CHAPTER ONE

INTRODUCTION
Objectives

By the end of this chapter students should be able to

-Define research
-State the purpose of research
-Outline motives what motivates doing a research
-State and identify Scientific and Non Scientific method of acquiring knowledge

RESEARCH DEFINED

1. Is a systematic, objective analysis and recording of controlled observations

that may lead to development of generations, principles or theories resulting in
prediction and possible ultimate control of events

2. Is an objective formal process of systematically collecting, analyzing and

interpretation of data for the purpose of providing functional information to managers
for decision making.

3. Is a fact finding activity

Conclusions which can be derived from the definition

1 Is an intensive and purposeful search for knowledge and understanding of a
phenomena and establishing of facts, theories and laws
2 Is a method of discovering the truth
3 Is not haphazardly but is done systematically and takes time.
4 Involves collection of data using acceptable research tools on which data is recorded
5 Is done with a purpose so as to solve a problem

PURPOSE OF RESEARCH
Is a fact finding activity and its purpose is to discover answers to questions through the use of
certain pre-defined procedures. These procedures were developed over time in order to increase
the likelihood that information that is gathered is relevant, reliable and unbiased.

Research helps managers to make better informed and less risky management decisions.
Accordingly the information obtained through research must be objective, impartial, current,
translatable and relevant.
MOTIVES (REASONS) FOR CARRYING OUT A RESEARCH
a) Desire to face challenges in solving the unsolved problems
b) Desire to get intellectual joy of doing some creative work
c) Desire to get a research qualification together with the subsequent benefits. In some cases
it may be a course requirement being studied e.g. HND, PHD etc. The research is not self
initiated by a requirement that a student should fulfill to get a full qualification.
d) Desire to offer service to the society
e) To fill the gap in knowledge
f) Government directive – Research can be directed by the government to try and get a
solution for the nation or society
g) Employment conditions – There are some established research stations that specializes
in research. In such research stations people are employed to carry out research e.g.
Grasslands Research Station which specializes in Agricultural Research
h) Organizational problems may motivate research e.g. a company may have high rate of
staff turnover

The research Process in brief

Basically, the research process follows a series of steps as follows:
-Defining the problem and set research objectives
-Developing the research plan(proposal)
-Collecting the data
-Analyzing the data and derive information
-Presenting of findings

*Research is one of the methods of acquiring knowledge.

METHODS OF ACQUIRING KNOWLEDGE

Ways of acquiring knowledge can be grouped into:

a) Non scientific methods

b) Scientific methods

a) NON SCIENTIFIC METHODS OF ACQURING KNOWLEDGE

-Are subjective methods of acquiring knowledge.
-Depend mostly on human judgment
-Issues addressed incorporate the existence of God, whether a person is good or
bad and issues relating to moral values.
- Issues addressed do not require investigation and are even difficult to verify or
investigate
-Mostly used in social research
*Social research is the collection of methods and methodologies applied systematically to
produce scientific based knowledge about social world
-no variables can be deduced to be tested

2
Non scientific methods include:

i) Appealing to custom and tradition

ii) Appeal to authority
iii) Mystical methods
iv) Rationalistic method/ use of personal experience
v) Empirical method
vi) Common Sense
vii) Media Myths

i) Custom and tradition

-Is the earliest human search for truth or knowledge which has been

passed from generation to generation since time immemorial. Specify how things
have been done since time immemorial
-Reference to custom and tradition is used to solve problems
-Determines or dictate people’s conduct in societies, living style, food, language,
articulating problems, making judgments, specify human food and speech etc
-Is a special case of authority of the past

People obey custom and tradition avoiding as much as possible to violet them and
seek explanations to various situations and problems through custom beliefs.

Advantages
-Is the simplest way of acquiring knowledge
-Knowledge acquired may be universally accepted
-May result in standard approach to problems

Disadvantages
-Traditional knowledge although once true may be come distorted as it is passed
on and may no longer be true
-People may clinch to traditional knowledge without real understanding it
-May instill development of new ideas

ii) Authority method

This method depends upon acquiring knowledge and wisdom from prominent
people who are considered to have vast social and environmental knowledge more
that an ordinary person
-Is mostly used by a child to acquire knowledge from parents and also by elders
from village heads, chiefs, priests, presidents etc
-Is based on faith inputted into a person

3
 *Man appeal to tribal wise- man and to gods in times of crisis. Tribal chiefs
remain the supreme authority and considered as all-knowing and all-competent
supermen
-Chiefs are sought for guidance and advice in social, religious, political and even
private affairs. Moreover, priests in churches assumed the role of authority,
people turning for knowledge, advice, explanations, domestic problems etc

-Authorities spend time and effort to learn more things (researching) to remain up
to date and others may benefit from their experience and work

Advantages of Authority
-Is a very cheap, quick and simple way of learning something
-Not seriously dangerous because people placed in authority will rely on
continuous researches and strategies to justify and preserve their positions, using
professional jargon and emphasizing the uniqueness of one’s e.g. It is true
because I the Minister …….
-Do not require carrying out a sophisticated research in order to understand the
world and events

Disadvantages of Authority
-May result in individuals hiding their superficiality of their knowledge until they
are in authority
-May create conflict in the society due to struggle for power
-Authorities may speak on fields they know little about explaining things in the
wrong way thereby misleading people.
-Authority may be frequently misused or abused
-May result overestimating the expertise of other people when they are not
-Overdependence on experts or authority keep others in the dark, loosing ability
to make own meaningful judgments
-Authorities may only promote ideas that strengthen their power and position

iii) Mystical method

Is when or where the correctness of the knowledge is assumed to reside in a
supernatural source.
-One’s ability to appeal to irrational feelings by using rituals, ceremonies,
emotionally loaded situations and explanations and using unusual languages is
necessary and remain from being questioned e.g. Medium spirits, traditional
healers, prophets etc

iv) Rationalistic Method/Personal experience/Self Knowledge

This methods is based on human reasoning, ability to think logically and
discovery of laws and principles using pure, abstract intelligence in solving
problems

4
 -Is when man thinks that he cannot refer his difficulties or problems to others but
should try to solve them by himself. Uses his experience and trials to solve
problems not outsiders
-What a person, personally sees or experience is accepted as true, has a strong
impact and is a forcefully source of knowledge. Based on seeing and
experiencing is believing.

-Observation of reality and collections of facts using the five human senses is
necessary

-Rationalistic may lead to astray due errors of personal experience which are:

a) Overgeneralizations Are statements that go far beyond what can be justified

based on the data or empirical observations that one has experienced. Occurs
when one has evidence that he/she can believe and assume that is applies to
many other situations. Is caused by misjudging.
b) Selective observation is making observations in a way that reinforces pre-
existing thinking rather than observing in a neutral and balanced manner. Is
determined or governed by preconceived ideas.
c) Premature closure Is making a judgment or reaching a decision and ending
an investigation before one has the amount or depth of evidence required by
scientific standards. Operates with and reinforces overgeneralization and
selective observation. Occurs when one feels has all the answers and do not
need to listen or seek more information or raise more questions. A person or
researcher take few pieces of evidence (information) and believe to have all
necessary information to make a decision
d) Halo effect Is allowing prior reputations of persons; places or things to
color (overshadow) one’s evaluation rather than evaluating all in a neutral,
equal manner. E.g. People do not scrutinize what is said or written by a
respected person, highly educated or from a prestigious university or college

v) Empirical method
-Is where facts observed in nature are the only elements to rely on
-Is not based on theories but experience and observable factors
-What is observable and can be perceived by human senses constitute knowledge
e.g. A wood floats on water but a metal does not but surprisingly a boat made
with metal cannot sink

vi) Common Sense

Is form of ordinary reasoning. Is when people rely on what everyone knows and
what makes sense to be the truth. What is believed by an ordinary reasonable
person to be good and true is taken without any further analysis.

Disadvantages

5
 -Can allow logical fallacies to slip into thinking eg if wind blows eastwards it will
rain
-Contain contradictory ideas that may go unnoticed because people use the ideas
at different times eg Unlike poles attracts and birds of the same feather flock
together.

vii) Media Myths

Television shows, movies, newspaper and magazine articles are important sources
of information about social life. The main advantage of this method is that what
is learned is permanent

Disadvantages
-What is portrayed distorts reality (do not accurately reflect social reality) e.g
movies (eg flying vehicles), crimes, sports(eg wrestling)
-Tend to perpetuate the myths of a culture
-People are misled by visual images more easily than other forms of lying (Has
powerful effect on people)
-Is a forum in which competing interests try to win public support (propaganda)

b) SCIENTIFIC METHODS OF ACQUIRING KNOWLEDGE

This is known as scientific research or scientific thinking. Therefore research is regarded
as scientific thinking

Scientific thinking is regarded as the men’s unending search for truth which has brought
him to the stage of science as the main source of evidence

-Refers to a systematic investigation of a question, a phenomena or a problem using

certain principles. Considered as the integration of empirical and rationalistic methods of
acquiring knowledge because is based on observations made during experiments
followed by reasoning ability.

Types of Research
Scientific enquiry methods are descriptive research, experimental research or quasi-
experiment research. This classification is based on the nature and approach of the
research activities

i) Descriptive research (Comparison research)

Is a research intended to produce an accurate description of a variable under
study ,relevant to the problem being faced without necessarily demonstrating exactly
the degree of relationship that exists between variables.

6
 Attempts to determine the frequency with which something happens or the extent to
which 2 or more variables are related, which would permit only inference to be
drawn about causation.

Ranges from simple data gathering to investigating possible relationships between 2

or more variables, but without determining to what extent are variables related, so
that predictions of future events can be possibly made

Is a research which the primary purpose is to “paint picture”, using words or

numbers and to present a profile, a classification of types or an outline of steps to
answer questions such as who, when, where, and how. But mainly focus on how and
who questions

Research is conducted to describe a well-defined subject accurately. Outcome is a

detailed picture of the subject eg 80% of subjects.

A great deal of social research is descriptive research.

Most data-gathering techniques are surveys, field research, content analysis and
historical- comparative research

*Descriptive and exploratory research have many similarities and are blurring
together in practice

Purpose of Descriptive Research

-To provide a detailed, highly accurate picture of a thing or problem
-To locate new data that contradict past data
-To create set of categories or classify type of situations
-To clarify a sequence of steps or stages
-To document a causal process or mechanism
-To report on the background or content or a situation

ii) Experimental research

Is a type of research intended to demonstrate the cause-end effect relationship
between variables being studied
An experiment entails a test which allows the researcher to determine the effect of an
independent variable on a dependent variable and controlling other external factors
that might influence the outcome. The researchers have direct control over at least
one or more independent variables and determine the effects of the independent
variable on the dependent variable

NB a) A variable is an empirical property that is capable of taking two or more

values, thus, can change in value or king eg water can be solid, liquid, or
gaseous

7
 *Attributes are the categories or levels of a variable

b) Constant is a property that cannot take more than one value e.g
citizenship of a specified country

c) An independent variable is the one which can influence the value or state
of another variable. Is a factor in an experiment over which the researcher
has some control and manipulates its value, this being expected to have
some effects on the dependant variable. Is a cause variable that produces
an effect or results on a dependent variable in a causal hypothesis

e) Dependent variables are factors whose values are influenced by the value of
other variables. This is a response measure understudy in an experiment
whose value is determined by the independent variable e.g. height of a child
depends on age
-Is the effect or result variable that is caused by the independent variable in a
causal hypothesis

d) Intervening variable is a variable that is logically or temporally after the

independent variable and before the dependent variable, and through which
their caual relation operates. is a variable which appears in more complex
causal relationships. Comes between the independent and dependant
variables showing a link or mechanism between them. Acts as a dependent
variable with respect to the independent variable and also acts as an
independent variable towards the dependant variable. Found in theories with
multiple independent, intervening and dependent variables.

NB Casual relationship requires only an independent and dependent variables

iii) Quasi-experimental research

Sometimes known as semi experimental research
Is an experiment which does not possess the pre-requisite of control. It lacks the
control attributes of a truly experimental research. It takes short-cuts in experiment
requirements thus it lacks valid demonstration of variables’ association but still it is
not descriptive because it does not yield some quantitative indicators of association
between variables

iv) Exploratory research

Is one in which the primary purpose is to have a little understanding of an issue or
phenomenon, to develop preliminary ideas, to have an insight into the problem and
move towards refined research questions by focusing on the “what” question. The
goal is to formulate more precise questions that future research can answer. Can be
the first stage in a sequence of research studies. Can be done to know appropriate
designs to use and execute a second more systematic and extensive study

Purpose of Exploratory research

8
 -To become familiar with the basic facts, setting and concerns
-To create a general mental picture of conditions
-To formulate and focus questions for future research
-To generate new ideas, conjecture or hypothesis
-To determine the feasibility of conducting a research
-To develop techniques for measuring and locating future data

v) Explanatory research Is a research in which the primary purpose is to elaborate,

extend or test thories. Its emphasis is to explain why something happens although the
issue is known and there is description of it. Looks for causes and reasons.

Purpose of Explanatory research

-To elaborate and enrich a theory’s explanation
-To extend a theory to new issues or topics
-To support or refute an explanation or prediction
-To link issues or topics with a general principle
-To determine which of several explanations is best
-To test a theory’s predictions or principle

STEPS IN THE SCIENTIFIC THINKING PROCESS

1 Location and definition of a problem through exploratory research techniques

2 Survey of past experiences with the problem and available data to get ideas about the
past and possible future solutions and methods of investigation (Review of related
literature)
3 Formulation of a hypothesis
This represent a tentative solution to a problem under investigation and employed as a
guide in the collection of additional data which may lead to an accepted solution to
the problem

Mental elaboration of the hypothesis checking for agreement with facts, verifiability
and logical consistency
4 Collection of data i.e. additional data by means of research tools, measurements,
observation, experimentations and interviews
5 Analysis , classification and summarization of the data collected
6 Formulation of new generalization representing observed uniformities, explanatory
principles or scientific law.

7 Formulation and communication of recommendations

CHARACTERISTICS OF RESEARCH/SCIENTIFIC THINKING

1 Research requires expertise, systematic logic and an accurate investigation

9
 Researcher should:
a) be an expert in the field of research before undertaking an investigation
b) plan procedures carefully and detailed eg no prediction before description and
explanation of observed phenomena
c) gather, record and analyze data with complete accuracy
d) standardize and ensure that valid data gathering tools are devised and used
e) ensure that mechanical means are employed to improve the accuracy human
observations, recordings, and computations of data

2 Is based on facts and empirical events

Framework is constructed basing on facts which are observable leading to the
formulation of important theories. Research is aimed at knowing reality and each
step is based on observation

3 Employs methods of analysis

This involves breaking down of a phenomenon into small manageable parts and
attending to each part separately to enable the investigator to give each part separate
and undivided attention. This ensures full appreciation of the problem as a whole

4 Employs hypothesis/ predictions to guide thinking process.

Experience is used to show that two phenomena are related and follows one another

5 Employs quantitative methods in the treatment of data

Scientist/researcher does not rely upon estimates, guess and general impression but
measures a phenomena under consideration as accurately as possible

Results are recorded in units that can be counted, added, subtracted, multiplied and
divided

Make use of new and exact language of mathematics to make the analysis,
classification, summarization of data more reliable

Fact are summarized into mathematical values such as means, medians, modes,
deviations etc which are less ambiguous than ordinary language terms

6 Employs objective measurement

Scientists measures, weighs, tests and accurately determine amount of phenomena
under investigation

7 Free from emotional bias

Research demands alertness, flexibility and open minded ie mind free from emotional
bias. The researcher needs to be alert and watch for new ideas and ready to modify
old ideas in the light of new evidence and never allow personal likes and dislikes
coloring the facts

10
 Personal feelings and preferences should be prevented from research activity. These
argues with a scope and relevance of data

8 Research gathers new knowledge or data from primary sources . Research endeavours
to reach first hand sources of data and not from second-hand sources

9 Requires patience, unhurried and courage

A researcher has to work patiently towards sound conclusions and need to realize that
significant findings do not come as a result of hurried and careless procedures

10 Research is highly purposive

Research deals with a significant problem demanding solution(s). Usually places
much emphasis upon the discovery of general principles and scientific generalizations
that can be applied to the solution of wide range of problems

11 Is carefully recorded and reported

Terms of references and unique technical terms should be clearly defined for easy
understanding by readers.
-Procedures are described in detail
-Limiting factors are recognized
-References are documented
-Results are objectively recorded
-Conclusions and generalizations are cautiously arrived at considering limitations,
data collected and errors minimized
-Non essential details are omitted

12 Is replicable and transmittable

Anyone placed in the same circumstance can observe the same event and make
reasoning leading to the same explanations and predictions because of logistic and
objectivity.

13 Is reductive (Reductionism)
Complexity of reality is reduced by grasping the main relationship of laws. No
essentials are omitted. Necessities and chances are distinguished

Disadvantages of Scientific Method of acquiring knowledge

-Is not possible to observe systematically every phenomena
-Is expensive in terms of time and resources ie money and instruments
-Some characteristics remain unknown so that explanations and predictions cannot be
accurate.

CLASSFICATION OF RESEARCH

A- Basic research or fundamental research

11
 -referred to as pure research
-Carried out to provide broad and general principles, theories, results or
information
-Aimed at discovering basic truth laws on which action research may be based on
-Not carried out to provide specific information or for a specific problem hence
not concerned with day to day solutions but to provide basic information in general eg
educational research on aims of education, motivation, environment etc
-Results are public
-Used to advance fundamental knowledge about the social world
-Is a source of most new scientific ideas, methods, theories, ideas used in applied
research.
-Ask questions that seem impractical eg causes of cancer in chickens
-Stimulates new ways of thinking about deviance etc
-Provides foundation for knowledge and understanding that are generalized to many
policy areas, problems and studies
-Demand higher level problems, procedures, tools and expertise
-Acts like an exploratory research aimed at gaining ideas and insights about a
research problem.
-Basic research can take 2 forms:
i) Discovery of new theories
ii) Development of the existing theory

B Applied research or field research

-Is a research designed to offer practical solutions to a concrete problem or address the
immediate and specific needs of clinicians or practitioners. Is that type of research work
of which its results find direct application in the field of work. It is devoted to problems
of field work and other affected individuals
-Used to answer a policy question
-conducted for a specific problem
-Results are private
-carried on a small sample
-is based on application of known theories derived from basic research
-is used to test theories and laws discovered through basic research and applying them
into practice i.e. application of known theories e.g. use of learning devices (aids),
textbooks, library
-Is sometimes known as decision making research because is used for decision making
-Rely on quick, small scale study that provides practical results that provides practical
results that people can use in the short term
-Primary audience or consumers are teachers, counselors, practitioners, decision makers
(managers) etc
-Its results may generate conflicts if the desired outcome is not expected

B Action research
-Is similar to applied research in many ways, hence regarded as another form of applied
research with the primary goal of facilitating social change or bring about a value
oriented social change.

