LDP 603 Reserach Methods
LDP 603 Reserach Methods
LDP 603 Reserach Methods
In collaboration with
Authored by:
Page 1 of 241
GENERAL INTRODUCTION TO THE COURSE MODULE
The Research Methods course is one of the first semester core courses for those learners
pursuing the Master in Project Planning and Management course. You are aware that
any good decision is based on facts. Facts are based on data. The data must be
systematically collected, processed, analysed and presented for use. The best-known way
of collecting empirical data is through scientific research methods. This is what this
Distance learning demands a great deal of perseverance on the part of the students and
there are a number of challenges you have to contend with. One of them is that you have Comment [PU1]:
to cope with the study material in a different way. Other students can refer to their
professor for questions and information at all times. You will be mostly alone. However,
you can contact your teacher by phone, email or by physically visiting him or her.
Time management and planning is a major issue for ODL students who are probably less. Other
students also have a number of fellow students and build a network. Contacts with fellow
Page 2 of 241
students are important for moral support and study support.
This course is structured in form of lectures. In each lecture you will find some activities or
exercises that you should go through before starting the subsequent unit. It is advisable that you
answer all the questions and check the answers from the various sources i.e., the library, the
Internet or any other relevant source. You should also consult with the university lecturers or any
other qualified staff. As you might be aware, you should read other sources to supplement the
materials in this course unit.
.
This being a distance-learning course, you are expected to spare at least two hours everyday to
read the course unit. You are also advised to set aside some reading space equipped with a chair
and a reading table in you house or any other place you choose to study from.
The course will be delivered by distance –teaching mode. However, you will have at least
twenty hours of face-to-face sessions when you will meet with the course tutor. The Department
will communicate the venue and the timetable to you.
Continuous Assessment Test = 30%. This will be comprised of a timed test (15%) and an
assignment (15%)
The final examination = 70%
Total = 100%
The course will adopt the University of Nairobi Postgraduate courses grading system,
which is indicated here below:
It important that you familiarise yourself with the general regulations and rules that guide
Postgraduate Studies in the University of Nairobi which are available on-line via
www.uonbi.ac.ke.
It is my hope that you will find the course beneficial to your study and career aspirations.
In case you require some assistance do not hesitate to contact the course tutor or the
Department.
Page 3 of 241
Course Objectives
This course module aims at equipping the learner with the knowledge and skills to
handle research methodology issues in any type or size of an organization or institution.
It is expected that by the end of this course you should be able to specifically:
Page 4 of 241
Lecture Series: LDP 603- Research Methods
Printed by the College of Education and External Studies, University of Nairobi, P.O.
Box 30197, Nairobi, 2010
Page 1 of 241
LDP 601: RESEARCH METHODS
COURSE OUTLINE
Page 2 of 241
16. LECTURE SIXTEEN: ETHICS IN RESEARCH
17. LECTURE SEVENTEEN: FORMAT WRITING RESEARCH
PROPOSALS AND REPORTS
Page 3 of 241
LECTURE ONE
THE NATURE OF RESEARCH
Lecture outline
1.1 Introduction
1.2 Lecture objectives
1.3 Ways of knowing
1.4 Types of research
1.5 Lecture summary
1.6 Activity
1.7 Suggestion for further readings
1.1 Introduction
Research takes many forms. In this lecture, we introduce you to the subject of research
and explain why knowledge of various types of research can be of value to researchers.
Research is only one way through which we obtain knowledge; we look at several other
ways of knowing. We also briefly discuss several research methodologies used in
research.
_____________________________________________________________________
Page 4 of 241
By the end of this lecture, you should be able to:
Over the years, human beings have generated knowledge using various methods. In this
lecture, we are going to discuss the five methods that human beings have used to
generate knowledge. There are basically five ways of knowing. Let us look at each one
of them.
Page 1 of 241
1.3.2 Agreement with other
The opinion of other is another source of knowledge. We can share our sensations with
others and also check on the accuracy and authenticity of these sensations. The problem
with such common knowledge is that it can be wrong. Majority shows of hands in a
meeting is no guarantee that the issue being agree on is correct or hold truth. Two groups
of eyewitnesses to on an assault may disagree as to which person began the aggression..
Hence we would require considering additional sources of information to obtain reliable Comment [PU2]:
knowledge. Therefore, agreement with other people does not necessarily assure the truth
and collaboration of other sources may be required.
1.3.4 Logics
We also get to know by logical thinking. That is by our intellect- the capability we have
to reason things out. This allows us to use sensory data to develop a new kinds of
knowledge. For example, we can logically have the following statements:
Note that the first statement (called the major premise) comes fromwe need only from
Page 1 of 241
our experience about the majority of individuals. We have never experienced anyone
who was not mortal, so we state that all human beings are mortal.
The second d statement (called the minor premise) is based entirely on sensory
experience. If we come in contact with Juma and classify him as human beings we then
can deduce that the third statement (called the conclusion) must be true. Logics tell us it
is. As long as the first two statements are true the third statement must be also true.
The scientific method essentially involves the testing of ideas in the public arena. This
means that any knowledge generated in a scientific method must stand a critical scrutiny
of the public. The public includes other experts in the same field of knowledge, other
researchers and the general public for its truth. This is because,
aAlmost all human beings are capable of making connections- of seeing relationship and
associations. These connections are called “facts”. Facts are items of knowledge about
the world in which we live. inIn many cases we guess or speculate about the world
around us. To be sure that our guesses or speculation s are true, we need to 0[rigorously
test to see if they hold up under more controlled conditions. To investigate our
speculations, we can observe carefully and systematically the whole process of the
generation of the knowledge.
However such investigations do not constitutes science unless they are made public.
This means that all aspects of the investigations are described in sufficient details so that
Page 2 of 241
the study can be repeated by any who question the results. This basically boils down to
seven distinct steps.
1. There must be a problem to be investigated. This can be something bothering Formatted: Numbered + Level: 1 +
Numbering Style: 1, 2, 3, … + Start at: 1 +
youus or disturbing other peopleus. It may also be an unexplained discrepancy in Alignment: Left + Aligned at: 0.25" + Indent
at: 0.5"
a researcher’s field of knowledge, a gap in knowledge that needs to be closed.
2. The second step involves defining more precisely the problem or the question to
be answered, to be clear about exactly what the purpose of the study is.
3. In the third step we attempt to determine what kinds of information would solve
the problem.
4. The fourth step involves going to the field to collect the data
5. In the fFifth step, we must decide as far as possible, how we will organize the
information that we obtain.
6. In the sSixth step , and after the information you have has been collected and
analyzed, it must be interpreted.
7. In the sSeventh step , you we must write and present the report ofn our finding.
Formatted: Indent: Left: 1.29", No bullets or
numbering, Tab stops: 0.5", Left
Page 3 of 241
However, it is important to note the following:
______________________________________________________________ Formatted: Numbered + Level: 1 +
Numbering Style: 1, 2, 3, … + Start at: 1 +
___ Alignment: Left + Aligned at: 0.54" + Tab
after: 0.79" + Indent at: 0.79"
1. In many studies, there are several possible explanations for a problem or Formatted: Numbered + Level: 1 +
Numbering Style: 1, 2, 3, … + Start at: 1 +
phenomenon. These are called Hypothesis and may occur at any stager of an Alignment: Left + Aligned at: 0.54" + Tab
after: 0.79" + Indent at: 0.79", Tab stops:
investigation. 0.5", Left
Formatted: Font: Italic
2. There are two features of scientific research: freedom of thought and public
Formatted: Font: Italic
procedure. At every step, it is crucial that the researcher be as open as Formatted: Numbered + Level: 1 +
Numbering Style: 1, 2, 3, … + Start at: 1 +
humanly possible to alternatives, - in focusing and clarifying the problem, in Alignment: Left + Aligned at: 0.54" + Tab
after: 0.79" + Indent at: 0.79"
collecting and analysing information, and in interpreting results. The process
must be as public as possible. It is not a private game to be played by a group
of insiders. The value of scientific research is that it can be replicated (i.e.
repeated) by anyone interested in doing so.
3. The essence of all research originates in curiosity that is -a desire to find out
how and why things happen, including why people do the things they do, as
well as whether or not certain ways of doing things work better than other
ways.
4. A common misperception of science fosters the idea that there are fixed,
once-and-all answers to particular questions, which contributes to a common,
but unfortunate tendency to accept, and rigidly adhere to oversimplified
solutions to very complex problems.
_____________________________________________________________________
The term “research” can mean any sort of “careful, systematic, patient study and
investigation in some field of knowledge, undertaken to discover or establish facts and
principles”. In scientific research, however, the emphasis is on obtaining evidence to
support or refute proposed facts or principles. There are many methodologies that fit this
definition. Let us now look at the various types of research.
Page 4 of 241
This is the most conclusive of scientific methods. The researcher has two groups: the
experimental group and the control group. The researcher actually establishes different
treatments and then studies their effects. The ; results of this type are likely to lead to the
clear-cut interpretations. In this type of research, we have the control group and the
experimental group. The researcher will administer some treatment to the experimental
group while denying the control group, and then he/she sees the effect.
Another form of experimental research is the single-subject research which involves the
intensive study of a single individual (or sometimes a single group) overtime. These
designs are particularly appropriate when individuals with special characteristics are
studied by means of direct observation.
This is a type of research that is done to determine relationships among two or more
variables and to explore their implications for cause and effect. Correlation research
seeks to investigate whether one or more relationships of some type exist. For example:
wealth and family background; wealth and education. In this approach no manipulation
or intervention on the part of the researcher other than that required administering the
instrument(s) necessary to collect the data desired.
Page 5 of 241
In general, this type of research would be undertaken when one wants to look for and
describe relationships that may exist among naturally occurring phenomena, without
trying in any way to alter theses phenomena.
This type of research is intended to determine the cause for or the consequences of
different treatment between groups of people. Suppose a teacher wants to determine
whether students from single –parent families do more poorly in the class than students
from two-parent families. To conduct this investigation, the teacher would systematically
select two groups of students and then assign each a single parent or two-parent family-
which is clearly impossible (and unethical).
To test this issue using a causal-comparative design, the teacher might compare two
groups of students who already belong to one or the other type of family to see if they
differ.
However, interpretations of this type of research are limited because the researcher
cannot say conclusively whether a particular factor is a cause or a result of the
behaviour(s) observed. In our example above, the teacher could not be certain whether:
a) Any perceived difference in achievement between the two groups was due to the Formatted: Numbered + Level: 1 +
Numbering Style: a, b, c, … + Start at: 1 +
differences in home situation. Alignment: Left + Aligned at: 0.25" + Tab
after: 0.5" + Indent at: 0.5"
b) The parents’ status was due to the difference in achievement between the two
groups (though this seems likely).
c) Some unidentified factor was at work.
Page 6 of 241
1.4 .4 Survey research
This is a type of research used to obtain data that can help determine specific
characteristics of a group. A descriptive survey involves asking questions (often in the
form of a questionnaire) toof a large group of individuals either by mail, by telephone or
in person. When answers to a set of question are solicited in person, the data collection
methodresearch is called an iInterview.
The main difficulties involved in survey research are mainly the following:
1. Ensuring that the questions to be answered are clear and not misleading Formatted: Numbered + Level: 1 +
Numbering Style: 1, 2, 3, … + Start at: 1 +
2. Getting respondents to answer questions thoughtfully and honestly Alignment: Left + Aligned at: 0.25" + Tab
after: 0.5" + Indent at: 0.5"
3. Getting a sufficient number of the questionnaires completed and returned so that
meaningful analyses can be made.
The main advantage of survey research is that it has the potential to provide us with a lot
of information obtained from quite a large sample of individuals.
While most such studies involve an analysis of written documents, some is conducted
using films, folk songs, ancient pottery etc. The method is applicable to any material that
does not come pre-organized for the researcher’s purpose. The major task of the
researcher is to locate appropriate materials and then find a way to analyze them.
Page 7 of 241
Qualitative involves obtaining a holistic picture of what goes on in a particular situation
or setting. There are two categories:
a) Ethnographic study: the emphasis in this type of research is on Formatted: Numbered + Level: 1 +
Numbering Style: a, b, c, … + Start at: 1 +
documenting or portraying the everyday experiences of individuals by observing Alignment: Left + Aligned at: 0.25" + Indent
at: 0.5"
and interviewing them and relevant others. For example, a researcher may want
to study the behaviour of students in a lecture rooman elementary classroom. This
can be done by observing on a regular basis, and also interviewing the teacher
and the student in an attempt to describe as fully and as richly as possible what
goes on in that lecture classroom. The data could include detailed prose
description by students of lecture classroom activities, audiotapes of classroom
discussions, examples of teacher lesson plans and students work, sociograms
depicting “power” relationships in the lecture roomclass and flows charts
illustrating the direction and frequency of certain types of comments.
b) Case studies: this is a well-detailed study of one or a few individuals or Formatted: Numbered + Level: 1 +
Numbering Style: a, b, c, … + Start at: 1 +
situation. The focus is to study the unit of analysis( the individual,the institution Alignment: Left + Aligned at: 0.25" + Indent
at: 0.5"
or the situation) in a holistic perspective.
In historical research, some aspects of the past is studied, either by perusing documents
of the period or by interviewing individuals who lived during the time. An attempt is
then made to reconstruct, as accurately as possible, what happened during that time to
explain why it did happen.
Page 8 of 241
1.5 Lecture summary
There are several ways in which knowledge is generated. They include: through
sensory organs; agreement with others; expert opinion; logics and the scientific
method.
The term “research” can mean any sort of “careful, systematic, patient study and
investigation in some field of knowledge, undertaken to discover or establish facts
and principles”. In scientific research, the emphasis is on obtaining evidence to
support or refute proposed facts or principles. There are several types of research
which include; experimental, correlation, survey, causal-comparative,content analysis,
historical and qualitative research.
1.6 Activity
Page 9 of 241
LECTURE TWO
PHILOSOPHICAL FOUNDATION OF RESEARCH
Lecture outline
2.1 Introduction
2.2 Lecture objectives
2.3 Schools of thought on research theory
2.4 Ethics in research
2.5 Deception in research
2.6 Lecture summary
2.7 Activity
2.8 Suggestion for further reading
2.1 Introduction
Page 10 of 241
There are three major schools of thought which underpin research theory. They are:
a. The positivism/ post positivism paradigm
b. The interpretive/ constructivism paradigm
c. The emancipatory paradigm.
Each of the schools has their distinct way of explaining what research is, its methodology
and its processes. We are now going to look at each of the four schools of thought.
The positivism and the post-positivism school of thought have guided much of the
research particularly in psychology and education. Positivism is based on the rationalistic
euphemistic philosophy that has originated with Aristotle, Francis Bacon, John Locke
August Comte and Emmanuel Kant. The underlying assumptions of positivism are:
The belief that the social world can be studied in the same way as the natural
world.
That is there is a method for studying the social world that is value-free and
That explanation of a casual nature can be provided.
This paradigm was in practise before the Second World War when it was replaced by
post positivism.
Ontology: The positivism hold that one reality exists and that it is the researcher’s job to
discover that reality (naïve realism) Guba & Lincoln, 1994). The positivists concur that a
Page 11 of 241
reality does exist but it can be known only imperfectly because of the researcher’s
human limitations (critical realism). Therefore, researchers can discover “reality” within
certain realms of probability. However, they cannot “prove” a theory, but they can make
a stronger case by eliminating alternative explanations (Reichardt & Ralli, 1994)
Epistemology: Positivists assume that the researcher and the subject of the study were
independent and that they did not influence each other (Guba & Lincoln, 1994). The post
positivists modified this belief by recognizing that the theories, hypothesis and
background knowledge held by the investigator can strongly influence what is observed
(( Reichardt & Ralli, 1994). The positivists hold that a researcher should strive to achieve
objectivity in research by remaining neutral to prevent values and biases from
influencing the work by following prescribed procedures rigorously.
Methodology: The positivists borrowed their experimental methods from the natural
sciences. The post positivists recognized that many of the assumptions required for
rigorous application of the scientific methods were not appropriate when studying
people. Therefore, quasi-experimental methods were needed. In other words, many times
it is difficult to randomly assign subjects to conditions (i.e. a plot of land for study of
fertilizer). In this case the researcher needs to devise modifications to the experimental
methods of the natural sciences in order to apply them to people.
