structure. Just as a building with a questionable
foundation is very likely to be unsafe for habitation, so also is a research effort
employing faulty tools of measurement will likely end as a defective research
endeavour. There are many specific techniques for estimating the reliability of a
measure, but they are all based on one or two principles – stability or equivalence.
Stability is the idea that a reliable measure should not change from one application to
the next, assuming the concept being measured has not changed. Equivalence is the
idea that all items that make up a measuring instrument should be measuring the same
thing and thus be consistent with one another. The first techniques for estimating
reliability, i.e. test-retest reliability, are based on the stability approach.
Activity 10.2
Before you will continue reading the concept of reliability, distinguish between face
and criterion validity.
123
Activity 10.2 Feedback:
Face Validity: This type of validity relies basically on the subjective judgement of the
researcher. It asks two questions that the researcher must finally answer in accordance
with his or her best judgement is the instrument measuring what it is supposed to
measure? And is the sample being measured representative of the behaviour or trait
being measured?
Criterion Validity: This type of validity is determined by relating performance on
one measure (e.g. a test) to performance on another measure called a criterion. The
essential component in criterion validity is a reliable criterion – a standard against
which to measure the results of the instrument doing the measuring.
Figure 10.2: Validity and Reliability (Photo from www.exploreable.com)
Assessment of Reliability
The assessment of reliability are discussed under the following subtopics:
Test-Retest; Split-Half Approach Formulae.
Test-Retest
The first and most generally applicable assessment of reliability is called test – retest.
This technique involves applying a measure to a sample of subjects and then
somewhat later, applying the same measure to the same subjects again. After the
retest, you have two sets of score on the same measure. These two sets of scores are
correlated by using appropriate statistical measure of association, because with scores
obtained from the two identical questionnaires, a high degree of association is
expected. According to Monettee et. al. (1994), a correlation coefficient of 0.8 or more
is normally necessary for a measure to be considered reliable.
124
Split-Half Approach
In the split-half approach to reliability, the test group responses to the instrument are
then divided into two halves. Each half is then treated as though it was a separate
scale, and the two halves are correlated by using an appropriate measure of
association. Because the split half approach divides the scale, each half is shorter than
the whole scale. To correct for this, the correlation coefficient is adjusted by applying
Kuder-Richardson Formulae
Kuder and Richardson developed a number of formulae for determining reliability
using scores from only one test administration. The correlation coefficient calculated
is a measure of internal consistency. A basic assumption is that the score are
homogenous and therefore possess inter-item consistency
Summary of Study Session 10
In Study Session 10, you have learned that:
The basic levels of measurement are: nominal; ordinal; interval and ratio levels of
measurement.
There are different methods of determining validity and reliability. Types of validity –
face, criterion, construct, content, internal, external and predictive validity; and
reliability – test-retest, split-half and Kuder-Richardson formulae.
Self-Assessment Questions (SAQs) for Study Session 10
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 10.1 (tests learning outcome 10.1)
How can we define the term ‘Measurement’?
Box 10.1: Assessment of Reliability
• In the split-half approach to reliability, the test group responses to the
instrument are then divided into two halves.
• Each half is then treated as though it was a separate scale, and the two
halves are correlated by using an appropriate measure of association.
• Because the split half approach divides the scale, each half is shorter than
the whole scale.
125
SAQ 10.2 (tests learning outcome 10.2)
Describe the different levels of measurement
SAQ 10.3 (tests learning outcome 10.3)
List the appropriate reliability methods to be used on a set of data.
SAQ 10.4 (tests learning outcome 10.4)
Validity is an important aspect in constructing instruments, explain three different
types of validity.
Notes on the Self-Assessment Questions (SAQs) for Study Session 10
SAQ 10.1: Measurement is “the assignment of numerals to objects or evens
according to rules”. Data may be characterised and classified in four
ways: nominal data, ordinal data, interval (continuous) data and ratio
data.
SAQ 10.2:
Nominal Level (Scale)
Ordinal Level (scale)
Interval Level (scale)
Ratio Level (scale)
SAQ 10.3: Test-Retest
The first and most generally applicable assessment of reliability is called
test – retest. This technique involves applying a measure to a sample of
subjects and then somewhat later, applying the same measure to the
same subjects again.
