Click Here & Upgrade

Expanded Features
PDF Unlimited Pages
Documents
Complete

1.0 INTRODUCTION
This paper aims at presenting the specifications required in the presentation of the research report in order to achieve
uniformity and consistency.

2.0 THE MEANING OF RESEARCH

Research refers to “the search for knowledge, or a scientific and systematic search for pertinent information on a
specific topic, an art of scientific investigation”. (Kothari: 1985). As such, it is considered a movement from the
known to the unknown. This means that research should add value or make contributions to the existing stock of
knowledge, and so it needs to be original

Basic Research is carried out for the sake of increasing knowledge. Applied research is carried out to solve a
problem. The scientific method of acquiring knowledge, also called scientific research, is a systematic investigation
of a question, a phenomenon/event, or a problem using certain principles. All different sciences are united not by
their different matter but by their common method, by the way knowledge is acquired – research methodology.

Properties of Scientific Research

• Research is empirical since the aim is to know reality.
• Research is systematic and logical.
• Research is replicable and transmittable. As the observation is objective and the explanation is logical,
anyone placed in exactly the same circumstances can observe the same event and make the same reasoning,
leading to the same explanation and prediction.
• Research is reductive. To grasp the main relationships of laws, the complexity of reality is reduced.
Reductionism is the method of considering only the essential and necessary properties, variables or aspects
of a problem.

Characteristics of a Good Research

• Purpose clearly defined.
• Research process detailed
• Research design thoroughly planned
• Limitations frankly revealed
• High ethical standards applied
• Adequate analysis for decision maker’s needs
• Findings presented unambiguously.
• Conclusions justified
• Researcher’s experience reflected.

3.0 ADMINISTRATIVE PROCEDURES FOR RESEARCH PROJECTS

3.1 Attendance of Research Methodology Lectures
The research project is an integral portion of every undergraduate programme at the National University of Science
and Technology. In order to maintain standards students are encouraged to attend the research methodology course
as a prerequisite in order to carry out the project smoothly.

3.2 Supervision
Students are allocated supervisors, who are normally members of staff within the National University of Science and
Technology. The onus is on the student to make the necessary arrangements with the supervisor for the schedule of
consultations. Regular consultations are critical and the student can only proceed to the subsequent stage only after
the preceding stage has been approved.

The student must take full ownership of the project and should not expect anything more than just guidance from the
supervisor. Submission of the project is subject to certification and approval of the supervisor. Once the project has
been submitted, it may not be retrieved for further editing.

Late submissions will not be accepted. Failure to submit a project on time is similar to missing an examination

3.4 Getting Started

1
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Decide on a topic that covers a field that you are passionate about from the courses that you have studied
during the programme. The topic should be researchable within the time-frame given for the study. It
should add value to the field of study and should have great scope for follow-up research. In short, the
important requirements of a research topic are: originality, topicality, replicability, and practical feasibility.
• Decide on the research design for the study.
• Consult with your supervisor for approval of the topic.
• Have access to library sources. It is necessary to consult a range of primary, secondary and statistical
sources at a fairly early stage of the research. Important questions are:
- Where are the sources located?
- Is access to the sources and/or archives available to bona fide researchers?
- What costs are likely to be incumbent upon the research?
- What photocopying facilities are available?
Ensure that you do don contravene any copyright law which applies to most of the materials that you would
wish to consult. It is also important to comply with all library and/or archive regulations.
• Read widely on the area taking the full details of the sources of information.
• Develop initial data analysis plans. While this aspect is covered in more detail in chapter 3, some
preliminary discussion need to be made at the beginning of the research process. If the intended data is
likely to be quantitative, it is very important to ensure that (1) there are statistical methods available for
dealing with the data, and (2) these methods are available on a standard computer package that is easily
accessible to the person undertaking the data analysis. If the intended data to be collected is qualitative,
some initial decision about data format (how it is to be presented) is necessary in order to decide how to
start the analysis.
• Remember that your study is being conducted on a small scale. Therefore it is important not to be too
ambitious in scope, but to make sure that the research makes an “original” contribution to knowledge.
• Refine your topic on the basis of the preliminary consultations plus the preliminary literature review.
• Write and submit your proposal.
• As you proceed with the draft of the main research, always make backup copies of your work on C-drive
and at least two floppy/memory sticks.

3.5 Ethical Issues

Any research involving human (or animal) subject is bound to have ethical implications. Ethics are norms or
standards of behavior that guide moral choices about our behavior and our relationships with others. The goal of
ethics in research is to ensure that no one is harmed or suffers adverse consequences from research activities. Most
professional organizations in the social sciences (e.g. AERA, BERA, American Psychological Association, British
Psychological Society, British Sociological Association, American Sociological Association) have drawn up ethical
guidelines which should be taken into account when carrying out research.

Broadly, research ethics cover the following issues:

Respect for Persons Involved. There are guidelines relating to the physical and psychological well-being of the
subjects. In particular, a person’s confidentiality must be respected. It should not be possible to identify a person
(or institution) from the completed report. Respondents who have provided information should have the opportunity
of checking that it is an accurate statement of what they have said or written.
Respect for Truth. This is particularly important with respect to data collection and analyses. In both processes
authenticity is required – bias should not be introduced.
Openness. Readers of the final report should be given as much information as possible about how the data was
collected (sample size, sample source, data collecting and analysis methods). However, it is still important that
confidentiality is maintained and if the two are in conflict, the confidentiality criterion should have a higher priority.
Record Keeping. It is essential that records be kept of data collected (when, where, how and from whom) to aid
validity and reliability checks. For funded projects records of expenditure (including receipts received) should be
maintained for auditing purposes.

In addition the following specific ethical considerations must be balanced:

• Protect the rights of the respondent or subject. Ensure the respondent does not suffer physical harm,
discomfort, pain, embarrassment, or loss of privacy.
• Ensure the sponsor receives ethically conducted and reported research.
2
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Follow ethical standards when designing research.

• Protect the safety of the researcher and team.
• Ensure the research team follows the design.

3.6 Originality in Research

One of the more complex issues with which a researcher has to grapple is that of originality. A researcher can
demonstrate originality in a number of ways:
1 The research might make a contribution to theory in some way. For example, by
(a) confirming or refuting the validity of the theory
(b) modifying existing theory by demonstrating that some variables were more important than others in determining
causes, or that there were additional variables from those originally formulated which the theory failed to take into
account.
(c) seeking to establish theory anew from the data that has been collected in the research (referred to as “grounded
theory”) (Strauss and Corbin, 1989).
(2) The research might demonstrate methodological advance either in terms of a new methodology per se an
innovative use of an existing methodology or in the development of a new or novel data-collecting instrument
(3) It might be an existing methodology applied to a group of respondents who had not previously been part of the
sample.
(4) It could be original in terms of being a re-evaluation of existing research, possibly using different methods of
data analysis.
(5) It could be updating in some way an earlier piece of work.

.THE RESEARCH PROPOSAL

The research proposal is an activity that incorporates decisions made during early project planning phases of the
study. Writing a proposal is a necessary initial stage of all research. The proposal clarifies precisely what your
research is investigating and how you intend carrying out the research. This does not mean that you will necessarily
stick exactly to the proposal. There may be need to make adjustments as new ideas are generated in the course of
the research.

This paper assumes students have already submitted their proposals, having followed earlier guidelines on “Writing
a Research Proposal”.

4 RESEARCH REPORT PRESENTATION

4.1 THE PRELIMINARY PAGES

The front pages of the research report are numbered using small Roman numerals and are arranged as follows:

4.1.1 The Title Page

This is considered to be page (i), but is left unnumbered. It shows:
• The Institution granting the degree
• Title of project (single spaced)
• Name of writer
• Purpose of research project
• Name of department
• City
• Year of award

4.1.2 The Release Form

This form grants the University permission to produce copies of the project and also reserves the author’s
publication rights.

4.1.3 The Approved Form

This serves as official acknowledgement and acceptance of the project as satisfactory. It is signed by the
supervisor(s).

3
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

4.1.4 Dedication (Optional)

Serves as a tribute to a specific individual or individuals.

4.1.5 The Abstract (Executive Summary)

This is a very brief summary of the report, used by potential readers to determine at a glance, the contents of the
project. It is a concise summary of
• The statement of the problem(s)
• The purpose of the study
• The methodology used
• Summary of major findings conclusions, recommendations and suggestions for further research.

This is usually written after the rest of the project is finished. It should be at most a page, but not more than two
pages.

4.1.6 Acknowledgements
This section enables the researcher to thank persons who contributed to the successful making of the project.

4.1.7 Table of Contents

The table of contents outlines the components of the research report. After the preliminaries, you then list chapters
and chapter headings and sub-headings showing the page(s) where these are located in the research report. Finally,
the back pages materials (i.e. references and appendices), are presented.

4.1.8 List of Tables

Using the lower case, show the table number, its title and reference page.

4.1.9 List of Figures

In the lower case, show the figure number, title and reference page.

4.1.10 List of Acronyms

4.1.11 List of Abbreviations

4.2 TEXT FORMATTING

In presenting the main text of the report attention should be paid on the following specifications:

4.2.1 Margins
The top, bottom, left and right hand margins should have the following specifications:
• 3.8 cm on the left margin to allow for binding
• 2.5 cm at the top and bottom
• 2.5 cm on the right hand.