This school of thought holds that reality is socially constructed. This paradigm grew out
of the philosophy of Edmund Husserl’s phenomenology and white Dilthey study of
interpretive understanding called hermeneutics (Eichelberger, 1989). Hermeneutics is the
study of interpretive understanding or meaning.
Interpretive/ constructivist researchers use the term more generally to interpret the
meaning of something from a certain standpoint situation.
The basic assumptions of this paradigm are: that knowledge is socially constructed by
people active in the research process researchers should attempt to understand the
“complex” world of lived experience from the point of view of those who live it”(
Schwandt,1994,p.118).
Page 12 of 241
They emphasize that research is a product of the values of researchers and cannot be
independent of them.
Ontology: They belief that reality is socially constructed. Therefore, multiple mental
constructions can be apprehended, some of which conflict with each other and
perceptions of reality may change throughout the process of the study. For example, the
term gender is socially constructed phenomena that mean different things to different
people.
Epistemology: the researcher and the research itself are interlocked in an interactive
process; each influences the other. This school of thought therefore opts for a more
personal, interactive mode of data collection.
Page 13 of 241
college students’ development.
The school-age population is becoming poorer and more racially and ethnic diverse. This
has contributed to the increased interest in multicultural education and ways to conduct
race-sensitive research
Some ethnic-minority psychologists believe that white researchers who study other
communities do so without an understanding or caring for the people who live
there.(Mio& Iwamasa,1993).
That research is conducted without due consideration of the disadvantaged people like
the disabled, hence ignoring genetic and biological factors.
Need for more culturally sensitive research.
A need for informed practitioners to form partnerships with researchers to plan and
conduct research and evaluation studies in a meaningful way.
Page 14 of 241
politics.
Epistemology: The relationship between the knower and the would-be known (i.e., the
researcher and participant) is viewed as interactive. According to Harding (1993), the
researcher should use a methodology that involves” starting off thought” from the lives
of marginalized people. This would reveal more of the unexamined assumptions
influencing science and generate more critical questions.
The relationship should be empowering to those without power. Thus, the research
should examine ways the research benefits or does not benefit the participants (Kelly
etal., 1994).
Ethics in research should be an integral part of the research planning and implementation
process, not viewed as an afterthought or a burden. There should be increased
consciousness of the need for strict ethical guidelines for researchers. Some of the ethical
issues touch on deception and invasion of privacy.
There are three main ethical principles that need to be considered:
Beneficence: Maximizing good outcomes for science, humanity, and the
individual research participants and minimising or avoiding unnecessary risk,
Page 15 of 241
harm, or wrong.
Respect: Treating people with respect and courtesy, including those who are
not autonomous (e.g., small children, people who have mental retardation or
senility)
Justice: Ensuring that those who bear the risk in the research are those who
benefit from it; ensuring that the procedures are reasonable, non-exploitative,
carefully considered and fairly administered.
Most professional associations prohibit the use of deception unless it can be justified and
the effect of the deception “undone” after the study is completed. The “undoing” of
deception is supposed to be accomplished by the following:
Debriefing the research participants after the research study, which means
that the research explains the real purpose and use of the research
Dehoaxing the research participants in which the researcher demonstrates the
Page 16 of 241
device that was used to deceive the participants. The researcher’s
responsibility is to attempt to allay a sense of generalized mistrust in
educational and psychological research.
Guarding the privacy and confidentiality of the research participants
obtaining fully informed consent.
You will note that the emancipatory paradigm emerged because of the dissatisfaction
with research conducted within other paradigms that was perceived to be irrelevant to, or
a misrepresentation of, the lives of people who experience oppression. There are three
characteristics of the emancipatory paradigm with ethical implications for
methodological choices:
Traditionally silenced voices must be included to ensure that the groups
marginalized in society are equally “heard” during the research process and
the formation of the findings and recommendations.
An analysis of power inequalities in terms of the social relationships involved
in the planning, implementation, and reporting of the research is needed to
ensure an equitable distribution of resources (conceptual and material)
A mechanism should be identified to enable the research results to be linked
to social action: those who are most oppressed and least powerful should be at
the canter of the plans for action in order to empower them to change their
own lives.
When the research is cross-cultural, it is important that cross-cultural ethical standards
are developed to guide researchers while conducting research in other communities.
Cross-cultural ethical principles require collaboration between the researcher and the
host community. It also requires that the researcher communicate the intended research
agenda, design, activity, and reports with members of the host community. The research
should be designed in such a way as to bring benefits to the host community and to foster
the skills and slf-sufficiency of the host community scientists.
The paradigms considered here are certainly not exhaustive. New paradigms might come
in the future. However, what is crucial is that researchers should be aware of their basic
beliefs, their view of the world (their functional paradigm), and the way they influence
their approach to research.
Page 17 of 241
2.6 Lecture summary
In this lecture we looked at the various philosophies that have guided research
overtime. We have stated that there are basically three schools: positivist/post
positivists; the emancipatory/ contructivism ; and the entrepretivism
2.7 Activity
1. For each of the three schools of thought on research, highlight their
main epistemological, ontological and methodological underpinnings
2. Which school appeals to you most and why?
Page 18 of 241
LECTURE THREE
VARIABLES, CONCEPTS AND CONSTRUCTS
Lecture outline
3.1 Introduction
3.2 Lecture Objectives
3.3 Concepts
3.4 Variables
3.5 Hypothesis
3.6 Theory
3.7 Models
3.8 Summary
3.9 Activity
3.10 Suggestion for further reading
3.1 Introduction
In research the words concepts constructs or variables are very important terms and need
to be understood clearly. Concepts and constructs are broad terms applied and used in
academics and research. In this lecturer we are going to define and give a general
understanding of the terms Concepts and Constructs from various scholars' point of
view. Outlining the functions of concepts is also key to further understanding these terms
in research and their relevance in research. Then we will go ahead to give the difference
between conceptual and operational definitions of concepts by giving relevant and
practical examples to demonstrate these two aspects.
_____________________________________________________________________
Page 19 of 241
3.2 Lecture Objectives.
By the end of this topic, you should be able to:
1. Explain what is meant by the term “variable”
2. Distinguish between the various types of variables
3. Explain how independent and dependent variables are related
4. Explain what a hypothesis is and formulate them.
5. Explain what is meant by the term “scientific method”
6. Name the advantages and disadvantages of stating research
questions as hypothesis
7. Distinguish between directional and non-directional
hypothesis
8. Define the term “ theory” and explain the role of theory in
research
9. Explain the functions of a “model” in research.
3.3 Concepts
It is common knowledge that we need to notice something before explaining what it is.
For example we see a dog first and then we are able to describe the dog in details. In this
case we have an idea (concept) of the phenomenon before it is explained.
He says that construct is a concept which has an added meaning of having been
deliberately and consciously invented or adopted for a special scientific purpose. He
explains saying that for example "intelligence" is a concept, an abstraction from the
observation of presumably intelligent and non intelligent behaviours. But as a scientific
Page 20 of 241
construct "intelligence" means both more or less than it may mean as a concept.
It means that scientists consciously and systematically use it in two ways. One, it enters
into theoretical schemes and is related in various ways to other constructs. We say for
example, that school achievement is in part a function of intelligence and motivation.
Two, intelligence is so defined and specified that it can be observed and measured. We
can make observations of the intelligence of children by administering some intelligence
test to them or we can ask teachers to tell us the relative degrees of intelligence of their
pupils.
A concept from the point of view of documented Wikipedia authors is an abstract idea or
a mental symbol, typically associated with a corresponding representation in language or
symbology, that denotes all of the objects in a given category or class of entities,
interactions, phenomena, or relationships between them. Concepts are abstract in that
they omit the differences of the things in their extension, treating them as if they were
identical. They are universal in that they apply equally to every thing in their extension.
Concepts are also the basic elements of propositions much the same way a word is the
basic semantic element of a sentence. Unlike perceptions which are particular images of
individual objects concepts cannot be visualized. Because they are not, themselves,
individual perceptions concepts are discursive and result from reason. They can only be
thought and designated by a name.
Concepts are bearers of meaning, as opposed to agents of meaning. A single concept can
be expressed by any number of languages. The concept of DOG can be expressed as dog
in English Hund in German as chzen in French perro in Spanish and mbwa in kiswahili.
The fact that concepts are in some sense independent of language makes translation
possible - words in various languages have identical meaning, because they express one
and the same concept.
Page 21 of 241
convey and transmit information. Concepts do not actually exist as empirical
phenomena/ but rather are symbols of phenomena/ not phenomena themselves.
Concepts introduce a perspective is a way of looking at empirical phenomena.
As Norman Denzin (1989) puts it "Through Scientific conceptualization, the
Perceptual World is given an order and coherence that could not be perceived
before conceptualization". The concept therefore enables scientists to relate to
some aspect of reality and identify it as a quality common to different examples of
the Phenomenon in the real world.
Concepts allow scientists to classify and generalize. That is, scientists structure,
categorize, order and generalize their experiences and observations in terms of concepts.
As John McKinney (19660 puts it, "All Phenomena are unique in their concrete
occurrence, therefore no phenomena actually recur in their concrete wholeness. To
introduce order with its various scientific implications, including prediction, the scientist
necessarily ignores the unique, the extraneous and the non-recurring and thereby departs
from perceptual experience.
In a nutshell, the four functions of concepts are:
Concepts provide a common language which enables scientists to
communicate with one another.
Concepts give scientists a perspective - a way of looking at
phenomena.
Concepts allow scientists to classify their experiences and to generalize
from them.
Concepts are components of theories - they define a theory's content and
attributes.
However, science cannot progress with ambiguous and imprecise language. Social
scientists have attempted to establish a clear and precise body of concepts to characterize
Page 22 of 241
their subject matter. To achieve clarity and precision in the use of concepts during
research, scientists employ two major types of definitions/ that is, conceptual and
operational definitions.
Defining a concept is not very different form from defining any word. The reason for
defining a concept is to make it very clear to some audience. A good conceptual
definition is that one that clearly distinguishes the properties or characteristics of a
concept from other concepts.
A conceptual definition defines a concept with other concepts. For instance/ we can
define weight by saying that it is the heaviness of objects. Or we can define anxiety as
subjectified fear. In this case we have substituted one concept for another.
Definitions have two segments: how the concept is similar to other concepts and how it
differs from them. An example is of a cat as an animal/ but unlike other animals it
"meows".
Another example is term “income”. Income can be defined as the money one gets after
engaging in an economically productive activity. A group of people may be involved in
different economic activities from which they get income. Some are farmers, other
traders, others civil servants etc. But at the end of the day each gets an income.
Better definitions are those ones that are more useful and used more often in scientific
theories and in research questions. Some concepts are also distinguished from other
concepts through the use of examples or analogy. Although a concepts name, such as
"self-esteem" may have wide usage in everyday use, it is generally a different concept
from the one that is carefully defined as technical term is scientific field. This is
important because by using an accepted technical term then a concept is given a single
definition.
A good conceptual definition not only expresses how the phenomenon is similar and
different from other concepts but also provides insight into the kind of variability one
might expect to find. For example the definition of family income would suggest that
Page 23 of 241
income could range from low to high and could be captured in some monetary unit such
as Kenya shillings and fractions thereof .It would indicate that the range is continuous.
These features of conceptual definition specify the kind of relationships between
categories that are envisioned that are referred to as level of measurement of the concept.
Operational definitions are also used to define system states of a specific, publicly
accessible process of validation testing which is repeatable at will. For example 100
degrees celsius may be crudely defined by describing the process of heating water until it
is observed to boil. An item like a brick may be defined in terms of how it is prepared
and baked (recipe is its operational definition).
Given the usual definition of gender as the sex role with which one identifies man or a
Page 24 of 241
woman one could determine a person's gender by asking what gender are the person is
and writing down the response. Usually there are many potential indicators of the same
concept. For example one could observe a person's dress, the form of address the person
prefers( Mr./ Ms./ Miss). Each of these indicators provides a basis for classifying the
person as man or woman. But it is important to note that not all these indicators of
concept “gender” are likely to be good measures in all situations.
One primary way in which operational definitions vary is in the extent to which they are
able to capture the concept the investigator has defined. This correspondence is called
“validity” of the operational definition or the extent to which it actually measures the
concept it is intended to measure. The temperature of a room is not a valid measure of
the room's ceiling height but on the other hand the answer to the question "what is your
gender?" is a valid measure of concept gender, assuming that the respondent also
understands the question in the way that investigator expects.
3.4 Variables
Most research studies involve looking for relationships among variables. The concept of
“variables” is one of the most important concepts in research. In this topic we look at
several kinds of variables. We will also discuss the concept of “hypothesis”. Hypotheses
express relationships between variables and they are based directly on the research
questions.
A variable is an empirical property that can take on two or more values. Any property
that can change, either in quantity or quality can be regarded as a variable. Foe example,
Page 25 of 241
the term “student” in the university is a variable because it can be differentiated into
several distinct values, for example, the academic year they are enrolled in. For example,
a student can be in first year, second year, third year fourth year, postgraduate, or
undergraduate etc. A students can take any one of the four academic years (variables, in
this case)
There are several ways in which we can categorise variables. Let us look at the
categories.
(a) Dichotomous versus discrete variables.
A dichotomous variable have only two values reflecting the presence or absence of a
property. For example, a male or a female; employed or unemployed; dead or alive. On
the other hand, discrete variables are those characteristics that take only one value. That
is a variable that takes only one value for example a person can have one religion say
Christianity or Islam. There is no situation of having half of it.
Page 26 of 241
( c) Moderating variables, Extraneous variables and Intervening variables.
An example:
Assume that you are to measure a person’s income. The first question that comes in
your mind is the variables that could adequately measure that person’s income.
For example if you only derive your income from a salary. Then it become the only
source of your income. However, the real income you get at the end of the month
will be determined by the your salary scale, the taxes you pay, your pension ,
bonuses and allowances, contribution to your cooperative etc.
Page 27 of 241
Salary= salary scale+ allowance+ taxes-pension- contribution to cooperative society
How do the CEOs of large companies feel about training their staff?
How does Kenyans feel about their economy?
In the two examples above, the researcher simply wants to identify the feeling of the
respondents and not the relationships between the respondent’s feelings with anything
else.
The main problem with purely descriptive research questions is that answers to them do
not help us understand why people feel or think or behave in a certain way. As a result
our understanding of a situation, group, or phenomenon is usually limited.
Page 28 of 241
3.5 Hypothesis
An example:
Research question: Does training staff in change management help staff to cope with
change in their organizations?
Hypothesis: Staff trained in change management cope easily with change in their
organizations. This hypothesis predicts that the staffs that undergo training in change
management can easily cope with changes that might occur in their organizations than
perhaps the staff that have not been trained in change management.
Activity:
Identify the dependent and independent variables from the research question given in the research question
given above. Note that we can formulate many different hypotheses from a give question.
Page 29 of 241
3.5.1 Characteristics of a good hypothesis
3.5.2 What are the advantages and disadvantages of stating hypothesis in research?
The following are the advantages and disadvantages of stating hypothesis in research:
Advantages
A hypothesis forces us to think more deeply and specifically about the
possible outcomes of a study. It enables us to understand what the question
implies and exactly what variables are involved.
If one is attempting to build a body of knowledge in addition to answering a
specific question, then stating hypothesis is a good strategy because it enables
one to make specific predictions based on prior evidence or theoretical
argument.
Hypothesis stating helps us to see if we are or are not investigating a
relationship.
Page 30 of 241
Disadvantages
Stating a hypothesis may lead to a bias, either consciously or unconsciously,
on the part of the researcher. This is because the researcher may be tempted to
arrange the procedures or manipulate the data in such a way as to bring about
a desired outcome. This depends on the honesty of the researchers.
Stating hypothesis may sometimes be unnecessary, or even inappropriate, in
certain research projects of certain types i.e., descriptive or ethnographic
studies.
Stating hypothesis may prevent researchers from noticing other phenomena
that might be important to study.
A directional hypothesis is one in which the specific direction (such as higher, lower,
more or less) that a researcher expects to emerge in a relationship is indicated. The
particular direction expected is based on what the researcher has found in the literature,
from personal experience, or from the experience of others.
Non-directional hypothesis on the other hand does not make a specific prediction about
what direction the outcome of a study will take.
3.6 Theory
Page 31 of 241
variables, while hypothesis tend to be simple, two-variable propositions involving
concrete instances.