SAQ 10.4:
Face Validity
Criterion Validity
Content Validity
126
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi:
Prentice Hall. PP 20 – 23.
Cresswell, J.W. (1994). Research Design: Quantitative and Qualitative Approaches
Thousand Oaks C.A.: Sagem in P.D. Leedy (1997) Practical Planning and
Design. (6th Edition) New Jersey: Merril, PP103 – 110.
Monette, D.R.; Sullivan, T.S. & Dejong, R.D. (1994). Applied Social Research: Tool
for the Human Services (3rd Edition) Chicago: Harcourt Brace College
Publishers, :3 – 8.
Oloyede, I.O. (2007). Measurement and Evaluation for Tertiary Institutions:
127
Study Session 11: Test Administration and Interpretation
Expected duration: 1 week or 2 contact hours
Introduction
This session exposes you to the concept of test administration and interpretation in a
research. The condition in which a test is administered is very critical to the
interpretations of the responses of subject. This section briefly discusses the more
important human and non-human conditions which affect test results. All the
conditions must be uniform in order for the responses subjects to be comparable. It is
therefore for the purpose of comparison among individual scores that the tester must
ensure that all testees take the test under uniform conditions.
Learning Outcomes for Study Session 11
When you have studied this session, you should be able to:
11.1 Explain the process of test administration (SAQ 1.1)
11.2 Describe the two aspects of test venue that are particularly important. (SAQ
11.2)
11.3 List the two activities that are expected of a tester during testing session.
(SAQ 11.3)
11.1 Process of Test Administration
Advance Preparation by Tester
In preparing to administer a test, a tester should pay attention to the following:
a. Test materials: Preparations with regards to test materials involve a listing of all
items together with the quality of each to be used. Also, test items should be sorted
out according to the type and order in which they will be used. All these helps to
save testing time and increase morale of testees.
b. Proctoring or invigilation: Arrangement should be made for adequate number of
proctors who should have been adequately briefed about their work. Effective
invigilation requires that there should be an invigilator to about 25 testees. Proctors
perform the following function: distribution of test materials, collection of
128
materials after use, provision of sharp pencils as required by testees, remaining
alert to problem arising from inadequacy of materials, insufficient grasp of
directions and abnormal behaviour of testees. Except security forbids, proctors
should study test and test directions before administration.
c. Recording of Proceedings: Arrangement should be made in advance for the
recording of the various processes taking place during testing. The following
records should be kept, time of starting of test, time test ended, and incident during
testing.
d. Sequence of Tests: For tests which are conducted in series, the administrator
should note the following guides; easy tests should be given first, the first set of
tests should not be personal like intelligence or personality tests, the first test
should be objective and have face validity, the middle tests may be difficult and
self-searching, the last tests should be short, pleasant and possibly speeded so all
testees could leave at the same time in high spirits.
11.2 Test venue
Two aspects of test venue are particularly important. These are stated below:
a. Freedom form distraction: A good test venue is one which testees are not
disturbed by people, incidents, noises, or views which could draw their attention away
from the tests. The effects of these elements will certainly vary across testess who are
presumably at various levels of motivation for the test. It is however advisable for the
tester to have the use of a venue through which there is no passage and to which no
one needs to have access during testing. Test venue should not be affected by noise in
adjacent rooms, corridors or place space. As much as possible the temperature in test
venues should be normal and reasonably constant.
Figure 11.1: Students Writing Test (Photo from www.mmnfpe.org)
129
b. Working space A good working space for the testees is essential in testing. For
the purposes of group testing with which this course is concerned, properties of a good
working space are worth emphasis. These are sufficiency of space and privacy of
work. Operationally, a flat topped table, which is sufficiently large for the testee to be
able to rest his elbows without touching his neighbour and to spread out his papers
without exposing them to the eyes of such neighbours constitutes a good working
space. Rather than violating the above condition, a tester may have to arrange for two
testing sessions while taking necessary precautions.
o Mr Shola just took up an appointment with a secondary school after graduating
as an education student. After series of lessons, he believed that the students
needed to be tested on how far they have gone with their learning. What are the
basic conditions Mr Shola should give attention to in administering the test?
