Research Methodology - dlc.ui.edu.ngdlc.ui.edu.ng/oer.dlc.ui.edu.ng/app/upload/GCE 209_1507036947.pdf · In conclusion, it is envisaged that the course materials would also be useful

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


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




Study Session 3: The Research Problems and Literature Review 27

Introduction 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




Study Session 4: Concepts, Constructs and Variables 43

Introduction 43

viii


4.1 Defining Concepts 43

4.2 Constructs 44




References 53

Study Session 5: Types of Research Approaches –Qualitative and Quantitative

Approaches 54


5.1 Qualitative Approaches 54

5.3 Longitudinal Research 57



Study Session 6: Survey Research and Research Tool 62

Introduction 62


6.1 Meaning of Survey Research 62

6.2 Questionnaire 65

6.3 Interview and Observations 69

6.9 Case Study 77




Study Session 7: Experimental and Quasi Experimental Research 80

Introduction 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





ix

Study Session 8: Research Design 91

Introduction 91


8.1 Definition 91

8.2 Experimental Vs Ex-post facto research 91

8.3 Method of controlling independent variables 93




Study Session 9: Population and Sample 102


9.1 Population and Samples 102

9.2 The Purpose of Sampling 103

9.3 Types, Methods and Procedures for Sampling 104




Study Session 10: Level of Measurement Validity and Reliability 117


10.1 Levels of Measurement 117

10.2 Validity and Reliability 120

10.2.1 Validity 120

10.2.2 Reliability 122





Study Session 11: Test Administration and Interpretation 127


11.1 Process of Test Administration 127

11.2 Test venue 128

11.3 The actual testing session 130




x

Study Session 12: Writing Research Report 133

Introduction 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





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:







analysis;


- 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.


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’?


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


Can you list at least four goals of a research?


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:







analysis;


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


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.



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.


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.


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.



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.






How will you define theory?


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


A good hypothesis should possess some characteristics, can you describe at least four

of these features?


Briefly enumerate the functions of hypothesis in research.


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


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.



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


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.


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.


Basically, there are two sources of information that students could use in reviewing

relevant literature. These are:

The Library


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.



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.






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


Assumption and significance are importance aspects of a research, how will you

distinguish between the two?


Can you enumerate the factors related to time constraint in a study?


Discuss three sources in review of the literature.


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


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


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.



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.


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.


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



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.






If you were told to define concept, how would you define it?


Can you differentiate between concepts and constructs?


How will you describe the term ‘variable’?


Discuss the two major types of variables.


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


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


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.



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.


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


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.



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.





Box 5.1: Cross-Sectional Research

The defining characteristics of cross-sectional research are:





60


How will you explain the meaning of qualitative approach?


Differentiate between qualitative and quantitative approaches


How will you describe longitudinal research?


State the general characteristics of qualitative studies.


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





61

References


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


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.



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.


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


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.


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.



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





79


How will you define survey research?


Survey research has many characteristics, describe at least three of these

characteristics.


Can you explain the term ‘questionnaire’?


Discuss at least five steps in constructing questionnaire


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


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.



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


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.


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



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.






How will you define experimental research?


Describe at least five characteristics of experimental research


Discuss the classifications of experimental research.


How will you explain the concept of quasi experimental designs


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


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



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


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


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



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;






How can we correctly explain correlation design.?


Define ex-post facto research..


Describe method of controlling independent variables.


Ex-post facto research has some limitations, discuss at least three of them


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


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


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.



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


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.


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.


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.



- 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.






Define population.


What do you understand by sample?


Describe at least three different types of sampling techniques.


State the various stages in sampling designs


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


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



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.


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


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



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.






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


Describe the different levels of measurement


List the appropriate reliability methods to be used on a set of data.


Validity is an important aspect in constructing instruments, explain three different

types of validity.


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




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


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.



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.


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



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





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


How will you explain the process of test administration?


Describe the two aspects of test venue that are particularly important


List the two activities that are expected of a tester during testing session.


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



Publishers, ::3 – 8.

133

Study Session 12: Writing Research Report


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.



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.


A research report has a relatively simple format. In general, it should achieve three

objectives thus:


explain its implications adequately enough so that they have a clear orientation

to the problem;



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.


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.


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.


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?



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






Can you correctly enumerate three objectives of research report?


List the format for research report


How would you differentiate between literature review and methodology?


SAQ 12.1:


explain its implications adequately enough so that they have a clear

orientation to the problem;



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




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.



Publishers, PP3 – 8. 201

Documents

Research Methodology - dlc.ui.edu.ngdlc.ui.edu.ng/oer.dlc.ui.edu.ng/app/upload/GCE 209_1507036947.pdf · In conclusion, it is envisaged that the course materials would also be useful