3.7 Models
Models differ from theories in that the role of a theory is to offer explanation while a
model’s role is of representation.
A model represents a structure of something. For example a researcher is expected to
develop a conceptual model, which structurally describes the relationship between the
variables of the study.
3.8 Summary
In this lecture we have discussed about concepts, variables, models and theories.
We have said that:
A concept is an abstract idea or a mental symbol, typically associated
with a corresponding representation in language or symbology, that
denotes all of the objects in a given category or class of entities,
interactions, phenomena, or relationships between them.
A variable is an empirical property that can take on two or more values.
Page 32 of 241
Any property that can change, either in quantity or quality can be
regarded as a variable.
A hypothesis is a prediction of some sort regarding the possible outcomes
of a study.
A model is defined as a representation of a system that is constructed to
study some aspects of that system as a whole.
Page 33 of 241
3.9 Activity.
Now that you have understood what hypothesis are attempt to provide some answers
to the following question in respect to a research you expect to undertake.
Page 34 of 241
Check whether there are extraneous variables that might affect your results. List
them down---------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
---------------------------------------------------------------------------
3.10 Suggestion for further research
Page 35 of 241
LECTURE FOUR
MEASUREMENT OF VARIABLES, VALIDITY AND RELIABILITY
Lecture outline
4.1 Introduction
4.2 Lecture objectives
4.3 Defining the term “Measurements”
4.4 Levels of measurements
4.5Measurement errors
4.6 Scaling techniques
4.7 Validity and reliability
4.8 Lecture summary
4.9 Self- evaluation test
4.10 Activity
4.11. Suggestion for further reading
4.1 Introduction
In your daily life, you carry out some measure when you use some yardstick to
determine weight, height, length, and time of any other feature of an object. You also
measure when you judge how well you like a person, a song, a place or an academic
course. We, therefore, measure physical objects as well as abstract concepts. We need to
appreciate that measurement is a relatively complex and demanding task, especially
when it involves qualitative or abstract phenomena. In this lecture, we will look at the
various measurement and scaling techniques a researcher can use.
_____________________________________________________________________
Page 36 of 241
4.2 Lecture objectives.
By the end of this lecture, you should be able to:
1. Define the term “ measurement” as used in research
2. Explain the various measurement scales
3. Discuss the sources of error in measurement
4. Define the term “scaling” as used in research.
5. Discuss the various scaling techniques.
6. Explain what is meant by the term “validity” in
research
7. Name the three types of validity in research
8. Explain what is meant by the term “reliability”.
9. Describe three ways to estimate the reliability of
the scores obtained using a particular instrument.
_____________________________________________________________________
Page 37 of 241
In measuring a property of an object, we devise some form of scale in the range (in terms
of set theory) and then transform or map the property of the objects from the domain.
Rules are the most significant component of the measurement procedure because they
determine the quality of measurement. Poor rules make measurement meaningless.
Measurement is meaningless when it is not tied to reality, when it lacks an empirical
basis. The functions of rules are to tie the measurement procedure to reality.
Therefore, we can say that indicators are specified by operational definitions. After a
researcher observes the indicators, they substitute numerals or numbers for the value of
the indicator and perform quantitative analyses.
4.4.1What is a scale?
A scale may be thought of as a tool for measuring. The most widely used classification
of measurement scales are; nominal, ordinal, interval and ratio. Let us discuss each of
them.
Nominal scales are the lowest level of measurement. Nominal scale is simply a system of
assigning symbols to events in order to label them. The numbers assigned to an object is
Page 38 of 241
only a symbol. For instance, we can use numbers “1” and “2” to represent male and
female respectively. As a rule, we should the categories should be exhaustive (that is,
with no case that include all cases of that type) and mutually exclusive (that no case can
be classified as belonging to more than one category). The numbers are just convenient
labels for the particular class of events and as such have no quantitative value.
Nominal scale is the least powerful level of measurement. It does not indicate order or
distance relationship and has no arithmetic origin. It simply describes differences
between things by assigning them to categories. The scale wastes all the information that
may have about varying degrees of the variable. The main statistics used for nominal
scale are the mode, measures of qualitative variation and appropriate measures of
association. Chi-square test is the most common test of statistical significance. Fore
measures of correlation the contingency coefficient can be worked out.
This is a level of measurement that shows the relative importance of variables in order of
magnitude, size and preferences. Ordinal scale emphasizes order, which is expressed in
degree of quality. The typical relations are, “ higher”, “ greater”, “More desired” and so
on. In most cases, ordinal scales indicate rank order.
This level of measurement is used where particular data and information collected has
quantifiable magnitude such as population size, weight and distances, which are
measured against an established criteria or standard. Examples of such measurements
include year calendar, temperature, time, and test scores.
This is the highest level of measurement that entails expressing the number of persons,
Page 39 of 241
and other attributes such as proportions of the total population. It is a scale that possesses
an actual, or zero point. Variables such as weight, time, length, and area have natural
zero points and are measured at the ratio level.
Any good scientific study should be precise and unambiguous. However, some errors
can occur in the process of measurements. There are four main sources of measurement
errors. They are:
It is therefore important for the researcher to ensure that they meet all the problems listed
above.
Page 40 of 241
4.6 Scaling Techniques
Their greatest disadvantage is that we do not have objective evidence that such scales
measure the concepts for which they have been developed. They rely on the researcher’s
insight and competence.
Under this approach, the selection of items is made by a panel of judges who evaluate the
items in terms of whether they are relevant to the topic area and unambiguous in
implication. The procedure entails the following:
Page 41 of 241
that express various points of view towards a group, institution, idea or
practice( i.e., statements belonging to the topic area).
These statements are then submitted to a panel of judges, each of who
arranges them in eleven groups or piles ranging from one extreme to another
in position. Each of the judges is requested to place generally in the first pile
the statement which he thinks are most unfavourable to the issue, in the
second pile to place those statements which he thinks are next most
unfavourable and he goes on doing so in this manner till in the eleventh pile
he puts the statements which he considers to be the most favourable.
This sorting by each judge yields a composite position for each of the items.
In case of marked disagreement between the judges in assigning a position to
an item, what item is discarded?
Fri items that are retained, each is given its median scale value between one
and eleven as established by the panel. That is the scale value of any
statement is computed as the median position to which it is assigned by the
group of judges.
A final selection of statements is then made. For this purpose, a sample of
statements whose median scores are spread evenly from one extreme to the
other is taken. The statement so selected, constitute the final scale to be
administered to respondents. The position of each statement on the scale is
the same as determined by the judges.
The Thurstone method has been used widely for developing differential scales which are
utilised to measure attitudes towards varied issues like war, religion etc. However, they
are difficult to develop and also expensive. The method is not completely objective; it
involves ultimately subjective decision process.
These are scales that are developed by utilizing the item analysis approach wherein a
particular item is evaluated on the basis of how well it discriminates between those
persons whose total scores is high and those scores is low. Those items or statements that
Page 42 of 241
best meet this sort of discrimination test are included in the final instrument.
4.7.1 Validity
Before you use a research instrument you must ensure that it has some validity. Validity
is the most important idea to consider when preparing or selecting an instrument for use.
For example, if a project manager want to know whether the people support the project
or not he or she will need an instrument to record the data and some sort of assurance
that the information obtained will enable him or her to draw the correct conclusions
about the peoples feelings or opinions. The process of drawing the correct conclusions
based on the data obtained from an assessment is what validity is all about.
Validation on the other hand, is the process of collecting evidence to support the
inference made. What is important to us is to realize that validity refers to the degree to
which evidence supports any inferences a researcher makes based on the data he or she
Page 43 of 241
collects using a particular instrument. It is important for us to note that it is the inferences
about the specific uses of an instrument that are validated, and not the instrument itself.
Therefore, the inferences made should be appropriate, meaningful and useful.
An “appropriate inference” is one that is relevant to the purpose of the study. For
example, if the purpose of a study was to determine what people know about the
importance of a project, it would make no sense to make inferences about this from a test
score about the most popular politician in the area.
A “meaningful inference”: A meaningful inference is one that says something about the
meaning of the information obtained through the use of the instrument. For example, if
you say that a person attitude towards something is high? What exactly does a positive
attitude score mean? What does such a positive attitude score allow us to say about an
individual who has it? In what ways is that individual different from one who receives a
negative attitude score? The important thing to remember is that the purpose of research
is not merely to collect data, but to use such data to draw warranted conclusions about a
people or a situation on which the data were collected.
Validity, therefore, depends on the amount and type of evidence there is to support the
interpretations the researchers wish to make concerning data they have collected. The
most important question we should ask ourselves as researchers is whether the results of
the assessment provided useful information about the research questions or the variables
being measured.
Page 44 of 241
There are three types of validity that are of interest to us as researchers. They are:
content-related evidence of validity, the criterion-related evidence of validity and the
construct –related evidence of validity. Let us briefly look at each one of them.
i)Content-related validity: this type of validity refers to the content and format of the
instrument. The mains questions a researcher should ask are:
How appropriate is the content of the instrument to the purpose of the study?
How comprehensive is the content in measuring all the constructs of the
variable being measured?
Does the content logically get at the intended variable?
How adequate does the sample of items or questions represent the content to
be measured?
Is the instrument format appropriate?
A researcher needs to provide answers to these questions before using the instrument to
collect data.
ii)Criterion-related validity: this refers to the relationship between scores obtained using
an instrument and scores obtained using one or more other instruments or measures. It is
expressed as the coefficient of correlation between test scores and some measure of
future performance or between test scores and scores on another measure of known
validity. What is important is to ask ourselves how well the scores estimate present or
predict future occurrences. The criterion validity must possess the following qualities:
Relevance: a criterion is relevant if it is defined in terms of what we judge to
be the proper measures.
Freedom from bias: a criterion is said to be free from bias if it gives each
subject an equal opportunity to score
Reliability: a reliable criterion is stable or reproducible
Availability: the information specified by the criterion must be available.
iii) Construct Validity: this refers to the nature of the psychological construct or
characteristic being measured. A measure is said to possess construct validity to the
degree that it conforms to predicted correlations with other theoretical propositions. It
measures the degree to which scores on a test can be accounted for by the explanatory
Page 45 of 241
construct of sound theory. In this case, we associate a set of other propositions with the
results received from using our measurement instrument. If the measurements on our
devised scale correlate in a predicated way with these other propositions, we conclude
that there is some construct validity.
4.7.3 Reliability
Reliability is another important measurement in research. Reliability refers to the
consistence of the scores obtained. That is how consistent the scores are for each
individual from one administration of an instrument to another and from one item to
another. It is important to note that reliable measuring instruments do contribute to
validity, but a reliable instrument needs to be a valid instrument. For example, a
measuring scale that consistently under weighs an object by one kilo is a reliable scale
but it is not a valid measure of weight!
Equivalency is the measure of how much error gets introduced by different investigators
or different samples of the items being studied. A good way to test for the equivalency of
measurement by two researchers is to compare their observations of the same events.
By standardising the conditions under which the measurements takes place. That
is by ensuring that external sources of variation to the measure are minimised
By carefully designing directions for measurement with no variation from group
to group, by using trained and motivated persons to conduct the research and also
by broadening the sample of items used. The aim here is to improve equivalency
Just like many situations in life, errors of measurement can occur in research. If an
Page 46 of 241
instrument is administered to the same group more than once, or when two different
forms of an instrument are used there is bound to be variation in the test score.
Researchers should strive to ensure that their instruments are reliable. They can do so by
calculating the reliability coefficient of the instruments. A reliability coefficient
expresses a relationship between scores of the same individual on the same instrument at
two different times or between two parts of the same instrument. There are three best –
known ways to obtain a reliability coefficient. They are: the test-retest method, the
equivalent-forms method; and the internal consistency methods. Let us briefly look at
how they are used in research.
i) Test-retest method
This method involves administering the same test twice to the same group after a certain
time interval has elapsed since the previous test. A reliability coefficient is then
calculated to indicate the relationship between the two sets of scores obtained.
Note that this coefficient will be affected by the length of time that elapses between the
two administrations of the test. The longer the time interval, the lower the reliability
coefficient is likely to be since there is greater likelihood of changes in the individuals
taking the test. However, the variable being tested should have some level of stability
within a given period of time.
This involves giving two different but equivalent forms of an instrument to the same
group of people or research object during the same time period. Although the items
(questions) are different, they should sample the same content and they should be
constructed separately from each other. A reliability coefficient should be calculated
between the two sets of scores obtained. A high coefficient would indicate strong
evidence of reliability. This would imply that the two forms are measuring the same
thing.
The two method so far considered (i.e., the test-retest and the equivalent methods)
Page 47 of 241
require two administration or testing sessions. However, there are other methods of
estimating reliability which requires only a single administration of an instrument. They
are; the split-half method, the Kuder- Richardson approaches and the alpha coefficient
method. Let us discuss each one of them separately.
This involves scoring two-halves of a test separately for each person and then calculating
a correlation coefficient for the two sets of scores. In most cases researchers will split the
instrument into the odd items and the even items. The resulting coefficient indicates the
degree to which the two halves of the test provide the same results, and hence describes
the internal consistency of the test.
It is possible to increase reliability by increasing its length if the items added are similar
to the original ones.
This is the most frequently used method by researchers for determining internal
consistency. It uses two formulas, the KR20 and KR 21. KR20 formula requires three
types of information: the number of items in the test, the mean, and the standard
deviation. It is important to note that this formula can only be used if we assume that the
items are of equal difficulty.
As you are aware by now, this is a coefficient and that a coefficient value of .00 indicates
Page 48 of 241
a complete absence of a relationship and hence no reliability at all. A coefficient of 1.00
on the other hand indicates a complete relationship. For research purposes, the rule of
thumb is that the reliability should be at least.70 and preferably higher.
vi)The Alpha coefficient (Cronbach alpha): This is a general form KR20 formula and it
is used to calculating reliability of items that are not scored right versus wrong.
4.8 Summary
We have discussed that validity as used in research refers to the appropriateness,
meaningfulness and usefulness of any inferences a researcher draws based on data
obtained through the use of an instrument. There are three types of validity: the content-
related validity, the criterion-related validity and the construct-related validity. On the
other hand, “reliability” as used in research refers to the consistency of scores or answers
provided by an instrument. There are three methods of estimating reliability: the test-
retest method, the equivalent forms method and the internal –consistency methods.
Whatever method a researcher decides to use, he or she must ensure that the results
represent the true picture of the situation.
Page 49 of 241
4.9 Self-Evaluation Test
In each of the statements presented here below, indicate what type of evidence
(Validity) would better represent the statement.
Ninety five of the respondents who scored high on an attitude test.
Do you think there is a relationship between reliability and validity in research?
Discuss the three types of reliability and give example in what research situations
you would use each of them.
What are the main sources of measurement errors in research?
Page 50 of 241
4.10 Activity
1.In my research project , I will use the following existing instruments--------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
--------------------------------The following is a summary of what I have learnt
about validity and reliability of the scores obtained with these instruments------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
------------------------------------------------------------------------------
2.If you are not going to use an existing instrument , indicate here below the
instruments you intend to use-----------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
--------------------------------------------------------------------------------------------
3. Indicate how you will ensure reliability and validity of the results obtained
with these instruments-------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
(a) indicate how you will collect evidence to check internal consistency of the
instruments you will use ----------------------------------------------------------------
----------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------
(b) indicate how you will collect evidence to check reliability over time
(Stability of the instrument)----------------------------------------------------------------
---------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------
--------------
(c )Indicate how you will collect evidence to check validity----------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
-------- Page 51 of 241
4.11 Suggestions for further readings
Page 52 of 241
LECTURE FIVE
SAMPLING DESIGN
Lecture outline
5.1 Introduction
5.2 Lecture objective
5.3 Definition of the term “Sample”
5.4 Aims of sampling
5.5 A population
5.6 A sampling unit
5.7 A sampling frame
5.8 Sampling design
5.9 Non- probability sampling technique
5.10 Lecture summary
5.11 Activity
5.12 Suggestion for further readings
5.1 Introduction
When a researcher wants to know something about a certain group of people, they
usually find a few members of the group and study them. After they have finished
studying the individuals they usually come up with conclusions about a larger group.
Researchers collect data in order to test hypothesis and to provide empirical support for
explanations and predictions. Once the researchers have constructed their measuring
instruments in order to collect sufficient data pertinent to the research problem, the
subsequent explanations and predictions must be capable of being generalized to be of
scientific value.