• In preparing to administer a test, a tester should pay attention to the following:
a. Test materials: Preparations with regards to test materials involve a listing
of all items together with the quality of each to be used.
b. Proctoring or invigilation: Arrangement should be made for adequate
number of proctors who should have been adequately briefed about their
work.
c. Recording of Proceedings: Arrangement should be made in advance for
the recording of the various processes taking place during testing.
d. Sequence of Tests: For tests which are conducted in series, the
administrator should note the following guides; easy tests should be given
first, the first set of tests should not be personal like intelligence or
personality tests, the first test should be objective and have face validity, the
middle tests may be difficult and self searching, the last tests should be
short, pleasant and possibly speeded so all testees could leave at the same
time in high spirits.
It holds that when a highly charged emotional consequence (C) follows a significant
Activating Agent (A) A, may seem to but actually does not cause C. Instead,
emotional consequences are largely created by B - the individual's belief system.
When an undesirable consequence therefore occurs, such as severe anxiety, this can
usually be quickly traced to the person's irrational beliefs, and when these beliefs are
effectively disputed at (point D) by challenging them rationally there will be change in
130
self verbalizations and change in behaviour and the disturbing consequences will
disappear.
Activity 11.1
Take a moment and briefly describe the conditions that are important in organising a
test venue.
Activity 11.1 Feedback:
Two aspects of test venue are particularly important. These are stated below:
a. Freedom form distraction: A good test venue is one which testees are not
disturbed by people, incidents, noises, or views which could draw their
attention away from the tests.
b. Working space A good working space for the testees is essential in testing. For
the purposes of group testing with which this course is concerned, properties of
a good working space are worth emphasis. These are sufficiency of space and
privacy of work.
11.3 The actual testing session
Two activities are expected to be performed by the tester before and during the testing
session. These are introductory talk and familiarization with testing instructions
a. Introductory talk: The tester should arrive early at the testing venue to ensure
that materials are ready for use before testing session actually begins. The
tester should also talk briefly to the testees about the purpose of the test, the
work required of them and the time to complete the task. If possible, the tester
should relate the testing exercise to something of interest to the testees.
b. The Scoring: The last process in having a set of psychological data at the
disposal of an investigator is test scoring. It is as important as any of the above
processes. For if the responses of testees are not properly, consistently and
uniformly scored the basis for comparing testees’ performances ceases to exist
just as if they did not take the test under uniform conditions.
By test scoring we mean a process of attaching specific number to the response of
testees. To most readers it may seem like killing a rat with an elephant gun to make an
issue resting. It is the process which yields the information which the investigator
have been looking for through all the other processes. It is not a casual matter. Any
131
error in scoring distorts the true performance of testee. Consequently scoring has to be
so objectively done that testees handwriting and tester’s disposition to be lenient or
otherwise do not affect the test result. Fortunately, psychological tests follow the
objective format
The objective test items are either hand scored or machine scored. Whichever the
scoring option that is used depends on the scope of the investigation in items of
number of testees and the number of items. The time available within which to
produce the result is also an important factor which is equally important as the
resources available to the investigator of either hire people or pay for machine time.
However, what is important for our purpose is that test scores are obtainable by both
machine and hand. In fact the leader may imagine that all the preceding processes
have
Summary of Study Session 11
In Study Session 11, you have learned that:
The learners have being exposed to various steps to be taken for effective
administration and interpretation of tests.
Students have been taking through the process of administration of test which includes
advance preparation by testers, working space, test venue, freedom from distraction,
the actual testing session, introductory talk and the scoring of the test
Self-Assessment Questions (SAQs) for Study Session 11
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
Box 11.1: Actual Testing Session
• The last process in having a set of psychological data at the disposal of
an investigator is test scoring.
• It is as important as any of the above processes.
• For if the responses of testees are not properly, consistently and
uniformly scored the basis for comparing testees’ performances ceases
to exist just as if they did not take the test under uniform conditions.
132
SAQ 11.1 (tests learning outcome 11.1)
How will you explain the process of test administration?
SAQ 11.2 (tests learning outcome 11.2)
Describe the two aspects of test venue that are particularly important
SAQ 11.3 (tests learning outcome 11.3)
List the two activities that are expected of a tester during testing session.