12
 -The difference is that applied research is carried out on a larger sample resulting in more
universal applicable findings while action research is conducted on an immediately
available small sample in order to solve the immediate problem for the same group
-Is a research conducted on a small sample to solve immediate problem for the same
group on which applied research is being done.
-Is the process by which practitioners attempt to study their problems scientifically in
order to guide, correct and evaluate their decisions and actions
-Primary concerned with field workers eg teachers to try to improve own teaching and
also by school administrators to improve administration behavior.
-Represents a more disciplined, objective and scientific approach to a problem
-Tries to keep problem solving in close touch in reality at every stage and focused on the
immediate application of know laws
-Deals with really practical problems

Characteristics of action research

-Those who are being studied participate in the research process
-Research incorporates ordinary or popular knowledge
-Focuses on power with a goal of empowerment
-Seeks to raise consciousness or increase awareness
-Is tied directly to political action
-Used to advance or improve conditions by expanding public awareness

*Researchers try to equalize power relations with research subjects, avoiding more
control, status and authority that those they study.

C) Evaluation research
-Is an applied research in which one tries to determine how well a program, theory or
policy is working or reaching it goals and objectives. Is the mostly used applied research
-Ethical and political conflicts may arise in this research because people may have
opposing interests in findings of a program or theory. Is part of the administration
applied by many organizations
-Types of evaluation research are formative or summative evaluation

Formative evaluation is built- in monitoring or continuous feedback on a theory or

program.

Summative evaluation is one that looks at the final program or theory or end result

D) Social impact research

Is an applied research that document areas of social life if a major new change is
introduced into a community. Estimate the consequences of a planned change.

*It should be therefore be concluded that Action, Evaluation and Social impact researches are
types of applied research.

13
Differences between Basic and Applied Research
ASPECT BASIC RESEARCH APPLIED RESEARCH

i) Primary Audience Scientific Community & Other Practitioners, Participants,

researchers Supervisors, Non- researchers

ii) Evaluators Researcher peers Practitioners , Supervisors

iii) Autonomy of High Low moderate

researcher

iv) Research rigor Very High Varies, Moderate

v) Highest Priority Verified truth Relevance

vi) Purpose Create new knowledge Resolve a practical problem

vii) Success Indicators Publications, Impact on Direct application to address

knowledge or Scientists specific problem or concern

Research can be classified as quantitative or qualitative research

Quantitative research is a research which looks at a large group of cases,people ,units and
measures a number of features. Can be a single point in time (cross-sectional) and multiple time
points (longitudinal research). Collects hard data, that is in form of numbers. Measure variables
and test hypotheses that are generalized to the entire population

Qualitative research is a study or research which focuses on one or few cases during a limited
time period. Is used in or as a case study. Collects soft data, that is, data in form of impressions,
words, sentences, photos, symbols e t c s.

Quantitative research distinguished from Qualitative research

Quantitative research Qualitative research

Measure objective facts Construct social reality, cultural meaning

Focus on variables Focus on interactive processes and events
Reliability is key Authenticity is key
Value free Values are present and explicit
Theory and data are separate Theory and data are fused together
Independent of context or situation Situational constrained

14
 Many cases or subjects are considered Few cases and subjects
Employs statistical analysis Employs thematic analysis
Researchers are detached Researcher is involved

CHAPTER TWO

OVERVIEW OF THE RESEARCH PROCESS

This is just a rundown of the process or procedures when doing a research

1) PROBLEM AND OPPORTUNITY

A problem arises when a decision maker faces uncertainty regarding which action to
adopt in a situation. Or is a situation where management needs to makes decision but
do not have enough information to make a sound decision. Posses a negative
connotation

An opportunity is a situation with a positive potential if recognized and capitalized, is

profitable to the enterprise. Possesses a positive connotation

Researcher must identify and define the research problem or opportunity, sometimes
known as research question. Distinction between the two is difficult, hence research
problem is used to refer to either problem or opportunity.

Research problem may come from the desire to solve a current problem or pursue an
opportunity

2) PRELIMINARY STUDIES
-Is exploratory research to gain insights and ideas about a problem
-Undertaken in order to:
a) Discover the problem and define it
b) Analyze the problem’s environment
c) Determine correct definition of the problem
Common preliminary research includes desk research, observation and discussion.
Data is collected through discussion and observations

Other necessary techniques are expert opinion, focus groups, secondary data analysis,
etc

3) DEVELOPMENT OF HYPOTHESIS
-Hypothesis is a suggested answer to a problem, which is tested, and empirical
-Gives direction
(For details refer to detailed notes on hypothesis

15
4) STATEMENT OF AIMS, OBJECTIVES, DELIMITATION AND
LIMITATIONS
a) Aims are broad goals of the research. Sometimes are called institutional
objectives
b) Objectives are the specific clearly goals that are developed from aims. Are the
precise and measurable goals to be achieved by the research process.. Are also
specific tasks the researcher carries out in order to achieve or test the aims
c) Delimitations are statements of limits or scope of the investigation. Determines
the study boundaries. Indicates what the researcher should include or exclude
d) from the investigation. Outline breadth and depth of issues to be covered e.g. the
entire industry or part of it. Defines radius of the research, timeframe of the
research process, and time to be covered by the research process.

5) REVIEW OF LITERATURE
-Is a summary and analysis of related literature, which provide background of the
research. Researchers look into theories put by other experts, and authors and
analyses their model and make opinions. Contain the following:
a) Conceptual view of experts
b) Comparisons of the views from experts
c) Nature of issues being researched on

6) DESIGNING A RESEARCH PROPOSAL

Is a research plan. Is a formal written document that discusses the problem for
research. Acts as a marketing in a research process

Purpose of research proposal

a) State the purpose of the study
b) State how the research will be executed by:
-Outlining research methodology to be used e.g. interviews
-Outlining in detail resources to be used that is time, finance, human, &
material
d) Provides the researcher with a guideline in carrying out the research work

7) DECIDING ON A METHODOLOGY AND JUSTIFYING IT

The researcher should decide on research methodology and justify why he or she
opted for each

8) PRESENTATION AND ANALYSIS OF DATA

- collected data is edited, coded, summarized and tabulated before the statistical
designs. Testing of hypothesis is done and relationship noted, with characteristics
revealed

9) INTERPRETATION OF DATA
Is done derive meaning from data after analysis. Findings are deduced. These should
be supported
10) CONCLUSIONS are deduced from findings in light of objectives and hypothesis

16
 11) RECOMMENDATIONS BASSED ON FINDINGS AND CONCLUSIONS
-Are based on findings and conclusions
- Are formulated course to action to solve the problem
-Should be specific, applicable and justified

12) PRESENTING OF RESEARCH FINDINGS

Is the reporting or communication of research results either orally or in writing

RESEARCH TOPIC OR RESEARCH QUESTION

A research topic/ problem is the starting point in any research. A topic is too broad for
conducting a study. This must be narrowed down to or focused on a specific research question
that can be addressed in the study. Topic can be narrowed or focused during exploratory stage,
planning and gathering data, therefore is an on going process. As already alluded a research
problem may result from an opportunity or problem.

Techniques for narrowing a topic into a research question

1 Examine the literature – published articles are an excellent source of ideas for research
questions. These articles suggest questions that focus on:
-Replicating of previous research project exactly or with slight variations
-Exploring unexpected findings discovered in previous research
-Following suggestions an author gives for future research at the end of an article
-Extending an existing explanation or theory to a new topic or setting
-Challenging findings or attempt to refute a relationship
-Specifying the intervening process and consider linking relation

2 Talk over ideas with others

The researcher should ask people who are knowledgeable about the topic for questions
that they have thought. Seek out those who hold opinions that differ from yours on the
topic and discuss possible research questions with them.

3 Apply to a specific context

Focus the topic onto a specific historical or time period. Narrow the topic to a specific
society or geography unit.

4 Define the aim or desired outcome of the study

-Will the research question be for an exploratory, explanatory or descriptive study, will
the study involve applied or basic research.

Sources of topics or ways to select topics

17
 1. Personal experience, knowledge and interest-One can choose a topic based on
something that happens to him or her. Also may choose a topic only because has
interest in that area or topic or thorough knowledge in that area e.g. a doctor can
research related to medicine

2. Curiosity based on something in the media. One can read a newspaper, magazine
article, see TV program and left with questions. What is read or seen raises questions
and suggest researches

3. The state of knowledge in a field – Basic research is driven by new research

findings and theories that push at the frontiers of knowledge.

4. Solving a problem- Applied research topics begin with a problem that needs a
solution

5. Personal values – people highly committed to a set religious, political or social

values select topics to further their knowledge and understanding of those values

6. Everyday life- topics can be found throughout everyday life in old saying, novels,
songs, statistics etc

7. Social premiums- topics can be selected because are “hot” or offer an opportunity eg
reward for conducting a research etc

Identification of a good research problem is one of the most difficult phases of any research
project. In the discovery of a problem, the investigator can take the help of the following
procedures so as to save from wild search and aimless wondering:

 Select field of specialization and become a scholar in that specialization as early as

possible
 Develop a concern for that field
 Study critically the available research literature on the field of specialization
 Accept the difficulties or obstacles of that field as a challenge and try to exercise own
originality and ability to tackle them
 Should be prepare a record of problems already studied by others in the area of
specialization
 Should acquaint him or herself with a research under way in that area
 Should analyse the trends in his/her field of specialization
 Acquaint himself or herself with the possibilities, commitments and prophecies in that
particular area
 Should widely discuss any problems that comes to his mind

Criteria of a good research problem/topic

Before the proposed research problem can be finalized, several conditions and considerations
have to be satisfied. Although there are no standard rules that will guarantee the suitability of a

18
research problem, a number of criteria in the form of conditions might be listed for guidance in
the selection of a topic.

1 Novelty
It should be sufficiently original so that it does not involve objectionable duplication.
Originality is the basic credit point of any research. Ignorance of prior studies may lead
to students spending time on a problem already investigated by some other workers.
Moreover, the study should employ the most recent data. However, the fact a problem
has been investigated in the past does not mean that it is longer fit for study. There is
constant need for verification of the findings of previous investigations using other
methods or newer or better devices

2 Interesting
Problem should be motivating and interesting for the investigator. If not interesting for
him or her, will not be able to face and overcome the obstacles which may come at every
stage in the research. If it seems to him dull and boring, there is little hope that he would
do justice to it. His interest should be purely intellectual and should not be there only for
a reward, material benefits, advancement in position, increased authority etc

3 Important
Research problem should be significant enough and involve an important principle or
practice. If it is not worthwhile, if it neither adds to knowledge nor leads to any
improvements in the current practices, it would be in vein (vain)

4 Feasibility or amenability
The suitability of a plan for a particular research worker is the matter of its feasibility.
Researcher should be able to carry it out to a successful conclusion. Should possess the
special qualifications, training and experience and should have the required competence,
knowledge and understanding
.
5 Availability of data
The research worker should ensure the availability of valid and reliable data, data
gathering devices and procedures, although confidential, sensitive and classified
information may not be possible to obtain

6 Availability of cooperation
The study may require cooperation from various institutions, authorities and individuals.
Researcher must make sure that necessary permission and cooperation will be readily
available

7 Availability of guidance
A research activity needs the patronage of a guide and the approval and sanction of a
competent authority , that is, sponsorship guidance from a qualified member of the
faculty

8 Availability of other facilities

19
 The should availability of necessary financial resources to carry out the study.
Reesearche should be able to meet the expenses involved in data gathering, equipment,
printing, test materials, travel, computerization, clerical help, postage, etc plus enough
time

9 Immediate application
The research should help in solving an agent problem

10 Aim of the research

The aim of the research also influence the selection of the problem

11 Level of research
The nature and scope of a study will be determined in the light of levels, like Master’s
Degree, HND, PHD etc. It may simply be an action research or a research to produce a
research paper or an experimental project.

12 Experience and creativity

Good research problem stem from a clear understanding of the theoretical, empirical and
practical aspects of the subject derived from personal experience and from thorough
review of the literature. Another major contributor to the wise choice of a problem is
creativity and the other personality factors that make for are originality, flexibility,
initiative, ingenuity, and foresight.

13 Courage and confidence

Researcher should have the courage and determination to pursue the study in spite of the
difficulties and social hazards that may be involved. Should be willing to risk the
criticism, suspicious or even opposition that a delicate or a controversial study may arise
etc

Common errors in selecting and formulating a research problem are:

a) Naming a broad field or area of study instead of a specific problem e.g. Higher Education
System in Zimbabwe is too broad
b) Stating it such that the investigation is impossible e.g. huge costs required
c) Narrowing or localizing a topic, such that it becomes too small or insignificant from
research point of view e.g. A critical study of the poems in book
d) Including it in terms of an unscientific, emotional or biased nature e.g. “The blessing of
teaching profession”
e) Lack of precision in the instruments. The tools, tests, or devices which are proposed to
be used in data collection may not be precise enough resulting in another error.

HYPOTHESIS
20
-Is a tentative assumption drawn from knowledge and theory used as a guide in the investigation
of other facts and theories that are yet unknown
-Acts as a temporary solution to a problem or suggested answer which is tentative, concrete and
testable

CHARACTERISTICS OF HYPOTHESIS
-Is a guess
-Is in agreement with the observed facts
-Does not conflict with any law of nature, which is true
-Be stated in simplest possible terms and clear
-Serves to explain facts
-Must be specific not generalize the problem
-Should be conceptually
-Guide the search for new truth
-Must have empirical events and or evidence
-Looks forward
-Must be testable with available techniques
-Must show expertise and stated in scientific and research terms

FUNCTIONS OR IMPORTANCE OF HYPOTHESIS

a) Provides direction by:
-Defining what is relevant and what is irrelevant in any research process
-Directing the collection of data and state what the researcher is looking for
-Prevents collection of useless or excess data and review of useless data
*Should precede the gathering of data

b) Organize the efforts by giving direction

Provide a framework for collecting, analyzing and interpretation of data
It is a guide to the thinking process and the process of discovery. Act as researcher’s eye
into the world of darkness

c) Places specific goals before the researcher, which enables the researcher to select
samples and procedures to meet the goals. Prevents blind research

d) Save as framework for drawing conclusions by making possible the interpretation

of data
e) Enables the researcher to understand the problem clearly and data required. Clarify
necessary procedures and methods in solving problems

f) Sensitizes the individual to facts . It focuses research

g) Link together facts and information and organize them into comprehensive whole

h) Guide the search for new truth and thinking process

i) Set boundaries and indicate limitations

21
SOURCES OF HYPOTHESS
i. General culture is source of ideas, theories and laws of existing problems e.g.
prejudices against women education.

ii. Personal experience. Hypothesis can emanate from personal experience eg.
Teacher’s punctuality enhances student’s punctuality

iii. Scientific theories and laws can be put into use. Sound mind in a sound body

iv. Analogies – formulated basing on likenesses and similarities between objects.

Similarities means something in common

THE DISCOVERY OF HYPOTHESIS

To discover or formulate hypothesis the researcher should have been:
i) Well read and informed about the problem
ii) Examined with care features of the objects. Be alert to new and incidental features of
the objects and obvious features. Look with care ordinary and extra-ordinary aspects
of phenomena object
iii) See beyond the superficial aspects of objects. Search for vital relationships and
underlying principle conditions and uniformities
iv) Recognize the importance of analogies in the formation of hypothesis
v) Recognize the importance of thinking about things

DIFFERENT FORMS (TYPES ) OF HYPOTHESIS

A) Directional and No-directional form

i) Directional form – Indicates the way in which the dependent variable will be
influenced by the independent variable. Is preferred because is more precise
and gives more information.
ii) Non- Directional form is opposite of directional

*Use of directional and non-directional depends on the ability to predict future

relationship between variables

B) Null and Alternative Hypothesis

i) questiona (HO)
-States that no relationship exists between variables being studied nor
significance difference will found between the treatment. Results of studied
population group will not show any differences whatever factors will be
measured.
-Assumes that all groups are equal
-Does not reflect the researcher’s expectations

22
 -Better fitted to statistical techniques and defines the mathematical model to
be utilized.
-Is the original hypothesis
-Is a positive hypothesis which assumes that everything is ok and not worth
investigating

ii) Alternative Hypothesis (HA)

-Reverse of HO
-Can be either directional or non-directional
-States that the differences shown in results reflect real differences between
population groups.

C) Prediction form
-Allows the researcher to state principles, laws, etc expected to merge from the
study or experiment
-Mainly used in action research

D) Declarative form
-States the relationship between the variables concerned

E) Question form
-Ask whether there is a relationship between variables concerned.

A RESEARCH PROPOSAL
Is a research plan, which is a formal written document that discusses the problem for research.
Acts as a marketing tool in a research process
Is a description of work to be performed
Helps to convince funding agencies to fund and support the research therefore is a persuasive
tool which should be factual, accurate and complete.
Reveals the image, characteristics, qualifications, and ability of the researcher in doing the
research process
Organize ideas and thoughts of a research process and communicate these effectively
Should be easy to understand and comprehensive, be in simple direct language and in writing.
Unique terms be clearly defined
Should be justified, be specific not vague
Reveal what should be expected from the research that is its benefits to the funding agency

Steps in Proposal writing

1. Identify the need for research , that is, define the problem through exploratory
research techniques
2. Critical analyze the need for the project
3. Plan and design the research
4. Estimate the research costs and time
5. Write a research proposal

23
Characteristics of a good research proposal
i) Should communicate the importance of the research
ii) Bears an idea that is worthwhile and account part of community problem
iii) Should make it clear that the applicant has made a suitable approach to the problem
and that the plan of action is likely to succeed
iv) Should indicate that the research falls within the area of interest of the sponsor by
being properly delimitated
v) Should indicate the probable results of the proposed project results so as to justify the
expected expenditure
vi) Should indicate `that the researcher has the required capabilities , knowledge of the
area of research
vii) Should clearly indicate the delimitation (scope) of the research project

Purpose of research proposal

a) State the purpose of the study
b) State how the research will be executed by:
-Outlining research methodology to be used e.g. interviews
-Outlining in detail resources to be used that is time, finance, human, &
material
a) Provides the researcher with a guideline in carrying out the research work

ELEMENTS OF A RESEARCH PROPOSAL

1. Title of the research project. This should be specific not general and not
too broad. Should be not more than 15 words

2. Background of the study: addresses the environment of the problem. It

outlines the history of the company, its objects, and partially introduce the
problem

3. Statement of the problem. The problem should be clearly defined,

quantified; effects on the operations of the organization should be
highlighted. The research proposal should clearly, specifically and
physically define the problem. The problem should be separated from
symptoms

4. Sub- problems. Are problems contributing to the main problem. These

can be framed in form of questions

5. Aim of the study- what is to be achieved by the research. Aim is

developed from terms of reference of the research. Is a broad goal of the
research process?

24
6. Objectives of the study
These should be specific, measurable, attainable (achievable), realistic,
and time bound, hence people talk of SMART objectives

Hypothesis
This is a tentative solution or answer to the problem. (Refer to detailed
notes)

7. Assumptions are conditions taken for granted and on which the research
depend on.

8. Importance or significance of the study- relevance of the research

should be highlighted both to the researcher and the client company. Are
the benefits of the study

9. Delimitations of the study-refers to boundaries or scope of the research.

Specifies what, who, when etc to be included in the research process

10. Limitations are obstacles or difficulties to be faced in the research

process.