Typically, generalizations are not based on data collected from all the observations, all
the respondents, or all the events that are defined by the research problem. Instead,
researchers use a relatively small number of cases (a sample) as the bases for making
inferences about all the cases (a population), Nachmias, 1996 pp178.
Page 53 of 241
Many commonsense observations, in fact, are based on observations of relatively few
people. In this lecture, we discuss the meaning of sampling and its purpose in research.
Later in the lecture, we will discuss the various sampling techniques.
_______________________________________________________________________
_
What is a sample?
One of the most important steps in the research process is to select the sample of
Page 54 of 241
individuals who will participate as part of the study.
Sampling refers to the process of selecting these individuals. Researchers would prefer
to study the entire population in which they are interested. However, this is difficult to
do. Most populations of interest are large, diverse, and scattered over a large geographic
area. Finding, let alone contacting all the members can be time- consuming and
expensive. For that reason, of necessity, researchers often select a sample to study.
For a researcher to accurately estimate unknown parameters from the known statistics,
they have to effectively deal with three major problems namely, first, the definition of
the population, secondly, the sampling design and thirdly, the size of the sample. Let us
now discuss each of them.
5.5 A population
A population can be referred to as the entire set of relevant units of analysis, or data. It
can as well be referred to as the “ aggregate of all cases that conform to some designated
set of specifications, Isidor Chein, 1982, pp 419{ Isidor Chein, “ An Introduction to
Sampling”, in Claire Selltiz,et al., Research Methods in Social relations, 4th ed.( New
York: Holt, Renehart and Winston,1981), p.419. For example, we can define a
population consisting of all the people residing in Kenya. We can even narrow this down
to a specific population of say university students in the University of Nairobi.
Page 55 of 241
in a given election year. An infinite population, on the other hand, consists of an endless
number of sampling units, such as an unlimited number of stars in the sky.
A sampling unit is a single member of a sampling population. For example, if you are
studying the University of Nairobi students, each single students becomes your sampling
unit. A good sampling unit must it must be relevant to the research problem. It is
important to note that a sampling unit need not be an individual. It can be an event, a
city, or a situation.
It is very important for a researcher to draw a sampling frame for the population of the
study. A sampling frame is a complete listing of the sampling units. The accuracy of a
sample depends largely on the sampling frame. Indeed, every aspect of the sampling
design- the population covered, the stages of sampling, and the sampling frame
influences the actual selection process-.
Page 56 of 241
5.8.1 Probability Sampling Techniques
This provides a scientific technique of drawing samples from the population according to
the laws of chance in which each unit in the universe has some definite pre-assigned
probability of being selected in the sample. The selection of the sample based on the
theory of probability is also known as random selection and sometimes the probability
sampling is also called Random Sampling. According to Simpson and Kafka, "Random
samples are characterized by the way in which they are selected. Randomness is not used
in the sense of haphazard or hit or miss".
When using probability sampling technique, the sampling units are selected according
to some probability laws. Some of these laws are that:
With a probabilistic sample, we know the odds or probability that we have represented
the population well. We are able to estimate confidence intervals for the statistic. Some
of the important types of probability sampling techniques include;
Page 57 of 241
In small populations such sampling is typically done "without replacement", i.e., one
deliberately avoids choosing any member of the population more than once. An unbiased
random selection of subjects is important so that in the long run, the sample represents
the population. However, this does not guarantee that a particular sample is a perfect
representation of the population.
Simple random sampling merely allows one to draw externally valid conclusions about
the entire population based on the sample. Although simple random sampling can be
conducted with replacement instead, this is less common and would normally be
described more fully as simple random sampling with replacement.
This type of sampling best suits situations where not much information is available about
the population and data collection can be efficiently conducted on randomly distributed
items. A simple random sample gives each member of the population an equal chance of
being chosen. It is not a haphazard sample as some people think! One way of achieving a
simple random sample is to number each element in the sampling frame (e.g. give
everyone on the Electoral register a number) and then use random numbers to select the
required sample.
Random numbers can be obtained using your calculator, a spreadsheet, and printed tables
of random numbers, or by the more traditional methods of drawing slips of paper from a
hat, tossing coins or rolling dice.
Page 58 of 241
It is ideal for statistical purposes.
It is free of classification error.
It requires minimum advance knowledge of the population
Disadvantages
Stratified sampling techniques are generally used when the population is heterogeneous,
or dissimilar or where certain homogeneous, or similar, sub-populations can be isolated
(strata). Simple random sampling is most appropriate when the entire population from
which the sample is taken is homogeneous.
Page 59 of 241
The following are the advantages and disadvantages of stratified sampling techniques.
Advantages
It assures the researcher of representation not only for the overall population, but also
key subgroups of the population, especially small minority groups. If you want to be able
to talk about subgroups, this may be the only way to effectively assure you'll be able to.
If the subgroup is extremely small, you can use different sampling fractions within the
different strata to randomly over-sample the small group (although you'll then have to
weight the within-group estimates using the sampling fraction whenever you want
overall population estimates).
When we use the same sampling fraction within strata, we are conducting
proportionate stratified random sampling.
Stratified random sampling will generally have more statistical precision than simple
random sampling. This will only be true if the strata or groups are homogeneous. If they
are, we expect that the variability within-groups is lower than the variability for the
population as a whole. Administrative convenience - in this case we have field officers
dealing with different parts of the population independently.
Disadvantages:
It can be expensive
Page 60 of 241
It requires accurate information about the population, or introduces bias.
This is a form of random sampling where the entire population is divided into groups, or
clusters and a random sample of these clusters are selected. All observations in the
selected clusters may be included in the sample or simple random sampling techniques
may be used to pick out the individuals to be included from each cluster. When all units
of the selected cluster are interviewed, this is referred to as "one-stage cluster sampling".
If the subjects to be interviewed are selected randomly within the selected clusters, it is
called "two-stage cluster sampling" (Caswell F 1989). Cluster sampling is a form of
random sampling where the units sampled are chosen in clusters.
This method of sampling is particularly useful where it is difficult to know the exact
numbers of individuals in a population, for example in developing countries where
official statistics are sparse. It is also applicable where the population is too large to carry
out simple random or stratified sampling and is commonly used in geography and
biology where; the survey area is covered with a grid of squares, A random sample of the
squares is then used for a complete investigation either by counting some physical or
manmade features in Geography or microbes, plant species etc in Biology. The results
are then generalized to the rest of the grids.
The following are the advantages and disadvantages of cluster sampling technique
Advantages
It helps to reduced field costs as a result of saving of travelling time and distance
covered
It is applicable where no complete list of units is available (special lists only
Page 61 of 241
need be formed for clusters).
Disadvantages
Units close to each other may be very similar and so less likely to represent the
whole population
Clusters may not be representative of whole population but may be too alike
Suppose that a survey is to be done in a large town and that the unit of enquiry is
the individual household. Suppose further that the town contains 20,000
households, all listed on convenient records, and a sample of 200 is needed. A
simple random sample of 200 could well spread over the whole town incurring
high costs and much inconvenience.
However one might decide to concentrate the sample in a few parts of the town.
Suppose for simplicity the town can be divides into 400 areas with 50 households
in each then one could select at random 4 areas (1/100) and include all
households in these areas. Constituencies, Wards, Districts etc may be used as
geographical demarcations.
Page 62 of 241
of all the doctors practicing in that particular city, and secondly it may mean
visiting most of the hospitals in that city to conduct the interviews. Therefore the
researcher could decide that each hospital in the city represents one cluster, and
then randomly select a small number, e.g. 20. He would then contact the doctors
in these 20 hospitals for interviews. Better still he may use another random
sample technique identify which doctors to interview
_________________________________________________________________
Multi-stage sampling is like cluster sampling, but involves selecting a sample within
each chosen cluster, rather than including all units in the cluster. Thus, multi-stage
sampling involves selecting a sample in at least two stages.
In the first stage, large groups or clusters are selected. These clusters are designed to
contain more population units than are required for the final sample.
In the second stage, population units are chosen from selected clusters to derive a final
sample. If more than two stages are used, the process of choosing population units within
clusters continues until the final sample is achieved. If we took the national elections as
an example, then a multi-stage sampling would involve, firstly, deciding on the electoral
sub-divisions (clusters) to be sampled from a city or state. Secondly, blocks of houses are
selected from within the electoral sub-divisions and, thirdly, individual houses are
selected from within the selected blocks of houses.
The following are the advantages and disadvantages of a multi-stage cluster sampling
technique.
Advantages:
Page 63 of 241
It is convenient.
It is economical.
It is more efficient than the simple random, cluster random sampling techniques.
Disadvantages
It has a lower accuracy due to higher sampling error.
The following are the advantages and disadvantages of systematic sampling technique.
Advantages
Very easy to operate and checking can also be done quickly.
More efficient than simple random sampling
Disadvantages
Works well only if the complete and up to date frame is available and if the units
are randomly arranged.
Gives biased results if there are periodic features in the frame and the sampling
interval is equal to or a multiple of the period.
Page 64 of 241
5.9 Non-probability sampling techniques
A sampling method in which elements are chosen based on purpose of the study.
Purposive sampling may involve studying the entire population of some limited group
(Diploma students at the University of Nairobi) or a subset of a population (Post
Graduate Diploma in Project Planning & Management). Purposive sampling does not
produce a sample that is representative of a larger population. It's a sample which is
selected by the researcher subjectively. It is also called judgment sampling.
Purposive sampling is the most popular in qualitative research and subjects are selected
because of some characteristic ( Patton, 1990).
Page 65 of 241
Advantages and disadvantages of Purposive sampling technique
The following are the advantages and disadvantages of purposeful sampling technique.
Advantages
Easy to undertake
Cheaper.
Disadvantages
Unable to generalize.
Page 66 of 241
have suffered closed head brain injuries in automobile accidents. This would be a
difficult population to find.
Used in political polling - districts chosen because their pattern has in the past
Quota sampling is the non-probability equivalent of stratified sampling. In this case the
population is first segmented into mutually exclusive sub-groups. This technique is one
of non-probability sampling, selection of sample is non-random. Judgment is used to
select the subjects or units from each segment based on a specified proportion
(Anderson,1966).
There are two types of quota sampling: proportional and non proportional. In
proportional quota sampling you want to represent the major characteristics of the
population by sampling a proportional amount of each. For instance, if you know the
population has 40% women and 60% men, and that you want a total sample size of 100,
you will continue sampling until you get those percentages and then you will stop.
So, if you've already got the 40 women for your sample, but not the sixty men, you will
continue to sample men but even if legitimate women respondents come along, you will
not sample them because you have already "met your quota."
The problem here (as in much purposive sampling) is that you have to decide the specific
characteristics on which you will base the quota. Will it be by gender, age, education
race, religion, etc? One need to have answers to these questions.
Non proportional quota sampling is a bit less restrictive. In this method, you specify the
minimum number of sampled units you want in each category. Here, you're not
concerned with having numbers that match the proportions in the population. Instead,
you simply want to have enough to assure that you will be able to talk about even small
groups in the population. This method is the non probabilistic analogue of stratified
random sampling in that it is typically used to assure that smaller groups are adequately
represented in your sample.
Page 67 of 241
Advantages and disadvantages of quota sampling technique
The following are the advantages and disadvantages of a quota sampling technique.
Advantages:
Useful for situations where you need to reach a targeted sample quickly and
where sampling for proportionality is not the primary concern.
Disadvantages
The problem is that these samples may be biased/ inaccurate because not
everyone gets a chance of selection.
As its name implies, convenience sampling refers to the collection of information from
members of the population who are conveniently available to provide it. It is a non-
probability sampling method, thus the elements in the population do not have any
probabilities attached to their being chosen as sample subjects.
This means the findings from the study of the sample cannot be confidently generalized
Page 68 of 241
to the population. In other words, the researcher has no way of estimating the
representativeness of convenience samples and therefore cannot estimate the population
parameters.
Advantages
It is not expensive
Disadvantages
It is very biased
Areas of application: It's important to note that, convenience sampling is most used
during exploratory phase of a research project i.e. during preliminary research efforts to
get a gross estimate of the results. The researcher will obtain some quick information to
get a feel for the phenomenon or variables of interest.
5.9.10 Snowball sampling
In the technique the researcher identifies a small number of individuals who have the
required characteristics. These people are then used as in formants to identify others
who qualify for inclusion in the sample. The second subjects also identify others
hence snowball. Snowball Sampling is used most when respondents are difficult to
identify and can best be located through referred networks. Snowball gathers subjects
Page 69 of 241
for the sample as it rolls, along.The Sampling has been used to study drug cultures,
power elites, teenage where respondents are difficult to identify or contact
Page 70 of 241
5.11 Activity
In respect to your proposed research problems fill in the bank spaces in the
following exercise.
1. In my proposed research study the following will be my intended
sample
Subjects of the intended study---------------------------------------------
--------
List down the specific sample and their respective numbers----------
----------------------------------------------------------------------------
----------------------------------------------------------------------------
---------------------------------------
2. Indicate the demographics (characteristics) of the sample in terms of
the following:
Age range------------------------------------
Sex distribution-------------------------------
Ethnic breakdown-----------------------
Their geographical location----------------------------------
Mention any other relevant characteristic not mentioned in the above
list
-------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------
------------------------------------------
3. Indicate, by ticking one, the type of sample from the list given here below:
Simple random? -----
Stratified random? ------
Cluster random? --------
Two-stage? -------
Convenience? -----
Purposive? ----
4. Indicate how you will obtain your sample-----------------------------------------
----------------------------------------------------------------------------------------------
Page 71 of 241
----------------------------------------------------------------------------------------------
------------------------------------------
5.11 Suggestion for further readings
Page 72 of 241
Page 73 of 241
LECTURE SIX
QUALITATIVE RESEARCH
Lecture outline
6.1 Introduction
6.2 Lecture objectives
6.3 Characteristics of qualitative research
6.4 Comparison of qualitative and quantitative research
6.5 Steps followed when conducting qualitative research
6.6 Issues of validity and reliability in qualitative research
6.7 Issues of generalization in qualitative research
6.8 Research strategies used in qualitative research
6.9 Data collection techniques in qualitative research
6.10 Lecture summary
6.11 Activity
6.12 Suggestions for further readings
6.1 Introduction
There has been much confusion over what qualitative research is. Some people
think that qualitative research is non- quantitative. This is not true. First, some
qualitative research results in some quantification (e.g., counting the numbers of
occurrences of a particular behaviour). Second, that qualitative research is based
on the phenomenological paradigm, which uses a variety of interpretive research
methodologies while quantitative research is based on the logical-positive
paradigm, which utilizes experimental research methodologies. Still some people
believe that qualitative research uses a unitary approach, when in reality it uses a
variety of alternative approaches.
Page 74 of 241
As Patton (1990) says, qualitative research uses different data than those used in
traditional research methods.
The data from interviews consists of direct quotations from people about their
experiences, opinions, feelings and knowledge. The data from observations consists of
detailed description of people’s activities, actions, and the full range excerpts, quotations,
or entire passages from organizational clinical or program records; memoranda and
correspondence; official publications and reports; personal diaries; open- ended written
responses to questionnaires and survey”
_____________________________________________________________________
Page 75 of 241
6.2 Lecture objectives
By the end of this topic you should be able to:
1. Explain what is meant by the term “qualitative research”.
2. Describe the general characteristics of qualitative research.
3. Describe at least three ways that qualitative research differs
from quantitative research.
4. Describe briefly the steps involved in qualitative research.
5. Explain how generalizing differs in qualitative research and
quantitative research.
_____________________________________________________________________
Patton proposes ten themes, which inculcate qualitative research. These themes make the
various qualitative research methods distinct from quantitative methods.
Page 76 of 241
interrelationships;
The research activity begins by exploring genuinely open
questions rather than testing theoretically derived (deductive)
hypothesis.
Page 77 of 241
stance towards whatever content may emerge
10. Design It is an open adapting inquiry process. The researcher’s
flexibility understanding of the phenomenon changes as the situations
changes.
It avoids getting locked into rigid designs that eliminates
responsiveness; pursues new paths of discovery as they emerge.