Notes on the Self-Assessment Questions (SAQs) for Study Session 11
SAQ 11.1: In preparing to administer a test, a tester should pay attention to the
following:
a. Test materials: Preparations with regards to test materials involve a
listing of all items together with the quality of each to be used. Also,
test items should be sorted out according to the type and order in
which they will be used. All these helps to save testing time and
increase morale of testees.
b. Proctoring or invigilation: Arrangement should be made for
adequate number of proctors who should have been adequately
briefed about their work.
SAQ 11.2: These are:
Freedom form distraction
Working space
SAQ 11.3:
Introductory talk
The Scoring
Reference
Monette, D.R.; Sullivan, T.S. & Dejong, R.D. (1994). Applied Social Research: Tool
for the Human Services (3rd Edition) Chicago: Harcourt Brace College
Publishers, ::3 – 8.
133
Study Session 12: Writing Research Report
Expected duration: 1 week or 2 contact hours
Introduction
This last session in the course will expose you to the rudiments of writing research
report after completing data generation and analysis. A research work will be a
worthless effort if after an excellent problem has been identified, a clearly stated and
comprehensive literature has been searched and reported with an appropriate
methodology, the final analysis and interpretation are now poorly presented. This can
only be likened to roofing a new building with perforated roofing sheets. To write a
report of one’s research that shows fidelity to the data and the ability to organise them
logically and present them clearly is to bring one’s research effort to its destined
conclusion: to share with others your findings and whatever nuggets of meaning they
may contain. A research report is a straightforward document that sets forth clearly
and precisely what a researcher has done to resolve a research problem. In structure, it
is factual and logical.
Learning Outcomes for Study Session 12
When you have studied this session, you should be able to:
12.1 Enumerate three objectives of research report (SAQs 12.1)
12.2 List the format for research report. (SAQ 12.2)
12.3 Differentiate between literature review and methodology (SAQ 12.3)
12.1 Planning a Research Report
A research report has a relatively simple format. In general, it should achieve three
objectives thus:
i. It should acquaint readers with the problem that has been researched and
explain its implications adequately enough so that they have a clear orientation
to the problem;
134
ii. It should present the data fully and adequately. The data within the report
should substantiate all the interpretation and conclusion that the report
contains; and
iii. It should interpret the data for the reader and demonstrate exactly how the data
resolved the problem that has been researched. A report that merely represents
raw data and uninterpreted fact is of little help to the readers in deriving
meaning from these data. For a thesis or dissertation report, it is advisable to
visit the department to find out whether the department has a prescribed set of
guidelines for writing theses. Such issues as paper size, width of margins, size
and style of typeface should be checked from the department. Often, what is
acceptable in one department may not be accepted in another.
12.2 Format for the Research Report
12.2.1 Title Page
This includes the title of your study, your name and the date, the title should
accurately reflect the nature of your study and should be brief and to the point.
12.2.2 Acknowledgement
You may wish to acknowledge the help given to you in the preparation of your report.
If so, acknowledge generally.
12.2.3 Abstract
In most cases, institutions demand an abstract. It is quite difficult to produce in few
words what your research set out to do, the methods employed and the conclusions
reached. Also, most institutions specify the length; an abstract should be at most two
A4 pages. Generally, most scientific journals, theses and dissertations contain
abstracts. An abstract is a summary of the study that allows the reader to learn enough
to decide whether to read the whole thing or not. In an abstract, the first sentence
should be clear statement of the problem that was investigated in the study. The
research methodology and sampling techniques are then indicated. A brief summary
of findings and conclusion completes the abstract.
135
Figure 12.1: Writing Research Report (Photo from www.ayuriinfo.wordpress.com)
12.2.4 Introduction and Problem Statement
The first part of the body of the report states the research problem and its importance.
This should include a brief literature review or the history of the problem in previous
research and theory. The objectives of the study, research questions and hypotheses all
constitute part of this aspect of the study. A guide should explain the research
problem in few sentences, state the aims/objectives/hypotheses, provide any
background to the study which is necessary to place the study in its context. Draw
attention to any limitation the study might have at this stage. This chapter also
accommodates the significance of the study and definition of terms.