11. Definition of key terms – only unique terms are defined. Ordinary words
or meanings are not necessary

12. Review of related literature – categories of books to be used should be

presented

13. Research methodology. This discusses how the research is to be done.

Discusses the processes the researcher uses to meet the stated research
objectives.

a) Research design is a programme to guide the researcher in

collecting, analyzing and interpreting observed facts. The research
designs to be used include::
i) Exploratory research designs. These are case studies,
depth studies, experience survey, secondary data analysis,
and background analysis
ii) Conclusive research designs, which includes descriptive
research design, experimental, quasi experimental,
evaluative research design
These should be fully discussed and justified
b) Types of data :required and to be sourced should be clearly
disclosed and identified. These include Primary, secondary and
tertiary. Data can be qualitative or quantitative
c) Sources of data. Researcher should identify and explain sources
of required data. These should be clearly specified and quantified

25
 d) Data collection methods or plan. This refers to how the data is to
be collected. Methods include interviews, observation and
abstraction.
-Interviews can be personal interviews (face to face), mail
interview (questionnaire) telephone or telephone interviews etc.
-Observation can be mechanical or personal observation, which
can be participant or non-participant
The researcher should identify and justify methods to be used.
Specify to whom these are to be applied, how many etc.
e) Research instruments or tools. The researcher should explain the
research instruments (tools) on which data is to be recorded during
collection. These should be quantified; method of distribution
(administration) cleared, and the subjects to who there are
administered identified and quantified. The tools include
questionnaire, interview schedule, observation form, and checklist,
action forms.
f) Population of study. The researcher should define the population
of study, which is refers to all elements or objects to be possibly
considered in the study. Is the totality of subjects of investigation.
Include both human beings and non-human beings
g) Sampling frame is the list of element of the population from
which the sample is to be drawn.
h) Sampling versus Census. The researcher defines whether is to use
census (all population elements) or sampling (part of the
population) should justify the reason for his or her option. For
population elements more than 30, sampling can be done.
i) Sampling methods. The researcher to select, explain and justify
the sampling designs to be employed. Sampling methods are
probability and non-probability. The objects (subjects) to which
these are to be applied should be specified. Should also explain
how these are to be done that is sampling steps
j) Determination of sample size. The researcher should specify how
the sample size is to be determined. This can be mathematical
approach, judgmental approach etc
k) Sample size is a list of sample elements. This should be indicated

14. Data presentation and analysis

The researcher should outline how data is to be presented (tables, graphs,
charts) and statistical designs (computations) and other tests to be
applied on summarized data to test hypothesis

15. Research Project budget

Budget is a plain expressed quantitatively. The proposal to include
estimated project costs, time for activities, human resources, and material
resources

26
 16. Prepared by, approved and supervised by. There should be an
indication of the person s who prepared, approved and supervised the
proposal

It should be noted that the research proposal is plan, should be in future

tense, single line spacing and at least 3 pages and not more than 5 pages

RESEARCH DESIGN
Is a plan of how to proceed in determining the nature of the relationship between variables.
Is a formal written set of specifications and procedures for conducting and controlling the
research project.
Is a form of the framework or strategies of the entire research process
Act as a programme to guide the researcher in collecting, analyzing and interpreting observed
facts relevant to the variables understudy
The researcher must decide and justify the research design after formulating the problem

What is the difference between research design and methodology?

Research design depends on:

-Value of the information provided by alternative courses of action
-Requirements of the research objectives and hypothesis

Characteristics of a good Research Design

-Should ensure that information obtained is relevant to the research problem
-Should be capable of managing or detecting errors
-Should collect data objectively and economically using less costing procedures

Types of Research designs

Research designs can be classified into:
a) Exploratory research designs
b) Conclusive research designs

DIFFERENCES BETWEEN EXPLORATORY AND CONCLUSIVE RESEARCH DESIGNS

ISSUES EXPLORATORY RESEARCH CONCLUSIVE RESEARCH
1 Objectives To provide insights and understanding -To test specific hypothesis and examine
of a problem relationships

-To provide information for decision making

2 Characteristics -Information needed is loosely -Information clearly defined

defined
-Research process is flexible and -Research process is formal and structured
unstructured
-Sample is small and may not be a -Sample is large and representative of the
representative of the population Population

27
 -Data analysis qualitative -Data analysis is quantitative

3 Findings -Tentative -Findings are conclusive

4 Outcome -Generally followed by exploratory -Findings used as input is decision making

or conclusive research

Exploratory research may define hypothesis which is then tested by conclusive research, but a
by-product conclusive research may be a suggestion of new opportunity or a new problem. The
following diagram suggests the process.

VAGUE PROBLEM

EXPLORATORY RESEARCH

DEFINITION OF HYPOTHESIS
AND RESEARCH PROBLEM

CONCLUSIVE RESEARCH NEW IDEAS

DECISION MAKING

In summary, research design, can be defined as the arrangement of conditions for the collection,
analysis and interpretation of data in a manner that aims to collect data relevant to the problem.

A EXPLORATORY RESEARCH DESIGNS

All research projects of any significant scope must begin with exploratory research. This
preliminary
phase is absolutely essential in order to obtain a proper definition of the problem at hand.
Exploratory research studies are not characterized by formal research design, nor are there
basically
very scientific in nature. The purpose is to bring the central problem into focus.

28
 -Used to gain ideas and insights. Insights is the capacity to understanding hidden truths
especially
of situations or characteristics
-Useful in breaking broad, vague problems statements into precise simple small and easy to
understand ones.
Objective is to provide insights and understanding of the problem, that is, to discover ideas,
insights
and tentative explanation of the problems

Characteristics of exploratory research designs

-Information is loosely defined
-Research process is flexible and unstructured
-Sample is small and not representative of the population
-Analysis of data is qualitative
-Often the front end of total research project
`
Findings or Results of Exploratory research designs
-Temporary or tentative
-Not applicable in decision making but in understanding the problem
-Outcome is followed by further exploratory or conclusive research

Exploratory designs are used for the following purposes

-Formulating or defining a problem for precise investigation
-To develop hypothesis.
-To identify alternative courses of action when solving a problem
-Establish priorities for further research
-Isolate variables and relationships for further investigation
-To gather information about the practical problems of carrying out research of particular issues
-To gain insights for developing an approach to the problem
-To clarify unclear concepts
-To familiarize the researcher with the problem

Encourages the development of creative, alternative research strategies. Provides guidance as to

how of the entire research programme, eg sampling methodology to use
Act as a preliminary and used to lay down principles and theories

COMMON TECHNIQUES OF CONDUCTING EXPLORATORY RESEARCH

1. Secondary data analysis

This refers to the process of searching for and interpreting existing information or
existing literature relevant to the research problem. It takes the form of abstractions from
research done by other researchers, publications in newspaper, journal and magazines,
company reports etc

29
 Secondary data analysis is the core of exploratory research because there are many
benefits of examining secondary data and the costs are less. Also known as literature
search/ review/ survey

2. Experience surveys or executive opinion or expert opinion

This refers to gathering of information from those thought to be knowledgeable in the

issues relevant to research problem. Includes the knowledge and experience from people
familiar with a general subject being investigated. It is in the form of interviews with
people having knowledge about the general subject being investigated. Where
discussions are held with company executives, executive opinion is obtained and where
discussions are held with experts, expert opinion is obtained. The term expert is used in
this context to include anyone with knowledge relevant to problem being addressed.

3. Case studies
This refers to a reveal of available information about former situations having similarities
to the research problem. Usually, there are few research problems that do not have some
similarities to some situations in the past. Even when the research problem deals with a
radically new product, they are often similar past experience which may be observed. By
studying a small number of cases, it is possible to obtain a good understanding of the
relevant characteristics that might be operating in a given decision, situation of the broad
inter-relationship that might exist in a specific decision situation.

Since a case study is an in-depth investigation based on a small number of cases, it helps
clarify the nature of the problem, identify relevant variables and show relationships
between variables. The analysis of selected cases involves examining existing records,
observing business activities and conducting unstructured interviews.

4. Focus groups
These are small groups of people, about 6 to 12, brought together and guided by a
moderator through an unstructured discussion for the purpose of gaining information
relevant to the research problem. While a focus group should encourage openness and
everybody’s participation no member must be allowed to dominate the group discussion.
The moderator’s task is to ensure that the discussion is focused on some general interest
and objective of the research, while the conversation is free-wheeling. The purpose of a
focus group is to learn what people think about some specific aspects of any business
under research.

Is a useful technique for gathering information from a limited sample of respondents.

The information can be used to gather ideas or to gain some insights into basic needs and
attitudes of respondents.

Screening questionnaires can be used to obtain information from focus group members.
These questionnaires set qualifications a person must possess to be in the focus group.

30
 A discussion guide being as used as an agenda establishes the plan of the focus group
interview, including topics to be covered and to some extent time allocation.

Advantages of focus groups

- Allow the client to get first hand, close-up picture of the largest population
- Provide more complex answers about the set of relevant issues than a structured
questionnaire
- Are flexible

Disadvantages of focus groups

- It is impossible and expensive to assemble and reach geographically spaced persons and
working class.
- Cannot provide hard and fast conclusions.
- Can result in quarrelling if not controlled
- Results cannot be projected to a whole population

4. Depth study/interview
Often referred to as “One on One”

Depth study is a form of exploratory research to gain insights and understanding of a

problem. This is done through an intensive unstructured, direct and personal interview in
which a single respondent is probed by highly skilled interviewer to uncover underlying
motivations beliefs, attitudes and feelings on sensitive issues or topics.

Useful when dealing with confidential, emotionally and embarrassing matters and when
group interviews are difficult to schedule for the target population. The depth interviewer
must follow the following rules:

- Avoiding appearing superior or condescending

- Use familiar words
- Ask questions indirectly and informative
- Do not accept “yes or No” answers
- Should remain detached and objective

6. Projective techniques

Is an exploratory technique in which participants are allowed to project themselves into a

situation and then respond to specific questions regarding the situations. Used when
respondents cannot can not or will not communicate their feelings and belief directly.
This provides a structured questions format in which respondents can respond indirectly
their own feelings and beliefs. The most common projective techniques are:

(a) Word association. Respondents are presented with a list of words, one at a time and
asked to indicate what word comes immediately to mind.

(b) Sentence completion. Respondents are asked to complete an incomplete sentence.

31
 (c) Unfinished scenario story completion. Respondents complete the end of a story

(d) 3rd person or role playing. Respondents are presented with situations and asked to
relate to feelings and beliefs of a third person to the situation, rather than to express
their own feelings and beliefs about the situation.

(e) Cartoon completion test. Respondents are presented with a cartoon of particular
situation and asked then to suggest a dialogue that cartoon character might make in
response to the comments of another cartoon character.

7. Situation investigation

Consists of scanning or over viewing both the internal environment and the external
environment of the organization to identify potential, political and practical consequences
of the research project.

Determine whether the research is being proposed to discredit on individual, to prove a

point, to support a decision, postpone a decision etc.

- Result in a description of possible research project boundaries on terms of broad time

and dollar resource requirements.
- Provides guidelines for further avenues of both formal and informal inquiry.
- Major purpose is to place a problem in perspective and this is accomplished by posing
a service of “ what if question”
- Should conclude with an impact statement which self form one potential impact of
one resulting information on one problem, organization and its general environments.

Back ground analysis

Is a type of exploratory research that aims at generating ideas and insights about the
problem.
Can be carried out prior to, during or after problem identification
Background analysis activities require the researcher to understand the topic by
undertaking extensive reading and participating in discussions with numerous
individuals both in and outside the organization. Case studies, experience surveys
situational
investigation and desk researches are considered to be background analysis activities.

Enables the researcher to determine why and how the problem arose

2) CONCLUSIVE RESEARCH DESIGN

Is a design with the major emphasis of demonstrating the true explanation of problem
Its objective is to test specific hypothesis and examine relationship between variables under
investigation

32
 Characteristics of Conclusive research design
-Information needed is clearly defined
-Research process is formal and structured ie is guided by principles and procedures
-Sample is large and representative of the population
-Data analysis is quantitative

Findings of Conclusive research

Are conclusive that is, final and used as input is decision making

Types of conclusive research designs

Conclusive is classified depending on objectives as Descriptive Research, Experimental,
Evaluative, Correlation and Historical research design

A) Descriptive Research design

Its objective is to have a clear and accurate description of characteristics and functions of
variables relevant to the decision or problem being faced without necessarily demonstrating
to what extends some relationship exists between or among the variables

Emphasis is on determining the frequency with which something occurs and extent to
which two variables co-vary, which would only permit inferences being drawn about
causation.

Descriptive research ranges from simple data gathering to investigating possible

relationships between two or more variables so as to make predictions about future events

It also includes the use of techniques such as comparisons, contrasting measurements,

classification and evaluation.

-Is guided by the initial hypothesis, that is the null hypothesis

-Require a clear specification of the who, what, when, where, or how of the research
-Is more of qualitative and subjective in nature

Is used for the following purposes

-Description of the characteristics or functions of variables relevant to the decision being
faced
-Estimating precisely percentage of the specified population units
-Determining how relevant variables are related or associated
-Making specific predictions

Descriptive research may be classified as time dimensi0n in research as follows:

True Panel
i) Longitudinal designs

33
 Omni Bus Panel

Field study
ii) Cross-sectional designs
Sample survey

i) Longitudinal designs
-Is a multiple time points measuring . Measures or examine features of subjects at more than
one time
Enable changes that took place over a time to be noted
Is an investigation involving repeatedly measuring of a fixed sample of elements
throughout a considerable time. The sample is called the panel

A panel is a fixed sample of elements which remains relatively constant through a time,
maintained by replacing drop outs

Types of panels are

a) True panel which is a fixed sample of elements which is measured repeatedly over
time in respect of the same variables
b) Omnibus panel is a fixed sample of elements which is measured repeatedly overtime
on variables that change from measurement to measurement. Measured variable
is not re- measured
Types of longitudinal research are time series , panel and cohort research.

a) Time series research is a longitudinal research in which information can be about

different cases or people in each of several time periods. The same type of
information is collected from a group of people or units across multiple time
periods. Information can be very revealing of time changes

b) Panel study is a longitudinal research in which information is about the same

identical cases or people in each of several time periods. The same people, group or
organizations are observed across time periods. Different information is gathered. Is
the most powerful type of longitudinal. Is formidable to conduct and costly.
Tracking the same people is difficult because people die or relocate. Results of
panel study may be very valuable.

c) Cohort study is a longitudinal research in which information about a category of

cases or people that shared a common experience at one time period is traced
across subsequent time periods. Is similar to the panel study but observes not the
same people but people who shared similar life experience in a specified time period
is studied.

(ii) Cross-sectional designs

34
 Is an investigation involving measuring the select sample once at a single point in time. Is a
single point time measurement. There is one time measurement of the sample. No need for
maintaining the sample. Is the simplest and least costly research design.
Can be a single cross sectional or multiple cross sectional

Cross-sectional designs can be:

a) A field study which is an in-depth investigation of a few cases of the targeted

population, emphasizing the interrelation of a number of factors

b) A sample survey or field survey which is investigation of the population sample in

which the emphasis is on the generalization of summary statistics such as averages
and percentages

Cross – sectional designs cannot capture changes over time. It can be exploratory or
explanatory design but is more of descriptive research design.

Techniques used in conducting descriptive research

a) Abstraction which is the use of secondary data
b) Survey is a cross-sectional investigation of a sample which is a representative of the
target population
c) Panel as discussed
d) Observations which a systematic watching and recording of behavioural patterns of
variables to obtain information about the issues of interest

B) Experimental research design

-Also known as a casual research
-Refers to observations under controlled conditions.
-Researcher manipulates and controls one or more variables and observes the dependent
variables for variation concomitant to the manipulation of the independent variables
-Determine cause and effects relationships dependent and independent variables
-Because of the control, provide more convincing evidence of causal relationships than
exploratory and descriptive research
-There is need to identify other conditions

Types of experiments

i) True experiment or laboratory experiment

Is an investigation in which the researcher creates a situation with exact
conditions so as to control some and manipulate other variables
Investigator can observe and measure the effects of the manipulated
variables while the effect of other factors are minimized

ii) Quasi experiment

35
 Is an experiment which lacks the control aspect of other variable. Is an
experiment which does not possess the pre-requisites of control. Lacks the
control attributes of truly experimental research and takes short cuts in
experiment requirements. Thus quasi experiment research lacks valid
demonstration of association of variables. It is not descriptive because it
does not yield some qualitative measures of association of variables

Ex-post facto research is a type of experimental research. This is whereby the

research has to use experiment information which is already exists

C) Evaluative research design

Is a conclusive design aimed at testing interventions and their effectiveness
-Represents an important means of linking actions and their outcomes
-Is very specific in its purpose of accessing the results of a programme
-Evaluative research may be used in social intervention as:

i) Diagnostic tool
Help the people implementing an intervention to identify neglected areas, neglected
targeted group and problems with organizations and programme

Diagnostic tool evaluations are designed to inform the researcher and project managers
about:
-the present situation within the organization
-highlighting current problems trends forces, resources and any possible
consequences of various interventions

Is a technique for gathering crucial information and is usually carried before a

project is designed

ii) Formative tool

-Designed to promote the effectiveness of a programme
-Relates to the development and implementation of a programme
-Aimed at shaping a programme to achieve greatest beneficial impact upon the
organization and the community
-Evaluate the programme in order to improve its performance
-Form part of the initial programme planning and is an on going process
-Formative evaluation can be used at regular intervals.
-Mostly used in pilot study or survey

NB Pilot study is the testing of the actual programme on a small sample taken from the
community for whom the programme is being planned. This enables the researcher to
identify the materials, methods, difficulties to be faced, accuracy and appropriateness of
instruments eg questionnaires

iii) Summative tool

36
 -Set out determine the extent to which programme meet their specified aims and
objectives. This is done at the end of the programme
-Used to gain credibility with targeted communities

Process of summative evaluation research occurs according to the following steps:

i) Identification of a programme’s aims and objectives
ii) Formulation of aims and objectives in measurable terms
iii) Construction of research instruments
iv) Evaluation of study designing and data collection
v) Reporting back

The use of Evaluative research is determined by

-Characteristics of the problem
-Initial level of knowledge of the researcher
-Properties of the variables
-Purpose of the investigation

D) Correlation research design

-A type of a conclusive research between descriptive and experimental research
-Researcher is able to use this design to state a hypothesis, express the relationship
between at least two variables
-Results obtained provide more than just a description of reality
-Used when relationship between variables cannot be determined precisely and co-
variation is expected
-Does not give an explanation of why variables vary
-Its purpose is to detect the existence of relationship between variables and suggest a base
for causality
-Is useful as a first step in experimental research
-Does not imply causation but causation implies correlation
-Can also used to assess the type of strength of the relationship between 2 variables

-The correlation is the value -1 < r < 1 inclusive , where

r = 1 indicates a perfect positive correlations
r = -1 indicates a perfect negative correlation
r = 0.5 indicates a positive correlation

0.5 < r >1 the relationship is said to be strong relationship

0<r > 0.5 represents a weak positive correlation
-0.5 < r > 0 is a weak negative relationship

r = 0 means no relationship

E) Historical research design

Is the application of scientific method of enquiry to historical problems. Is viewed in 3
ways

37
 i) Historical approach refers to the organization of historical facts in support
of the new concepts to be developed
ii) Historical subjects refers to the biography of great man, places, sketches
of ideal, thoughts and traits
iii) Historical technique refers research which is conducted on the basis of
historical records and documents

Purposes of historical research design

-To gain and arrive at an accurate, clearer perspective of the past and the present problem
-Greater appreciation of the culture in the progress of the society
-Effects of certain past practices may suggest future actions
-To avoid past mistakes
-To understand to the deep rooted causes of present day problems
-Develop ability to locate, analyze and approve historical events and understand
limitations of evidence

Characteristics of a good research design

-Adopt a strategy for managing errors
-Supply accurate and error free information
-Should be economic in terms of time and costs

Errors that may affect Research designs

Errors affecting research designs may be classified into
a) Sampling errors are errors resulting from the chance specification of the population from
elements accorded to the sampling plan
b) Non- Sampling error refers to all other errors associated with the research project.