It is important to note that all qualitative research methods have one thing in common:
The use of qualitative data,
Sensitive to the context,
Emphasis on researchers neutrality, and
It focuses on inductive analysis.
i) Context sensitivity: qualitative data are so powerful because they are sensitive to
the social, historical, and temporal context in which the data were collected. The
particular importance of context sensitivity is that data are not generalized to
other contexts, socially, spatially, or temporally. The logical-positivistic
paradigm on the other hand, purposefully pursues research findings that can be
generalised to other settings, persons, and times.
ii) Inductive analysis enables the researcher to explore the data without prior
hypotheses. This permits the researcher to discover reality without having to fit it
into a preconceived theoretical perspective. This is obviously the antithesis of the
logical-positivistic approaches, which insists that research be based on hypothesis
generated from theory, prior research, or experience.
Page 78 of 241
throughout the data collection process. This permits the researcher to adjust the
direction of the inquiry based on the ongoing experience of collecting and
thinking of the data.
Page 79 of 241
complex phenomenon into specific of complex phenomena
parts for analysis
11. Willingness to manipulate aspects, Unwillingness to tamper with
situations, or conditions in studying neutrality occurring phenomena
complex phenomena
The steps involved in conducting a qualitative research study are not as distinct as they
are in quantitative research; they often overlap and are sometimes even conducted
concurrently. However, they have a starting and ending point.
There are several steps that are followed in qualitative research. They are:
i) Identification of the phenomenon to be studied. Before any study can begin, the
researcher must identify the particular phenomenon he or she is interested in
investigating
ii) Identification of the participants in the study. The participants in the study
constitute the sample of individuals who will be observed (interviewed, etc.). In
other words, the subjects of the study. In almost all-qualitative research, the
sample is a purposeful. Random sampling ordinarily is not feasible, since the
researcher wants to ensure that he or she obtains a sample that is uniquely suited
to the intent of the study.
Page 80 of 241
iv) Data collection. There is no “treatment” in a qualitative study, nor is there any
“Manipulation” of subjects. The participants in a qualitative study are not divided
into groups, for example, as in experimental research, with one group being
exposed to a treatment then measured in some way. Data are not collected at the
end of the study; rather the collection of data is the research goes on. The
researcher is continually observing people, events, and occurrences, often
supplementing his or her observations with in-depth interviews of selected
participants and the examination of various documents and records relevant to the
phenomenon.
As you are already aware, validity refers to the appropriateness, meaningfulness, and
usefulness of the inferences researchers make based on the data they collect, while
Page 81 of 241
reliability refers to the consistency of these inferences overtime.
Page 82 of 241
l) Observing the setting or situation of interest over a period of time. The length of
an observation is extremely important in qualitative research. Consistency over
time with regards to what researchers are seeing or hearing is a strong indication
of reliability. Furthermore, there is much about a group that does not even begin
to emerge until some time has passed, and the members of the group become
familiar with, and willing to trust, the researcher.
In qualitative studies, on the other hand, the researcher may also generalize, but it is
much more likely that any generalization to be done will be by interested practitioners-
by individuals who are in situations similar to the one(s) investigated by the researcher. It
is the practitioner, rather than the researcher, who judges the applicability of the
researcher’s findings and conclusions, who determines whether the researcher’s findings
fit his or her situation.
The choice of qualitative strategy depends on the focus of the research and the desired
time frame for the study. The main and mostly used strategies are the ethnographic
studies, case studies, content analysis, and field study.
Page 83 of 241
a) Documents or content analysis
When using documentary sources, one must bear in mind that data appearing in print are
not necessarily trustworthy. Documents used in descriptive research must be subjected to
careful criticism. The documents must be authentic and valid. The researcher must hence
establish the trustworthiness of all the data.
Page 84 of 241
b) The Case Study.
The case study is a way of organising social data for the purpose of viewing social
reality. It examines a social unit as a whole. The unit may be a person, a family, a social
group, a social institution, or a community. The main purpose is to understand the
lifecycle or an important part of the life cycle of the unit. The case study probes deeply
and analyses interactions between the factors that explain present status or that influence
change or growth. It is a longitudinal approach, showing development over a period of
time.
The focus of such a study is the typicalness rather than uniqueness. According to
Bromley (1986), “A ‘case’ is not only about a ‘person’ but also about that ‘kind of a
person’. Thus the selection of the subject of the case study needs to be done carefully in
order to assure that he or she is typical of those to whom we wish to generalise.
Case studies are not confined to the study of individuals and their behavioural
characteristics. They also include groups and organizations.
There are several precautions that one should consider when using case study as a
methodology:
The method may look deceptively simple. To use it effectively, the researcher
must be thoroughly familiar with existing theoretical knowledge of the field of
inquiry, and skilful in isolating the significant variables from many that are
Page 85 of 241
irrelevant. There is tendency to select variables because of their spectacular
nature rather than for their critical significance.
Subjective bias is a constant threat to subjective data –gathering and analysis.
The danger of selecting variables relationships based upon preconceived
convictions and the apparent consistency of a too limited feeling of certainty
about the validity of his or her conclusions.
Effects may be wrongly attributed to factors that are merely associated rather
than cause-effect related.
c) Ethnographic Studies.
Ethnography is a method of field study observation that becomes popular in the later
parts of the 19th century. It is alternatively called, cultural anthropology or naturalistic
inquiry. It studies cultural features as language, marriage and family life, child-rearing
practices, religious beliefs and practices, social relations and rules of conduct, political
institutions, and methods of production.
The data gathered consists of observation of patterns of action, verbal and nonverbal
interactions between members of the tribe as well as between the subjects and the
researcher and his or her informants, and the examination of whatever records or
artefacts available.
In most cases, the researcher is integrated into the group he or she is studying. Using the
method of observation, the researcher observes, listens to, and sometimes converses with
the subjects in as free and natural an atmosphere as possible. The assumption is that the
most important behaviour of individuals in groups is a dynamic process of complex
interactions and consists of more than a set of facts, statistics, or even discrete incidents.
The strength of this kind of study lies in the observation of natural behaviour in a real-
life setting, free from the constraints of more conventional research procedures.
The second assumption is that human behaviour is influenced by the setting in which it
occurs. The researcher must understand that setting and the nature of the social structure;
its traditions, values, and norms of behaviour. This is because it is important to observe
Page 86 of 241
and interpret not just as an outside observer but also in terms of the subjects-how they
view the situation, how they interpret their own thought, words, and activities, as well as
those of others in the group. The researcher gets inside the minds of the subjects; while at
the same time interpreting the behaviour from his or her own perspective.
The relationship of the researcher to their subjects is based upon trust and confidence. A
researcher should avoid aligning with either the authority or the subjects. He or she
should take a neutral position. This will help in objective observation.
There are two main techniques of collecting data in qualitative research. They are
observation and interviews.
i) Observations
The observer’s role may vary from full participation to complete outsider.
The observer may conduct the observations covertly with the full knowledge of
those being observed or with only some of those being observes aware of the
observation
Those being observed may be given full explanation, partial explanations, no
explanations, or given false explanation.
The observation may take place over the course of an entire duration or a brief
duration
Page 87 of 241
Observations may vary in breadth. Some may be broad while others may be
narrow.
It is important to note that the observations can be of the setting or physical environment,
social interactions, physical activities, nonverbal communications, planned and
unplanned activities and interactions, and unobtrusive indicators. The observer should be
alert for non-occurrence- the things that should have happened but did not.
ii) Interviews
Interviews range from quite informal and completely open-ended to very formal with
questions predetermined and asked in a standard manner (for example, the question may
read to the interviewee to assure the same wording with all those being interviewed).
Page 88 of 241
informal interview is to find out what people think and how the views of one
individual compare with those of another.
Page 89 of 241
situational responses
Standardized open-ended The exact wording Respondents answer the Little flexibility in
interview and sequence of same questions, thus relating the interview to
questions are increasing particular individuals and
determined in comparability of circumstances;
advance. All responses; data are standardized wording of
interviewees are complete for each questions may constrain
asked the same person on the topic and limit naturalness and
basic question in addressed in the relevance of questions
the same order. interview. Reduces and answers.
Questions are interviewer effects and
worded in a bias when several
completely open- interviewers are used
ended format .permits evaluation
users to see and review
the instrumentation used
in the evaluation.
Facilitates organization
and analysis of the data.
Closed, fixed-response Questions and Data analysis is simple; Respondents must fit
interview responses responses can be their experiences and
categories are directly compared and feelings into the
determined in easily aggravated; many researcher’s categories;
advance. Responses questions can be asked may be perceived as
are fixed; in a short time impersonal, irrelevant,
respondents and mechanistic. Can
chooses from distort what respondents
among these fixed really mean or have
Page 90 of 241
responses experienced by so
completely limiting their
response choice.
Patton has identified six basic types of questions that can be asked of people. They are:
i) Background or demographic questions. They are routine sorts of questions
about the background characteristics of the respondents. They include questions
about education, previous occupations, age, incomes, etc.
ii) Knowledge questions: they are questions researchers ask to find out what factual
information (as contrasted with their opinions, beliefs, and attitudes) respondents
possess.
iii) Experience or behaviour questions: they are questions a researcher asks to find
out what a respondent is currently doing or has done in the past. Their intent is to
illicit descriptions of experience, behaviour, or activities that could have been
observed but were not.
iv) Opinion or value questions: are questions researchers ask to find out what
people think about some topic or issue. Answers to such questions call attention
to the respondent’s goals, beliefs, attitudes, or values.
v) Feeling questions: they are questions that a researcher asks to find out how
respondents feel about things. They are directed towards the emotional responses
of people to their experiences.
vi) Sensory questions: They are questions a researcher asks to find out what a
respondent has seen, heard, tasted, smelled, or touched.
Interviewing Behaviour
Fetterman has identified a number of elements common to all interviews. They are:
Respect the culture of the group being studied.
Respect the individual being interviewed
Be natural
Page 91 of 241
Ask the same questions in different ways during the interview.
Ask the interview to repeat an answer or statement when there is some doubt
about the completeness of a remark.
Learn how to wait.
6.10 Lecture summary
6.11 Activity
a) What are the strengths and weaknesses of qualitative research?
b) Explain the characteristic of qualitative research
c) Describe the steps followed in qualitative research
d) How do you ensure validity and reliability in qualitative
research
e) How would you handle generalization in qualitative research
f) Describe the various types of interviewing questions
g) Describe the various interviewing behaviour
Page 92 of 241
6.12 Suggestions for further research
Page 93 of 241
LECTURE SEVEN
QUANTITATIVE RESEARCH METHODOLGY
Lecture outline
7.1 Introduction
7.2 Lecture objectives
7.3 Descriptive versus inferential statistics
7.4 Frequency distribution
7.5 Lecture summary
7.6 Activity
7.7 Suggestions for further readings
7.1 Introduction
In the previous lecture we discussed the meaning and process of qualitative research
approach. In the next series of lectures, we are going to discuss the quantitative research
methodologies.
Quantitative research methods are usually used in an attempt to establish general laws
and principles . This kind of approach to science is often termed nomothetic. It assumes
that social reality is objective and external to the individual.
On the other hand, qualitative analysis regards social reality as a creation of individual
consciousness, with meaning and the evaluation of events seen as a personal and
subjective construction this is a naturalistic approach to research and is termed an
ideographic approach.
What we should note is that each of these two perspectives on the study of human
behaviour has profound implication for the way in which research is conducted.
In this section, we are going to discuss the quantitative approach and its implications to
the way we conduct research.
Page 94 of 241
7.2 Lecture objectives.
By the end of this lecture, you should be able to:
1. Differentiate between descriptive statistics and inferential statistics
2. Explain what is meant by a “normal distribution” and a” normal curve”
3. Discuss how to develop a sampling design
4. Discuss the levels of measurements.
5. Explain how to design a quantitative research
6. Describe how to test hypothesis
7. Describe how to measure reliability and validity
8. Describe how a researcher can make predictions using linear regression.
Descriptive statistics consist of graphical and numerical techniques for summarizing data
.It enables a researcher to reduce a large mass of data to simpler, more understandable
terms. This makes it easier for an observer to understand the data.
The major advantage of descriptive statistics is that they permit researchers to describe
the information contained in many scores with a few indices such as the mean, median
and mode.
On the other hand, inferential statistics consists of procedures for making generalizations
about characteristics of a population based on information obtained from a sample taken
from the population.
It is important for us to fully understand what we mean by the terms “statistics” and
“parameters”
Page 95 of 241
A parameter is an indices calculated from the entire population.
Quantitative data are the data obtained when the variables being studied is
measured along a scale that indicates how much of the variable is present. They
are reported in terms of scores. Higher scores indicate a higher presence of the
variable while lower scores indicate a presence of the variable. Good examples of
such scores are such as weight, height, length and academic ability.
Categorical data is the data that indicate the total number of objects, individuals
or events a researcher finds in a particular category. In this case, the researcher is
looking for the frequency of certain characteristics in the variables.
Page 96 of 241
How do researchers summarize quantitative data?
There are several techniques used by researchers to summarize quantitative data. Let us
look at two main techniques: the frequency distributions and the normal curve
.
7.4 Frequency distributions
This is a tabular method of showing all the scores obtained by a group of individuals.
This is done by listing in a rank order from high to low with tallies to indicate the
number of subjects receiving each score .
Page 97 of 241
51 2
38 4
36 3
34 5
31 5
29 5
27 5
25 1
24 2
21 2
17 2
15 1
6 2
3 1
n=50
In some cases a researcher might find it easier to present the data in a grouped frequency
distribution. The researcher will then group the data into intervals. In our example above
, we can group the data into intervals of five. In this case, the grouped frequency
distribution will appear in table 1.2
Page 98 of 241
20-24 4
15-19 3
10-14 0
5-9 2
0-4 1
n=50
A researcher can use graphs to describe frequency distributions. Let us look at the
various types of graphs used to describe the distributions.
Some people find it difficult to red and understand numerical tables. For such people. a
researcher may provide graphical representations as an alternative. Method of displaying
the information organized in frequency distributions. This helps to create a visual
impression of the data that might be more effective in communicating the information.
There are three types of graphs used by researcher: the pie-chart, the bar chart and the
histograph.
The pie chart is used to show difference in frequencies or percentages among categories
of nominal or ordinal data. Such categories of data are displayed as segments of a circle.
The segments are either differently shaded or differently patterned to differentiate among
them and they sum up to either 100 percent or the total frequencies.
Let us look at the examples in figure 1.1. Showing students’ attitude towards a schools
spending on sports and Music festivals as indicated in table 1.3.
Table 7.3 Students’ attitude towards a school spending on sports and music festival
Page 99 of 241
Too little 54.7% 23.5%
About 41.0% 36.9%
right
Too 4.3% 39.6%
much
Total 100% !00%
Like the pie chart, a bar chart provides a researcher with a tool for displaying nominal or
ordinal data. Bar charts are constructed by labeling the categories of the variables along
the horizontal axis and drawing rectangles of equal width for each category. The height
of each rectangle is proportional to the frequency or percentage of the category. Let us
use the data in Table 1.3 to describe how to construct a bar chart shown in Figure 1.2.
Note that it is important to shade the rectangle representing each variable differently to
facilitate comparisons.
(c ) The Histogram.
Let us use table 1.4 showing a hypothetical case of HIV infection by Age..
7.7 Activity
8.1 Introduction
In this lecture we shall reflect on the data that we use in research. In most cases, research
generates masses of data from which we are able to make sense of our world. The data
comes in different forms depending on the problem, objective and variables of interest
from the target population. We make measurements like colour, age, position and many
other characteristics from the variables of interest. We also decide to use one or many
variables depending on the information being sought. For example, we may be interested
in establishing the relationships between the location of a hotel and the occupancy rates
leading to the use of bivariate measures to determine whether location has an impact on
When data comes to us from the field, it comes as raw data which does not make a lot of
sense. We need to reorganization it in a way that we can understand and manipulate. This
lecture unravels the ways used to process and classify such data.
We divide data into two main types depending on how it was aquired by the researcher.
The two types are primary data and secondary data. We shall look at each of the them in
the following sub-sections.
Attributes are also variable which are based on characteristics like colour, sex or attitude.
Colour has values like black, white, blue and so on. Sex has two values: male and
female. The value of attributes is based on qualities and is therefore defined as
qualitative as they mainly deal with the quality of phenomena or the presence and
absence of it.
We can therefore conclude that our population parameters come from variables and
values which are either quantitative or qualitative. We shall talk about quantitative and
qualitative data later. Variables are divided into three main categories as discussed
below.
Note that a variable is a concept that has values which can be measured like cost,
colour or sex. As a variable like sex has two values: male and female.