12.2.5 Review of Literature
A thesis or dissertation should have a chapter which deals exclusively with the review
of literature. This should include the history of the problem (if any). It also indicates
how the current study flows from previous studies. Simply put, the review of literature
provides the context and background of the study. Selection has to be made and only
books and journals articles which relate directly to the topic should be used. You
should be able to select material which is related to your study, group the information
into categories, comment on features which are of particular importance and also
compare the results of different investigations. A good review of literature sets the
scene, places the work in context and prepares the reader on what is to follow. The
literature review consists of theoretical and empirical sections
136
Activity 12.1
Before you will continue reading the formats of a research, it is important to reflect on
what you have read up till this point. Briefly describe the three general objectives a
research report should achieve.
Activity 12.1 Feedback:
A research report has a relatively simple format. In general, it should achieve three
objectives thus:
i. It should acquaint readers with the problem that has been researched and
explain its implications adequately enough so that they have a clear orientation
to the problem;
ii. It should present the data fully and adequately. The data within the report
should substantiate all the interpretation and conclusion that the report
contains; and
iii. It should interpret the data for the reader and demonstrate exactly how the data
resolved the problem that has been researched.
12.2.6 Methodology
The methodology section describes the sample that was studied and the research
techniques employed. It also shows how concepts are operationalised and what
measurement devices such as scales were used, choice of variables and controls. This
section is very important because it provides the basis for any future replication
efforts, so such section must be written with sufficient detail so that it can perform
these two functions. Readers of this section of your thesis or dissertation must be able
to tell precisely what was done in the study and who participated.
12.2.7 The Statement of Results
This is the heart of the report and will consist of tables, figures and text, depending on
the nature of the thesis and dissertation. The way results are presented is important.
Tables, charts, graphs and other figures should illustrate and illuminate the text. The
text should be written after the results are prepared and should not duplicate
information in the tables, but should highlight significant aspects of the findings. You
should ensure that all relevant facts are presented in a way that draws the readers’
attention to what is most important. This requires some practice and in most cases,
you need to repeat the draft severally before you produce something acceptable to
137
your supervisor. You should recognise that other students have produced research
reports before you, so you will do well to look through their work.
12.2.8 Analysis and Discussion
It is often advisable to start this section with a restatement of the problem before
discussing how the result affects existing knowledge of the subject. If your research
aimed at certain hypotheses then, this section demonstrate whether they were or were
not supported by the evidence. Any deficiencies in the research design should be
mentioned, with suggestion about different approaches which might have been more
appropriate. Implications for improvement of practice, if any, should be stated under
this section. Often, research results raise new questions as they answer others. Any
suggestions for future research should be stated in this section.
12.2.9 Specification of Statistical Analysis
In research it must be done using computer but we must know two types of statistic.
Descriptive: contains a number of categories research include measures of central
tendency (mean, medium and model) measures of dispersion (range, semi inter-quater
SD, variance). All these are suppose to be used in describing the characteristics of
variables.
Inferential statistic: deal with whether the result taken is due to chance. It tests
whether the result is genuine
(i) For comparing two groups, we used the student’s t-test, for three groups we
need analysis of variance, for comparing two or more groups when there is a pre-test
and post-test we need analysis of covariance e. g. JSS II and III anxiety scale we
want to compare we can use student’s t-test for related or interdependent.
(ii) Measures of relationship (a) Rank difference correlation-coefficient is used for
finding relationship. When the data available in form of ranks not in form of scores
A B
1st 31st
3rd 32nd
2nd 33rd
At times, the scores are available we use the product moment correlation
(Coefficient called Pearson “r” e. g. 40%, 50% etc)
There is a non-parametric statistic used when we have frequency instead of scores e.
g. number of male pass and number of female pass it is called chi square Χ2 it is used
138
for finding relationship. When the numbers are in form of frequencies and not scores.
Scores can be reduced to percentages and we have to use chi square
12.2.10 Statement of How to use Results
Relate the statistic to hypothesis e. g. there will be no difference in student who has
been counselled and those not counselled in academic performance. Take two groups
randomly selected e. g. experimental and control. Take academic performance with
the counselling scores. If one is higher than the other we reject the null hypothesis that
is, hypothesis we propounded at the beginning. We compare the scores using t-test
and come out with conclusion.