Sampling errors

1) Population specific error

Result from the selection of an inappropriate population or wrong population to obtain
data

2) Sampling error
-Arise when the probability is used to select a sample and the sample may not be a
representative of the population concerned
-Is affected by the homogeneity of subjects. The more the homogeneity, the smaller the
sampling error. Increase in sample size, sample error is reduced

3) Selection error
-Also known convenience error
-Results from a sample selected by a non probability method
-Investigators select only respondents who are mostly accessible and agreeable or friends
especially if there is a gratitude for being included in the sample or only available at the
time of research. Result in bias error.

38
4) Sampling frame error results from failure to account for all the elements of the
population or including other twice. Results from poor definition of a sampling frame

5) Non response error

Is failure to get valid response from the sampled respondents. Affects validity of the
survey because some segments of the population might be omitted from being counted
or heard Also exists where obtained sample differs from the originally selected sample
or expected sample. Types of non-response errors or response rates are:

a) Non location error is failure to find a sampled respondent. Results in location

rate which is percentage (%) of located respondents in the sample who are
located over sampled elements. Location rate can be improved by better
sampling frames, use of maps and telephone directories
b) Non Contact error – happens when respondents were not at home or nor reached
after many attempts. Researcher not able to contact all or some sample
members Contact rate is the percentage of located respondents who are
contacted. Can be improved by many repeated calls and varying time of calls
and lengthening periods of call

c) Ineligible error is when respondent was reached but was not the proper person
due to age, sex, citizenship etc fort he survey. Eligibility rate is the % of
contacted respondents who are eligible. Can be improved through careful
screening of respondents, better sampling frame definition and multilingual
interviewers

d) Refusal error Researcher is able to contact but respondents refuse to answer all
or part of question items. Result when some respondents were not willing to
be interviewed. This error is also known as cooperation response error.
Cooperation rate is the percentage (%) of contacted eligible respondents who
complete the survey. Refusal can be reduced by sending letter in advance of
an interview (appointments), rescheduling interviews, offering incentives,
adjusting interviewer behavior and statements, using alternative interviewers
in terms of age, sex etc. Can also be reduced by using alternative methods,
review benefits of the survey to respondents (salient to respondents), clear
introductions and instructions to respondents etc

e) Incomplete participation error result when respondent stopped answering be

the end of the questionnaire or began answering every question with “Do no
know” or “No opinion or idea”. Completion rate is the percentage (%) of
cooperating respondents who completed the survey.

Can be reduced by prepaid calls, pre-testing of questionnaires, more incentives

for fully completed questionnaires.

39
 NB Total response rate is the percentage (%) of all respondents in the initial
sampling frame who are located, contacted, eligible, agreed to participate and
completed the entire questionnaire. Is a product of all other response rates.

Question
A researcher began with 1000 respondents in a sampling frame, located 950 by
telephone and address, can contact 800 by interview, established that 750 were
eligible due to language and mental problems, 700 respondents cooperated but only
690 completed the entire questionnaires. Calculate all the possible six response rates.

Location rate = 950/1000 = 0.95 = 95%

Contact rate = 800/950 = 0.84 = 84%
Eligible rate = 750/800 = 0.975 = 97.5%
Cooperation rate = 700/780 = 0.897 = 89.7%
Completion rate = 690/700 = 0.9857 = 98.57%

Total response rate = 0.95x0.84x0.975x0.987x0.9857 = 0.688 =68.8%

Non – sampling errors

1 Response errors are errors made by the researcher or respondents

These can be:
a) Surrogate which is the provision of substitute information for what is required,
arising from inability or unwilling to provide needed information and even failure
to understand the question
b) Measurement error result from using the wrong measurement processes
c) Experimental error arises from may sources eg measuring what is not the effect of
the independent variable or failure to manipulate and control variables

2 Data analysis errors

Occurs while raw data from questionnaire are being transformed into research findings eg
incorrect interpretation of data and findings. This may result from response errors made
by either interviewer or respondents.

Response errors made by the interviewer

a) Questionnaire error or bias is made from questioning to the respondents eg

using wrong wording, ambiguous questions, improper sequence or phrasing of
questions
b) Recording error arises from lack listening skills, interpreting and recording
answers given by the respondents e.g. neutral to mean positive or negative

40
 c) Cheating error results from the interviewer fabricating answers to a part or
whole of the interview e.g. not asking sensitive questions when it is necessary to
do so

Response errors made by respondents

a) Inability error results from inability to provide accurate answers because

unfamiliarity, fatigue, boredom, questioning format or content. Questions may be
outside the respondent’s personal knowledge or interest.
b) Unwillingness errors
Unwillingness by respondents to provide answers. Respondents may misreport
answers because of undesired or avoid to embarrassment or to please the
interviewer

Other error affecting research designs are:

a) Response by a non targeted population
b) Inter-respondent bias
c) Yea saying error
d) Respondent fatigue
e) Incompetent interviewing

REVIEW OF LITERATURE
-Is a summary and analysis of related literature, which provide background of the
research. Researchers look into theories put by other experts, and authors and
analyses their model and make opinions. Contain the following:
Conceptual view of experts
Comparisons of the views from experts
Nature of issues being researched on

Is reviewing the accumulated knowledge about a question (Research topic) Is based on the
assumption that knowledge accumulates and that people learn from and build on what others
have done(collective effort)

Goals of a literature review

i). To demonstrate a familiarity with a body of knowledge and establish credibility. A
review tells that a researcher knows the research
ii) To show the path of prior research and how a current project is linked to it. Places the
research in a body of knowledge
iii) To integrate and summarize what is known in an area. Points out areas where prior
studies agree, disagree, and what major questions remain unanswered. Indicate the
direction for future research.
iv) To learn from others and stimulate new ideas. Reviews divulges procedures, techniques
and research designs worth copying and improving on to gain new insights

Types of literatures review

41
 a) Meta analysis
Is a type of literature review in which a researcher or writer organizes the results from
many studies and uses statistical techniques to identify common findings in them.

b) Context review
A type of review in which the author links a specific study to a larger body of knowledge.
Often appears at the beginning of a research report. Introduces the study by situating it
within a broader framework. Show how it continues or builds on a developing line of
thought or study

c) Historical review
A review in which the author traces an issue over time. Can be merged with a theoretical
or methodological review to shoe how concept, theory or research method developed
over time.

d) Integrative review
A review in which the author presents and summarises the current state of knowledge on
a topic, highlighting agreements and disagreements with it.

e) Methodological review
Is a special type of integrative review in which the author compares and evaluates the
relative methodological strength of various studies and show how different
methodologies (Research designs, measures, sample) account for different results.

f) Self-study review
A review in which an author demonstrates his or her familiarity with a subject area.
Often part of an educational program or course requirement.

g) Theoretical review
A review in which the author presents several theories or concepts focusing on the same
topic and compares them on the basis of assumptions, logical consistency and scope of
explanations.

Where to find Research literature

i) Periodicals/Publications such as newspapers, popular magazines, TV and radio
broadcasts, Internet, encyclopedias

ii) Scholarly journals is a type of periodicals filled which peer-reviewed reports of research
studies. Citations (details of a scholarly publication’s location that helps people to find it
quickly) may be necessary

iii) Books e.g. textbooks, short story

iv) Dissertations – original research produced by students doing degrees etc

42
v) Government documents
vi) Policy reports and presented papers

Conducting a systematic literature review

Literature review should be conducted systematically for it to achieve its purpose. The following
steps may be followed.

a) Define and refine a topic

Literature review should begin with a clearly defined, well focused research question and
a plan. A review topic should be as focused as a research question. NB Review helps to
bring greater focus to the research question

b) Design a search strategy or plan

Is after choosing a focused research question for review. It Involves
- deciding on the type of reveiew and its extensiveness and
- deciding on the types of material to include
- Setting parameters on the search eg, time to devote it, how far back in time to
consider, minimum number of research reports(books) to examine and libraries to
visit
- Deciding how to record the bibliography
- Developing a schedule of visits

c) Locate research reports(published articles)

This depends on the type of report or of research being searched. Major ones are
scholarly journals, periodicals, dissertations, government documents, policy reports and
presented papers.

d) Evaluating research articles

Is done to determine relevance of articles. Evaluation of articles includes:
-Evaluating the title. A good title is specific, indicates nature of research, describes the
topic, tells about the setting or people being studied
-Reading the abstract. Good abstract summarizes critical information , gives purpose of
study, methods used, highlights major findings
-Scanning introduction and conclusion sections. Introduction section has 3 purposes:
-Introduce a broad topic and show a transition to a specific research question, that
is, primacy focus of the study
-Establish the significance of the problem
-Outline the theoretical framework and define major concepts being used
Amount of time required for reading an source of review is influenced by:
-Article’s high quality. Well defined purpose, clear writing and smooth and
logical organization of work
-Reader’s sharply focus on a research issue or question
-Reader’s solid background on the theoretical issues, knowledge of
substantive topics, and familiarity with multiple research methodologies
NB A good literature review is selective, comprehensive, critical and current.

43
e) Decide what to record about an article, book or other sources
Is governed by hypothesis being tested, major concepts measured, the main findings,
basic design of the research. Is advisable to record too much than too little.
Photocopying relevant articles or reports may save time than recording notes and may
result in:
-having an entire report, which is good
-Notes can be made on the photocopy

Disadvantages of photocopying
-Can be expensive for a large literature search
-Should obey copyright laws
-May end up photocopying entire article, and forget citation information
-Organizing entire articles can be cumbersome especially for several different parts of a
single article
-May result in re-reading the entire article later

f) Organize notes
Gathered references and notes should be organized. Organizing scheme(method)
depends on the purpose of the review and type of review.:
i) Context review implies organizing recent reports around a specific research
question
ii) Historical review – organizes by major theme and date of publication
iii) Integrative review- organizes around core common findings and the main
hypothesis
iv) Methodological review- organize studies by the topic and within a topic by the
design or method used
v) Theoretical review implies organizing studies by the theories and major thinkers
being examinee

SAMPLING PROCEDURES

Sampling is the process of selecting a representative sub-set of elements from the population to
determine the characteristics of the variable understudy and to draw inference about the larger
group or population . Is the process of selecting a representative of population to determine the
characteristics of the variables of population understudy. The larger group is known as the
population being the totality of cases being studied

JUSTIFICATION FOR SAMPLING or REASONS OR PURPOSE OF SAMPLING

Some of the advantages of sampling as compared to the collection of data on the whole
population or census
i) Gathering data on a sample is less time consuming
ii) Gathering data on a sample is less costly since the costs of research are proportional
to the number of hours spent on data collection

44
 iii) Sample study permits concentration more on study rather than worry about
collecting data from the field. If the entire population is to be studied, a great deal of
effort is exhausted, simply in pooling up the data rather than on its analysis and
interpretation.
iv) Sample study permits higher overall level of accuracy. More checks and test for
accuracy can be afforded at all stages.
v) Sampling may be the only practicable method of data collection especially in studies
where the characteristics under investigations necessitate the destruction of the
object.
vi) Sampling is the only practical way of collecting data when the population is infinity
or extremely large
vii) Only way of collecting information where the population is widely distributed
viii) Offers easy control of procedures

What are the disadvantages of sampling?

TERMS USED IN SAMPLING

a) Population/ Universe
The entire set of objects and events which is the object of research, about which the
researcher wants to determine some characteristics e.g. all college students in Zimbabwe.
Is the large pool of sampling units.

Target population is a specified large group of many cases from which a researcher
draws a sample and to which results from a sample are generalized. Is a specific pool of
cases the researcher wants to study. eg students at Kushinga Polytechnic

b) SAMPLE FRAME
Is a list of all the elements of the defined target population from which the sample is to be
selected or drawn e.g. a list showing all college students at Kushinga Phikelela Poly,
telephone directories, driver’s licence records etc

c) SAMPLE
The sub—set of the whole population which is usually investigated by the research and
whose characteristics will be generalized to the entire population. Is a small proportion of
the population selected for observation and analysis

d) CENSUS means the same as population. Counting all the elements of the population

e) Sampling element is the unit of analysis or a case in a population or a unit to be sampled

f) Sampling ratio is the number of cases in the sample divided by the number of cases in
the population or the sampling frame or the proportion of the population in the sample

45
g) Population parameter is a characteristic of the entire population that is estimated from a
sample. Is the true characteristics of the population eg % of residents who smoke
cigarettes
h) Sampling error is the deviation between the sample results and a population parameter
due random process
SAMPLING PROCEDURES /STEPS

Below are common sampling procedures or steps:

i) Population definition is a large group of many cases from which the researcher
draws a sample and to which results from the sample are generalized.
This is the target population the researcher wishes to make an inference. The
population must be defined in terms of time, ages and characteristics. First there is
need to understand the research topic and objectives because objectives specify
the specific information to be sort, hence identification of who can provide
necessary and important information

ii) Census Vs Sampling

There is need to decide where the survey is to be conducted after definition of
population i.e. on the whole or sub-set of the population. Choice depends on:
a) Size of the population
b) Size of the precise boundaries of the population
c) Accuracy required
d) Administration consideration
e) Ability to control sampling and non-sampling errors
f) Costs and time

Census may be called for when:

-Population is small
-Where cost of error is high
-Fixed costs of sampling are high
-Where the variance in the characteristics being measured is high

iii) Determine and select the appropriate sampling frame

Sample frame is a list of study objects from which the sample is to be drawn. The
list can exist in physical and non-physical form and should contain all
characteristics of the population once and information is both complete and
accurate
-Can be acquired through commercial research

iv) Select sampling methods

The researcher should choose between probability and non probability sampling
methods
a) Probability sampling methods is where all elements of the population have
equal chances of being included in the sample. Is an objective and
scientific method of sampling. Make use of randomness when selecting
sample elements

46
 b) Non-probability sampling methods is where sample elements are selected
using other methods other than randomness. Elements do not have equal
chances of being included in the sample. Elements are purposively
included or excluded from the sample at the discretion of the researcher
hence is a subjective method of sampling. Sometimes known as
purposeful sampling method

vi) Determine the most adequate sample size

Number of elements or size of sample should be decided using either
mathematical approach or judgmental approach among other methods

vii) Select the sample and gather data

viii)Validate the sample by comparing the characteristics obtained from the sample
with those present in entire population
Attributes of sample so as to meet requirements of management and researcher

i) The sample should be a representative of the population. The larger the sample the
more representative of the population
ii) Sample should be sufficient, accurate and provide stable results
iii) All research resources should be used as efficiently as time permits

TYPES OF SAMPLING DESIGNS

Sampling is the process of selecting a representative sub-set of the population to determine the
characteristics of the variable under study. There are two basic methods of sampling techniques
which are non-probability sampling methods and probability sampling methods.

The method to use depends on:

-Objectives of data collection
-Cost of sampling
-Time factor
-Expertise of the researcher
-Accuracy required or expected. Probability produces accurate sample, results etc

Non- Probability Sampling Designs

- Are sampling methods in which the sample units are not selected randomly.
- Criteria other than randomness are the basis for selecting elements from the population.
Rely heavily on personal judgment. Prohibit estimating or chancing. Useful in
exploratory research to probe ideas and insights. More subjective or biased method
- Mostly used by qualitative researchers
- May not require determination of sample size in advance before sampling

Disadvantages of Non-Probability Sampling

47
 - The sample may not be representative of the population from which it is drawn. This is
likely to produce biased results. Influences based on an unrepresentative sample are
likely to be misleading
- The sampling error cannot be predicted and quantified. This means that statically
inferences based on a non-probability sample data would be invalid.

What are the advantages of Non-Probability Sampling?

*Despite the above weakness, non- probability sampling designs are useful tools in the
exploratory phase of research, a phase in which ideas and insights are more important than
scientific objectivity.

Non-profitability sampling procedures include convenience sampling, judgment sampling and

quota sampling.

1. Convenience Sampling
- Also known as haphazard or accidental sampling
- This represents a sample drawn to suit the convenience of the researcher e.g. it may be
convenient to interview only employees within one company instead of selecting
employees from a number of companies or those available during the study or readily
accessible with minimum effort and costs or friendly to the researcher
- The researcher may also selects anyone he or she happens to come across

- Saves costs and time

- Systematic errors are unpredictable and not measurable on the resultant sample.
- The sample may not be a representative of a the defined population

Convenience sampling is a useful tool in the exploratory phase of research, phase in

which h ideas and insights are more important than scientific objectivity.