4.2.2 Typing Rules

• Use A4 page and type on one side of the page only using font size 12 points and double line spacing
• All chapters should be numbered in Roman numerals, centered at the top of the page and typed in the upper
case.
• Chapter titles should be in capitals, centered and 2 spaces below the chapter number heading.
• Sub-headings should start from the left margin and can be underlined, highlighted or done in bold, using
capitals and lower case combined. When the bold font is used, do not underline.
• Sub-headings that take more than one line should be single-spaced
• Each chapter begins on a separate new page.

4.2.3 Text Spacing

Begin 3 spaces below the last line of the chapter title. Use 2 lines to divide paragraphs or when beginning a new
sub-division. Be consistent in the mode of typing.

4
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

4.2.4 Quotations
• Long quotations should be indented (4 spaces from left and right margins), be single-spaced, in italics and be
enclosed in quotation marks.
• Quotations less than 3 lines are enclosed in quotation marks and typed within the general text.

4.2.5 Pagination
• Preliminary pages are numbered in the lower case of the Roman numerals.
• Arabic numbering is used for the rest of the project, beginning on chapter one up to the end of appendices.
• Numbering should be done at the bottom center of each page.

4.2.6 Back Page Materials

References should come immediately after the last chapter. The references should be listed alphabetically and not
numbered. They should be single spaced, leaving a line in between each reference. Use italics for titles of books and
Journals. Titles of articles should be in inverted commas.

References should reflect works consulted and appearing in the text. All sources of information referenced in the
report must be acknowledged and traceable. The reader must be able to verify these sources without having to
contact you first. All articles cited must appear in your reference list.

The references sections should start on a separate page. If the same person is author of more than one reference in
your list, then these are ordered chronologically for that person. If more than one references are supplied for the
same author for the same year, then these are labeled “a”, “b”, “c”, etc.

For further details on referencing styles, download the APA Crib Sheet PDF.1

Appendices should be paged normally as an extension of the research document. Each appendix should have a title
and numbered appropriately.

4.2.7 Dissertation Length

The length of the final report/project/dissertation should be between 10 000 and 12 000 words for undergraduate
level and between 12 000 and 15 000 words for MSc level. The word count tool on your computer should assist you
in providing an update of the number of words and pages as you word-process your text.

5 THE MAIN RESEARCH REPORT

This refers to chapters one to five and constitutes the main research report. The main of the project is written in the
past tense, using reported speech.

Each chapter should begin with an introduction and end with a summary. The introduction gives the researcher the
opportunity to spell out in brief the main concerns and focus of the chapter. The introduction must be clear, concise
and to the point. Avoid generalizations. The chapter summary provides a brief summary of the chapter at the same
time providing a statement to prepare the reader for the next chapter.

5.1 CHAPTER I - INTRODUCTION

This sets the foundation of your study by providing background information, which enables others to understand
your research. After introducing the chapter, the main sub-headings are arranged as follows:

5.1.1 Background to the Study

This should situate the research into relevant context, from general to specific. In the final analysis, you should
identify the gap that must be filled by the present study.

5.1.2 Statement of the Problem

1
This style is documented in the America Psychological Association Publication Manual (5th ed., 2001).
5
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Provide a brief and clear statement of the problem to be solved. The problem statement should be clear, without the
use of idioms. It is important to separate symptoms from problems and focus on the problem. The problem should be
researchable, carefully fitting into the broader context of current theory and relevant research and clearly and
logically related to its sub-problems/research questions or hypotheses.

5.1.3 Purpose of the Study

Clarifies the aims or objectives of the study, what the study sought to accomplish. The objectives flow naturally
from the problem statement, providing concrete, and achievable goals. The research objectives are crucial as they
provide the basis for judging the remainder of the final report. (for more details refer to the section on purpose and
objectives of the study in the “Research Proposal” module)

5.1.4 Research Questions/Sub-problems

Research questions must be developed from the research problem and study objectives and should yield responses,
which can be reconstituted, to make up a complete answer to the main research question or problem. They are
foundations for creating the research data collection instrument. The research questions must be precise and must
specify variables. [Variables in research may be classified as objects or properties. Objects include the things of
ordinary experience, such as tables, people, books, and cars. They also include things that are not as concrete, such
as genes, attitudes, neutrons, and peer-group pressures. Properties are the characteristics of the objects. A person’s
physical properties may be stated in terms of weight, height, and posture. Psychological properties include attitudes
and intelligence. Social properties include leadership ability, class affiliation, or status].

In general, research questions should be positive statements which are capable of being proven or not proven. This
is why research questions can be stated in the format of hypotheses. Each research question must be stated as
explicitly as possible. It should be sufficiently specific to enable both the methodology and data collecting
techniques relevant to the methodology to be formulated (Birly and Moreland, 1998).

As the research project proceeds, is always useful to return periodically to the research question(s), asking yourself
“why am I doing this and what am I trying to find out?”

5.1.5 Statement of Hypotheses

These are tentative answers to research questions (or sub-problems). A hypothesis is a proposition that is
formulated for empirical testing. (a proposition is a statement about concepts that may be judged as true or false).
The role of a hypothesis is to
• Guide the direction of the study
• Identify facts that are relevant and those that are not.
• Suggest which form of research design is likely to be most appropriate.
• Provide a framework for organizing the conclusions that result.
They should be stated in a testable format and can be written in null or alternate form. Hypotheses should be related
to the principal problem and sub-problem. Hypotheses can be relational, correlational or explanatory (causal).

A null hypothesis is an assumption of a neutral (non-causal) or interactive relationship between items – e.g. one
might with to examine if there is any difference between rural and urban households in their consumption patterns.
The null hypothesis might therefore be:
There is no difference between urban and rural households in terms of their
consumption patterns.

A hypothesis guides the direction of the study, identifies facts that are relevant and those that are not, suggests which
form of research design is likely to be appropriate, and provides a framework for organizing the conclusions that
results.

If the researcher is not quite confident of statistical testing of hypotheses, research questions can be used instead.

Important points regarding research questions and/or statements of hypotheses

• A statement which is value judgment will not do.
• Similarly, a statement for which empirical evidence could not be obtained would not be suitable.
6
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• A research question/hypothesis which presupposes the result would also not be appropriate.

5.1.6 Significance of the Study

This section allows the researcher to indicate how the solution to the problem or the answer to the research questions
can influence theory and practice. There is need to demonstrate why it was worth the time, effort, and expense
involved to carry out the research, who would benefit from the study and how?

5.1.7 Assumptions
These are statements of what the researcher believes to be fact but cannot be verified. For you to be able to carry
out the study you should hold certain facts about the study as given. The assumptions should be clear and
defensible.

5.1.8 Definition of Terms (if not part of preliminaries)

Define all important terms in order to avoid any misinterpretations. These terms should be employed consistently
throughout the report. Dictionary meanings do not serve adequately in defining terms of a research project.

5.1.9 Scope (Delimitation) of the Study

This section specifies conceptual and practical (physical) boundaries of the study. It gives the researcher the
opportunity to focus on the main concerns of the study at the same time clearly pointing out what it did not focus on.

5.1.10 Limitations
The researcher should state those conditions beyond his/her control that inhibited his/her ability to conduct the
research as planned, thus affecting his/her ability to interpret findings and make conclusions on the study. The
researcher should also suggest compensatory factors that assisted in making the research valid and reliable.

5.2 CHAPTER II – LITERATURE REVIEW (REVIEW OF RELATED LITERATURE)

5.2.1 Purpose
This chapter discusses theoretical and empirical literature within the area of study in a comparative and
argumentative form. The following thesaurus of the word review provide clues as to what a literature review means:
Analysis Perusal
Discussion Study
Judgement Writing
Round up Criticism
Theme Investigation
Breakdown Recount
Exploration Summary

In summary, the above words provide three aspects to what literature review means.
(1) To review means to peruse or provide a round up. This implies that a literature review should be:
• comprehensive
• relevant.
(2) A literature review should involve critical study and investigation. This implies that it should be
• an accurate account of each separate piece of literature, noting its good and bad features, and particularly its
relevance to the research concerns that have been identified
• Compare and contrast the different books and articles both with each other and with your own perspectives on
your topic of investigation
• Involve judgments of literature, mainly in terms of your own concerns and needs.
(3) A literature review should aim to be a comprehensive summary of the relevant literature. This implies that it
should :
• be focused
• be inclined to pithiness
• provide a round up of literature with some connection to your own research

7
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• clearly identify any literature that may have been relevant but could not be obtained during the completion of
the research process.

The purposes of literature review are:

• To sharpen and deepen the theoretical framework of the research.
• To familiarize the researcher with the lasted developments in the area of research. The researcher must be
acquainted with the problems, hypotheses and results obtained by previous research in order not to duplicate
efforts but to widen and deepen them. To use research as a starting point for new research.
• To identify gaps in knowledge, as well as weaknesses in previous studies. To determine what has already been
done and what is yet to be studied or improved.
• To discover connections, analogies (similarities) or other relations between different research results by
comparing various investigations. To highlight points of divergence or convergence among the different
authors.
• To identify variables that must be considered in the research, as well as those that prove irrelevant. This finding
is often a result of the comparison of different investigations.
• To study the definitions (conceptual [theoretical} and operational), as well as the characteristics of the
population/samples if applicable, used in previous work with the aim of adopting them for the new research.
• To study the advantages and disadvantages of the research methods used by other researchers. The researcher
can copy or improve the method.
• To find a link between your study and the accumulated knowledge in your field of interest.
• Helps you to limit your problem or research questions and to clarify and define concepts of the study.
• Provides insights into useful methodologies.
• Helps avoid unintentional replication of previous studies.
• Puts the researcher in a better position to interpret the significance of his/her results.