Activity 12.2
Before you will continue reading the formats of a research, it is important to reflect on
what you have read up till this point. Briefly describe the statement of results in a
research.
Activity 12.2 Feedback:
This is the heart of the report and will consist of tables, figures and text, depending on
the nature of the thesis and dissertation. The way results are presented is important.
Tables, charts, graphs and other figures should illustrate and illuminate the text. The
text should be written after the results are prepared and should not duplicate
information in the tables, but should highlight significant aspects of the findings.
12.2.11 Results
Always in chapter four research preamble. Take the hypothesis one by one; one must
be very careful because there is a well known pattern for describing hypothesis. Tell
us what you did e. g. used of t-test.
i. Re-state the hypothesis
ii. State how you test it
iii. Give us the result or provide a table.
Put the table. There are well known form of presenting table research must have a
title.
139
Title
Comparison of JS 1and3 – on anxiety
Variable Group 1 Group II t P
Anxiety N × N × 4.68 <.01
30 20.6 2.4 30 19.4 2.1
The t- is always small t-
iv. Describe the table that is briefly mention what is there. As could be seen from
this table group 1 performed higher than group 2. Quote the t (t=4.68, P < .01
in the bracket)
v. Conclude as to whether the hypothesis is rejected or accepted since group 1
perform higher than group 2 we reject the null hypothesis. These five will be
done for every hypothesis.
There are two more things in the result section.
a. The result section must be ended with summary of findings and conclusions.
Findings are difference from conclusion e. g. two groups experimental and
control group – Group A perform significantly better than B that is “finding”.
The conclusion is that the method is a superior one or that counselling is
effective in fostering academic performance.
b. Diagrammatic and tabular forms of presenting results. Tables and figures can
be used to present result photography, bar graph, diagram, pie graphs, line
graphs are figures. These are not necessarily numerical.
Tables are actual numerical figures e. g. t-test table, factor analysis table. They must
all have titles and keys.
Some tables have (one*) show level of significant P=0.005 if these is (2**) it means it
is significant at 0.01 level which should be done for all the hypothesis.
At the end of chapter three, there is summary of result and conclusion.
A result is exactly what we find e. g. group I perform better than group II conclusion
is the meaning of that result e. g. if we say that one method is effective and the other is
not.
At the end of hypothesis we need to summarize all these.
140
12.2.12 Discussion
This is usually chapter 5. There should be introduction, one tries to explain the
findings and compare findings from previous years.
Discussion is the real area that shows the researchers ingenuity. In explaining
findings.
1. We explain why we obtain such results e. g. if counselling improves the
motivation of students to read, our task is to explain what it has done to make it
effective. The researcher use his/her own entire knowledge of the area (topic)
to answer why the counselling is effective.
Discussion shows total expertise in the area (topic) in addition, the researcher
compares his/her finding research existing one in other studies. If is contradict
the existing one, the researcher has to explain why if it has time, technique,
method etc.
2. Describe the implication of the finding especially for practice e. g. Counselling
improves performance therefore student failing should be expose to counselling.
Discussion should also follow the hypothesis. First hypothesis should come first under
discussion.
There should be a section that the discussion should give room for short coming of
researcher’s studies (limitation). The research is expertise in research design would be
shown here e. g. small number of subject time, certain occurrences like riot in the
school money affect study.
The final section of discussion is suggestion for further research. The researcher
should bring out suggestion which will further extend the frontier of the knowledge.
Firstly, suggest what will correct the short comings noticed in the study e. g. more
subjects, time expansion.
Abstract
Although it is written last, it should go to the front of the project. An abstract is the
summary of the entire dissertation of not more than 2 ½ pages.
It should contain the obligation of the study. The method use in collecting data, the
findings and conclusion-the number of hypothesis, statistic used all must be stated.
141
12.2.13 Summary and Conclusion
The main issues arising from the analysis and discussion should be summarised
clearly in this section. Only conclusion that can be justifiably drawn from findings
should be made. It is advisable to read the entire report before writing this section of
the research report.