There are three types of variables as follows:
Independent variable: In a statement like ‘A student’s performance in
tourism is dependent on sex’. Here, sex is independent.
Dependent variable: In the statement above, performance is a dependent
variable.
Intervening (or extraneous) variable: If, in the statement above, the
relationship is influenced by the university where the programme is taken,
then university becomes an intervening variable.
We may also talk of our gastronomic tastes being shaped by the types of food available
The data discussed in a bivariate relationship can be faulted by ignoring another variable
that could interfere with the direct relationship between the two. For example, in the case
of age and disease, particularly on humans, you may find the medical scheme of a
country will influence the pattern. The environment could also have an influence. We
could decide to these variables as independent or dependent to make elaborate statement
like: In areas where health facilities are poor, the older you are the more susceptible to
diseases you become.
Intext Question
Think about the following examples:
Life expectancy in Australia is 80 years
Life expectancy in Kenya is 45 years
Life expectancy among the !kung is 32.5 years.
In all cases, women tend to live longer than men
What variables are responsible for all these differences?
By now, I know you have encountered computers and appreciated their capacity. You
will definitely need them more in research and data analysis. What you also need to
know is that when provided with the right information, computers are able to generate
accurate results in speed that is far much higher than manual computations.
Data can also exported to more powerful statistical packages like SPSS which have data
editing and manipulation capacity. SPSS is an abbreviation of the Statistical Package for
the Social Sciences. The package is normally used in the computation of both descriptive
and inferential data. For those of us who have used it for a long time, the most important
asset of this package is its flexibility. The spreadsheet is spacious and user friendly.
Data stored in SPSS is easy to manipulate through coding and recoding and all possible
statistics in the social sciences can be used on the same data. The data is usually stored in
raw form and this make is easy to attempt all forms of informative and relevant
computations.
The strength of SPSS has been dealt with comprehensively by Felbinger and Schwelgien
(2005:503-547). Here, I will only mention that:
It is available in different versions to suit ones working environment.
It is well integrated to windows and therefore easy to use
It allows for data editing, classification, coding and recoding during computation
process.
It hand multidimensional tables
It has high resolution graphics
It has data transformations and capabilities
8.13 Summary
We have seen that data can be defined in many ways and classified
appropriately for statistical manipulation. We have also seen the role of
computers and statistics data for manipulation which enables us to
provide more orderly and professional information summaries.
Statistics also helps us to present information in pictorial or graphic
forms which are easy to understand from a mass of data.
We also talked about classification of data into measurements like ratio,
interval, ordinal and nominal data. This classification enables us to
decide how to handle data as their manipulation levels vary considerably
due to their inherent characteristics.
8.14 Activity
1. Discuss the characteristics of raw data.
2. Make a clear distinction between qualitative and quantitative data
3. Why are computers important in research?
4. What do you understand by nominal level of data measurement?
Cite 5 areas in tourism where you can use this type of data.
5. Make a distinction between ratio and rank data and cite example
from the tourism industry.
8.15 References
Lecture outline
9.1 Introduction
9.2 Lecture Objectives
9.3 Defining a research problem
9.4 Formulating research questions
9.5 Constitutive, operational and conceptual definitions
9.6 Formulation of hypothesis
9.7 Lecture summary
9.8 Activity
9.9 Suggestion for further research
9.1 Introduction
Any good research begins with a research problem. A research problem is the focus of a
research investigation. A research problem is the problem the researcher wishes to
investigate. In most cases, research problems are stated as research questions. In this
topic we discuss the nature of research problem and describe its characteristics.
_____________________________________________________________________
Usually a research problem is initially posed as a question, which serves as the focus of
the researcher’s investigation.
Some example of research problem:
How do parents feel about private schools?
Do NGOs in Kenya have good governance guidelines?
Do Kenyans support the governing party?
Does an increase in salary increase a lecturer’s output?
The second question is a question of value .It implies notions of right and wrong, proper
and improper and therefore does not have any empirical (or observable) referents. There
is no way to deal empirically with the verb “should”. However if we changed the
question to read “ Do people think mathematics should be included in the school
curriculum”? The question would be researchable because now we can collect data to
enable us answer the question.
Exercise
Here below are some ideas for research questions. Indicate which ones
you think are researchable or unresearchable.
a) Is God good?
b) Are students happier when taught by a lecturer of the same gender?
c) Does a firm performance influenced by the firm’s corporate strategy?
d) What is the best way of motivating staff?
e) What would be the world today without the September 11 bomb attacks in USA?
It is very important that a researcher evaluates whether the research questions he or she
has formulated are good. To achieve this, the researcher weighs the research question
against the following four characteristics.
The question should be feasible (i.e., it can be investigated without an undue
amount of time, energy or money.
The question should be clear (i.e., most people would agree as to what the key
words in the question mean)
The question should be significant (i.e., it should be worth investigating it
because it will contribute important knowledge to humanity)
The question should be ethical (i.e., it will not involve physical or psychological
harm or damage to human beings or, to the nature or social environment of which
they are part).
It is always important to define the terms the researcher is using particularly in the
research questions. This helps to give clarity to the research questions. There are
essentially three ways to clarify important terms in a research question.
ii) Conceptual definition: this involves attempts to describe as fully as possible the
terms used in the research question. For example a term “job interest” needs to be
defined conceptually. This involve the researcher showing the relationship
between the two concepts, i.e., “job” and “interest”
It is important to note that good research questions frequently( but not always) suggests a
relationship of some sort to be investigated. A suggested relationship means that two
qualities or characteristics are tied together or connected in some way or there is some
sort of association between characteristics.
1. State a research problem that you expect to investigate in you project. ----------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
--
2. State the research questions that you can formulate from your research
problem. -----------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
-
3. State the key terms in the research problem or research questions that are not
clear and thus need to be defined-------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------
4. Write down the constitutive definitions of these terms------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
-------------------------------------------------------------
5. Write down the conceptual definitions (if any)---------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
-
6. Write down the operational definitions of these terms------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
Page 118 of 241
7. Give the justifications for investigating this research problem---------------------
----------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------
9.9 Suggestion for further readings
10.1 Introduction
The conclusions of any research study are based on the analysis of data collected.
Therefore, data collection is extremely important in all research activities. A researcher
must consider with absolute care the kind(s) of data collected, the method(s) of data
collection used and the scoring of the data. In this lecture, we are going to discuss the
“data” collection methods used in research.
_____________________________________________________________________
_______________________________________________________________________
The term “data” refers to the kinds of information researchers obtain on the subjects of
their research. An example of data includes: demographic information such as age,
gender, ethnicity, religion; responses to the researcher’s questions in an oral interview or
written replies to survey questionnaire etc. Every researcher must make the decision on
what kind (s) of data he/she intends to collect.
Sources of information can be classified into primary and secondary types. Primary data
comes from the original sources and are collected specifically to answer the research
questions. Secondary sources of data come from other sources, for example, other studies
conducted by other persons for other purposes.
The data can be found more quickly and cheaply than primary data. Collecting
primary data can be costly and time-consuming. Data about distant places can
be collected more cheaply through secondary sources
(ii) Disadvantages.
The information may not meet the specific needs of the research in question. This is
because others have collected the material for their own purpose. Definitions will differ,
units of measure are different, and different times may be involved. It is difficult to
assess the accuracy of the information because one knows little about the research design
or the conditions under which the research occurred. Secondary information is often out
of date because of time (time may have elapsed since it was conducted).
There are basically two sources of secondary data; internal sources and the external
sources.
Internal Sources are those sources within the organization itself. In a business
organization this would include, the accounting and information systems,
research and development, planning and marketing functions reports.
External sources are those sources of data found outside the organization. For
example, published sources, periodicals and special collections.
The following are the questions researchers need to ask themselves in the process of data
collection.
Where will the data be collected? The researcher must decide on the location of
the data collection.
When will the data be collected? That is the time the data will be collected.
How often are the data to be collected? That is the frequency of the collection of
the data.
Who is to collect the data? That is who will administer the instrument.
These questions are important to be answered before the researcher begins to collect the
data. A researchers’ decision about location, time, frequency and administration are
always affected by the kind(s) of instrument to be used. Every instrument no matter what
kind, if it is to be of value must allow researchers to draw accurate conclusions about the
capabilities or other characteristics of the people to be studied.
An instrument must be valid, reliable and objective. A valid instrument is one that helps
the researcher make defensible inferences. A reliable instrument is one that gives
consistent results. An objective instrument is the one that enables the researcher to make
judgment that is not subjective.
A researcher should have satisfactory answers to these questions. If not so, he /she might
waste a lot of time and other resources doing unnecessary work.
You already know that there are three main ways in which a researcher can obtain
information.
By collecting it themselves with very little or no involvement of other people.
Directly from the subjects of the study.
From others, frequently referred to as informants, who are knowledgeable about
the subject.
Researchers can collect data that require a written or marked response or a more general
evaluation of performance on the part of the subjects of the study. What are they?
There are basically two ways for a researcher to acquire research instruments:
To find and administer a previously existing instrument
To administer an instrument the researcher personally developed or had
developed by someone else.
There are several problems associated with the use of an instrument developed by the
researcher; first, it is not easy to develop it. Secondly, it takes a fair amount of time and
effort to develop it. Thirdly, it requires a considerable amount of skills.
This is the reason why the use of an already developed instrument when appropriate is
preferred. Such instruments are developed by experts who possess the necessary skills.
There exists already developed, quiet good, instruments and they can easily be located by
means of a computer. You only need to go to the relevant search engines like ERIC to
get them.
11.1 Introduction
11.2 lecture Objectives
11.3 Survey method
11.4 Observation method
11.5 Experimental method
11.6 Lecture summary
11.7 Activity
11.8 Suggestion for further readings.
11.1 Introduction
A researcher needs to decide which data collection method to use in the study. Many
amateur researchers confuse between data collection methods and data collection tools.
In this lecture we are going to discuss the three main data collection method and the data
collection tools that can be used for each method. There are three main methods of
collecting data. They are:
The survey.
Observation.
Experimentation.
11.3 The survey method
When is survey method most applicable? A survey method is most appropriate where the
respondents are uniquely qualified to provide the desired information.
There are three main techniques that can be used to get information using the survey
methods. They are;
Personal interview
Telephone interview
Mail interview/ self-administered questionnaires.
Advantages:
There are advantages and clear limitations of personal interviewing. They are as
follows:
This technique is more deep and detailed in terms of the information collected.
This is because the interviewer can control the process hence probing more by
adding questions that help to add more information unlike in an observation
method.
The interviewer has more control than other kinds of interrogation. They can
prescreen to assure the correct respondent is replying and can set up and control
interviewing conditions.
The interviewer can use special scoring devices and visual materials.
The interviewer can adjust to the language of the interviewee because they can
observe the problems and effects the interview is having on the respondent.
Disadvantages:
It is relatively expensive, particularly if the study covers a wide geographic
area or has stringent sampling requirements.
Interviewers are usually reluctant to visit unfamiliar neighbourhoods alone.
The results can be affected adversely by interviewers who alter the questions
asked or in other ways.
How can a researcher ensure success of the personal interview? There are three broad
Advantages:
It is relatively cheaper. This is because of saving that comes from cuts in travel
administrative costs (training and supervision).
Responses are received immediately.
Unlike the personal interview, the use of telephone brings a faster completion of
the study.
Disadvantages
The respondent must be available by phone. In cases where such services are
scarce, it can be difficult and even expensive.
The discussion is relatively limited because of the time one can spend on a
telephone line.
It is not possible to use maps, illustrations or other visuals. The medium also
limits the complexity of the questioning and the use of sorting technique
In some situations, the response rate is lower than for comparable face-to-face
interviews. This is because the respondents find it easier to terminate an
interview.
Telephone interviews can result in less thorough responses and that those
interviewed by phone find the experience less rewarding to them than a personal
interview.
Advantages
They typically cost less than the personal interviews. The more dispersed the
sample, the more likely it will be the low-cost method.
Using mail is possible to get to the respondents who can otherwise be
inaccessible.
It allows the respondent to take more time to collect facts, talk to others or
consider replies at length than is possible with the telephone or personal
interview.
Mail survey is perceived as more impersonal, hence providing more anonymity
than the other communication modes.
Disadvantages
The non-response rate is high. This makes it difficult to know how their answers
might differ from those who do not answer.
In most cases the respondents do not provide adequate information. Usually,
there are many questions that are never answered.
How do you improve questionnaire return rate? There are several ways in which a
researcher can improve the rate of return of the mail survey. They include the following:
Follow-ups or reminders.
Preliminary notifications by telephone that a mail survey is on the way to the
respondent and request for response.
Use of Concurrent techniques.
This includes the following:
Ensuring the questionnaire is not too long
A researcher should attempt to ensure that the rate of return of the survey is maximized.
The main contributing factors to low rate of mail survey return include:
The wrong address and a low-rate postage can lead to non-delivery or non-return
The letter may look like junk mail and be discarded without being opened
Lack of proper instructions for completion leads to non-response
The wrong person opens the letter and fails to call it to the attention of the right
person
A respondent finds no convincing explanation for completing the survey and
discards it.
A respondent temporarily lays the questionnaire aside and fails to complete it.
The return address is lost so the questionnaire cannot be returned.
Activity
In the following situations, would you use a personal interview, telephone survey or
mail survey?
i) Question content
In deciding on the question content, a researcher should ask themselves the following
questions:
Should this question be asked?: a good question should contribute significantly
towards answering the investigative question
Is the question of proper scope and coverage? That is whether the question
includes so much content that it should be broken into several questions. It
important to avoid double-barreled questions ( two questions in one). For
example: have you deposited or saved money in your account in the last two
weeks?
Can the respondent answer the question adequately? The ability of the
respondent to answer adequately is often distorted by questions whose content is
biased by what is included or omitted.
Will the respondent answer willingly? In some cases the respondent may have
the information but they may be unwilling to give it either because the topic is
too sensitive to discuss with strangers or it is embarrassing.
There are three approaches that are used to overcome these problems;
Motivate the respondent to provide appropriate information
Change the design of the questioning process, or
Use methods other than questioning to secure the data.
Response structure refers to the degree and form of structure imposed on responses. The
options range from open (free choice of words) to closed (specified alternatives). Free
responses range from those in which the respondents express themselves extensively to
those in which their latitude is restricted to choosing one word in a “fill-in” question. On
the other hand, closed responses typically are categorized as dichotomous or multiple
choice.
iv)Question Sequence
A good questionnaire will be designed in such a way that the questions are related to
each other. Therefore, question sequencing is particularly important. The principle used
to guide question sequence decision is; the nature and need of the respondent must
determine the sequence of questions and the organization of the schedule. To achieve
this it is important to ensure the following:
That the question process must quickly awaken interest and motivate the
respondent to participate in the interview.
That the respondent should not be confronted by early requests for information
that might be considered personal or ego threatening.
That the questioning process should begin with simple item and move to the more
complex and from general items to the more specific.
That change in the frame of reference should be small and should be clearly
pointed out.
Activity
1. Describe a survey method of collecting data.
2. Distinguish between:
a. Direct and indirect question
b. Open and closed question
c. Research, investigative and measurement question
d. Question and response structure
(b ) Behaviour observation
Nonverbal analysis
Linguistic analysis
Extralinguistic analysis
Spatial analysis
Record analysis. This involves historical or current records and public or private
records. They may be written, printed and sound-recorded photographed or
videotaped.
Process or activity analysis includes the analysis of processes like traffic flow,
distribution systems banking system etc
Extra- linguistic behaviour: This includes the communication attributes like the
vocal (pitch, loudness etc); temporal (rate of speaking, duration of utterances, and
Advantages
Disadvantages
The observer must be at the scene of the event when it takes place, yet it is often
impossible to predict where and when the event will occur.
It is slow and expensive process that requires either human observers or costly
surveillance equipment.
Its most reliable results are restricted to information that can be learned by overt
actions or surface indicators to go below the surface, the observer must make
inferences.
The research environment is more suited to subjective assessment and recording
of data than controls and quantification of events.
Observation is limited as a way to learn of the past. It is similarly limited as a
method by which to learn what is going on in the present at some distant place.
In the observation method, we need to answer the following question if we are to gather
the required data:
1. Assume you wish to analyze the car traffic that passes the University of Nairobi
gate along the University Way. You are interested in determining how many vehicles
pass by the gate and you would like to classify them on various relevant dimensions.