2. Judgment Sampling
- The sample consists of sampling units deliberately selected from the population on the
basis of the experience, judgment and expertise used by the researcher to select the best
sampling units to include in the sample e.g. a sample of four of the most influential
economists is asked to estimate the next year’s rate of inflation
- Researcher uses his or her judgment to select what he or she thinks to be the best
satisfactorily
- This type of sample may be subject to error which although relatively, can still not be
measured or predicted.
- Pilot studies are based on such samples

Advantages of judgmental sampling

-Greater elimination of costs and time

48
 -Non-essential are dropped immediately
-Since the researcher is aware of objectives hence only relevant elements and data is
gathered

Disadvantages of judgmental sampling

-Bias in estimation
-Strong assumption based upon thorough knowledge of the population may
mislead
-Elimination of the use of inferential parameters and statistical tools increase
sampling errors

3. Quota Sampling
- Is a non random sampling in which the researcher first identifies general categories into
which cases or people will be fall and then selects cases from each category to reach a
predetermined number of cases in each category.
- The population is divided into a number of segments and the researcher arbitrarily select
a quota e.g. a certain percentage of sampling units of each segments eg under 10, 20 etc
- Chosen in a way that the proportion of sample elements possessing a certain
characteristics is approximate the same as the proportion of the elements with the
characteristics of the population
- Elements are assigned a quota that specifies the characteristics of elements to be
contacted believed by the researcher to be more realistic

Advantages of quota sampling

- Permits substitution in case of inability or refusal
- Less expensive and quick sampling method
- Suitable for a population on which no stable population frame is possible

Disadvantages of quota sampling

There is high increase in errors in that:
- Characteristics are not the same characteristics of the population and the sample
- Sample size is not determined using a consistent method
- danger of an error of omission
- Supervision of the interview work is impossible

4 Snowball sampling
Used where respondents are difficult to identify and are based on referrals network.
An individual is initially discovered mostly using a non-probability method, although
probability can be used, and is used to locate others who posses similar characteristics
through referrals, who in turn identifies others until the desired sample size is attained.
This method is used on issues which individuals are unwilling to come up openly; hence
incentives can be used to encourage elements to participate

5 Purposive sampling method

49
 -Considered a form of judgmental sampling
-Selection of the cases is governed by some criterion acting as a secondary control
-A group is selected by the researcher to constitute a sample because is considered
to mirror characteristics in question
-There is purposively selecting and leaving some elements of the population
-Selected sample is to yield quickly averages and proportions of characteristics to
be studied e.g. an editor can be selected to represent the public
-Units are restricted to units considered to be a typical of the population

Advantages
-Only cases considered judgmentally make the sample quite representative
-Unnecessary elements are dropped out
-Purpose of the study can be fulfilled even if the sample is small
-The investigator clearly knows the objective of the research hence only necessary
elements are included

Demerits
-Selection can be biased and the resultant sample not representative of the
population
-Method cannot be used by an average investigator to select sample units
-Difficult to detect mistakes and inaccuracy till end

6) Deviant case sampling also called extreme case sampling. Is a nonrandom sampling in
which a researcher selects unusual nonconforming cases purposely as a way to provide
greater insight into processes or a setting. Used to seek cases that differ from
predominant characteristics of other cases

7) Sequential sampling is a non random sampling in which a researcher tries to find as

many relevant cases as possible until time, financial resources or his or her energy is
exhausted or until there is no new information or diversity from the cases. Is similar to
purpose sampling in that it tries to find as many relevant cases as possible

8) Theoretical sampling is a non-random sampling in the researcher selects specific

times, location or events to observe in order to develop a theory or evaluate theoretical
ideas. A growing theoretical interest guides the selection of sample cases

Probability Sampling Designs

- Probability sampling includes all selection methods where the sampling units to be
included in a sample are selected on a purely random or chance basis from the
population.
- All elements of the population have equal chances of being selected
- Is a more objective method of sampling, that is, not subjective and not biased because is
not dictated by human feelings
- The sample is thus representative of the population

50
- The sampling error can be measured and predicted
- Valid statistical inferences can be applied to data collected from probability sample
- The common type of probability sampling designs are sample random, systematic,
stratified, cluster and multi- stage sampling

1. Simple Random Sampling

- Also known as accidental sampling
- Each sampling unit in the entire population has an equal chance of being selected. This
method of sampling is used when it is assumed that the population is relatively
homogenous in respect of the variables or characteristics understudy.
- Uses the lottery system. The sampling units can be selected using numbered pieces of
paper put in a drum or hat and the numbers are selected one at a time until the required
size is reached or by using random dig its or numbers. One unit at a time is picked form a
pool of units. Lucky ones are picked and included in a sample. A picked element is not
returned into the pool or drum

1 2 3 45 = 1 4

2. Systematic Sampling
- Follows a systematic or uniformity process when selecting sample element. Sampling
begins by randomly selecting the first sampling units, thereafter subsequent sample units
are selected at a uniform interval, relative to the first sampling unit.

Procedures on systematic sampling

i) Population elements are numbered and put on a list on ascending order
ii) Determine sample size
iii) Divide the population by the sample size to get the sampling interval(population
segments). Sampling interval is the inverse of sampling ratio
iv) An element is picked randomly from the first segment of the population
v) Subsequent sample units are selected at a uniform interval relative to the
first selected sampling unit being governed by sampling interval

3. Stratified Sampling
- Used when the population is believed to or regarded as heterogeneous with respect to the
various variables or characteristics understudy
- The population is divided into mutual exclusive segments or strata where the sampling
units in each strata are relatively homogenous
- Thereafter random sampling can be done from each stratum. The researcher needs to
estimate the average mass of a large group of people. First divides the group into strata
e.g male and female and then selects a sample from each stratum.

Steps for stratified sampling

i) Divide the population into mutual exclusive and exhaustive segments or strata
which are homogeneous
ii) Determine the sample size

51
 iii) Determine the proportion of elements in each stratum to the size of the
population
iv) Determine the number elements to be picked from each stratum
v) Pick randomly elements from each stratum until the required number from
stratum is reached
Heterogeneous Homogeneous Sampled
Population Stratas Elements

1,4,5,6,7,8,, 1,4,5, a,b,c, 1,5,8,a,

a.b.c.d,e,f, 6,78 d,e,f, c, e

4. Cluster Sampling
- The population is divided into clusters where each cluster is similar in profile to every
other cluster, which is, containing the required number of elements and characteristics of
the population. A cluster is a unit that contains final sampling elements but can be treated
temporarily as a sampling element.
- Used when researchers lack good sampling technique for a dispersed population and the
cost to reach a sampled element is very high
- A cluster is randomly selected from other clusters for investigation. The sampling units
within this randomly selected cluster maybe then be randomly selected to provide a more
representative sample from the population.
1, 2, 2,
A, B, C,
II III I
A
5. Multi- Stage Sampling
- Multi-stage sampling is an extension of cluster sampling.
- It involves the process of selection at two or more successive stages
- The population is divided into clusters and a sample cluster is randomly selected.
- The randomly selected clusters are divided into smaller clusters and are randomly
selected. The procedure is continued until the clusters are small enough and the required
sample size for investigation is attained.

ZIMBABWE

PROVINCE
A B C

DISTRICTS
1 2 3 4 5

6 Multiphase or Double sampling

52
 -Is where the same elements are involved at each phase but with other units (elements)
asked to give more information or different information than others. This result in
differentiating details of questions to asked to respondents.
-The process also involves collecting data from a sample using a previously defined
technique basing on the information found from the previous sample and hence a sub
sample is selected for further study.

Other statistical sampling methods are:

Block sampling
Discovery sampling
Acceptance sampling

DETERMINATION OF SAMPLE SIZE

Is done before sampling to know the size of the sample
In order that the study be logical and meaningful the sample size should be adequate to fulfill the
objective of the research. Although various methods of determining sample size are described
separately below, rarely, will the researcher rely only on one approach.

1. JUDGEMENT/SUBJECTIVE APPROACH
-The judgmental approach involves the researcher using informed intention to
determine the most adequate sample size. Determination of sample size is
arbitrary, subjective and doesn’t consider the likely precision of the survey results
or the cost of obtaining.
-Is a qualitative approach or subjective method depending on individual’s feelings
-Is a convenient sample size determination technique, which is based on the
researcher’ judgment
-Is a cheaper and faster to obtain the required sample size and information required
although the sample might not be a representative of the population

2. MATHEMATICAL APPROACH
- This is a quantitative approach based on statistical precision which relies on the
traditional statistical formulae for determining sample size.
-Is objective, reliable, transmittable and measurable
-The sample size depends on 3 factors:
a) The level of confidence desired, obtained from the values of Z scores
b) Maximum allowable error
c) The standard deviation of the population.

3. BAYESIAN DECISION ANALYSIS

This is a process that bases a decision on sample size on the expected value of the
information obtained by the sample and the cost of taking the sample. The approach
involves computing difference known as expected net gain from sampling for various
sample size and choosing the sample size with the largest positive net gain.

4. COST LILMITATION

53
 - This cost limitation methods determine sample size on the basis of budget allocated to
the research project. This approach involves deducting from the available budget all
non-sampling related cost such as fixed cost of designing the survey questionnaires
preparation, data analysis and report generation to arrive at the desired sample size.
This approach is unsatisfactory because its emphases on cost to the exclusion of all
other factors especially precision. Time may also be considered.

5. INDUSTRY STANDARDS
- It refers to those rules of thump developed from experience that have become
standard industry guidelines for determining sample size. Convectional guidelines on
sample size vary with the type of research study.

Size of sample is also determined by

-Kind of data analysis the researcher plans
-How accurately the sample has to be for a researcher’s purpose
-Population characteristics
-The degree of accuracy required
-The degree of variability or diversity in the population elements
-The number of different variables examined simultaneously in data analysis

CHARACTERISTICS OF A GOOD SAMPLE

a) It must be representative of the population, that it should contain the majority of the
characteristics of the universe, if not all.
b) It should be free from bias. Bias may creep into selection of the sample due to reasons
like convenience substitution, mistake in the adoption of the sampling method and
failure to cover the whole of the chosen population.
c) The sample should be adequate to yield reasonable conclusions regarding the
population from which the sample is drawn.
d) The sample must enable the researcher to measure the sampling errors.
Representative-ness of the sample and the validity of the conclusions depends to a
large extend on these errors. The results of a good sample must be applicable to all
items of the population with a reasonable level of confidence.
e) Should yield results applicable to all items of the population with a reasonable confidence
f) Should be economic, that is, cost saving.

DISCREPANCIES (ERRORS) ARISING FROM THE SAMPLING FRAME

1 OMISSION DISCREPANCY
Result when the defined population is extremely larger than the sampling frame.
Some objects in the population have no chance of being in the sample because do not
appear in the sampling frame although having the required characteristics. This result
in a less representative sample

Sampling frame Population

54
 2 COMMISSION DISCREPANCY
Occurs when a sampling frame is larger than the defined population. Results in
objects being included more that once and also including of unnecessary objects.
Arises due to lack or incomplete information about potential study objects. This can
be reduced by screening before data collection by using qualifying questions.

Sampling frame Population

3 COMMISSION-OMISSION DISCREPANCY
Is a combination of omission and commission discrepancies. Caused by, not
including all objects in the population in the sampling frame although these
essentially containing the required characteristics or not all objects in a sampling
frame are in the population. Other elements in the population are excluded while
other objects in the sampling frame are not from the population
Sample

Population Sampling frame

METHODOLOGY

Types and Sources of data of Data

TYPES OF DATA

Data can be primary and secondary Data

1) Primary Data

Is data which is captured by the researchers at the point where it is generated, that is, at the
point of origin. The data is captured for the first time and specifically for the purpose of
investigation at hand.
Some times called source data

Some of the advantages are:-

(i) the data is directly relevant to the problem at hand

(ii) the researcher generally offers greater control over data accuracy.
(iii) Because of (i) and (ii) above, the data is considered more meaningful and reliable.
Some of the disadvantages are:-

(i) the data could be time consuming to collect

(ii) the data is generally more expensive to collect

55
Characteristics of Primary data
It’s accurate, sufficient, current, relevant and reliable

Types of Primary data

Primary data can be Overt and Non Overt data

a) Overt data

Is data that is collected verbally by asking people questions e.g. - data through
telephone, mail and personal interviews

It requires consent and co-operation of the study participants. Responses are given
orally or in writing. Respondents take an active role during the collection of data

Moreover overt data can be:

i) Socio, economic and demographic data
This include personal characteristics of study participants e.g age,
income, education, status, occupation etc
-The characteristics can be verified and as profiling information

ii) Psychological data

This includes internal or mental characteristics of the participants e.g.
attitude, opinions, personalities, motives, behaviour,
perceptions, preferences, etc
-Used as independent variables to classify study individuals
-Necessary to understand behaviour occurrence

b) Non – Overt data

Is data collected by other means other than asking questions e.g. data collected by
traffic counts sales, audits or observations methods.
Refers to all data other than that derived through oral or written responses given by
the individuals being studied

Do not require person interactions with study participants since it do not involve
questioning study participants.

2) Secondary data
Is Data collected and processed by others for other purposes not for the problem at
hand. Is data extracted from other sources as a result of the research or work done by
other researchers. Originally, it was not for the study at hand but is only tailored to
meet or related to the problem being faced or solved.

56
 Mostly used in exploratory research to have an insight into the problem understudy, to
supplement the definition of the problem or developing a statement of hypothesis

Such data is already in existence either within or outside the organization- that is, one can
get both internal and external secondary data.

Advantages
(i) The data is already in existence therefore readily available
access time is relatively short.
(ii) The data is generally less expensive to acquire
(iii) May be adequate to solve the problem at hand

Disadvantages
(i) The data may not meet specifically the needs of the current problem and may be
difficulty to be fitted into it.
(ii) It maybe difficult to assess its accuracy, reliability and sufficiency
(iii) The data maybe outdated and not relevant to the problem at hand
(iv) The data may not be subject to further manipulation
(v) Combining various sources could lead to errors of collation and introduce bias.

NB A researcher can purchase data from a firm that specializes in providing research data.
Such data is called commercial data, since its source is commercial.

Considerations when using secondary data

i) Privacy and integrity of the source
ii) Primary sources of data
iii) Integrity of the researcher who collected the data
iv) Purpose of publication
v) Quality of the data collection methods
vi) Data presentation methods used
vii) Period , that is, when the data was collected
viii) Relevance considering the characteristics of the data

Evaluation of secondary data

Is done to determine the extent to which the data meet the criteria of accuracy, current,
sufficiency, reliability availability and relevance

Questions to be answered by evaluation of secondary data

i) Who was responsible for collecting the data
ii) Why was the data collected
iii) Is the data available, accurate, reliable, relevant, complete etc.
iv) Which methods and techniques were employed to collect the data

SOURCES OF DATA

57
1) Primary data
Is data extracted through data gathering methods like, surveys and observations
constitutes are the principal sources of primary data .

Examples of primary data and respective sources are:

i) Financial data- Sales vouchers, credit notes, accounts receivables,
accounts payables, asset register etc.
ii) Production data - Production cost records and stock sheets etc.
iii) Human resources data – time sheets, wages and salaries schedule,
employee personal final records etc.
iv) Marketing data – sales data, advertising expenditure etc.
v) Survey data – Personal surveys, salary surveys, research surveys etc.
2) Secondary data
Publications constitute the principal sources of secondary data. It should be noted that
results of using primary data, if used for another problem becomes secondary data

Sources of secondary data can be internal, external and literature review

i) Internal sources of secondary data are records found within the

organization for which the research is being done eg;
-Aged market research figures
-Company reports like financial statements for previous years
-Past sales reports
-Employees records
-Internal company publications
-Reports, memos etc.

ii) External sources of secondary data is when data is obtained from other
organizations’ publications, periodicals, industry and research institutions
egg-Commercial and industrial Associations like CZI, ZNCC,

58
 Zimtrade, Zimbabwe Investment Centre(ZIC) and SEDCO
-Research organizations such as Africa Research (Pvt) ltd,
-Central Statistical office which is the government’s data capturing and
dissemination instrument
-Publications in newspapers, trade journals, and trade magazines etc
-Libraries, universities
-Reserve Bank of Zimbabwe and other banks economic reports

iii) Literature review

-Text books are used
-Information is derived from the author’s point of view
-Researchers identify texts relevant in:
a) Problem definition
b) Building the statement of hypothesis
c) Suggesting recommendations

*Mostly secondary data constitutes basic research and be used to

understand the problem and develop statement of hypothesis

QUALITATIVE AND QUANTITATIVE DATA

Both primary and secondary data can be qualitative or quantitative data

(a) Qualitative data

Qualitative data refers to data which cannot be subjected to mathematical
computations, that is, addition, subtraction, multiplication, and division
In qualitative data the values of variables represent categories. The categories can
be coded using either numbers or letters –eg- gender is a qualitative random
variables where the categories male and female can be represented by using letters
M for male and F for female.

Another example is rating of “communication skills”, which is also a qualitative

variable, where the categories can be represented by numbers:- 1 for good , 2 for

59
 average, and 3 for poor for a person’s communication skills. These numbers are
used as labels and are meaningless as quantities.
(b) Quantitative data
In quantitative data the values of variables are always numeric(numbers) and
represent a measure. The data can be subjected to mathematical
computations, that is, addition, subtractions, multiplication and division

There are two types of quantitative data – Discrete and Continuous data.

Discrete data refers to data to which the attributes can be measured with only a
limited number of distinct separate categories. Data values have distinct values
with no immediate points and usually arise from a counting process. Have
relatively fixed set of separate values or variable attribute.

This means that the elements of discrete quantitative variables can only take
certain specific values. Data attributes can be measured with only a limited
number of distinct, separate categories

Usually discrete quantitative variables have an integer values –eg- the number of
employees in a department may be 0 or 1 or 2 or 3 etc, but not 0,5 or 7,31 etc.

Not usually measured over a time

It is possible for discrete quantitative variables to have non integer values –eg-
ladies shoe sizes can be size 3; 3 ½; 4; 4 ½ etc but not 3,17; 4, 231 etc

However discrete data can be conceptualized as a continuous data eg

i) Temperature can be continuous variable ( eg degrees, fraction of a degree) and
can be measured with discrete categories as hot or cold (qualitatively
measured)
ii) Age can be continuous eg 2 years, 2.5 years , months, days, hours and minutes
or can be in discrete categories (qualitatively) eg infants, adolescence, young,
adulthood, old
age etc

Continuous data refers to data which can have any values over a range either as
whole number or as fraction. Arise from measuring process. Measuring is done
over time. Have an infinite number of values or attributes that flow along a
continuum.

This means that the elements of a continuous quantitative variable can take any
value over a range or internal eg if weights are recorded, there might be any value
–eg- 11,27kg ; 23,05kg ; 19kg ; 7,002kg etc. The values can be divided into many
smaller increments, that is, infinite number of increments.

60
 Age is also a continuous quantitative variable since it is measured over a time.

However, age is usually rounded to the nearest year

Primary data can be either discrete or continuous, and likewise secondary data.

LEVELS OF MEASUREMENT

Measurement is the assignment of numbers (by quantitative researchers) or other symbols (by
qualitative researchers) to characteristics of variables according to certain pre-specified rules.
Note that what are measured are not the variables but some characteristics of the variable.
Measurement extends human sense and helps people observe what is otherwise invisible. Levels
of measurements are significant when collecting data.

There are four levels of measurement – ie nominal, ordinal, and internal and ratio. The level of
measurement depends on :
-how a construct is conceptualized, that is, assumptions about whether it has particular
characteristics
-the type of indicators measurement that a researcher use

(a) Nominal scale

Is where data of a qualitative random variable such as marital status, ender etc, is
assigned to one of a number of categories of equal importance

Is the lowest least precise level of measurement for which there is a difference in
type only among the categories of a variable. Only indicates that there is
difference among categories but cannot indicate to what extend the categories
differ. Consists of putting responses or observations into classes such as yes or
no, red or green or blue, or other qualitatively interpreted names for each
category.

It is based on classification and is not often subject to numerical interpretation

except from a frequency of a response observation of view.
Nominal scaled data is associated mainly with qualitative random variables such
as gender, type of car, marital status etc.

There is no implied ranking or ordering between the groups of the random

variables. No or limited statistical measures can be applied on the data.

b) Ordinal scaled data

Like nominal scaled data, ordinal scaled data is assigned to only one of a number
of coded categories, but there is now a ranking implied between the categories in
terms of better, longer, older taller or stronger etc.

61
 Is a level of measurement that identifies a differences among categories of a
variable and allows the categories to be rank ordered.

Ordinal measurement simply means the ranking of the characteristics of a variable

according to some criterion such as preferences, importance etc. While there is an
implied difference between the categories, this difference cannot be exactly
measured, that is, the distance between categories cannot be quantified nor
assumed to be equal. Eg agree, Strongly agree, Disagree, Strongly disagree.