Activity 12.3
It is important at this point to reflect on what you have read so far. Briefly describe the
abstract of a research.
Activity 12.3 Feedback:
Although it is written last, it should go to the front of the project. An abstract is the
summary of the entire dissertation of not more than 2 ½ pages.
It should contain the obligation of the study. The method use in collecting data, the
findings and conclusion-the number of hypothesis, statistic used all must be stated.
12.2.14 References
All the materials referred to in the text and only those listed alphabetically in the
reference section of the manuscript should be written. Some examples are shown
below.
Books:
Busari A O . (2011). Essentials of Guidance and Counselling Practices. Ibadan: Arab
Ventures PP73 – 111.
Oloyede, O.I. (2007). Measurement and Evaluation for Tertiary Institutions, Kaduna:
Euneeks and Associates, PP37 – 53.
Associations:
Science Teachers Association of Nigeria, STAN (1998). The History of STAN.
Ibadan: University Press Plc. PP90 – 110.
Journals:
Otuka, J.O.E. (2000). Think and Do Activity-Based Science for Nigerian Primary
Schools. Journal of the Science Teachers Association of Nigeria, 80, PP1 – 10.
142
Thesis:
Ezenwa, V. (1998). Children’s Ideas about Chemical Bonds. Unpublished doctoral
thesis, Ahmadu Bello University, Zaria.
Chapters in Edited Books:
Otuka, J.O.E. and Atadoga, M. (2000). How Children Learn Scientific Concepts, in
M.J. Shuaibu and M.F. Bandele (eds.) History of Science in Nigeria, Kaduna:
Bode Press, PP211 – 219.
Appendices:
Should include copies of questionnaire, interview schedules, raw data etc.
Principles of Writing
Remember that your report is you; your report is a powerful psychological weapon.
Without your meaning to do so, a report can say more about you to your supervisor
than you possibly can imagine.
Below are some tips to note:
Neatness and Clarity: Your report should be clean and clear. Each sub-problem
should be a clear stepping stone in the resolution of the main problem. Little things
leave lasting impressions. Your page should be easy to read, double-spaced lines and
clear crisp letters, placed attractively on the page.
Precision: In addition to neatness and clarity, a research report should be crisp with
precision. There should be no room for guessing or for ambiguous or foggy terms.
Your document should be clear-cut and present its information plainly. Writing
research reports effectively is not an easy task. Good reports are not written in a hurry.
Even experienced and skillful writers revise their report severally before they submit
their final report.
Step 1:
The Problem
- Is the problem clearly and concisely stated?
- Is the problem adequately narrowed down into a research effort?
- Is the problem significant enough to warrant a formal research effort?
- Is the relationship of the identified problem to previous research clear?
143
Step 2:
Literature Review
- Is the literature review logically organised?
- Does the review provide a critique of the relevant studies?
- Are gaps in knowledge about the research problem identified?
- Are important relevant references omitted?
Step 3:
Theoretical or Conceptual Framework
- Is the theoretical framework easily linked with the problem, or does it seem
forced?
- If a conceptual framework is used, are the concepts adequately defined, are the
relationships among these concepts clearly identified?
- Are the independent and dependent variables identified?
- Is a predicted relationship between two or more variables included in each
hypothesis?
- Are the hypotheses clear, testable and specific?
- Do the hypotheses logically flow from the theoretical or conceptual framework?
Activity 12.4
Before you will continue, do this activity. Write three examples of an acceptable
reference.
Activity 12.4 Feedback:
Busari A O . (2011). Essentials of Guidance and Counselling Practices. Ibadan: Arab
Ventures PP73 – 111.
Oloyede, O.I. (2007). Measurement and Evaluation for Tertiary Institutions, Kaduna:
Euneeks and Associates, PP37 – 53.
Otuka, J.O.E. (2000). Think and Do Activity-Based Science for Nigerian Primary
Schools. Journal of the Science Teachers Association of Nigeria, 80, PP1 – 10.
Step 4:
Research Variables
- Are the dependent and independent variables operationally defined?
- Are any extraneous or intervening variables identified?
144
Step 5:
Hypotheses
- Is a predicted relationship between two or more variables included in each
hypothesis?
- Are the hypotheses clear, testable and specific?