WARNING
• The researcher may be influenced by the results of previous research.
• The researcher may accept without criticism the characteristics and explanations and fail to remain objective
and to discover new possibilities.
• The researcher may develop the tendency to work within an already established framework, instead of exploring
new approaches.

5.2.2 Sources of Literature

• Summaries of theses and dissertations
• Indexes of research journals
• Bibliographies
• Dictionaries
• Encyclopedia
• Internet
• Primary documents, e.g. circulars, reports

5.2.3 General Hints

• Always begin with the most recent publication, working back to earlier publications
• The sub-topics should be well-structured and show logical sequence of ideas
• Always refer to the original source to avoid errors of interpretation or transcription
• The literature review should address relevant methodological issues. Highlight the weaknesses of the design and
show how your study would avoid similar problems.
• Pay attention to correct methods of referencing. Avoid academic dishonesty and cheating by continuously
acknowledging authors.
• Paraphrase information and be precise. Use your own words to explain what the author is saying.
• Use short directional quotations where necessary. Avoid long quotations unless otherwise
• To be relevant, literature review must focus on:
- the theory from which the research topic is derived

8
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

- stated problem, hypotheses and/or research questions

- identifying the gap in research to be filled by the study
• Avoid outdated literature
• At the end of the literature review you need to highlight in brief how your study will be the same or different
from what has been discussed by others in the material that you have consulted.

5.2.4 Making Notes

Start by working out a system for recording your notes. This can best be done using either a computer or a set of
index cards. Using a computer from the outset can save you a lot of time and problems of presentation. Putting
notes on to a to a computer also short-circuits your having to type them in from cards when you are constructing
your report or research article.

If what you present is a summary of the position, viewpoint or research findings of an author in general, then it is
often acceptable just to provide the author and date (the Harvard system), though it is strongly recommended to
include the page number as well. Page referencing can be very useful if someone contests your understanding of
what an author has written. Page numbering demonstrates your confidence in your understanding of the literature.
Page numbering also makes sure that yu are ot guilty of plagiarism. Plagiarism is the process of passing someone
else’s work off as your own.

5.2.5 Comparing and Contrasting Authors

There will usually be more than one perspective on the topic(s) you are researching. It thus becomes very important
to be able to compare and contrast authors. The following guidelines are useful:
• Which authors (more or less) say the same, and which say different things?
• What do they say differently? Do their assertions start from the same, similar or different assumptions?
• Do their arguments or propositions follow logically and rationally from each other?
• Are their arguments or logic rational and acceptable (an if so, why?0
• What does each author say about the other?
• What is the relevance or impact of their points of view on your research topics(s) and concern(s)?

5.2.6 Referring to others in the text

In the Havard system (which is recommended by the Department) at every point in the text at which reference is
made to other writers the name of the writer and the year of publication should be included. If the surname of the
author is part of the sentence, then the year of publication will appear in brackets, e.g.:
Sithole (2008) describes this….
If the name of the author is not part of the sentence then both the surname and the year of publication are in
brackets, e.g.
In a recent study (Sithole, 2008) it is described as……
If there are two authors then both family names should be given, e.g.
Sithole and Moyo (2008) suggest that….
If there are three of more authors the first author’s family name should be given, followed by et al., e.g.
Sithole et al (2008) suggest that…..
If the same author has published two or more works in the same year then each work should be referred to
individually by the year followed by lower case letters (a, b, c, etc). (These different references should be included
in the bibliography). For example:
Sithole (2008a) shows how….
When you need to refer to two or more different parts of a work or are quoting directly you should include the page
numbers after the year of publication.
Sithole (2008, p112) gives a model to explain…..

5.2.7 Presentation of the Bibliography/References

The bibliography comes at the end of your project report or dissertation. The citations are listed as either
bibliography or references, which are then arranged alphabetically by authors’ names and then, where necessary, by
date. Single-author books and articles appear before co-authored articles and books by he same author, with the
earliest dated first. Where there is more than one reference for one author in any one year, the order is alphabetical
by title, and lower case letters added to denote more than one publication in that year. The presentation should be:
9
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Author’s family name, comma, initials of first name(s), year of publication in brackets, title of the book in upper and
lower case in italics, comma, edition (if applicable followed by comma), publisher, comma, place of publication, full
stop. e.g.
Wegner, T (1999) Applied Business Statistics (1999), Juta & Co, Ltd,, Cape Town

Books and articles by more than two authors should be given in the order presented by the authors themselves,
linked by the work “and” for the last name. e.g.
Campbell, J Y, Lo, A W and MacKinlay, A C (1997) The Econometrics of Financial Markets, Princeton
University Press, Princeton, New Jersey.

Citing Articles in Journals

The author’s family name, comma, initials of the first name(s), comma, year of publication in brackets, title of
contribution in upper and lower case and in quotation marks, comma, title of the periodical or journal in italics,
comma, volume number in bold type, issue number, comma, the page numbers of the article prefaced by pp, full
stop.
Clare, A D and Thomas, S H (1995) “The Overreaction Hypothesis and the UK Stock Market”, Journal of
Business Finance and Accounting, 22 7, pp961-73

Citing Contributions in Books

Where books are made up of contributions from a variety of authors, cite the author’s surname, comma, initials of
first name(s), comma, year of publication in brackets, then the title of the contribution in upper and lower case and
in quotation marks, comma, followed by “in”, the editor’s surname followed by a comma, the initials then (ed), the
title of the book in upper and lower case underlined or in italics, comma, publisher, comma, place of publication,
comma, page numbers, full-stop.
Sithole, W (2005) “Short-Run forecasting of Stock Prices”, in Jones, P A (ed) Investment Management,
NUST Press, Bulawayo, pp47-63

Government Publications
Name of the responsible department, year of publication in brackets, title of the publication, in upper and lower case
underlined or in italics, comma, publisher, comma, place of publication. If there are official series codes these
should be cited after the title, and if there is a chairperson then this name should be included at the end. E.g.

Theses
Author’s surname, comma, initial(s), year of award in brackets, title in upper andlower case underlined or in italics,
comma, degree awarded followed by Thesis, comma and the name of the awarding institution, e.g.
James, P J (2004) Approaches to Active Portfolio Management, with Reference to Zimbambwe, PhD thesis,
National University of Science and Technology.

Mimeographed Materials
Mimeographs refer to duplicated materials, such as papers presented to conferences but not formally published. Cite
the name of the author(s) and initials, the date, the title of the paper, the circumstances under which the paper was
presented, and the fact that it was mimeoed (duplicated). The title of the paper is in quotation marks and has initial
capitals, e.g.
Johnson V G (2003), “The Effects of Relaxing Trade Barriers on the Zimbabwean Economy”, paper
presented to the 2003 Annual Conference of the World Customs Organisation, mimeo.

5.3 CHAPTER III – RESEARCH METHODOLOGY

Research methodology describes how the study was carried out. It focuses on methods/techniques used in
conducting the research, the logic behind the methods employed, and justifications for each of the
approaches/techniques. The review of related literature should have assisted in deciding on the suitable methodology
for the study. The chapter is concerned with research design, research instruments, data collection procedures, data
presentation and analyses plans.

10
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

5.3.1 Research Paradigm [SEE APPENDIX I]

A perspective based on a set of assumptions, concepts, and values that are held by a community or researchers. A
theoretical model within which the research is being conducted, and organizes the researcher’s view of reality
(though they may not be aware of it). There are 3 research paradigms:
Quantitative research – relies primarily of collection of quantitative data. Focuses on measuring. Can be positivism
or empiricism.
Qualitative research – relies on collection of qualitative data. Focuses on issues.
Mixed research – involves mixing quantitative and qualitative methods or paradigm characteristics.

5.3.1 Research Design (Plan) [APPENDIX II]

The design constitutes “the blueprint for the collection, measurement, and analyses of data” (Phillips B S, 1971:93).
It is the “plan and structure of investigation so conceived as to obtain answers to the research questions” (Kerlinger
F N, 1966:279).

The researcher should justify the chosen paradigm and method by providing the advantages and disadvantages.

The main concerns of this section are:

• To identify and justify the design. The design should be appropriate to the solution of the problem. It should be
adequately described. Appendix I gives a summary of the classifications of designs using eight different
descriptors.
• To describe the theoretical framework of the design so as to bring out its strengths and weaknesses
• To justify the choice of the design in relation to the context in which the study was done

5.3.2 Population and Sampling Procedures

The target population and sample used in the study should be described. The sampling techniques used should be
clearly defined.

A population is the total collection of elements about which we wish to make inferences. A census is a count of all
elements in the population. It is feasible when the population is small and necessary when the elements are quite
different from each other. The basic idea of sampling is that by selecting some of the elements in a population we
may draw conclusions about the entire population.

5.3.3 Types of Sampling Designs [APPENDIX II]

A variety of sampling techniques is available. They may be classified by their representation basis and element
selection techniques as shown in Table 1.

Table 1. Types of Sampling Designs

Representation Basis
Element Selection Probability Nonprobability
Unrestricted Simple random Convenience
Restricted Complex random Purposive
Systematic Judgment
Cluster Quota
Stratified Snowball
Double (Sequential, Multiphase)

Probability Sampling is based on random selection – a controlled procedure that ensures that each population
element is given a known chance of selection.
In developing a probability sample, 6 procedural questions need to be answered:
1 What is the relevant population?
2 What are the parameters of interest?
3 What is the sampling frame?
4 What is the type of sample?
5 What size sample is needed?
6 How much will it cost?