In research, ordinal scaled data is generated from ranked responses. In ranking,

one cannot say that the item ranked as one is twice as preferred as that ranked two
or that the difference between ranks is constant. There is limit on statistics on this
data.

c) Interval scaled data

-Is a level of measurement that identifies differences among variables, attributes,
ranks categories and measures distance between categories, but there is no true
zero origin
-Is data associated with quantitative random variable.
-differences can be measured between values of a quantitative random variable.
Thus interval scaled data possess both order (implied ranking) and distance
properties.
-An interval scale is one where an item is assigned a score and the distance
between scores is constant e.g. the distance between an item scored fourth and
that scored second is the same as that between one scored tenth and that scored
eighth. Interval scaled data does not possess an absolute origin; therefore the ratio
of values cannot be meaningfully compared for interval scaled data. Only absolute
differences make sense when interval-scaled data has been collected.
Do not have zero origin eg Temperature can be 5 degrees, 45 degrees

d) Ratio scaled data

Is the highest, most precise level of measurement. Variable attributes can be
rank ordered, the distance between them precisely measured and there is an
absolute zero
-Is data associated mainly with quantitative random variables such as income,
time, age, distance, prices etc. It is numeric data with a zero origin. The zero
origin indicates the absence of the attribute being measured and zeros are there to
help keep a score
Thus ratio scales have intrinsically physical interpretation, in absolute terms
rather than being measurement which are relative to an arbitrary zero point.
-Ratio scaled data is the strongest form of data which can be collected and lends
itself to the widest range of statistics methods.

-However, it should be noted that if ratio scaled data is grouped into categories,
the data type becomes ordinal scaled data. This then reduces the scope of
statistical analysis on the random variable

62
METHODS OF EXTRACTING /COLLECTING DATA OR DATA COLLECTION
TECHNIQUES

Data need to be systematically collected so that can be able to answer research questions in
conclusive way. Data collection methods include:
a) Interviews these can be personal, telephone, mail or internet interviews
b) Observations can be personal, mechanical, content analysis, trace analysis,
audit, and abstraction

Interview is a method of collecting overt primary data. Collection of data requires cooperation
and interaction of study objects. The study objects take an active role.

a) PERSONAL INTERVIEWS/ FACE TO FACE INTERVIEWS

-Used for collecting overt primary data
-data is collected by means of face to face communication between the
interviewer (researcher) and the respondents or study participants
-Respondents are contacted, asked questions and the responses are recorded,
therefore, the interviewer’s tasks are to meet the respondents, ask questions and
record the responses on the interview schedule.
-Interviews can be either structured or unstructured.

Structured interview is a method of interviewing where questions are predetermined.

Interviews are purposefully standardized so that each respondent is exposed to the same
questions or questionnaire (interview schedule) as well as the questioning process. This
means that the interviewer cannot alter the interview either by:
-adding or deleting questions from the interview schedule or questionnaire
-changing the sequence of the questions
-changing the wording of the questions

Highly structured interviews contain structured questions, that is, questions which require
short answers like “Yes” or” No”

Have a low degree of flexibility. Questions are fixed.

Useful when the respondents are large
Data from such interviews are is to present and analysis

An unstructured interview is a method of interview where questions are not completely

pre-determined and the interviewer is free to probe for all details and underlying feelings.
Such interviews are used when the researcher wants to the flexibility of being able to ask
questions in different orders to determine the working of specific questions for different
respondents or to be able to probe respondents to clarify certain answers to questions

63
There is high degree of flexibility. Wording can be changed and also order of questions.
Additional questions not, included on the questionnaire, may be asked to gain more
information. Interview schedules are only used as a guide in questioning

Frequently used in exploratory studies and when the researcher has little understanding or
knowledge of the problem.

Questions are open ended. Respondents are not restricted in their responses

Advantages of personal interviews

i) Higher response rate is generally achieved because respondents tend to be more
cooperative in personal interviews
ii) Questioning allows probing for reasons
iii) Greater data accuracy is generally ensured because the interviewer can help the
respondent to understand the questions
iv) Data collection is immediate, that is, there is immediate feed back
v) Non-verbal responses and visual aids can be observed and noted. Face to face
contact can tell the interviewer a lot about the respondent therefore more data can
be collect at one goal. Is a two in one method of collecting data
vi) Generally more questions can be asked, permits the longest questionnaire, open-
ended and complex questions to be asked
vii) Responses are spontaneous, that is, spontaneous answers can be obtained
viii) The use of aided- recall questions is possible
ix) Difficult in one question does not result in the loss of answers to all of them
x) The respondent does not know what other coming questions when answering so
these cannot influence replies

Disadvantages of personal interviews

i) Time consuming and very cost especially when the responses are many
ii) Respondent anonymity is lost
iii) It may be difficult to find convenient time for interviewing certain busy people
iv) Respondents do not have time to consider replies or look up for information
therefore pre- mature or incomplete data can be collected
v) Possibility of gathering biased data is introduced by the interviewer’s interest or
respondents answering questions to please the interviewer or the interviewer may
influence the respondent’s answers. Appearance, tone of voice etc of the
interviewer may influence responses.
vi) The method is expensive if the interviewers/ees have to be paid. Interviewers may
need to be trained and paid. In addition, the interviewee may require an incentive
if to be involved in the process.
vii) Very difficult in case of geographically spaced respondents
viii) Other respondents may not be interested in well-coming unfamiliar visitors in
doors
ix) Generally fewer interviews are conducted because of time and cost constraints.
This means that the sample size is usually restricted by high costs and may not be
a representative of the population

64
Personal interviews are conversations, but differ from ordinary conversations in many
ways

Ordinary Conversation Structured Conversation(Interview)

1 Questions and answers from each Interviewer asks and respondent answers
participant are equally balanced most of time

2 Open exchange of feelings and Only respondents reveal feelings and

opinions opinions
Interviewer is non-judgment and no
3 Judgments’ are stated and attempts attempt to change respondent’s opinions or
made beliefs
to persuade the other’s point of view
Interviewer tries to obtain direct answers to
4 Person can reveal deep inner feelings to specific questions
gain sympathy or as a therapeutic
release Interviewer avoids ritual responses and
seeks genuine answers, not ritual
5 Ritual responses are common eg “Uh”, responses.
Shaking head, fine
Respondent provides almost all
6 Participants exchange information and information and interviewer does not
correct the factual errors that they are correct a respondent’s factual errors
aware of.
Interviewer controls the topic, direction
7 Topics rise and fall. Any person can and pace. Respondent is kept on task and
introduce new topics. Focus can split irrelevant diversions are contained
directions or digress to less relevant
issues
Interviewer attempts to maintain a
8 Emotional tone can shift from humor, consistently warn but serious and objective
to tone throughout
joy, to affection, sadness, anger etc
Difficult or impossible to evade and ignore
questions. Should give truthful and
9 People can evade or ignore questions thoughtful answers.
and give flippant or non-committal
answers

Stages of a face to face interview

1 Introduction and entry stage

65
 Includes getting into the doors, welcoming remarks, showing authorization,
reassuring and securing co-operation from respondents

2 Interview process is the second stage is the main part of the interview which
consists of asking and recording answers. There is also probing, that follow-up
questioning asking respondents to clarify or elaborate on an incomplete or
inappropriate answers

Forms (types) of probing are:

-Non verbal communications eg raised eyebrows, tilt of head etc
-Repeating the question or repeat a reply and then pause
-Asking a neutral question eg Any other reasons, Can you tell me more about that,
what do you mean? etc

3 Exit stage is the third and final stage: Interviewer thanks the respondent and
leaves.

Interviewer bias may chip in and affect the interview. Interviewer bias is when an
interviewer influences the responses from the respondents. Interviewee may also result
in interviewee bias

Categories of interview bias

i. Errors by the respondent eg forgetting, embarrassment, misunderstanding or

lying because of the presences of others. Is an interviewee bias

ii. Unintentional errors or interviewer sloppiness is caused by contacting the

wrong respondents, misreading a question, omitting questions, reading
questions in the wrong order or recording the wrong answer to a question or
misunderstanding the respondent

iii. Intentional subversion by the interviewer –is the purposeful alteration of

answers, omission or rewording of questions or choice of alternative
respondent
iv. Influence due to the interviewer’s expectations about the a respondent’s
answers based on the respondent’s appearance, living situation etc

v. Failure of an interviewer to probe or to probe properly

vi. Influence on the answers due to the interviewer’s appearance, tone, attitude,
reaction to answers or comments made outside of the interview schedule

b) TELEPHONE INTERVIEWS
-Is means of collecting overt primary data
-data is collected by means of telephone communication between the researcher or
interviewer and the respondents or study participants

66
 -Involves phoning a sample of respondents asking them a series of questions and
recording the responses on questionnaire or interview schedules

Advantages of telephone interviews

i) Allow quicker contact with geographically dispersed respondents
ii) Call backs can be arranged and made if the respondent is not initially available
thereby increasing response rate
iii) The cost is relatively lower as compared to personal interviews
iv) People are more willing to talk on telephone from the security and comfort of
their own homes or offices, even when not seeing the person on face to face
v) Interviewer probing is possible
vi) Interviewer has control of sequence of questions, time of questioning, and can use
contingency questions effectively
vii) Clarity on questions can be provided by the interviewer
viii) The use of aided-recall questions is possible
ix) A larger sample of respondents can be reached in relatively short space of time
x) Is a flexible method of collecting data and contacting respondents

Disadvantages of telephone interviews

i) Respondent anonymity is lost and there is element of interviewer bias
ii) Non-verbal responses cannot be observed
iii) Respondents may terminate the interview prematurely by putting the telephone
down or switching is off
iv) Sampling errors are compounded if significant numbers of the target population
do not have telephones. These cannot be reached or contacted
v) Trained interviewers are required and this increases costs
vi) There is no possibility of check backs to check on the validity of responses after
termination of the interview when doing data analysis if data was incorrectly
captured/recorded.
vii) Calls may come at an inconvenient time eg meetings
viii) Open ended questions are difficult to use

c) MAIL INTERVIEWS (Postal questionnaire method)

-Is method of collecting overt primary data
-data is collected by means of mail communication between the researcher and the
respondents
-Questionnaires are mailed (posted) to pre-selected potential respondents
-A complete mail interview package consists of the out- going envelope, cover
letter, questionnaire, return stamped envelope and possibly an incentive for
participation
-Respondents complete and return the questionnaires to the interviewer
-There is no verbal oral interaction between the researcher and the respondents
-Is appropriate when the target population from which the data required is large
and geographically dispersed
-Questionnaires may personally hand administered to respondents and re-
collected

67
 Advantages of mail interviews
i) A larger sample of respondents can be reached. Questionnaires can be distributed
widely and to many people
ii) More cost effective, that is, cheaper than personal interviews and can be
conducted by a single researcher
iii) Respondents have more time to consider their responses. They have time to
consider their opinions and check on accuracy and current information or records
iv) Anonymity of respondent is assured, generally, result in more honest responses
hence avoid interviewer bias
v) Respondents can complete the questionnaire at their own convenient spare time
vi) Questionnaires can be sent out periodically to update the data

Disadvantages of mail interviews

i) Response rate is very low unless there is incentive or legal obligation to reply
ii) Respondent cannot obtain clarity on misunderstood questions and such questions
may not be answered or wrongly answered or answers may be meaningless
iii) Postal or mail questionnaires must be shorter and simpler to complete hence not
all questions may be asked
iv) The possibility of probing or investigating further are limited
v) Data collection takes a long time. It is difficult to or no control over how long
people take to reply
vi) There is no control over who actually answers the questionnaire. The answers
may not be entirely the respondent’s own. No control over conditions under
which the questionnaire is completed
vii) Data cannot be collected from illiterate people.
viii) There is no possibility of check backs to check on the validity of responses.
ix) Knowledge of what questions are to come, later may influence answers to earlier
ones
x) There is no spontaneity feedback
xi) Postal questionnaires are difficult to design
xii) Non verbal responses cannot be observed

What can be done to increase mail interview / questionnaire response

1 -Address the questionnaire to a specific person not “Occupant or office” and send it first
class
2 -Include a carefully written, dated cover letter on letterhead stationery. Request
respondent co-operation, guarantee confidentiality, explain purpose of the survey, and
give the researcher’s name and contact details
3 -Include a postage paid, addressed return envelope
4 -Questionnaire should have a neat, attractive layout and reasonable page length
5 -Questionnaire should be professionally printed, easy to read and with clear instructions

68
6 -Send two follow-up reminder letters to those not responding. First should arrive about
one week after sending a questionnaire, the second a week later. Gently ask for co-
operation again and offer to send another questionnaire in case the first got lost
7 -Do not send questionnaires during major holidays periods
8 -Do not put questions on the back page (on both sides of a page)
9 -Leave enough blank space after each question for responses. Mostly ask the respondent
for general comments
10 -Sponsors that are local and seen as legitimate eg gvt, universities, get a better response
11 -Include a small monetary incentive and promise more for fully completed questionnaires
12 -Avoid ambiguous questions, wording etc

Comparison of personal, telephone and mail interviews

Characteristics Personal Telephone Mail interviews
interviews interviews
1. Unit cost of data collected Most expensive Moderately Least
2. Speed of data collection Moderately slow Fastest Slowest
3. Quantity of data collected Most least Considerable
4. Control of data collection environment Very good Good None
5. Control of data collection process Very good Moderately good Potentially difficult

6. Size of staff required Large Moderate Small

7. Diversity of questions possible Good Limited Moderate
8. Flexibility of data collection Very flexible Moderately flexible Least flexible
9. Use of facilitating aids Very good Limited Moderate
10. Geographical dispersion of respondents Limited Good Very good
11. Refusal & Non response problems Least Moderate Greatest
12. Perceived anonymity of respondents Low Moderate high
13. Interviewer bias High Moderate None

d) Internet Interviews also known as web survey or email survey. Are becoming the most
fast due to technological advancement. Interviews are done the network.

Advantages of web survey

-Are very fast and inexpensive
-Allow flexible designs. Can use visual images, audio or video communications
-Very efficient. Eg completely eliminate paper, postage, data entry costs.
-Overcome international boundary limitations

Disadvantages of web surveys

-Coverage. Many people especially less educated, lower income and rural people do not

69
 have access to internet.
-Privacy of respondent’s responses not 100% assured
-Design issues- requires the researchers to check and verify compatibility of software and
hardware combination used by computers of respondents

2) OBSERVATIONS
Used for collecting non- overt primary data
-Are methods of data collection in which the situations of interested are
systematically watched or scrutinized and relevant facts, actions or behaviours
are recorded on the observation form or schedule.
-Observer does not question or communicate with the people being observed
-Data is recorded as the events occur or from records of past events
-Observations can be applied on both human beings and non-human beings

*Observation methods can be classified by the mode of administration as personal

observation, mechanical observation, audit, content analysis and trace analysis

a) Personal observations
Used to collect non-overt primary data
-The researcher observes the events as they occur and does not attempt to control
or manipulate the phenomenon being observed
-The observer merely records what takes place on the observation form or
Schedule

-Personal observation can be participant or non-participant observation

i) Participant observation is when the researcher will involve him or

herself in the activity being observed, that is, doing the work being
done by those being observed. This form enables the researcher to
assess demands of the activities being done and also how those
experienced in the work are doing. Can even get data by doing the
activity under investigation. No undivided attention in recording data.
Researcher can be absorbed in the work not recording

ii) Non Participant observation is when the researcher may not take part
in the activities being observed. Observation is done from a distance or
taking a passive role in the activities being done. Allows undivided
attention in the recording of data. Participants may react to the observer

b) Mechanical observation
-Used for collecting non-overt data
Mechanical devices rather than human beings record the phenomenon
being observed. The devices may or may not require the respondents’
direct participation. Example of mechanical observation devices include
devices used to record weather changes, robots, over speed detecting
devices used by police

70
 c) Content analysis
-Used for collecting non-overt data
Is appropriate method when the phenomenon being observed is
communication rather than events or physical object.
-is defined as the objective, systematic and quantitative description of the
manifest content of communication
-It includes observation as well as analysis. Unit of analysis may be type
of words ,characters (individuals), things, space, time and measurements
(length or duration) of the message or topic(subject of the message)

d) Trace Analysis
-Used for collecting non-overt data
-In trace analysis, data collection is based on physical trace or remains or
evidence of past behaviour.
-The traces may have be left intentionally or unintentionally by the respondents
-Eg. The number of different finger prints on a page can be used to gauge the
readership of various topics in a text book, Bushmen paintings used to understand
their living style

e) Audit
-Used for collecting non-overt primary data
-In an audit, the researcher collects data by examining physical records or
performing inventory analysis eg auditing of financial statements by auditors so
as to verify authenticity of financial statements

f) Abstraction
-Used for collecting non-overt primary data
-Is applied on secondary data
-Abstraction or desk research is the collection of secondary data already available
that is closely related to the investigation on hand. The data is collected from
sources of secondary data already discussed.
-Recording of collected data is done on a check list

Advantages of observations
-Can give accurate additional information than any other means of data
collection
-Can collect information on sensitive issues
-Information can be collected from non-human beings eg toilets

Disadvantages of observations
-Time consuming
-Appropriate on small groups
-Participants may react to the observation resulting in collecting wrong data

71
TRIANGULATION IN RESEARCH
Is the use of 2 or more methods of data collection in the study of some aspect of human
behaviour or an aspect from which data is being collected. Is observing something from different
angles or view points to get the true location. The idea is that looking at something from multiple
points or view improves accuracy
Is used for verification or validation purposes to ensure correct and accurate data is collected

Types of triangulation

i) Time triangulation
-Take into consideration time factor of changes and process by utilizing cross-
sectional and longitudinal designs over time
-The same phenomenon is measured repeatedly over time to note any changes

ii) Space triangulation

-Attempts to overcome the limitations of studies conducted within one culture or sub-
culture.
-This involves testing of theories among different people or differences between
populations by using several different measuring instruments (cross-cultural studies)

iii) Combined levels of triangulation

-Uses more than one level of analysis

iv) Theoretical triangulation/Triangulation of observers

-Is when a researcher uses multiple theoretical perspectives in the planning stages of
research or when interpreting the data
-This draws upon alternative or competing theories in preference to utilizing one
viewpoint only. The researcher should be more active in designing his or her
research so that competing theories can be tested.

v) Investigators triangulation
-This engages more than one observers, interviewers or investigators recording data
from the same subjects being investigated in respect of the same characteristics.
-Is used to ensure that corrected data is collected by comparing data from different
investigators.
-Discrepancies can be analyzed if any and investigated
-Prevents the limitations of a single observer to affect the entire study

vi) Triangulation of measures is when researchers take multiple measurements of the

same phenomena either using the same measuring instrument or different instruments

vii) Methodological triangulation

72
 -This type uses either the
(a) The same method on different occasions or
(b) Different methods on the same objects of study
(c) Mixing quantitative styles and qualitative styles of research and data

SURVEY DIFFERENTIATED FROM CASE STUDY

Survey method
-Is a quantitative research in which the researcher systematically asks a large number of people
the same questions and records their answers. Written questionnaires or formal interviews are
used to gather data/information
-Is associated with the deductive approach
-Used to answer who, what, where, how much and how many questions
-Tend to be used for exploratory and descriptive research
-Allow collection of large amount of data from a sizeable population in a highly
economic way
-Allows collection of quantitative data which can be analyzed quantitatively using
descriptive and inferential statistics
-Data collected can be used to suggest possible reasons for particular relationships
between variables and to produce models of these researches
-Researcher has more control over the research process but does not manipulate a situation or
condition to see how people react but only ask and record answers
-Sample is a representative of the entire population. Sample elements are randomly sampled
using random sampling technique
-Data is collected through questionnaires, structured observations and structured
Interviews and then summarized in charts, graphs or tables and analyzed statistically
-Surveys are mostly used in descriptive and explanatory research
* is the collection of information on a wide range of cases, each case being investigated only on
a particular aspect under consideration

Information or data which can be sourced through survey include

- behavior - Attitudes, beliefs and opinions -Characteristics -Expectations
-Self – classification

Factors which contributed to growth in survey

1) Computers- made the sophisticated statistical analysis of large-scale survey

data sets feasible and easy
2) Organisations- new research centres with an expertise and interest in
quantitative research were established.
3) Data storage- Collection, storage and sharing of information on hundreds of
variable for thousands of respondents expanded the use of surveys
4) Funding by governments, NGOs etc

73
 5) Methodology – survey technique advanced as errors were identified and
corrected through prior researches. Improved statistics for data analyzing data
were devised.