- Do the hypotheses logically flow from the theoretical or conceptual
framework?
Step 6:
Sampling
- Is the sample size adequate?
- Is the sample representative of the defined population?
- Is the method of selection of the sample appropriate?
- Are the sample criteria for inclusion into the study identified?
- Is there any sampling bias in the chosen method?
Step 7:
Research Design
- Is the research design adequately described?
- Does the research design control for threats to internal and external validity of
the study?
- Are the data collection instruments described adequately?
- Are the reliability and validity of the measurement tools adequate?
Step 8:
Data Collection Methods
- Are the data collection methods appropriate for the study?
- Are the data collection instruments described adequately?
- Are the reliability and validity of the measurement tools adequate?
Step 9:
Data Analysis
- Is the result section clearly and logically organised?
- Is the type of analysis appropriate for the level of measurement for each variable?
- Are the tables and figures clear and understandable?
- Is the statistical test the correct one for answering the research question?
145
Step 10: Interpretation and Discussion of the Findings
- Are the interpretations based on the data obtained?
- Does the investigator clearly distinguish between actual findings and
interpretations?
- Are the findings discussed in relation to previous research and to the
conceptual / theoretical framework?
- Are unwarranted generalisation made beyond the study sample?
- Are the limitations of the results identified?
- Are recommendations for future research identified?
- Are the conclusions justified?
Summary of Study Session 12
In Study Session 12, you have learned that:
how to plan a research report;
- the format for a standard research report but institutions have variations of this
standard format;
- how to write a comprehensive research report and takes care of all aspects of
such a report, of the need to revise your research
- report draft severally before you arrive at the final report; that your research
report provides a mirror of your person to the society, you must therefore do it
very meticulously.
Box 12.1: Checking Hypotheses before Final Report
• Is a predicted relationship between two or more variables included in
each hypothesis?
• -Are the hypotheses clear, testable and specific?
• -Do the hypotheses logically flow from the theoretical or conceptual
framework?
146
Self-Assessment Questions (SAQs) for Study Session 12
Now that you have completed this study session, you can assess how well you have
achieved its Learning Outcomes by answering these questions. You can check your
answers with the Notes on the Self-Assessment Questions at the end of this Module.
SAQ 12.1 (tests learning outcome 12.1)
Can you correctly enumerate three objectives of research report?
SAQ 12.2 (tests learning outcome 12.2)
List the format for research report
SAQ 12.3 (tests learning outcome 12.3)
How would you differentiate between literature review and methodology?
Notes on the Self-Assessment Questions (SAQs) for Study Session 12
SAQ 12.1:
i. It should acquaint readers with the problem that has been researched and
explain its implications adequately enough so that they have a clear
orientation to the problem;
ii. It should present the data fully and adequately. The data within the report
should substantiate all the interpretation and conclusion that the report
contains; and
iii. It should interpret the data for the reader and demonstrate exactly how the
data resolved the problem that has been researched.
SAQ 12.2: 12.2.1 Title Page
Acknowledgement
Abstract
Introduction and Problem Statement
Review of Literature
Methodology
The Statement of Results
Analysis and Discussion
147
SAQ 12.3: Simply put, the review of literature provides the context and background
of the study. Selection has to be made and only books and journals articles which
relate directly to the topic should be used. The methodology section describes the
sample that was studied and the research techniques employed. It also shows how
concepts are operationalised and what measurement devices such as scales were used,
choice of variables and controls.
References
Best, J.W. & Khan, J.V. (1995). Research in Education (7th Edition). New Delhi:
Prentice Hall. PP 20 – 23.
Cresswell, J.W. (1994). Research Design: Quantitative and Qualitative Approaches
Thousand Oaks C.A.: Sagem in P.D. Leedy (1997)
Practical Planning and Design (6th Edition) New Jersey: Merril, PP105 – 107.
Leedy, P.D. (1997). Practical Research: Planning and Design (6th Edition). New
Jersey: Morill, PP103 – 110.
Monette, D.R.; Sullivan, T.S. & Dejong, R.D. (1994). Applied Social Research: Tool
for the Human Services (3rd Edition) Chicago: Harcourt Brace College
Publishers, PP3 – 8. 201