11
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

The simplest form of probability sampling is simple random sampling. In this design, each member of the
population has an equal chance of being included in a sample.

How to Choose a Random Sample

Selecting a random sample is accomplished with the aid of a computer software, a table of random numbers or a
calculator with a random number generator. Drawing slips out of a hat or ping-pong balls from a drum serves as an
alternative if every element in the sampling frame has an equal chance of selection. Mixing the slips (or balls) and
returning them between every selection ensures that every element is just as likely to be selected as any other.

A table of random numbers is a practical solution when no software program is available. Random number tables
contain digits that have no systematic organization. Whether you look at rows, columns, or diagonals, you will find
neither sequence nor order. The following procedure can be followed when selecting a sample using random
numbers:
1 Assign each element within the sampling frame a unique number.
2 Identify a random start from the random number table.
3 Determine how the digits in the random number table will be assigned to the sampling frame to choose the
specified sample size. This can be done horizontally right to left, bottom to top, diagonally across columns, and so
forth.
4 Select the sample elements from the sampling frame. If a digit appears twice the second instance will be omitted .

Generally, random sampling is employed when quantitative methods are used to collect data (e.g. questionnaires). It
allows the results to be generalized to the larger population and statistical analysis performed if appropriate.

Complex sampling is used when conditions make simple random samples impractical or uneconomical.
The following procedure is followed in systematic sampling:
• Identify the total number of elements in population.
• Identify the sampling ratio (k – total population size divided by size of the desired sample)
• Identify the random start.
• Draw a sample by choosing every kth entry.

The distinction between stratified sampling and cluster sampling is summarized below:
Stratified Sampling Cluster Sampling
1 We divide the population into a few subgroups, each 1 We divide the population into many subgroups, each
with many elements in it. The subgroups are selected with a few elements in it. The subgroups are selected
according to some criterion that is related to the according to some criterion of ease or availability in data
variables under study. collection.

2 We try to secure homogeneity within subgroups and 2 We try to secure heterogeneity within subgroups and
heterogeneity between subgroups homogeneity between subgroups, but we usually get the
reverse.
3 We randomly choose elements from within each 3 We randomly choose a number of subgroups, which
subgroup. we then typically study in depth.

The probability sampling techniques are summarized in appendix II.

Non-Probability Sampling consists of convenience, purposive and snowball techniques.

Convenience Sampling is a low-cost but less reliable non-probability sampling technique where element selection is
unrestricted or left to those elements easily accessible by the researcher. It is often used to test ideas or even gain
ideas about a subject of interest and can be used in the early stages of exploratory when you are seeking guidance.

Purposive Sampling consists of judgment sampling and quota sampling.

Judgment sampling occurs when a researcher arbitrarily selects sample members/elements to conform to some
criterion. It is appropriate in early stages of an exploratory study and when one wishes to select a biased group for
screening purposes.
12
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

With quota sampling relevant characteristics are used to stratify the sample in an attempt to improve the
representativeness of the sample. Although there is some danger of systematic bias, quota sampling can be
satisfactory in cases where predictive validity has been checked (e.g. election polls).

Snowball sampling is a nonprobability sampling procedure in which initial sample elements, which may or may not
have been chosen by probability techniques, refer to additional sample elements based on similar characteristics. In
the initial stage individuals are discovered and may or may not be selected through probability methods. This group
is then used to locate others who possess similar characteristics and who, in turn, identify others. This design is
useful in applications where respondents are difficult to identify and are best located through referral work. For
example, it can be used to study insider trading where respondents are difficult to identify and contact.

5.3.4 Important Points Regarding Samples and Sampling

• A good sample has both accuracy and precision. An accurate sample is one in which there is little or no bias or
systematic variance. A sample with adequate precision is one that has a sampling error that is within acceptable
limits for the study’s purpose.
• The specifications of the researcher and the nature of the population determine the size of a population sample.
• The greater the dispersion or variance within the population, the larger the sample must be to provide estimation
precision.
• The greater the desired precision of estimate, the larger the sample must be
• The narrower the interval range, the larger the sample must be.
• The higher the confidence level in the estimate, the larger the sample must be.
• The greater the number of subgroups of interest within a sample, the greater the sample size must be, as each
subgroup must meet minimum sample size requirements.
• If the calculated sample size exceeds 5 percent of the population, sample size may be reduced without
sacrificing precision.

5.3.5 Sources of Data

Primary Data
Original research where the data being collected are designed specifically to answer the research question. These
are data collected by the researcher and used for the purpose of the research. It could be collected using a
questionnaire administered on respondents included in the sample. Such data is original.

Secondary Data
Consists of studies done by others and for different purposes than for which the data are being reviewed. The
exploratory phase of the research process uses secondary data to look for ways others have addressed and/or solved
problems similar to the research question, and gather background information on the topic to refine the research
question. Sources include: encyclopedias, textbooks, handbooks, magazines and newspaper articles, most
newscasts, web-based sources, publications complied by different organizations that collect statistical information,
such as the Central Statistical Office (CSO). The research thus adopt data which may initially not have been
collected for use in that particular research but for something else. Whatever source, the trick is to browse and still
stay focused on the topic at hand.

5.3.6 Research Instruments / Data Collection Methods

[Link] Research Instruments are tools used for collecting information and data needed to find solutions to the
problem under investigation. There are a wide variety of data collecting methods available to researchers in the
social sciences; deciding which to use will depend on:
• the particular methodology that has been selected (quantitative or qualitative, or mixtures of both)
• what is feasible in a given research situation
• what is likely to yield the most appropriate information.

All these items are decided upon in the context of the particular research problem.

[Link] Some of the more important data collecting methods are listed below:
Questionnaires Dialogue records
13
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Observations Diaries and field notes

Documents Focus groups
Accounts and protocols Repertory grid methods
Tests Attitude scales
Sociometry Evaluation research
Action research Written and other works of various kinds (e.g. books, essays)
Interviews

Many of these data collecting methods have alternative forms. Observations, for instance, may be participant or
non-participant, while interviews may be structured, semi-structured or unstructured. Despite this variety there are
some basic principles which apply to all data collecting methods. The identified research instruments should be
appropriate for the study and should be clearly described so as to bring out their strengths and weaknesses as a way
of justifying their selection and suitability to the research. Measures taken to control the weaknesses identified
should be spelt out so as to ensure the validity and reliability of these instruments and the data to be collected. The
chosen instruments should reflect variables in the problems.

[Link] Basic Principles of Data Collection

(1) Any method must produce data that is relevant to the research question(s), and able to
provide answers or illumination on the topic.
(2) The technique must be convenient and relatively easy to use in the circumstances, for the
researcher has to be sure that the expenditure of time (and resources) is warranted by the project.
(3) Most importantly, issues of validity and reliability of the research techniques used should be
observed. The person undertaking the research needs to be sure that the data collecting instrument is both valid
and reliable.

Validity
If the research technique is valid it is “sound, cogent, well grounded, justifiable, or logically correct” (Schwandt,
1997). In simple terms, validity ensures that data sets collected or items used are pertinent or relevant to the
research.

There are different types of validity. Miller and Wilson (1983) list the following forms of internal validity:
• face
• concurrent
• content
• predictive
• criterion related
• construct
• convergent
• discriminant.

In the main, validity can be:

• internal or face validity (the techniques directly relate to the intended outcomes and concerns of the research)
• concurrent validity (the data collected is valid for the much wider population from which that sample is drawn)
• predictive validity (provides data that is useful for making predictions about the future behaviour of the research
subjects in a causal or correlational manner).

In validity the main concern should be to reduce the amount of interference by non-relevant or non-valid aspects,
such as language used. In research techniques such as interviews and questionnaires the language should not be
overly complex and arcane, or hinder understanding and answering (responses).

Ways of ensuring validity are:

• Carrying out an initial investigation (a pilot study) using intended data collecting instrument to check
authenticity and relevance of the data produced.
• A panel of experts can be used to assess that the planned instrument really does measure what it is supposed to
be measuring.
14
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Where a valid test or instrument already exists, it is open to the researcher to use the existing instrument. It is
not necessary to reinvent the wheel all the time!
• Factor analysis can be employed to check validity (Oppenheim, 1992 and Kline, 1994)
• The researcher can use a convenient approach known collectively as triangulation. Here the researcher repeats
the (pilot) study using alternative:
- data collecting techniques
- investigators
- methodologies
- units of analysis
- theoretical perspectives.

Reliability
Miller and Wilson (1983) define reliability as:
“The extent to which a test would give consistent results if applied more than once to the
same people under standard conditions”.

Methods of checking or assessing reliability are:

• Using the “Test-retest” method. This involves using an instrument (eg interview schedule) with a group on two
separate occasions and analyzing how closely the two sets of results conform to each other (provided the two
occasions are not separated by an excessive time-scale). A significant correlation should be observed between
the two sets of results, with minor differences as respondents are likely to change.
• Split the results of a test or questionnaire into two halves and then measure how well the one correlates with the
other. There are computer packages that carry out this procedure, such as SPSS, which as Cronbach’s Alpha
which works on the above principle.
Problems may still arise, however, for a test can be reliable but not valid. It is possible that a test may measure
something other than what it is believed the test is measuring, in which case it is reliable but not valid because it
would not be measuring what the researcher supposed it to be measuring.

Ease of Use
It is important that the data collecting instruments are easy to administer. One can have an instrument that is valid
and reliable but, if a single researcher is unable to use it because of its complexity, the results yielded may still be
suspect. Examples are structure observation schedules which involve complex activities and interactions.