Experiments are studies involving intervention by the researcher beyond that required for
measurement. This intervention involves the manipulation of some variables in a setting
and observing how it affects the subjects being studied. The researcher manipulated the
independent or explanatory variable and then observes whether the hypothesized
dependent variable is affected by the intervention.
The selection of measures for testing requires a thorough review of the available
literature and instruments. The measures must be adapted to the unique needs of the
research situation without compromising their intended purpose or original meaning.
The level of the independent variable is the distinction the researcher makes between
different aspects of the treatment conditions. The levels assigned to an independent
variable should be based on simplicity and common sense.
There are two main levels: the control group and the experimental treatment group.
The experimental method has a problem of controlling the effects of the extraneous
variables. They have the potential for distorting the effects of the treatment on the
dependent variable and must be controlled or eliminated. More so the researcher needs to
control the physical environment of the experiment. The introduction of the experiment
to the subjects and the instructions would likely be videotaped to assure consistency. The
arrangement of the room, the time of administration, the experimenter’s contacts with the
subject, etc. must all be consistent across each administration of the experiment.
There are other forms of control that involves the subjects and the experimenter. They
are:
When subjects do not know if they are receiving the experiment, they are said to
be blind.
When the experimenter does not know if they are giving the treatment to the
experimental group or the control group, the experiment is double –blind.
The two approaches helps to control unwanted complications such as subject’s reaction
A researcher must be judicious in selecting the experimental design to employ. There are
several designs that can be used. They are:
Pre-experimental design
True experimental designs
Field experiments
Let us look at each one of them in more depth.
(a) Pre-experimental designs: In this category we have three types:
The one –shot case study: In which case there is treatment or manipulation of
independent variable and observation or measurement is done on the dependent
variable.
Example: A company would like to initiate a health and safety campaign about
improving working conditions without prior measurement of the knowledge the
employees currently have. The experiment would only reveal the knowledge the
employee would acquire but it would be difficult to evaluate the effectiveness of the
campaign. The lack of pre-test and control group makes this design inadequate for
establishing causality.
The one-group pretest-posttest design: In this case we have a pretest
(O),manipulation(X),post-test(o)
The Static Group Comparison: This design provides for two groups: one which
receives the experimental treatment while the other serves as a control.
The main advantage of the field experiment is that they permit the investigation of
complex interactions, processes and change in natural settings.
Their main weakness is the fact that experiments cannot control intrinsic and extrinsic
sources of validity as systematically as in laboratory experiments.
The main issue to consider in field experiment is the ethical issues: In most cases, the
individuals are not aware that they are participating in research. Therefore, the
researcher has to ensure that the privacy of the affected individuals is not violated and
that they will be protected from undue embarrassment or distress.
LECTURE TWELVE
Page 145 of 241
DATA PROCESSING AND PRESENTATION
Lecture Outline
12.1 Introduction
12.2 Objectives
12.3 Editing
12.4 Coding
12.5 Tabulation
12.6 Classification
12.6.1 Attributes
12.6.1 Class-Intervals
12.7 Presentation
12.7.1 Percentages
12.7.2 Tables
12.7.3 Histogram
12.7.4 Bar Charts (Bar Graphs)
12.7.5 Frequency Polygon
12.7.6 Pie Chart
12.8 Summary
12.9 References
12.1 Introduction
Data comes to us in a complex, usually incoherent, form. It has to be organized and
presented in a computer friendly mode for manipulation purposes unless the information
was directly recorded in the computer. Data processing involves data conversion
involving four primary methods. C. R. Kothari in his book, Research Methodology
(2008) has given a detailed description of the processing methods. In this lecture, I will
cover the main processing methods which are editing, coding and re-coding, tabulation
and classification. I will also discuss the basic methods of data presentation.
In this lecture, I shall discuss data processing and presentation. This important stage
usually follows the data collection process. At this stage, we go through the data
gathered and organize it for manipulation through a computer package like SPSS. It also
involves extensive editing of the data so that mistakes that we often make in the field are
We also look at our data afresh and decide on appropriate codes to be used depending on
the statistical package chosen. In most cases, we code our data during the data collection
stage but depending on the data collected, one may need to recode the information
afresh. We also deal with classification, tabulation and the various methods of
presentation.
12.3 Editing
Data editing involves some level of proof-reading which eliminates common mistakes. It
can be done in the field or in a more centralized environment like the researchers office
when field work is complete. At this stage, it is possible to eliminate common problems
like duplication of information, vague responses and other information that might
interfere with the outcome of computer analysis.
12.4 Coding
Coding is the process in which variables are noted in the form of symbols or numeric
characters. This helps to reduce the amount of data entry required particularly where the
information sought comprises attributes. An example is sex of animals in a given animal
species which is either male or female. One could use numerical values to represent
observations like 1 for male and 2 for female. In entering this information into the
Take Note
Coding is the act of assigning numbers or symbols to variables and values
for convenience.
12.5 Tabulation
12.6 Classification
Classification is the organization of similar objects into same classes in order to maintain
cohesion. If we are dealing with hotels as our source of data, we could classify them
according to facilits or according number of stars. When dealing with gender, we could
make two broad classes: one of males and another of females. Classifications in the
social sciences are based on attributes and class intervals.
12.6.1 Attributes
This is classification of categorical variables and values. For example, we may be
interested in the height of students in LDP 603. We have a choice of measuring the
height of students and giving exact measures but we could to use terms that describe
their height like: short, medium and tall. The latter terms are referred to as attributes or
descriptive characteristics. If we are to classify them, we would put all the tall students
together, the short ones together and so on. We can also do the same for colour, taste,
opinion, age (young, mature and old) and so on. Such classes are based on shared
characteristics of quality.
12.6.1 Class-Intervals
When we use continuous values like weight of students, we can decide to break the
values into convenient classes and compute the occurrence of each class. Let’s assume
that the weight of students in kgs is as flows:
45.5, 50, 55.4, 57, 60, 67.7, 77.2, 80.1, 85.2
Classification of this kind of data into class intervals will make it easier for us to
compute general variations in the population.
12.7.1 Percentages
In percentages, data is computed with a common base of 100 for comparative
12.7.2 Tables
Tables also represent summary information in an organised form which is easy to
visualise. The table can be made up of percentage or row data in rows and columns. For
Example, we can classify 5 hotels according to their number and type (domestic
/international) of tourists as follows:
The table gives a visual summary of the data on different types of visitors. From this
table, one can easily conclude that hotel H3 is popular than others and, like H2, it attracts
a large no of international tourists. On the other hand, H5 registered the lowest no of
domestic tourists.
12.7.3 Histogram
A histogram represents data in visual blocks with the frequencies being provided on the
‘y’ axis or vertical axis and the categories (observation variables) on the ‘x’ or horizontal
axis. The units on the horizontal size are similar in size with variations being observed
on the vertical axis.
The information on occupancy presented in the table above can be converted into a
histogram as follows:
Figure 1
An example a one week occupancy rate of HI tabulated below is represent in a Bar Chart
below:
70
60
50
40
30 HI
20
10
0
mon tue wed thur fri sat sun
80
70
60
50
40 HI
30
20
10
0
mon tue wed thur fri sat sun
fri
sat
sun
The methods given above are quite good for visual representation. They are however,
12.13 Summary
We have seen that data must be processed in order to ensure that the
succeeding stages of analysis are smooth. It is particularly important to
ensure that thorough editing of the data is a continuous process in order to
eliminate spurious information which could impact our results.
Activity
i) Why is editing of research data important?
ii) What differentiates class-interval from attribute based classification
iii) Outline the methods of data presentation in research.
iv) Why is coding important in data collection?
12.14 References
1. Bohrnstedt, G. and Knoke D. 1988. Statistics for Social Data Analysis.
2. Walsh A. 1990. Statistics for the Social Sciences.
3. Durkheim, E. 1964. The Rule of the Sociological Method.
4. Freedman, P. 1960. The Principals of Scientific Research
13.1 Introduction
Data analysis refers to the computation of certain measures along with searching for
patterns of relationship that exists among data-groups. In the process of analysis,
relationships or differences supporting or conflicting with original or new hypothesis
should be subjected to statistical tests of significance to determine with what validity
data can be said to indicate any conclusions.
Data analysis is essentially the main stage in research. Analysis helps us in interpreting
data, drawing conclusions and making decisions. In descriptive statistics, we are able to
present our finding in a concise manner and in inferential statistics we are able to
develop generalisations from the sample to the population.
Many methods of data analysis are available and our decision to use any of them depends
on the nature of problem being investigated, the nature of data, the measurements used
Broadly speaking, data analysis falls into two categories namely descriptive and
inferential analysis. Descriptive analysis describes the phenomena in statistical terms as
it happens or in an ex- post-facto sense. No attempts are made to make predictions or
inferences. It helps us summarise data and manipulate with ease.
Take Note
The term descriptive statistics stands for the procedures used in the
description of data.
Inferential data analysis moves a notch higher by drawing implications from hypothesis
testing and making estimations or inferences in terms theory. The ultimate aim is come
up with general laws that explain the phenomena of interest on the basis of a sample.
What happens to the sample can be applied anywhere else on earth where the population
bears the characteristics of interest.
Data comes to us as ratio, ordinal, nominal and interval. These measurements have an
impact on how data is analysed and the reliability of the conclusions reached. Ratio data,
with its absolute ‘0’, can be manipulated using a wide array of methods while nominal
data is comparatively weaker as we saw in the previous lectures.
In this lecture, we shall look at the different methods used in the manipulation of
univarite statistics in the search of differences within attributes. Bivariate analysis and
relationships will be explored in the succeeding lecture.
Arithmetic mean is also referred to as arithmetic average. In this module, and unless
specified, the term mean has been used interchangeably with arithmetic mean to mean
the same. It is the value that results from dividing the total value in a distribution by the
total number of items. In this section, we shall compute the mean for grouped and
ungrouped data.
Ungrouped data
For examples if you had a sample of 5 students in a class of tourism aged: 17, 17, 20, 21,
30 the arithmetic mean is: 17+17+20+21+30 = 105 =21
5 5
Grouped data
steps
1 Obtain the midpoint of each class
2 Multiply midpoint by its frequency
3 Sum up frequency (f)
4 Sum up fx
5 Compute as follows
Formula x= ∑fx
n
Where
x = the mean,
n = the number of observations,
x = the midpoint,
∑ = the summation notation
13.3.1.2 Mode
13.3.1.3 Median
The value of the middle item in a sequential distribution is normally known as a median.
Median normally divides the distribution into two equal parts; one part being higher than
the median and the other lower than it. If you are to arrange the ages of students in a
sequential descending order, it would appear like this: 31, 21, 20 17 and 17.
The middle value (median) in the distribution is 20. However, if the observations are
even like: 33, 31, 21,20, 17, 17 and, then the median is determined as follows:
Note
1. The median has a value of 22(.5) =11
2. The cumulative frequency shows that it lies between the intervals 20-22 and 23-24.
3. There are only 10 observations preceding class interval 23-24 (fbelow)
4. The class containing 11 is 23-24
Where
Md = median
L = Lower Limit of interval containing the median
f= frequency of the interval containing the medium
n = total number of observations
i = class interval range
fbelow= cumulative frequency below the class interval containing the median)
We obtain L=23
i=3
n=22
fbelow =10
f=2
Computation:
Measures of dispersion deal with variations in the distribution of data i.e. they measure
variability within data. Measures of dispersion vary considerably from the measures of
central tendency. The latter depends on single cluster points while in the former, each
value is weighted against the mean, mode or median to determine its distribution from
the centre either negatively or positively. Most researchers compute deviation from the
mean. The most commonly used measures of dispersion are:
Range
Interquartile range
Mean deviation
Standard deviation
Variance
In the following sub-sections we discuss each one of these measures using examples
from tourism to illustrate how they work.
13.3.2. 1 Range
Range computes the spread of data from the lowest to the highest number. It is
essentially the difference between extreme values.
Formula
Range = H-L
Where H=Highest value and L= Lowest value
The distribution of conference rooms among five hotels is:
1, 2, 5,10, 22
Compute the range:
The range is H(22)-L(1) =22-1=21
Interquartile range is centred on variations within the middle 50% of the values which
are more stable than the extreme values. Consider the following set of marks (out of 10):
4,4,6, 7,8,8,9 and 10
The range is 10-4=6.
The cumulative frequency of the values is 56.
The two quartiles are:
Formula
Q¹ The lower quartile (n+1) th value where is the cumulative frequency
4
Q² The upper quartile 3(n+1) th value or 3(n+1)/4th
4
Mean deviation is made up of deviation of sum of scores from the mean arrived at by
subtracting the mean from the observed scores. Because the summation of such values,
as we shall see, is zero we normally use absolute figures i.e. |x- x |.
Where | | allows one to treat figures as absolute
x = observations
x = mean
In mean deviation, every score set is utilized and the overall variation from the mean
is computed. The deviation scores i.e. difference between the individual score and the
mean (x- x ) are considered.
For heterogeneous population the difference between the mean and the actual score is
large.
A combination of the deviation scores is indicative of the spread of scores which is a
stronger measure than the cluster points characterising the measures of central
tendency.
The sum of deviations from the mean is zero hence the need to take absolute scores
using the symbol│ │ meaning that│-1 │ and │1 │ are the same i.e. negatives values
are considered as positives.
Example
In a course assessment test (CAT) marked out of 10, the performance of 8 students is
given below and organised in a sequence from the highest to the lowest. What is the
spread of the scores? Use mean deviation, standard deviation and variance to support
your answer.
From this distribution, we shall compute the mean deviation, the standard deviation and
the variance.
In mean deviation, the difference between each value and the mean is computed and all
the differences are summed up and divided by the number of observations. This
difference varies from the earlier measures in that all the values are computed
independently and then included in the final computation.
Formula ∑| x- x |
N
=14/8=1.75
Where
x = each observation
x̄ = arithmetic mean
N= the total no of observations
∑= notation of summation
In this formula, all the variations of values from mean have been considered and divided
by the total number of observations.
13.3.2.3 Variance
S = ∑(x - x )2 s= 40 = 5 = 2.24
n 8
Where
s= standard deviation of the sample
We should note here that the standard deviation is generated by squaring the mean
deviation and computing the square root. The figure generated gives us an indication of
the variation between observed values.
94.56% of values
Note: There are standard proportions of fixed values between the mean and units of the
standard deviations as shown above as follows:
34.13% of the values are represented within one standard deviation from the mean –
right (+s)
34.13% of the values are represented within one standard deviation from the mean –
left (-s)
Merging the two statements indicates that 68.26% of the distribution falls within the
mean ± 1s (± means plus or minus)
95.46 falls within the mean ± 2s
Example:
Assuming that the student performance discussed earlier takes a normal curve, we noted
that the arithmetic mean was 5 and the standard deviation was 2.24. Therefore, 68.26%
Z =x- x
s
Where
Z= number of standard deviation units
x = observation
x = arithmetic mean
s= standard deviation
13.4 In this lecture, we have looked at two primary methods of data analysis. In the first
Summary case we talked about the measures of central tendency which gave us single
measures explaining the relationship between data. But this misses out the
underlying differences.
This shortcoming is addressed in the measures of dispersion which looks at data
variations around the arithmetic mean, median and mode. The variation takes
cognisance of the sum of average deviation of each value. The lecture also looked
at the standard deviation and variance which are frequently used.
Lecture outline
14.1 Introduction
14.2 Objectives
14.3 Measures of Relationship
14.3.1 Regression Analysis
14.3.2 Spearman’s Rank Correlation
14.3.3 Pearson’s Moment Product Correlation
14.4 Multivariate Relationships
14.4.1 Control
14.4.2 Interpretation
14.4.3 Prediction
14.4.4 Cross-Tabulation
14.5 Summary
14.6 Activity
14.7 References
14.1 Introduction
In this lecture, I wish to take you through bivariate and multivariate data analyses. I will
particularly concentrate on the methods that are commonly used in research. I also wish
to mention that the number of methods used in statistical analysis is large. We cannot
cover everything in this lecture alone. For those of you who may have done a statistics
course unit, I believe that you have an advantage but the course unit, as a research
method, is aimed at equipping students with a good understanding of the range of
information available without necessarily converting them into a statisticians. You do not
In addition, I will mention hypothesis testing by highlighting some of the key concepts
directing inferential research.