Case study
-Is a research that is an in-depth examination of an extensive amount of information about very
few cases or units for one period or across multiple periods of time. Many features are examined
in depth of few cases over a duration of time

Is a strategy for doing research which involves an empirical investigation of a particular

contemporary phenomenon within its real life context using multiple sources of evidence. Is the
detailed and thorough investigation of a few cases.

-Elements are not a representative of the population

-Non probability sampling methods are used to select cases
-Cases can be individuals, groups, movements, geographical units
-Data can more detailed, varied and extensive
-Researcher intensively investigate 1 or 2 cases focusing or several factors
-Uses the logic of analytic instead of enumerative induction
-Is a or use a qualitative approach
-Mostly used in exploratory research.

RESARCH INSTRUMENTS/ RESEARCH TOOLS

- Are data collection forms on which data is recorded as it is collected. As a tool to collect
data, a research instrument must be a reliable and validly measure what the researcher
seeks to measure. Its designs must ensure that the correct research issues are addressed
and that accurate and appropriate data for statistical analysis is gathered.
- The common research instruments are questionnaires, interview schedules,
observation forms, check lists, document schedules and evaluation schedule are
discussed below.

QUESTIONNAIRE
- This is a formalized set of questions to obtain and record specified and relevant data with
tolerable accuracy and completeness. It is a set of questions which will help in meeting
the objective of the research through directing the questioning process and promoting
clear and proper recording.
- Is a data collection tool on which written questions are presented to be answered by the
respondents.
- Can be defined as a systematic compilation of questions to assemble desired information
- Is an important tool in normative survey for gathering widely scattered data from
scattered sources where seeing the respondents personally is impossible for verbal oral
communication

74
 - Is flexible data collection tool for quantitative and qualitative data
- Suitable if large respondents are to be reached

OJECTIVES OF A QUESTIONNAIRE
- To translate the data needed into a set of specified questions that the respondents can and
will answer.
- To uplift, motivate and encourage the respondent to become involved in the interview, to
co-operate and to complete the interview.
- To minimize response error which is the error that arises when respondents give
inaccurate answers as their answers are mis-recorded or misanalysed.

TYPES (FORMS) OF QUESTIONNAIRES

a) Structured questionnaire
Is a closed form of questionnaire containing closed ended questions. Contain definite,
concrete and direct questions. Contain questions requiring short check responses and or
restricted answers. Answers are fixed and the only best answer is chosen. Are in form of
multiple choose questions. Can be dichotomous ( eg Yes or No) or multi-chotomous
(with more than alternatives) multiple choose questions. Restricts respondent’s response
by selecting a response from the supplied alternatives or providing short answers without
any further explanations

Mostly questions are presented in the same wording so that respondents reply to the same
questions

Easy to fill, requires less time, keep respondents on the subject, more objective and
acceptable.

Easy to administer, tabulate and analyse

b) Unstructured questionnaire
Are open form, open ended or unrestricted questionnaires. Respondents are not restricted
in their responses. Can respond in their own words (free response). No clues are
provided (Is non-directive). Calls for depth interview or depth response to solicit
attitudes and motives
Mostly used in motivational research.

Rearrangement of the form and questions as well addition or subtraction of questions can
be made

May contain partially completed questions or statements

Both structured or unstructured questionnaire can be disguised or undisguised. Disguise

refers to the amount of knowledge concerning the purpose of the study or data
communicated to respondents by the data collection method.

75
 Disguised Questionnaires- Is a questionnaire which attempts to hide the purpose of
study, that is, purpose of the study is not made known to respondents. Therefore there
can be a structured disguised questionnaires or Unstructured disguised
questionnaires

Undisguised Questionnaire- Is a questionnaire in which the purpose of the study is made

obvious. The researcher can use projective methods to conceal the truth subject of the
study. Similarly can be unstructured undisguised questionnaire or unstructured
undisguised questionnaire.

NB A questionnaire containing both closed questions and open ended questions is a

semi structured questionnaire. This happens because it may be impossible to have a
completely closed or open ended questionnaire

c) Fact questionnaire require factual information not opinions or attitudes

d) Opinion questionnaire (Opinionnaire) requires opinions, attitudes or preferences

of respondents. Attempts to measure the attitude or belief of an individual i.e. opinionaire
or attitude scale. How people feel, or what they believe as their attitude. Attitude in the
inner feeling is an individual which is difficult, impossible to describe. Opinion is what a
person says on certain aspects of the issue under consideration. It the outward expression
of an attitude holds by an individual.

An opinionaire is defined as a special form of inquiry used by researchers to collect the

opinions of a sample of population on certain facts or factors of the problem under
consideration. The opinions will then be be further quantified, analyzed and interpreted.
Opinionnaire makes use of statements or questions on different aspects of the problem
under investigation. Solicits responses on either a three point or five point scale eg Agree,
disagree, undecided agree or strongly agree, agree, uncertain, disagree and strongly
disagree.

Mostly used in researches of the descriptive type which demand surveys of opinions of the
concerned individuals

NB Sociometry is a questions technique for describing social relationship that exists between
individuals in a group. It is a tool to measure social distance. It is the study of measurement of
the social choices that people make. Is used to reveal group structures by showing preferences
that persons in the group have for each other and the way in which individuals persons are
chosen or preferred or ignored or rejected, e.g. of questions : “Whom would you like to be a
monitor of your class?”

DESIGNING THE QUESTIONNAIRE

The process can be presented as a series of steps as shown below:

1. DETERMINE SPECIFIC DATA TO BE SOUGHT

76
2. DETERMINE THE TYPE OF INTERVIEW that is administration method

3. DETERMINE CONTENT OF INDIVIDUAL QUESTIONS

4. DECIDE ON THE QUESTION STRUCTURE

5. DETERMINE QUESTION WORDING

6. DETERMINE ORDER OF QUESTONS

7. IDENTIFY THE FORM AND LAYOUT

8. REPRODUCE THE QUESTIONNAIRE

9. ELIMINATE BUGS BY PRE- TESTING

10 FINAL DRAFT

DISCUSSION STEPS WHEN DESIGNING A QUESTIONNAIRE

1. SPECIFY THE DATA BE SOUGHT (NEEDED)

The first step in questionnaire design is to specify the data needed. The researcher must
determine exactly what is to be measured.

2. SPECIFY THE TYPE OF INTERVIEW

The researcher should determine the type of interview to use, in which the
questionnaire is the be used, that is personal, telephone, or mail. The type of interview
influences questionnaire design and how the questionnaire will be administered.

In personal interview, respondents may not see the questionnaire. There is face to face
interaction with the interviewer, therefore length, complex and varied questions can be asked.
The interviewer records the responses on the interview schedule.

In telephone interview the respondents interact with the interviewer but they do not see
the interviewer. This limits the types of questions that can be asked to short and simple
ones.

In mail interviews, the questionnaires are self administered thus the questions must be
simple and detailed instructions must be provided. The interviewee records his or her
responses

3. DETERMINE CONTENT OF INDIVIDUAL QUESTIONS

In determining what to include in individual questions, the researcher must ensure that
every question in the questionnaire contributes to the exact requirements of some
specific required responses, and collection of required data for the objective of the study

77
4. DECIDE ON THE QUESTIONS STRUCTURE
A question may be structured or unstructured
.
(a) Unstructured questions – are open ended questions, that is, respondents answer in
their own words. These are also called free answered questions. Open ended
questions are good as first questions on a topic. They enable the respondents to
express general attitudes and opinions which can help the researcher interpret their
responses to structured questions

(b) Structured questions –are close ended questions that pre-specify the set of
response alternatives and the responses format. Close ended questions may be
multiple, dichotomous or scale.

(i) Multiple Choice/ Multichotomous questions- In multiple choice questions,

the researcher provides a choice of answers and the respondents are asked to
select one or more of the alternatives given.

(ii) Dichotomous – this is the extreme form of a close –ended question and has
only two response alternatives e.g. yes or no, agree or disagree male or female
etc often the two alternatives of interest are supplemented by a neutral
alternative e.g. no opinion, don’t know, both or none.

(iii) Scale – are close ended questions that employ primary scale of measurement
(such as nominal, ordinal, interval and ratio) or non- cooperative scales to
capture the responses e.g. Do you intend to buy a new car within the next six
months?

Definitely Probably Undecided Probably Definitely

will buy will not buy will buy not will not buy
1 2 3 4 5

5. DETERMINE THE QUESTION WORDING

- Question wording is the translation of the desired questions content and structure into
words that respondents may clearly and easily understand.
- Poorly worded questions can result in non- response errors i.e. respondents answering
incorrectly. Unless the respondents and researcher assign exactly the same meaning to
the questions the result will be seriously biased.

Points to note when formulating questions

a) A question should clearly define the issue being addressed
b) Ordinary words should be used in a questionnaire and they should match the vocabulary
level of the respondent.
c) The words used in a questionnaire should have a single meaning which is known to the
respondents i.e. avoid ambiguous words.

78
 d) Avoid leading questions e.g. Question which gives respondents clues as to what the
answer should be.
e) Avoid implicit alternatives i.e. alternatives that are not explicitly expressed.
f) Avoid implicit assumptions.
g) Avoid generalizations and estimates.
h) Avoid barrel (double) questions., that is, two questions in one

(6) DETERMINE THE ORDER/SEQUENCE OF QUESTIONS

Questions should be asked in a logical order. All questions that deal with a
particular topic should be asked before beginning a new topic. The questions should
be grouped into logical and meaningful categories. Easier questions should be
sequenced before difficult and complex questions.

Approach on order of questions

a) Starting with simple to difficult and complex ones. This is the normal order.
b) From complex to simple

Question order or sequence falls into three(3) categories

a) Overall questionnaire organization
b) Question order effects
c) Context effects

a) Organisation of questionnaire
Sequence of questions should minimize discomfort and not confuse respondents. A
questionnaire should start with an introduction fro instructions, identification
information, explaining the survey (in a different font and colour) followed by opening,
middle and ending questions

Opening questions should be pleasant, interesting and easy to answer for

respondents to feel comfortable about the questionnaire. Avoid boring
background and threatening questions

Middle questions
Questions should be organized into common categories of topics. Each category
of questions should have a short introductory statement. Questions topics should
flow smoothly and logically

Ending questions
Avoid ending with highly threatening questions. Questions should summarise,
ask opinions and suggestions. End with a thank you.

b) Question order effects is an organization in which a topic or some questions asked

before others influence respondent’s answers to later questions

79
 c) Context effects. Wording of questions should be simply and in line with the vocabulary
level of the respondent.

(7) IDENTIFY THE FORM AND LAYOUT

The format, spacing and positioning of questions can have a significant effect on the
results, especially self-administered questionnaires. It is a good practice to divide
questionnaire into a several logical parts. Poor layout may result in non-response

Good layout improves accuracy, completeness and helps the flow of the questionnaire. Is
most important and may determine response rate by persuading respondents

-Questionnaire should be clear, neat and easy to follow, have identifying information (the
introduction) The introduction should:
-explain purpose of the survey
-Provide identification information
-have instructions on how to answer the questions
-have element of confidentiality
*Should be a different font style, colour etc different from the questions’ font
style, colour etc
-Each question in a questionnaire should be numbered
-Never crumb the questionnaire
-Make a cover sheet or face sheet for each questionnaire
-Put the time, date, the interviewer’s name, respondent identification number (not name) spaces
on questionnaires or observation form

*Mail interviews may include a polite, professional cover letter on letterhead stationery,
identifying the researcher contact details and purpose of the study

(8) REPRODUCE THE QUESTIONNAIRE

The questionnaire should be reproduced on good quality paper, on side of a page, on
a single or double paged paper.

(9) ELIMINATE BUGS BY PRE TESTING

Pre testing refers to testing of the questionnaire on a small sample of respondents for
the purpose of improving the questionnaire by identifying and eliminating potential
problems. This is done prior to administering to the targeted population sample. Is a pilot
test aimed at improving the questionnaire/interview schedule

Methods of improving questionnaire through pilot tests

a) Think aloud interviews

Allow respondents to explain his or her thinking out loudly during the process of
answering each question

80
 b) Retrospective interviews and targeted probes
After completing a questionnaire, the respondent explains to researchers the process used
to select response or answers

c) Expert evaluation
Having an independent panel of experienced survey researchers reviews and critiques of
the questionnaire

d) Behaviour coding
Researchers monitor interviews, using audio or videotapes for misstatements, hesitations,
missed instructions, non-response, refusals, puzzled looks, answers that do not fit any of
the response categories etc

e) Field experiments
Researchers administer alternative forms of the questionnaire items in field settings and
compare results.

f) Vignettes and debriefing

Interviewers and respondents are presented with short, invented “lifelilke” situations and
asked which questionnaire response category they would use.

(10) Printing the final draft for administration

Points to consider when designing a questionnaire

 Should deal with significant topics only
 Seek data which cannot be obtained through other means or tools
 Should be attractive, neat in appearance and items clearly categorized
 Directions or instructions should be clear, complete and unique terms clarified
 Simply worded. Each question dealing with one idea and contain unambiguous
questions
 Questions should be objective and not leading
 Questions should be in good order, mostly from general to more specific ones
 Avoid offending, annoying and embarrassing questions
 Avoid descriptive adjectives and adverbs
 Avoid double questions
 Avoid double barrel questions, that is, two in one
 Should carry adequate number of responses and spaces
 Should be easy to interpret, tabulate, summarize and analyze
 A questionnaire should be self-administered

PRINCIPLES OF A QUESTIONNAIRE
 BREVITY – research questions should be written or asked as briefly as possible subject
to their being understood by respondents

81
  RELEVANCY – unless a question is relevant to the purpose of the research, it should not
be included in a questionnaire. Likewise unless a word is relevant to a particular
question, it should not be included in the question.
 OJECTIVITY – unless a question is worded objectively, it should not be asked. The
answer to a non-objectivity question is already known and there’s no need for asking.
 CLARITY – research questions should be as non-ambiguous as possible and technical
jargon should be avoided
 SPECIFICITY – a question must be specific and address one issue only.

INTERVIEW SCHEDULE
This is a data collection form with a list of questions on which data is recorded as it is
collected through personal(face to face) interview by the researcher. This is the other term
used to describe the personal interview questionnaire and the telephone interview
questionnaire.

Designing of an interview schedule as discussed under questionnaire design

The difference between a interview schedule and a questionnaires

All interview schedules are questionnaires but not all questionnaires are interview schedules.
i) Questionnaires are self administered while interview schedules are administered by
the interviewer. Questionnaires clearly highlight their purposes, introduce themselves
and give instructions to the interviewee but interview schedules everything is done
by the interviewer
ii) A questionnaire is completed by an interviewee while the interview schedule by the
interviewer
iii) Room for including other questions on the interview during interviewing is possible
which is not so with mail interviews(questionnaire)

Types of questions which can be included in both interview schedules and questionnaire

a) Open –ended questions

Are unstructured, free response questions. Are questions in which respondents are free to
offer any answer they wish in the question in their own wording.

Advantages of open ended questions

-Permit an unlimited and unexpected number of possible answers
-Respondents can answer in detail and can qualify and clarify responses
-Unanticipated findings can be discovered
-Permit adequate answers to complex issues
-Permit creativity, self-expression and richness of detail
-Reveal a respondent’s logic, thinking process and frame of reference

Disadvantages of open ended questions

-Comparisons and statistical analysis may be very difficult because different respondents
give different degree of detail in answers. and other responses may be irrelevant or

82
 buried in useless detail
-Coding responses is difficult
-Articulate and highly literate respondents have an advantage
-Questions may be too general for respondents who lose direction
-Responses are written verbatim, which is difficult for interviewers because of difference
in level of vocabulary
-A greater amount of respondent’s time, thought and effort is necessary which may lead
to boredom and non responses
-Respondents can be intimidated by other questions
-Answers take up a lot of space in the questionnaire

b) Closed ended questions

Are structured, fixed response questions in which respondents must choose from a fixed
set of answers. Questions can dichotomus, multichotomus or scale questions

Advantages of closed questions

-Easier and quicker for respondents to answer
-Answers of different respondents are easier to compare
-Answers are easier to code and statistically analyse
-Response choices can clarify question meaning for respondents
-Respondents are more likely to answer about sensitive topics
-There are fewer irrelevant or confused answers to questions
-Less articulate or less literate respondents are not at a disadvantage
-Replication is easier

Disadvantages of closed questions

-Can suggest ideas that the respondent would not otherwise have, therefore answers not
purely respondent’s own idea.
-Respondents with no opinion or no knowledge can answer anyway
-Respondents can be frustrated because their desired answer is not a choice
-It is confusing if many similarly response choices are offered
-Misinterpretation of a question can go unnoticed
-Distinctions between respondents’ answers may be blurred
-Clerical mistakes or marking the wrong response is possible
-Force respondents to give simplistic responses to complex issues
-Force people to make choice they would not make in the real world.

c) Partially open questions are semi structured questions containing both open and closed
ended characteristics or questions

d) Sleeper questions are questions about non-existent people or events to check whether
respondents are being faithful

e) Contingency questions are a two part survey question in which a respondent’s answer
to a first question directs him or her either to the next item or to a more specific and
related second question.

83
 Principles of good question writing
-Avoid jargon, technical terms, slangs and abbreviations
-Avoid ambiguity, confusion and vagueness
-Avoid emotional language and prestige bias eg words like murderous freedoms, peace
and any associate with prestigious or very respected person
-Avoid double-barreled questions eg Does your company have a pension and health
insurance policy? Do you support or oppose the idea?
-Avoid leading (loaded) questions eg You don’t smoke, do you?
- Avoid asking questions that are beyond respondents’ capabilities and what can be
difficult to recall
-Avoid false premise questions. Eg “The Post Office is open too many hours. Do you
want it to open four hours or less?
-Avoid asking about distant future intentions
-Avoid double negatives “I haven’t got no job” “ Do you disagree with those who do not
want to build a new swimming pool in the city” Are grammatically incorrect and
confusing. Second negative is used for emphasis.
-Avoid overlapping or unbalanced response eg Did you find the service at our hotel to
be :outstanding, excellent, superior or good

OBSERVATION FORM/SCHEDULE
Is a data collection form with a list off clearly defined observable factors, on which data is
recorded as it is collected by the researcher
After each observable factor, a space is provided for the observer to record the observations.
Recording can be either by writing a few descriptive words or by check mark to indicate the
presence, absence or frequency of the phenomenon being measured or a direct mark against
each factor

Designing an observation Schedule

The researcher should design an observation form that identifies the required data clearly,
makes it easy to record the data accurately and simplifies an observation form as a consistent
measuring instrument. The researcher needs to answer the following questions:
1) Who should be observed?
2) What should be observed?
3) When should the observation take place?
4) Where should the observation take place?