[Link] Questionnaires and Interview

Two of the most commonly use data collecting techniques are questionnaires and interviews. Some of the
advantages and disadvantages that may help a researcher decide which method is more appropriate in a given
instance are as follows:

Questionnaires Interviews
Useful for simple topics Useful for more complex topics
Possible to have large numbers of respondents Time consuming
Anonymous Not suitable for embarrassing issues
Often poor response rate Usually good response rate
Emphasis on writing ability Less emphasis on writing ability.

Someone other than the researcher should scrutinize a questionnaire or interview schedule to eliminate unsuitable or
ambiguous items. In addition it is essential that the schedule or questionnaire is piloted on respondents who are as
similar as possible to those in the group who will eventually be asked to participate. It is often helpful to go through
a questionnaire verbally at the pilot stage, for this can provide information on irrelevant or poorly constructed items.
At some point, however, it is also necessary to ask the pilot respondents to complete the instrument under conditions
similar to the “real” thing.

15
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

[Link] Planning Questionnaires

The instrument design process starts with a comprehensive list of investigative questions drawn from the research
questions. Three general classes of information are obtained:
1 Administrative questions identify the respondent, interviewer, and interview location and conditions
2 Classification (or demographic) questions concern respondent characteristics and allow respondents’
answers to be grouped for analysis
3 Target questions address the investigative questions and are the most important. They make up the major
portion of the questionnaire and consist of all the questions that will extract data from respondents to
address the research objective. This section may also be sub-divided to group questions into categories
dealing with different aspects of the study. Target questions may be structured (they present the
respondents with a fixed set of choices, often called closed questions) or unstructured (they do not limit
response but do provide a frame of reference for respondents answers, sometimes referred to as open-ended
questions)

Question construction involves 3 crucial decisions:

1 Question content
2 Question wording
3 Response strategy.

Question Content
Four questions guide the instrument designer in selecting appropriate question content.
• Should this question be asked?
• Is the question of proper scope and coverage?
• Can the respondent adequately answer this question, as asked?
• Will the respondent willingly answer this question, as asked?

Question Wording
The diligent question designer will put a given question through many revisions before it satisfies the following
criteria:
• Is the question stated in terms of shared vocabulary? (i.e. vocabulary that is understood equally by the
interviewer and the respondent). The use of simple, unambiguous and precise words is important.
• Does the question contain vocabulary with a single meaning?
• Does the question contain unsupported or misleading assumption?
• Does the question contain biased wording? Bias is the distortion of responses in one direction.
• Is the question correctly personalized?
• Are adequate alternatives presented within the question?

Response Strategy
The various response strategies offer options that include unstructured response (or open-ended response, the free
choice of words) and unstructured response (or closed response, specified alternatives provided). Free responses,
in turn, range from those in which the respondents express themselves extensively to those in which respondents’
latitude is restricted by space, layout, or instructions to choose one word or phrase, as in a “fill-in” question. Closed
responses typically are categorized as dichotomous, multiple-choice, checklist, rating, or ranking response strategies.
Situational Determinants of Response Strategy Choice. Several situational factors affect the decision of whether to
use open-ended or closed questions. The decision is also affected by the degree to which these factors are known to
the interviewer. The factors are:
• Objectives of the study. If the objective of the question is only to classify the respondent on some stated point of
view, then the closed question will serve well. E.g. assume you are interested only in whether a respondent
approves or disapproves of a certain corporate policy. A closed question will provide the answer. If the
objective is to explore a wider territory, then an open-ended question (free-response strategy) is preferable.
Open-ended questions are appropriate (a) when the objective is to discover opinions and degrees of knowledge
(b) when the interviewer seeks sources of information, dates of events, and suggestions, or when probes are
used to secure more information, (b) when the topic of a question is outside the respondent’s experience, open-
ended questions may offer the better way to learn his/her level of information.

16
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Respondent’s level of information about the topic. It may by better to use open-ended questions when the when
the interviewer does not have a clear idea of the respondent’s frame of reference or level of information
(especially in exploratory research or in pilot testing). On the other hand, closed questions are better when there
is a clear frame of reference, the respondent’s level of information is predictable, and the researcher believes the
respondent understands the topic.
• Degree to which respondent has thought through the topic. If the respondent has developed a clear opinion on
the topic, a closed question does well. If an answer has not been though out, an open-ended question may give
the respondent a chance to ponder a reply, then elaborate on and revise it.
• Ease with which respondent communicates. Open-ended questions require a stronger grasp of vocabulary and a
greater ability to frame responses than do closed questions.
• Respondent’s motivation level to share information. Closed questions typically require less motivation and
answering them is less threatening to respondents. But the response alternatives sometimes suggest which
answer is appropriate; for this reason, closed questions may be biased.
While the open-ended question offers may advantages, closed questions are generally preferable in large surveys.
-they reduce the variability of response
-they make fewer demands on interviewer skills
-they are less costly to administer, and
-they are much easier to code and analyze.

Alternative Response Strategies

The characteristics of the respondents, the nature of the topic(s) being studied, the type of data needed, and your
analysis plan dictate the response strategy.
• Free-Response Strategy – free-response or open-ended questions. Data generated can be nominal, ordinal, or
ratio data).
• Dichotomous Response Strategy – Dichotomous questions suggesting opposing responses. (E.g. yes-no
questions)
• Multiple-Choice Response Strategy – Multiple-choice questions are appropriate where there are more than two
alternatives or where we seek gradations of preference, interest or agreement; the later situation also calls for
rating questions. Multiple-choice questions generates ordinal data.
• Checklist Response Strategy – when you want the respondent to give multiple responses to a single question.
Relative order of response is not important. Generates nominal data.
• Rating Response Strategy – Checklist questions ask the respondent to position each property or object on a
companion verbal, numeric, or graphic scale without making a direct comparison to another object. Generates
ordinal or interval data.
• Ranking Strategies – The ranking question is ideal when the relative order of the alternative is important. Asks
the respondent to compare and order two or more objects or properties using a numeric scale. Generates ordinal
or interval data.

Summary of Characteristics of Response Strategies

Characteristics Dichotomous Multiple Checklist Rating Rank Ordering Free Response
Choice
Type of data Nominal Nominal, Nominal Ordinal or Ordinal Nominal or
generated ordinal, or ratio interval ratio
Usual number 2 3-10 10 or fewer 3-7 10 or fewer None
of answer
alternatives
provided
Used to Classification Classification, Classification Order or Order Classification,
provide…….. order, or distance order, or
specific specific
number number of
ideas.

All types of response strategies have their advantages and disadvantages. Several different strategies are often found
in the same questionnaire, and the situational factors are the major guides in this matter. Rating or ranking scales can
be used to quantify dimensions that are essentially qualitative.
17
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Question sequence can drastically affect respondent willingness to cooperate and the quality of responses.
Questions should be sequenced so early questions do not distort replies to latter ones.

Appendix IV provides a number of specific pointers to be borne in mind when designing a well-structured and
undisguised questionnaire to ensure that accurate, unbiased and relevant data of the correct type is gathered in line
with the research objective(s). In addition, the following hints are also essential:
• The use of capital letters (upper case) only should be avoided, as this format is hard to read.
• Questions should be numbered and clearly grouped by subject.
• Clear instructions should be given and headings included to make the questionnaire easier to follow.
• Avoid double-barrelled questions (two or more questions in one)
• Avoid questions with double negatives and leading or ambiguous questions.

Pretesting the instrument is recommended to identify problems before the actual collection of data begins.

5.3.7 Data Collection Procedures

These are steps taken to administer instruments and collection of data from subjects under study. These steps
include:
1 Making appointments with research subjects by telephone, letters, fax, etc
2 Distribution and administering of instruments on the sample, e.g., by hand, by mail or through research
assistants
3 Retrieval of instruments.

These steps need to be clear, orderly and justified.

5.3.8 Data Presentation and Analysis Procedures

This section spells out the overall procedures to be used to organize, describe and analyze the collected data. The
process entails suggesting the manner in which findings are to be (a) presented to logical themes (e.g. tables, graphs
and/or diagrams), (b) described to reveal their meaning.

These procedures should be clear, logical and justified. If the research has used a particular model, it should be
described at length and justified.

5.3.9 Conclusion
The methodology chapter is the blueprint or roadmap that will shape the outcome of the entire project. Therefore, it
is important to ensure that the right tools are used to ensure that correct results that are relevant to what the
investigator sought out to discover, are generated.

5.4 CHAPTER IV – DATA PRESENTATION ANALYSIS AND DISCUSSION OF RESULTS (OR,

DATA PRESENTATION AND DISCUSSION OR DATA PRESENTATION, ANALYSIS AND
INTERPRETATION)

The chapter focuses on presentation techniques, discussion and interpretation of research findings.

5.4.1 Data Presentation and Analysis

Before the data is presented it goes through the following process
1 Data preparation, which includes editing, coding and data entry. These activities ensure the accuracy of
the data and their conversion from raw form to reduced and classified forms that are more appropriate for
analysis.
2 Preparing a descriptive statistical summary leading to an understanding of the collected data.
3 Summarize the data using tables, graphs, figures, etc. Always choose the form that provides a clear
presentation of your data. Always refer to the table or figure that you are discussing and do not leave a
table or figure “hanging”. Tables and figures should be numbered and labeled.

18
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Editing, the first step in data analysis, detects errors and omissions, corrects them when possible, and certifies that
minimum data quality standards are achieved. The editor’s purpose is to guarantee that data are:
• Accurate
• Consistent with intent of the question and other information in the survey
• Uniformly entered
• Complete
• Arranged to simplify coding and tabulation.