14.2 Objectives At the end of this lecture, you should be able to:
1. Outline the main measures of relationship
2. Discuss linear regression analysis
3. Explain the working principles of the Pearson’s Moment Product
Correlation
4. Compute data on the basis of the Spearman’s Rank Correlation
5. Explain the meaning of a hypothesis
Y axis Y
B
A
X axis x
Y Y
D
C
x x
Key
Outliers
Xy coordinates
Best fitting line through the x,y coordinates
These kinds of relationship are common in research. In this study we shall concentrate on
The cause and effect relationships can be measured using the linear regression analysis.
Predictions can also be made.
Application
The relationship between the age of hotels and their daily maintenance.
The relationship between age of a person and their IQ.
The relationship between salary and service in years.
Relationship between economic decline and disposable income.
A maintenance officer employed by the Blue hotels in Kisumu wanted to establish the
relationship between the age of hotels and their daily maintenance expenses. This would
help the company in making a decision to purchase a 90 year old hotel which had been
advertised for sale in the local press. The question in his mind was what it would cost to
maintain such a hotel per day. Since there were hotels belonging to the same company,
they were analysed to determine the relation between age and maintenance cost as
illustrated below.
Because the relationship between the age of a hotel and its maintenance costs are likely
to be linear, a decision to use the simple linear regression was made using the formula
below:
Formula
y=a+bx
b = n∑xy - ∑x ∑y
n∑x2 - (∑x)2
a= ∑y - b∑x
n n
Critical information
N = 10
∑x = 300
∑y = 2,970
∑xy = 97,650
2
∑x = 10,050
Computation
(i) Lets find b
= 976500 – 891,000
100,500 – 90,000
= 85,500 = 2.8033
30,500
B = 2.8033
a = ∑y - b∑y
n n
a= 2970 - 2.8033x300
10 10
a = 297 – 84.099
a = 210.90
Formula
Rs = 1 – 6∑d2
n(n2-1)
Example 1
Critical Information
N=8
d2 = 12
Computation
Rs = 1– 6∑d2
n(n2-1)
Rs = 1- 60 = 1- 72
8(64-1) 504
Rs = 1-0.1429 = O.857
Example II
i) Compute Rs
ii) Comment on the strength of association
iii) Are the students consistent in their performance in the two course units
In the Pearson’s product moment correlation, we are interested in the degree of scatter in
a relationship and its strength. The less scattered the variables, the stronger the
relationship. We use r to represent the product moment coefficient.
As we mentioned and demonstrated earlier, the value of r lies between -1 and +1. If the
correlation is positive and the points lie in a straight line with a value of 1, then we call
this a perfect positive correlation.
Formula:
By looking at the formula we can see that we need the following items to calculate r
using the raw score formula:
Critical Information
N = 12 (Number of students)
X = 985 (Overall Score in Cultural Tourism)
Y = 849 (Overall Score in Eco-tourism
∑X2 = 83465
∑y2 = 63693
Formula
r= 867192-836265
1001580-970225 764316-720801
r = 0.8373 or 0.84
This shows a strong positive correlation in the performance of students in the two course
units and it indicates consistency in the teaching and examination of the two course units.
14.4.1 Control
Control is better achieved in the natural sciences than in the social sciences. However,
some extraneous variables are controllable in the social sciences. Control is usually
achieved through comparing different sub-groups and using cross tables to determine the
contribution of the extraneous to the main group. A good example would be the
relationship between the occupancy rate of a hotel and its geographic location. The
reputation of the hotel could interfere with the expected relationship.
Good location
The methods which we normally use to deal with this anomaly are known as control
methods. Control methods help us against spurious results. For example, in the
relationship between occupancy and location one could still get results without
14.4.2 Interpretation
14.4.3 Prediction
Prediction is the ability to foretell how relationships will be under different conditions.
Do you remember the example we used in the regression analysis? We were able to
predict the cost of maintaining a 70 year old hotel daily based on projection from other
similar hotels. Predication takes many forms depending on our observations
14.5 Summary We have looked at the various methods used in the computation of
bivariate data. I have particularly mentioned that the characteristics of a
variable will determine the suitability of the method used in its analysis.
However, it is also important to note that some methods are multipurpose.
At this point, I would also like to think that each one of us has a method
which s/he can use to get through statistical data analysis.
Most of the computational methods are better done in the computer
through an appropriate package like SPSS. But you must keep the
principles of each method at your fingertips in order to ably interpret the
computation. This was the main importance of this particular lecture.
I have also mentioned hypothesis testing which is a more advanced level of
data manipulation which you may not need at this stage. The aim was to
alert you to its existence. I will continue with it in the next lecture.
14.7 References Cooper, C.P. Faulkner, B. Fredline, and L. Jago. 2003. Processing
tourism research.
Freedman, P. (1960) The principles of Scientific Research.
Mendenhall W. (1981). Introduction to probability and statistics
Chase C.I. (1984). Elementary statistical procedures.
Fisher, R.A. (1958) Statistical Methods for Research Workers.
15.1 Introduction
Before a researcher engages in the details of their study, they usually search for the
literature to find out what has been written about the area or topic of their study. The
researcher will look for what experts in the field of study have found out. This kind of
reading is referred to as a “review of the literature”. In this lecture we are going to
discuss the steps a researcher goes through in conducting a literature review.
_______________________________________________________________________
15.2 Lecture Objectives
By the end of this lecture you should be able to:
1’Describe briefly the value of literature review.
2 State and describe the steps a researcher goes through in conducting a literature
review.
3.Explain the difference between a primary and a secondary source of literature
review
4. Describe and conduct both a manual and a computer search of the literature.
There are three main sources of literature review; the general references, the primary
sources and the secondary sources. Let us describe each of them.
i) General references: the general references are the first type of source a
researcher refers to. Such references help the researcher to know where to look
for other sources i.e. articles, monographs, books and other documents- that deal
directly with the research question. Such references are either indexes, which list
the author, title and place of publication of articles and other materials, or
abstracts which give a brief summary of various publications, as well as their
author, title and place of publication. For example, Psychological Abstracts is an
index commonly used in education.
ii) Primary sources: They are publications in which researchers report the results of
their studies. Most of the primary sources are journals, which are usually,
published monthly, quarterly, bi-annually or annually. The articles in them
typically report on a particular research study.
iii) Secondary sources: Secondary sources refer to publications in which authors
describe the work of others. The most common secondary sources are textbooks,
encyclopedias, research reviews and yearbooks.
There are two ways to do a literature search; manually and electronically. Let us look at
each of them.
There are several steps involved in a literature review. We are going to look at each one
of them:
i) Define the research problem as precisely as possible
A researcher should state the research question so that it focuses on the specific
issues for investigation. This will help the researcher to focus his search for the
needed information.
ii) Look at relevant secondary sources
After stating the research question in specific terms the researcher needs to look
through one or two secondary sources to get an overview of the previous research
that has been done on the problem.
iii) Select and pause one or two appropriate general reference works.
Once the researcher has reviewed the secondary sources to get a more informed
overview of the problem, he or she should have a clear idea of exactly what to
investigate. If the researcher is satisfied he or she can select one or two general
reverences to help identify particular journals or other primary sources related to the
problem.
There are very many general references a researcher can consult. However, it is also
important for the researcher to be very clear on the broad area of interest, for
example, education, psychology, economics etc.
Each academic discipline has its general source of information. Examples are:
current index to Journal in Education (CIJE); Reader’s Guide to Periodical
Literature; Social Science Citation Index (SSCI); Psychological Abstracts; Resources
in Education (RIE); Sociological Abstracts etc.
iv) Formulate search terms (key words or phrase) pertinent to the problem or
The researcher should list the key words alphabetically and then consult the general
reference work to see what articles are listed under these descriptors. The researcher
would then select the article that is relevant to the research topic.
vi) Obtain and read relevant primary sources, notes and summarize key points in
the source.
Once the search in the general reference has been done, the researcher will have
generated a pile of bibliographic cards. The next step is to locate each of the sources
lists on the cards and then read and take notes on those relevant to the research
problem.
Reports are also an important primary source. Many research projects produce a final
report of their activities and findings. These, reports may not be published, but are a
valuable source of information.
You will find out that most primary source materials are located in journal articles and
reports. If you are looking for them in a library, then it is important that you go to the
section dealing with journal articles in the library.
After you have gathered all the articles you want to refer to, then the literature review
process begins in earnest. It is good to begin with the most recent article and reports
backwards.
Though there is no perfect way of reading an article, the following are the basic steps
you should follow:
Read the abstract or the summary first to ascertain whether it is worth reading
the full article.
Record the bibliographical information at the top of the note card.
Take notes on the article or photocopy the abstract or the summary.
Be as brief as possible in taking notes. Note the most important points only.
A computer literature search has almost the same steps like in a manual search. The
following are the steps:
i) Define the problem as precisely as possible: Like in the manual search, one
should state the research problem as specifically as possible so that the
relevant descriptors can be identified.
ii) Decide on the extent of the search: It is important for the researcher to
decide on the desired number of references to obtain. The level of the article
being prepared will determine this. For an ordinary journal article, a review of
20 to 25 articles would be adequate. For a master’s degree, 30 to 40 articles
will do. But for a very exhaustive review for Doctoral dissertation, one may
iii) Decide on the Database: There are many databases available for literature
search. The most commonly used in education is ERIC. For effective use of a
database, one need have clear descriptors.
iv) Select the descriptors: The descriptors are the words the researcher uses to
tell the computer what to search for. Too general a descriptor may lead to too
many references, many of which may be irrelevant. Too narrow a descriptor
may lead to too few references.
v) Conduct the Search: After you have determined which descriptors to use the
next step will involve entering the descriptors into the computer. The
computer will give very many references and it is upon you to decide the
number you want.
The World Wide Web (www) is part of Internet and it is a vast reservoir of information
on all sorts of topics in a wide variety of areas. A researcher can use a Web browser (the
computer program that lets you gain access to the www.) to find information on almost
any topic with just a few clicks of the mouse button. You will find that some of the
information on the web has been classified into directories, which can be easily searched
by going from one category to another. There is also the search engine available.
i) What is a directory?
Directories will group websites together under similar categories such as Universities,
Pharmaceutical companies etc. The result of a directory search will be a list of websites
related to the topic being searched.
Directories often provide an excellent starting point for a review of literature.
After a researcher feels that he/she has read and reviewed enough of the literature, it
becomes imperative that a final review is prepared. The format will typically involve the
following:
The Introduction: It briefly describes the nature of the research problem and states
the research question. The researcher will state what section of the literature
review led him/her to investigate the problem, and why it is an important
question to investigate
The body: This section reports what others have found out or thought about the
research problem. It is good to discuss all the related studies together under sub-
headings. In most cases, several studies that reported similar results are grouped
In this lecture, we have discussed the following points: That there are six (6) essential
steps involved in the literature review:
defining the research problem as precisely as possible
perusing the secondary sources
selecting and perusing an appropriate general reference
formulating search terms
searching the general references for relevant primary sources
obtaining and reading the primary sources and noting and summarizing key
points in the sources.
There are three basic sources of information: general references, primary sources and
secondary sources. “Descriptors” are the key words researchers use to help them locate
primary sources.
There are five essential points that researchers should record when taking notes:
Problem
Hypotheses
Procedures
Findings
Conclusions
The literature review report consists of:
An introduction
The body of the review
A summary
The researcher’s conclusions
A bibliography
1. Which of the general references, would you consider on each of the following?
Marriage and family counselling
Secondary school management
Small school discussions
PhD thesis dissertations
2. Which of the secondary sources would you recommend for the following topics?
Recent research on the integration of ICT in education.
A brief overview on poverty eradication in Korogocho slums in
Nairobi.
A survey of students’ attitude towards mathematics.
3. List down the steps you would take to review an article in a given journal
publication.
Lecture outline
16.1 Introduction
16.2 Lecture objectives
16.3 Ethics in research
16. 4 Ethics in Research
16.5 Deception in research
16.6 Summary
16.7 Activity
16.8 Suggestion of further readings
16.1 Introduction
Ethics in research should be an integral part of the research planning and implementation
process, not viewed as an afterthought or a burden. There should be increased
consciousness of the need for strict ethical guidelines for researchers. Some of the ethical
issues
touch
on
decepti
on and
invasio
16. 2 Lecture objectives n of
By the end of this lecture you should be able to: privacy.
i) Define the term “ ethics”
ii) Describe how you would deal with deception in research There
are three main ethical principles that need to be considered:
Beneficence: Maximizing good outcomes for science, humanity, and the
Most professional associations prohibit the use of deception unless it can be justified and
the effect of the deception “undone” after the study is completed. The “undoing” of
deception is supposed to be accomplished by the following:
Debriefing the research participants after the research study, which means
that the research explains the real purpose and use of the research
You will note that the emancipatory paradigm emerged because of the dissatisfaction
with research conducted within other paradigms that was perceived to be irrelevant to, or
a misrepresentation of, the lives of people who experience oppression. There are three
characteristics of the emancipatory paradigm with ethical implications for
methodological choices:
Traditionally silenced voices must be included to ensure that the groups
marginalized in society are equally “heard” during the research process and
the formation of the findings and recommendations.
An analysis of power inequalities in terms of the social relationships involved
in the planning, implementation, and reporting of the research is needed to
ensure an equitable distribution of resources (conceptual and material)
A mechanism should be identified to enable the research results to be linked
to social action: those who are most oppressed and least powerful should be at
the canter of the plans for action in order to empower them to change their
own lives.
When the research is cross-cultural, it is important that cross-cultural ethical standards
are developed to guide researchers while conducting research in other communities.
Cross-cultural ethical principles require collaboration between the researcher and the
host community. It also requires that the researcher communicate the intended research
agenda, design, activity, and reports with members of the host community. The research
should be designed in such a way as to bring benefits to the host community and to foster
the skills and slf-sufficiency of the host community scientists.
The paradigms considered here are certainly not exhaustive. New paradigms might come
in the future. However, what is crucial is that researchers should be aware of their basic
beliefs, their view of the world (their functional paradigm), and the way they influence
16.Actvity
Have ever participated in research as a subject or a researcher assistant? If so what
ethical behaviour did you like from the researchers? List down some behaviour that you
think did not conform to ethical standards of research? If you were to be the principal
researcher of the project what specific ethical behaviour would you observe?
17.1 INTRODUCTION
_____________________________
______________________________
_________
______________________________
______________________________
i) Letter of Transmittal
The letter refers to the authorization
for the project and any specific
instructions or limitations placed on
the study. It should briefly state the
purpose and the scope of the study.
v) Table of contents
This section lists the main sections
and their respective page number(s)
vi) Introduction
The introduction prepares the reader
for the report by describing the
various parts of the project. These
include:
The problem statement- the main
focus of the investigation.
The purpose of the study - It states
very clearly what the researcher
ix) Findings
This is a very important section and
its objective is to explain the data
rather than to draw interpretations or
conclusions. It is important that the
findings are presented in numbered
paragraphs. Use of tables, charts,
and graphics is encouraged.
x) Summary and Conclusions and
recommendations:
xi) Budget
xii) Appendices
This section deals with complex
tables, statistical tests, supporting
documents, copies of forms and
questionnaires, detailed descriptions
of methodology, instructions to field
officers and any other evidence
important to the report.
xiii) Bibliography
If secondary sources of information
17.5 Summary
In a nutshell the final report is organized as follows:
Introductory Section:
Title page
Table of Contents
List of figures
List of Tables
Declaration
Abstract
Acknowledgement
Dedication
Acronyms
Abbreviations
Procedures:
Description of the research design
Description of the sample
Description of the instruments
Explanation of the procedure followed (the what, when, where, and how of the study)
Discussions of internal validity
Description and justification of the statistical techniques or other methods of analysis
used
Findings:
Description of findings pertinent to each of the research hypothesis or question
Every institution has its own preferred structure of a research proposal of report. The
standard structure followed by most universities would take the following format.
Preliminaries
Cover page
Declaration
Abstract
Table of contents
List of figures
List of tables
Acronyms
Abbreviations
CHAPTER ONE: Introduction
Background to the study
Statement of the problem
Purpose of the study
Objectives of the study
Research questions/ Research hypothesis
Justification of the study
Significance of the study
Scope of the study
Assumptions of the study
Limitations / Delimitations of the study
REFRENCES
APPENDICIES
TIME FRAME
A timeframe provides each activity and the estimated time it would take to complete.