Coding involves assigning numbers or other symbols to answers so the responses can be grouped into a limited
number of classes or categories. Coding can be alphanumeric (when numbers are combined with symbols) or
numeric (when numbers are used exclusively). Coding helps the researcher to reduce several thousand replies to a
few categories containing the critical information needed for analysis.

Data entry converts information gathered by secondary or primary methods to a medium for viewing and
manipulation. Keyboarding remains a mainstay for researchers who need to create a data file immediately and store
it in a minimal space on a variety of media. Other programs available are optical scanning instruments, voice
recognition and response systems or the bar code. A full-screen editor, where an entire data file can be edited or
browsed, is a viable means of data entry for statistical packages like SPSS, or SAS. The same software makes
accessing data from databases, spreadsheets, data warehouses, or data marts effortless. For large projects, database
programs serve as valuable entry devices. For data that need organizing, tabulating, and simple statistics,
spreadsheets provide an easy-to-learn mechanism.

Descriptive Statistical Summaries provide ways of describing the data. The tools used to clean the data and
summarize distributions include: measures of location or central tendency (mean, median, and mode), measures of
spread or dispersion (variance, standard deviation, range, interquartile range, and quartile deviation), and measures
of shape (skewness or kurtosis). The discovery of miscoded values, missing data, and other problems in the dataset
are enhanced with descriptive statistics.

Techniques used for displaying data include: frequency tables, bar charts, pie charts; histograms; stem-and-leaf
displays; and boxplots. Process control systems make use of control charts. The evaluation of relationships
involving categorical variables employs cross-tabulation and this can be made more efficient with the use of
computer packages such as Minitab, SAS, and SPSS.

Simple Quantitative Data Analysis

There a number of statistical data analysis techniques with which researchers ought to be familiar. There are two
ways that researchers can draw conclusions from quantitative data.
• Data may be used to describe the situation using descriptive statistics.
• The researcher may wish to infer something about the much wider population from which that sample was
drawn. In this case one would require the help of inferential statistics.

Types of Quantitative Data

Figure 5.1 illustrates the various types of data that could be obtained in a piece of research.

19
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Figure 5.1 Types of Research Data

Category Quantity

Discrete
variables
Nominal Ordinal

Continuous

Ordered
categories Ranks
Ratio
Non-ratio

Inferential Statistical Tests

A range of inferential statistical tests are available, provided the researcher is sure about what they wish to measure
and compare; and the type of data they have. Most of the tests are also available in the standard computer packages
such as Minitab, SPSS and SAA. Figure 5.2 provides details of some of the more commonly used tests.

Tests of difference are used where it is suspected that groups differ significantly on or as a result of a particular
variable.
Correlations (or mutual connections) can be used if it is suspected that there may be a relationship between two or
more variables within a group. It must be remembered that correlations do not necessarily mean that there is a
causal connection between the variables.
Contingency table analysis assists the researcher in determining whether an observed distribution is significant.
Contingency tables can be used to help with the processing of questionnaire data by collating data in tabular form.

20
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Figure 5.2 Some Basic Inferential Statistical Tests

Significant Contingency table Correlation

difference

Matched or -Chi square

independent -Fisher exact
Parametric
-Pearson
product
moment
correlation
- Related t -Unrelated t coefficient
-Sign test -Mann-
-Wilcoxon Whitney

Non-p
Non-p Parametric -Spearmen -Kendal
-Sign -Related t Non-p Parametric rank order Tau correlation
-Wilcoxon -Mann-Whitney -Unrelated t correlation coefficient
coefficient

In summary, the questions that need to be asked about statistical tests in quantitative data analysis are:
1 What sort of data has been obtained?
2 Is the data:
• parametric?
• non-parametric?
3 Which of the following is being sought
• significant differences
• correlations
• contingency table analysis?

4 What level of significance is required?

5 Is the test to operate under one-tailed or two-tailed conditions?
6 Are the subjects:
• matched/repeated measures
• independent?

Complex Quantitative Data Analysis

In quantitative data analysis there remain a raft of statistical approaches that go beyond simple tests. If the
researcher thinks such approaches might be applicable to them it is often beneficial to carry out exploratory data
analysis.
Exploratory data analysis
Exploratory data analysis is a preliminary examination of data undertaken in order to ascertain if there are any
patterns in the data that warrant a more detailed examination with more specialist techniques. Marsh (1988) suggests
a number of methods of performing this task. One of the easier methods is the “stem and leaf plot”, a means of
looking at the general shape of the data.
Cluster analysis
This is a computer-based method which groups individuals on the basis of measured characteristics. It might well
be that one can identify common characteristics of a group participating in the informal foreign exchange
transactions (by address, age, gender, socio-economic group, level of education, ethnicity, etc).
Factor analysis

21
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Factor analysis seeks to describe a range of variables (e.g. questionnaire responses) in terms of a more basics set of
underlying factors. This technique is used often with responses derived using a Likert scale. Principal components
analysis is a more sophisticated form of factor analysis which can be necessitated with the use of a computer
software package such as SPSS, where the manual will provide all the necessary help (Kline, 1994 and Child, 1980)
Discriminant analysis
Discriminant analysis seeks to link a number of possible outcomes statistically to variables that appear to be related
to these outcomes (Klecka, 1980).
Regression methods
This is where the outcome (dependent variable) is related to various independent variables that are weighted to show
their relative importance. Regression methods take a variety of forms and are useful in modeling social processes.
Path analysis
Path analysis is a method of assessing the extent to which a correlation between two or more variables is a causal
relationship. It is commonly used in association with the more sophisticated correlation methods.
Correlation methods
Some complex correlational tests and techniques that researchers can use are:
Pearson product moment Tetrachoric
Spearman rank order Partial correlation coefficients
Kendall test Canonical correlation coefficients
Point biserial r Multiple correlation coefficients
Phi

Analyzing Qualitative Data

The analysis of qualitative data does not present the same neat and predictable patterns typically presented by
quantitative data. Qualitative data is much richer and more diverse and therefore amenable to a range of analyses
which may at times appear to conflict with one other. Nonetheless, all qualitative data require there to be coding of
either a descriptive or categorical type.
Content analysis
Content analysis is a method of analyzing qualitative data that almost converts it back into quantitative data. The use
of content analysis is a systematic procedure using the following stages:
1 Select representative samples of the text.
2 Select appropriate categories; these may be phrases, words or “units of meaning” or amounts of
text which define an issue, i.e. racism, sexism, egocentricity, individual or organizational goals.
3 Select appropriate recording units; these may be:
-single words, eg “I”, “We”, “Us”
- themes, e.g. an ideology, developing world, technology, language
- a characteristic, e.g. “working class”
- a particular topic or idiom.
4 The context in which the text was produced needs to be known.
5 Select suitable enumeration systems (key works, frequency, space used, intensity of statements,
and apply them (count or measure).

The results can be tabulated, presented graphically and analyzed for regularities and/or balance or absence of items
(Paulston (1977).

Methods based on categorization

These are probably the most valuable ways of analyzing the data, by constructing and using an analytical
framework. This can be either an open coding approach, using the “scissors and paste” method, or a categorical
approach, where preconstructed categories (a are placed over the data or used as a sieve to bring out key aspects.

Network methods
Network data analysis methods construct a network of processes and relationships from the text data. This analysis
is particularly useful in the analysis of verbal protocols (streams of explanation).

Ethnographic data analysis

22
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

If the object of the analysis is to generate grounded theory (research that interrelates theory with practical situational
observation and understanding), we use open, axial and selective coding (Strauss and Corbin, 1989)

Career paths
Life histories can be a valuable way of making sense of the lives of individuals. This can be achieved through
narrative analysis, lifeline diagrams, etc.

Discourse analysis
Many forms of observational study take som account of the amount and form of discourse occurring within a social
setting (Stubbls, 1983 and Coulthard, 1985).

NB
Only facts should be presented in sequence, and in terms of sub-problems/research questions. After presenting data
on the background of the subjects, the researcher should move on to data presentation focusing on specific sub-
problems/research questions of the research study. These should be answered logically and separately. Make use of
sub-headings and these should emerge from the sub-problems/research questions. Analysis of data should focus on
accurate descriptions, clarity and ability to condense.

5.4.2 Discussion of Results (or Findings)

This is where you interpret the meaning of the research results. Tie your results to literature review, showing the
points of similarity and departures with theory and previous research findings on the topic. Where there are
deviations give reasons for such deviations and suggest the modifications that are needed, and if possible, suggest a
new theoretical framework.

If the study was set up to test hypotheses, the outcome of each hypothesis should be reported. The statistical test
used must also be appropriate.

Ensure that all the results of the sub-problems/sub-questions and/or hypotheses are discussed.

Practical implications of your results should always be stated.

In terms of language and expression, be definite about the data and statistics, but be tentative about interpretations
and conclusions.

The chapter summary should highlight the main findings of the study, and prepare the reader for the next chapter.

Hypothesis Testing
The purpose of hypothesis testing is to determine the accuracy of your hypotheses due to the fact that you have
collected a sample of data, not a census. Two kinds of hypothesis are used in classical tests: the null hypothesis and
the alternative hypothesis. The null hypothesis is used for testing. It is a statement that no difference exists between
the parameter and the statistic being compared to it. If we reject a null hypothesis, then we are accepting the
alternative hypothesis.

Hypothesis testing can be viewed as a six-step procedure:

1. State the null hypothesis.
2. Choose the statistical test.
3. Select the desired level of significance.
4. Compute the calculated difference value.
5. Obtain the critical value.
6. Interpret the test.

There are two general classes of significance tests: Parametric or nonparametric. Parametric tests are more
powerful because their data are derived from interval and ratio measurements. Nonparametric tests are used to test
hypothesis with nominal and ordinal data. Fig 2 provides a classification of the major parametric and nonparametric
tests and measures.

23
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Table 2. Major Statistical Techniques for Significance Testing

One-Sample Two-Samples Case k-Samples Case
Measurement Case Related Samples Independent Related Independent
Level Samples Samples Samples
Nominal Binominal McNemar Fisher exact test 2 Cochran Q 2
for k samples
2
one-sample Two-samples
test
Ordinal Kolmogorov- Sign test Median test Friedman two- Median extension
Smirnov one- Wilcoxxon Mann-Whitney U way ANOVA Kruskal-Wallis
sample test matched pairs Kolmogorov- one-way
Runs test Smirnov ANOVA
Wald-Wolfowitz
Interval and t-test t-test for paired t-test Repeated- One-way ANOVA
ratio Z test samples Z test Measures n-way ANOVA
ANOVA

The above descriptions emphasize testing hypotheses of difference. However, some research questions frequently
revolve around a study of relationships between two or more variables. This necessitates the need for a relational
hypothesis. The most commonly used measures of association are summarized in Appendix IV. Scatterplots provide
a visual inspection of the direction, magnitude and shape of relationships. Coefficients for more than two variables
can be displayed in a table called, a correlation matrix. Simple and multiple predictions are made with regression
analysis.

Multivariate analysis can also be applied to research problems. This focuses on bringing out the “structure of
simultaneous relationships among three or more phenomena.” (Sheth, J N, ed (1977) p.3.)

5.5 CHAPTER V – CONCLUSION AND RECOMMENDATIONS

The chapter focuses on 3 areas.

• The summary
• The research conclusions, and
• The recommendations.

5.5.1 Summary
This gives a brief summary of the research problem, methodology, and implications of findings in terms of research
objectives. Revisit your original objectives that you set out to achieve in chapter one and show how these have been
achieved.

5.5.2 Conclusions
These are summed up answers to the sub-problems/research questions and are drawn directly from data reported in
chapter 4. Conclusions should be based on the findings. It is important to note that: while findings state facts;
conclusions represent inferences drawn from the findings. There is also need to indicate whether hypotheses are
accepted or rejected. Whether or not your conclusions agree with the theory or previous researches, you have to
report them as they are. Do not “doctor” your findings to suit your preconceived ideas.

5.5.3 Recommendations
In the light of the conclusions drawn by the study, the researcher should clearly state recommendations on actions
that should be taken to solve problems identified. In addition, suggestions for future research on the study should be
made.

Finally, the chapter summary should tie the whole study together into a neat and comprehensive package.

24
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

REFERENCES

American Psychological Association (2001) APA Publication Manual, 5th ed. Washington D.C.

Child, D (19990), Factor Analysis, Holt, London.

Donald, R and Schindler, P. S (2000) Business Research Methods, 7th ed. Boston, New York: McGraw Hill.

Kerlinger, F. N. (1966), Foundation of Behavioural Research, 3rd ed. New York: Holt, Rinchart & Winston.

Klecka, W R (1980) discriminanti Analysis, Sage, London

Kline, P (1994) An Easy Guide to Factor Analysis, Routledge, London.

Kline, P (1993) A Handbook of Psychological Testing, Routledge, London.

Kothari, C. R. (1985) Research Methodology: Methods and Techniques, New Delhi: Wiley Eastern

Leedy, P. D. (1994), Practical Research Planning and Design, 3rd ed. New York: Harcourt Brace Jovanovich Inc.

Phillips, B. S. (1971) Social Research Strategy and Tactics, 2nd ed. Macmilan Publishing.

Sheth, J N ed (1977) Multivariate Methods for Market and Survey Research. Chicago: American Marketing
Association.

Strauss, A and Corbin, J (1989) Basics of Qualitative Research, Sage, London.

Stubbs, M (1983) Discourse Analysis, Blackwell, Oxford.

Wegner, T (1993) Applied Business Statistics, Kenwyn: Juta & Co Ltd.

APPENDIX I

RESEARCH METHODOLOGY PARADIGMS

Quantitative, Qualitative, and Mixed Research

A paradigm is a perspective based on a set of assumptions, concepts, and values that are held by a community or
researchers. For the most of the 20th century the quantitative paradigm was dominant. During the 1980s, the
qualitative paradigm came of age as an alternative to the quantitative paradigm, and it was often conceptualized as
the polar opposite of quantitative research. Finally, although the modern roots of mixed research go back to the late
1950s, it truly became the legitimate third paradigm with the publication of the Handbook of Mixed Methods in
Social and Behavioral Research (2003, by Tashakkori and Teddlie). At the same time, mixed research has been
conducted by practicing researchers throughout the history of research.

Characteristics of the Three Research Paradigms

There are currently three major research paradigms: quantitative research, qualitative research, and mixed research.
Here are the definitions of each:
• Quantitative research – research that relies primarily on the collection of quantitative data. (Note that pure
quantitative research will follow all of the paradigm characteristics of quantitative research shown in the
left column of Table 2.1.)
• Qualitative research – research that relies on the collection of qualitative data. (Note that pure qualitative
research will follow all of the paradigm characteristics of qualitative research shown in the right column of
Table 2.1.)

25
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Mixed research – research that involves the mixing of quantitative and qualitative methods or paradigm
characteristics. Later in the lecture you will learn about the two major types of mixed research, mixed
method and mixed model research. For now, keep in mind that the mixing of quantitative and qualitative
research can take many forms. In fact, the possibilities for mixing are almost infinite.

26
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Quantitative Research Methods:

Experimental and Nonexperimental Research

The basic building blocks of quantitative research are variables. Variables (something that takes on different values
or categories) are the opposite of constants (something that cannot vary, such as a single value or category of a
variable).

Many of the important types of variables used in quantitative research are shown, with examples, in Table 2.2.

Here is that table for your review:

27
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

In looking at the table note that when we speak of measurement, the most simple classification is between
categorical and quantitative variables. As you can see, quantitative variables vary in degree or amount (e.g., annual
income) and categorical variables vary in type or kind (e.g., gender).

The other set of variables in the table (under the heading role taken by the variable) are the kinds of variables we
talk about when explaining how the world operates and when we design a quantitative research study.

As you can see, independent variables (symbolized by "IV") are the presumed cause of another variable. Dependent
variables (symbolized by "DV") are the presumed effect or outcome. Dependent variables are influenced by one or
more independent variables. What is the IV and DV in the relationship between smoking and lung cancer? (Smoking
is the IV and lung cancer is the DV.)

Sometimes we want to understand the process or variables through which one variable affects another variable. This
brings us to the idea of intervening variables (also called mediator or mediating variables). Intervening variables are
variables that occur between two other variables. For example, tissue damage is an intervening variable in the
smoking and lung cancer relationship. We can use arrows (which mean causes or affects) and draw the relationship
that includes an intervening variable like this:
Smoking---->Tissue Damage---->Lung Cancer.

Sometimes a relationship does not generalize to everyone; therefore, researchers often use moderator variables to
show how the relationship changes across the levels of an additional variable. For example, perhaps behavioral
therapy works better for males and cognitive therapy works better for females. In this case, gender is the moderator
variable. The relationship be type of therapy (behavioral versus cognitive) and psychological relief is moderated by
gender.

Experimental Research
The purpose of experimental research is to study cause and effect relationships.
Its defining characteristic is active manipulation of an independent variable (i.e., it is only in experimental research
that “manipulation” is present). Also, random assignment (which creates "equivalent" groups) is used in the
strongest experimental research designs.

Nonexperimental Research
Remember that the defining characteristic of experimental research was manipulation of the IV. Well, in
nonexperimental research there is no manipulation of the independent variable. There also is no random assignment
of participants to groups.

What this means is that if you ever see a relationship between two variables in nonexperimental research you cannot
jump to a conclusion of cause and effect because there will be too many other alternative explanations for the
relationship.

Qualitative Research Methods

There are five major types of qualitative research: phenomenology, ethnography, case study research, grounded
theory, and historical research. All of the approaches are similar in that they are qualitative approaches. Each
approach, however, has some distinct characteristics and tends to have its own roots and following.

Here are the definitions and an example of the different types of qualitative research:
• Phenomenology – a form of qualitative research in which the researcher attempts to understand how one or
more individuals experience a phenomenon. For example, you might interview 20 widows and ask them to
describe their experiences of the deaths of their husbands.
• Ethnography – is the form of qualitative research that focuses on describing the culture of a group of
people. Note that a culture is the shared attitudes, values, norms, practices, language, and material things
of a group of people. For an example of an ethnography, you might decide to go and live in a Mohawk
communities and study the culture and their educational practices.

28
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

• Case study research – is a form of qualitative research that is focused on providing a detailed account of
one or more cases. For an example, you might study a classroom that was given a new curriculum for
technology use.
• Grounded theory – is a qualitative approach to generating and developing a theory form data that the
researcher collects. For an example, you might collect data from parents who have pulled their children out
of public schools and develop a theory to explain how and why this phenomenon occurs, ultimately
developing a theory of school pull-out.
• Historical research – research about events that occurred in the past. An example, you might study the use
of corporeal punishment in schools in the 19th century.

Mixed Research Methods

Mixed research is a general type of research (it’s one of the three paradigms) in which quantitative and qualitative
methods, techniques, or other paradigm characteristics are mixed in one overall study. Earlier we showed it major
characteristics of mixed research in Table 2.1. Now the two major types of mixed research are distinguished: mixed
method versus mixed model research.
• Mixed method research – is research in which the researcher uses the qualitative research paradigm for one
phase of a research study and the quantitative research paradigm for another phase of the study. For
example, a researcher might conduct an experiment (quantitative) and after the experiment conduct an
interview study with the participants (qualitative) to see how they viewed the experiment and to see if they
agreed with the results. Mixed method research is like conducting two mini-studies within one overall
research study.
• Mixed model research – is research in which the researcher mixes both qualitative and quantitative research
approaches within a stage of the study or across two of the stages of the research process. For example, a
researcher might conduct a survey and use a questionnaire that is composed of multiple closed-ended or
quantitative type items as well as several open-ended or qualitative type items. For another example, a
researcher might collect qualitative data but then try to quantify the data.

The Advantages of Mixed Research

First of all, we advocate the use of mixed research when it is feasible. We are excited about this new movement in
educational research and believe it will help qualitative and quantitative researchers to get along better and, more
importantly, it will promote the conduct of excellent educational research.
• Perhaps the major goal for researcher who design and conduct mixed research is to follow the fundamental
principle of mixed research. According to this principle, the researcher should mix quantitative and
qualitative research methods, procedures, and paradigm characteristics in a way that the resulting mixture
or combination has complementary strengths and nonoverlapping weaknesses. The examples just listed for
mixed method and mixed model research can be viewed as following this principle. Can you see how?
• Here is a metaphor for thinking about mixed research: Construct one fish net out of several fish nets that
have holes in them by laying them on top of one another. The "new" net will not have any holes in it. The
use of multiple methods or approaches to research works the same way.
• When different approaches are used to focus on the same phenomenon and they provide the same result,
you have "corroboration" which means you have superior evidence for the result. Other important reasons
for doing mixed research are to complement one set of results with another, to expand a set of results, or to
discover something that would have been missed if only a quantitative or a qualitative approach had been
used.
• Some researchers like to conduct mixed research in a single study, and this is what is truly called mixed
research. However, it is interesting to note that virtually all research literatures would be mixed at the
aggregate level, even if no single researcher uses mixed research. That's because there will usually be some
quantitative and some qualitative research studies in a research literature.

Our Research Typology

Putting the three research paradigms and their subtypes together, we get the following picture of research typology:

29
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

Figure 2.3 Research Typology

Level 1: 1 2 3
Quantitative Mixed Qualitative
Research research research

Level 2: a b a b a e
Experimental Nonexperimntal Mixed Mixed Phenomenology c
Historical
method b Case study d
Ethnography Grounded
Theory

APPENDIX II
DESCRIPTORS OF RESEACH DESIGN

Category Options Discussions

The degree to which • Exploratory study Tend to be loosely structured and immediate purpose is to develop hypotheses or
the research question research questions for further study.
has been crystallized.
• Formal study
Begins where the exploration leaves off – with a hypothesis or research question and
involves precise procedures and data source specifications. The goal is to test the
hypotheses or answer the research questions posed.
The method of data • Monitoring Researcher inspects the activities of a subject or the nature of some material without
collection attempting to elicit responses from anyone.
Researcher questions the subjects and collects their responses by personal or
• Interrogation/Comm impersonal means. The data may result from (1) interview or telephone conversation
unication study (2) self-administered or self-reported instruments sent through the mail, left in
convenient locations, or transmitted electronically or by other means, or (3)
instruments presented before and/or after a treatment or stimulus condition in an
experiment.
The power of the • Experimental Researcher attempts to control and/or manipulate the variables in the study.
researcher to produce Appropriate when one wishes to discover whether certain variables produce effects in
effects in the other variables.
variables under study
(or the ability to Investigators have no control over the variables in the sense of being able to
manipulate variables) • Ex post facto design manipulate them. They can only report what has happened or what is happening.
The purpose of the • Descriptive Concerned with finding out who, what, where, when, or how much.
study
Concerned with learning why – how one variable produces change in another – i.e.
• Causal tries to explain relationships among variables.
The time dimension • Cross-sectional Carried out once and represent a snapshot of one point in time.
studies
Are repeated over an extended period – therefore can track changes over time
• Longitudinal studies
The topical scope- • Case studies Place more emphasis on a full contextual analysis of fewer events or conditions and
breath and depth-of their interrelations for a single subject or respondent.
the study
Attempts to capture a population’s characteristics by making inferences from a
• Statistical study
sample’s characteristics. Involves hypothesis testing and is more comprehensive than
a case study.
The research • Field setting or The actual environmental conditions where the dependent variable occurs.
environment conditions
Studies that occur under conditions that do not simulate actual environmental
conditions.
• Laboratory research
or conditions
A study in which the conditions of a system or process are replicated.
• Simulation
The subjects’ • Actual routine Subjects perceive no deviations from everyday routines.
perceptions of
research activity Subjects perceive deviations as researcher-induced.
• Modified routine

30
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

APPENDIX III
COMPARISON OF PROBABILITY SAMPLING DESIGN

Type Description Advantages Disadvantages

Simple random Each population element Easy to implement with Requires a listing of
has an equal chance of automatic dialing (random population elements.
being selected into the digit dialing) and with Takes more time to
sample. Sample drawn computerized voice implement.
using random number table response systems. Uses larger sample sizes.
generator.
Systematic Selects an element of the Simple to design. Periodicity within the
population at a beginning Easier to use than simple population may skew the
with a random start and random. sample and results.
following the sampling Easy to determine If the population list as a
fraction selects every kth sampling distribution of monotonic trend, a biased
element. mean or preparation. estimate will result based
Less expensive than on the start point.
simple random.
Stratified Divides population into sub- Researcher controls Increased error will result
population or strata and uses sample size in strata. if subgroups are selected at
simple random on each Increased statistical different rates.
strata. Results may be efficiency. Expensive
weighted and combined. Provides data to represent Especially expensive if
and analyze subgroups strata on the population
Enables use of different have to be created.
methods in strata.
Cluster Population is divided into Provides an unbiased Often lower statistical
internally heterogeneous estimate of population efficiency (more error) due
sub-groups. Some are parameters if properly to subgroups being
randomly selected for done. homogeneous rather than
further study. Economically more heterogeneous.
efficient than simple
random.
Lowest cost per sample,
especially with
geographic clusters.
Easy to do without a
population list.
Double Process includes collecting May reduce costs if firstIncreased costs if
(sequential or data from a sample using a stage results in enough indiscriminately used.
multiphase) previously defined data to stratify or cluster
technique: based on the the population.
information found, selecting
a subsample for further
study.
Source: Donald, R and Schindler, P. S (2000) Business Research Methods, 7th ed. Boston, New York: McGraw Hill.

31
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

APPENDIX IV

CHECKLIST IN QUESTIONNNAIRE DESIGN

• A question must be included only if it contributes to the research objectives. Redundant questions must be
avoided.

• Ambiguous questions must be avoided

• Leading questions generate biased responses and must be avoided.

• Wherever possible, especially with self-administered questionnaires, fixed-alternative questions are

recommended. Open-ended questions are difficult to analyze.

• Questions should be arranged in a logical sequence.

• Always pilot a questionnaire before carrying out the full survey. This often highlights weaknesses in a
questionnaire which can be corrected in time.

• The questions should be kept as short and as simple as possible.

• Technical jargon should be avoided.

• Questions should not require calculations.

• Filter questions should be used to guide the respondent through various sections of the questionnaire.

• Instructions must be clear and explicit.

• Questions must be specific and address only one insure. Vague questions result in “noisy” and meaningless
data.

• Emotive language should be avoided to prevent biased responses.

• Questions on sensitive issues should be carefully worded so as not to offend or encourage non-response.

• The ability of the target population to understand and respond adequately to the questionnaire should be
considered when wording questions.

• Questions must be carefully worded to ensure that the appropriate type of data is captured for the envisaged
statistical analyses.
Source: Wegner, T (1993) Applied Business Statistics, Kenwyn: Juta & Co, Ltd

32
 Click Here & Upgrade
Expanded Features
PDF Unlimited Pages
Documents
Complete

APPENDIX V
COMMONLY USED MEASURES OF ASSOCIATION

Measurement Coefficient Comments and Uses

Interval and Ratio Pearson (product moment) For continuous linearly related variables.
Correlation Coefficient
Correlation ratio (eta) For nonlinear data or relating a main effect to a
continuous dependent variable.
Biserial One continuous and one dichotomous variable with an
underlying normal distribution.
Partial correlation Three variables: relating two with the third’s effect
taken out.
Multiple correlation Three variables: relating one variable with two others.

Bivariate linear regression Predicting one variable from another’s scores.

Ordinal Gamma Based on concordant-discordant pairs: (P-Q);

proportional reduction in error (PRE) interpretation.
Kendall’s tau b P-Q based: adjustment for tied ranks.
Kendall’s tau c P-Q based: adjustment for table dimensions.

Sommer’s d P-Q based: asymmetrical extension of gamma.

Spearman’s rho Product moment correlation for ranked data.

Nominal Phi Chi-square (CS) based for 2 x 2 tables.

Cramer’s V CS based: adjustment when one table dimension > 2

Contingency coefficient C CS based: flexible data and distribution assumptions.

Lambda PRE-based interpretation.

Goodman & Kruskal’s tau PRE-based with table marginals emphasis.

Uncertainty coefficient Useful for multidimensional tables

Kappa Agreement measure.

33