Characteristics of a good observation form

o Should identify clearly the required data
o Should record all the data correctly
o Should simplifying the coding of data
o Should be brief, not too long
o Factors to be observed should be logically categorized

84
The defined observable factors that are relevant to the problem should be grouped into logical
categories and worded. This enables the researcher to record many different observations
quickly and ensure that does not overlook any relevant evidence.
The form and layout as well as the reproduction of observation forms should follow the same
guidelines discussed for questionnaires. Like questionnaires, observation forms require adequate
pre testing.

OTHER RESEARCH INSTRUMENTS

a) Checklist – Is a data collection form with a list of clearly defined observable factors, on
which data is recorded as it is collected by the researcher during observation or
abstraction. Used for collecting secondary data
b) Document Schedule – Is a data collection form with a list of questions, on which data is
recorded as it is collected from various documents such as files and books etc
c) Evaluation Schedule – Is data collection form with a list of question, on which data is
recorded to measure the performance of a particular activity such as rural development
programme

Threats to the validity of collected data

Compbell and Stanley (1963) coined the phrases, “Threats to validity” for alternative
explanations of data that might explain a finding as well as or better than the explanation
intended by the researcher.
1. Descriptive mortality.
Mortality refers not only to death of subjects but also to individuals dropping out of a
group making it no longer representative of the population of study. Mortality is a
serious problem in survey research where the researcher starts a representative sample but
selectively loose subjects who do not return the questionnaire. The ones who do return
the questionnaire are probably comfortable dealing with the topic, where as those who are
uncomfortable with it do not return it, the mortality factor.

8. Instrument decay is due to a change in the measuring instrument over the course of
the study. Refers to, for instance, rules made up in the course of using an
observation checklist where, on encountering a situation not provided for by rules for
checklist, the researcher makes up other rules or forgetting the decision and make up
a different rule for the same next instance. Inconsistency is using the instrument may
be a reasonable alternative explanation rather that some characteristics of the group as
originally intended.

In qualitative research, the observer is the instrument. We have to wonder how the
observer’s changes in the perception of the observed over the period of the study and
how that affects the observations. The most serious change of this type is “going
native” losing perspective on the group observed and becoming entirely one of them.

Sadler (1981) in an article describing the mind’s cognitive limitations as they affect
qualitative data gathering, notes that there is long history of efforts to identify the
sources of distortion that cause the mind to make errors of judgment and inference.

85
 Listed a number of information processing limitations, man y of which will be
similar to qualitative researchers but all of should be of concern to both of them.
These are :

a) Data overload
b) Availability of information
c) Positive and negative instances- research shows that people tend to ignore
information that conflicts with already held hypothesis even intelligent
individuals adhered to their own hypothesis with remarkable tenacity when they
could produce confirming evidence for them.
d) Correlation and co-occurrence
e) Internal consistency, redundancy and novelty of information
f) Base rate proportion is the frequency with which a behavior occurs. There is
great difficulty judging base rate from small samples
g) Uneven reliability of information or data- people tend to treat data from an
unreliable source almost the same way as data from reliable sources
h) Confidence in judgment – once an assessment is made people have been shown
to have an almost unshakable confidence in the correctness of their decision,
even in the face of considerable relevant, contrary evidence.
i) First impression

DATA PREPARATION ,PRESENTATION AND ANALYSIS

1 Data Preparation – Is the process of translating or transforming raw data into a form
suitable for analysis. The process takes place after data collection but prior to data
analysis. Its purpose is to prepare the data for analysis. Act as a preliminary step to
data analysis. Some common data preparation activities are classification, editing,
coding, and tabulation are discussed below

A) CLASSIFICATION OF DATA
- Is the process of arranging data into groups or classes on the basis of common
characteristics. Data having common characteristics are placed in one class.
Classification of data is usually done according to attributes or class intervals.

Classification on the basis of attributes

Attributes refer to the particular characteristics of the population Attributes can be
descriptive or numerical

Descriptive attributes such as sex, education, marital status etc.

are qualitative and cannot be measured in numerical terms e.g. Zimbabwean population
can be classified as follows:

TOTAL POPULATION

86
 Rural Urban
Male Female Male Female
Literate Illiterate Literate Literate Illiterate Literate

Numerical attributes are quantitative and measurable attributes. Are expressed and
expressible in numbers e.g. age, height, income etc

Classification on the basis of class intervals

Is numerically based, that is, based on statistical variables
- Unlike descriptive characteristics, numerical characteristics refer to quantitative
phenomenon which can be measured through some statistical units. Data relative to
income, production, expenditure, investment etc can be classified according to these
statistical variables. When the goal of data classification is to arrange a set of data into a
useful form, frequency distribution provides a general approach to data classification.

In frequency distribution raw data is represented by distinction groups called classes.

The number of measurement in each class is called the class frequency. In this way the
data set which may be very large is condensed into smaller, more manageable set of
numbers.

B) EDITING OF DATA
Editing of data is the review of data collection form with the objective of increasing
accuracy and precision. Researchers edit data to discover errors and omissions, and
where, possible correct them. Editing is usually done in two stages, that is, field edit and
central office edit.

i) Field Edit
Is a preliminary edit, typical conducted by a field supervisor which is designed to detect
the most glaring omissions and inaccurate in a completed data collection form.
Items commonly checked for are completeness, legibility, comprehensibility, consistency
and uniformity. This is mostly for verification of data so as to identify conformity to
physical traits.

ii) Central office edit

Is a thorough, more complete, complicated scrutiny and correction of competed data
collection forms. Unlike the field edit, central office edit depends less on follow-ups
procedures and more deciding what to do with the data. It focuses on validation of data.

C) CODING OF DATA
Coding is the assignment of a code(symbol) to represent a specific response to a question
along with the data record and column position (or field) that code will occupy. A field
represents an item of data, such as sex of the respondent. A record consists of related
fields such as sex, marital status, house hold size, age, occupation etc.

87
 Coding at act as a technical procedure by which data are categorized. Involves
specifying the alternatives categories/classes into which the responses are to be placed
and assigning codes(symbols) to these classes. The codes can be numerical, alphabetical
or alpha-numerical.

Steps in coding

1- Determining and specifying categorises or classes to which the responses are to

be placed. The number of classes will depend on the response format (choices)
expected. The categorize should be mutually, exclusive, and exhaustive so that
every response falls into one category.

2- Assigning code numbers to the classes eg 1 for yes , 2 for no answer

Rules to be observed when assigning codes:

i) Locate only one character in each column
ii) Mostly use numeric codes and not letters of the alphabet
iii) Have as many columns as are necessary to capture all the variables
iv) Use other codes for no information eg “I do not know” can be given code
8, “No” can be coded 9 and none response coded 10
v) Code in a respondent’s identification number on each record.

Coding of closed questions is simple because the coding is established when the
data instrument is being designed. Respondents code themselves and also their
responses

Coding open ended questions can be very difficult as the coder has to determine
category basing on answers that are not always anticipated but which may be
different

3 Preparing a separate code book, is the final step, which contain the general
instructions indicating how each item of data was coded. Should list the code for
each variable and the categories included in each code.

Coding can either be pre-coding or post coding.

Pre-coding
- Pre-coding is determining the codes at the time of designing the data collection
instruments and therefore the codes can be printed on forms. This form of coding is most
appropriate for questions that are dichotomous, multiple choice and scale (closed
questions)
- For these forms of questions the researcher knows the potential symbols which will be
assigned to the different responses. This greatly reduces the task of later data preparation
and also tends to reduce coding errors and is less time consuming. E.g. question 3 on a
data collection instrument is about the marital status of the respondents with two possible
answers i.e either yes or no. The researcher may give codes 1=yes and 2=no.

88
 Post coding
- is determining the codes after data collection forms have been completed. This form of
coding is most appropriate for open-ended questions. One approach is for the
researcher to sample about 20% of the completed data collection forms and examines the
specific responses to specific questions. From this sample then establishes the necessary
categories that are appropriate to summarize the data for coding purposes.

After categories have been established each data collection form must be reviewed and
particular responses put into the appropriate categories. The codes used to identify the
appropriate category should be written in the margins of the appropriate data collection
instruments.

D) TABULATION
- Is a method of presenting statistical data in a table. A statistical table is an orderly and
systematic presentation of numerical data in columns and rows. Data is tabulated
according to quantitative characteristics and nature of variables. The main object of a
statistical table is to arrange data into a physical presentation of numerical facts to attract
and direct the attention of the reader automatically to relevant information. It facilitates
comparison by bringing related items of information close to one another.

Statistical table on putting and counting the number of cases that fall into the various
categories
Tabulation can be simple tabulation or cross tabulation

Simple tabulation involves counting a single variable. It may be repeated for each of the
variable in the study but the tabulation for each variable is independent.

Cross tabulation is when two or more variables are treated at the same time, that is,
counting the number of codes that fall into each of the covered categorises when the
categorises are based on more variables considered at the same time eg. One male and
female are earning 150 per month

In tabulation the researcher should be able to:

i) Determine the degree of non-response
ii) Locate a blunder which is an error that arises when editing, coding or
tabulating data
iii) Locate an out lier. These are observations which are different in magnitude
from the rest of observations
iv) To determine the empirical distribution of variables in question
v) To calculate summary statistics

PRINCIPLES OF GOOD TABULATION

 Every table should be numbered to facilitate reference to it.

89
  Every table should have a brief but descriptive title. At the end there should be a note of
the source of the data, which the researcher has used. Whether it is based on the
researchers own survey data or secondary data.
 Give brief descriptive headlines to rows and columns showing clearly the units which
quantities are recorded. If necessary the units of measurements should be defined in a
foot note.
 Use different lines to break up a large table. Double lines or thick add a great deal for the
a table to be understood.
 Don’t overcrowd tables. If the volume of the data is large, two or more simple tables are
better than one cumbersome table.
 Whenever it is useful, insert both row and row totals.
 Enclose the table within a book frame to make it look attractive.
 Should be neat, a messy table conveys nothing

COMPUTER AID
 Computer data processing is faster and more accurate than manual data processing. The
coding of data permits the transferring of data from data collection form into a computer
readable form which can then be processed by the computer.
 Computer processing is depended on what is referred to as data structure. A data
structure is simply the way the data is positioned in a storage medium. The researcher
must organize coded data into the appropriate data structure.
 Computers have direct data transfers capability through terminal, keyboard or optical
reader, to either disk or tape storage. With direct data entry, the researcher organizes
coded data into fields, records and files that constitutes the components of data base
system.
 The collection of characters (ie code) that represents a sign items of data is referred to as
a field. A group of related field is treated as a unit is referred to as a record.
 A collection of related records that are stored together called a file.

DATA ANALYSIS
- Is the application of statistical techniques to refine and manipulate data in order to make
references about variables or is the application of analytical methods to manipulate data.
May result in understanding the inter-relationship between variables and the quantitative
meaning of data relating to them becoming evident.
- Data analysis is aimed at testing the formulated hypothesis either for acceptance or
rejection.

TESTING OF HYPOTHESIS
- The hypothesis that were formulated need to be tested for their acceptance or rejection.
A hypothesis is tested by making use of a pre-decision rule which is applied on sample
data and which guides the researcher in deciding whether to accept or reject the
hypothesis on the basis of the outcome of the sample drawn. Hypothesis testing may take
the following form (terms) :-

1 The study of hypothesis for logical consistency Is a phase of thinking. The study consists
of:

90
 Firstly it consists of checking the logical reasoning by which the consequences
of the hypothesis are deduced for verification.

Secondly it involves checking the hypothesis for agreement with the already
known laws of nature. The hypothesis must not disagree with those principles of
science which are considered valid beyond reasonable doubt.

2 The study of hypothesis for agreement with facts

Is the study of hypothesis for agreement with facts the researcher organizes. Researcher
argues truthness or falseness of hypothesis under given conditions. Probes into whether
these conditions are present. If the conditions are not present, then he should restart his
hypothesis and collect fresh data for making the study worthwhile.

After testing a hypothesis may be accepted or rejected. Sometimes the facts and data may
suggest the revision or modification of the hypothesis for the purposes, verifying statistical
techniques are available for manipulating data to test different kinds of hypothesis. The
technique to be adopted depends upon convenience, availability and reliability of data,
knowledge of the researcher and the level of sophistication required. Hypothesis testing is a
procedure of arriving at a decision of either accepting or rejection a hypothesis.

These are standard steps which should be followed when testing the validity of hypothesis,
which are:
1. Stating the Null Hypothesis (Ho)
2. Stating the alternative hypothesis (H1)
3. Stating the level of significance and determine the Z values.
4. Stating the decision rule.
5. Calculating the test statistic.
6. Making a conclusion.

Data analysis should be done to refine data so as to get information for testing hypothesis.
Various data analysis techniques can be used

DATA ANALYSIS TECHNIQUES

Manual and statistical techniques can be used but statistical techniques are widely used in data
analysis. Manual techniques are mainly used on qualitative data and are based on human
judgement.

Various statistical techniques are used for manipulation of data to test hypotheses. The technique
to be adopted depends upon convenience, availability and reliability of data, knowledge of the
researcher and the level of sophistication and required accuracy.

The following are some of the reasons why statistical techniques are considered important in
research:-

91
 1. Statistics are the tools to judge the significance of the data and make maximum
use of the data collected.
2. Statistics techniques permit the most exact kind of description of the observed
phenomenon. They force the researcher to be definite and precise in the thinking
and procedures, this helps in the formulation of hypothesis for testing.
3. Statistical tools enable the researcher to test theories empirically. If the data is not
consistent with the theoretical predictions, the theory should be revised or
rejected. If the theory is judged valid, further empirical analysis may be needed to
measure relationships.
4. Statistical techniques help in summarizing the result in a meaningful manner.
5. Statistical tools help the researcher in drawing general or broad conclusions.

Statistical techniques widely used in research are :-

A. Measures of central tendency
B. Measures of dispersion
C. Time series analysis
D. Measures of relationships

1. MEASURES OF CENTRAL TENDENCY

Define the centre of any given data set. Three common measures of central tendency are
mean, mode and medium.

i) Mean refer to the average of the data set. Types of averages are arithmetic
(simple), geometric and harmonic mean.

Arithmetic mean is obtained by summing up given values and then dividing by

the number of values. Is the widely used mean. Can calculated for both grouped
and ungrouped data

Mean for ungrouped data = +(x) x = values or numbers ,+ = summation

n n = number of values

Mean for grouped data = +(fx) + = summation, f = frequency

+f x = class mid point

Geometric mean is the Nth root of the product of a given N scores obtained as a
result of measurement. Mostly used for averaging ratios
3
Geometric mean = AxBxC

Harmonic mean is used for averaging rates of changes eg speed

= n
+(1/x)

Advantages of mean

92
 -Consider every score in the data set
-The result is more stable as compared to other measures of location
-Easy to calculate and understand
-Very useful when comparing different data set

Disadvantages of mean
-Can be affected by extreme scores e.g. in 1, 4, 5,6,28. 28 affect the mean
upwards
-Cannot be used in a incomplete distribution

ii) Median is the middle number found when scores are put in array (ascending or
descending order). It splits the data set into two equal halves above and below the
median number. Similarly can be calculated both for grouped and ungrouped data.
On ungrouped data, the scores should be noted whether are odd or even.

Median for ungrouped data

i) 8, 5, 6, 10, 13 = 5, 6, 8, 10, 13 = 8

ii) 8, 4, 5, 6, 10, 13 = 4, 5, 6, 8 10, 13 = 6+8 = 7

2

Median for grouped data = L + i (n/2 – cf (m)

fm
L = lower class boundary of the median class
i = class interval or width
n = sample size or (sum of all frequencies)
cf = the cumulative frequency prior to the median class
fm = frequency of the median class

Advantages of median
-Is insensitive to extreme scores
-Useable in an incomplete distribution (open ended distribution)
-Can be calculated from any data set excluding nominal data
-Easy to obtain by mere inspection

Disadvantages of median
-Is less stable
-Consider only one or two scores not considering all values
-Applicable to few operations
-Is time consuming because of need to arrange scores in array.

93
 iii) Mode is the most frequent occurring score in a given data set. There can be more
than one mode for a given data set or no mode at all. Can be deduced from
grouped and ungrouped data

For ungrouped data is seen by grouping is array

For grouped data L + i (D1)

(D1+D2)
L = Lower class boundary of the modal class
i = class interval
D1 = frequency of the modal class less frequency of the class above it
D2 = frequency of modal class less frequency of the class below it

Advantages of mode
-Useable on nominal data, that is, qualitative variable data eg sex
-Locates the highest concentration of scores
-Not affected by extreme values
-Can be calculated even if all the values in a data set are unknown

Disadvantages of mode
-Do not necessitate comparison
-Is less stable

Relationship between mean, median and mode

Three relationships exists which are normal distribution, positively skewed and
negatively skewed distribution.

a) If the mean = median= mode, the distribution is said to be normal distribution and
that’s, it has got a line of symmetry, hence is not skewed to any side. The
distribution is bell shaped as below:

b) If the mean is greater than the median, greater that the mode, the distribution is
positively skewed. Is more heaped to the left of the line of symmetry. Its
skewness is greater than zero. Mean is a true representative of the distribution.

c) If the mean is less than the mode, less than the median, the distribution is
negatively skewed, that is, more heaped to the right of the line of symmetry.
Skewness is less than zero. Mean is not a representative of the distribution.

*Skewness measures the extended deviation of data set from the line symmetry and
is represented by the formula = Mean – Mode
Standard deviation
2. MEASURES OF DISPERSION
Dispersions is a measure of scattered ness of items around the central values. It shows

94
how the observations are deviated from the central values.
The methods of measuring dispersion are range, quartile deviation, mean deviation,
standard deviation, coefficient of variation and Lorenz curve.

Range is the difference between the highest and the lowest values in a given distribution.
Mostly used with statistical quality control

Advantages
-Is simple to calculate and understand

Disadvantages
-Is influenced by extreme values and hence not reliable
-Do not indicate the degree of clustering

Quartile deviation also known as semi-inter quartile range

Is similar to the range but differ in that a middle quarter range is found thus leaving out
the effect of unusual extreme scores, thus, eliminating the extreme value effect as on the
range. The bottom and top quarters are not considered. Quartiles divide the given
distribution into four (4) equal parts that is Q1, Q2, Q3 and Q4.
The quartile deviation = Q3 - Q1
2
Mean deviation Is the average deviation of all values ina given data set from the
distribution mean (arithematic mean). Is found by adding up deviations of all values from
the distribution mean and dividing by the number of items. Negative signs are ignored.
The mean deviation should be +x

Ungrouped data = +(x –X) X = arithematic mean

n x = value in the data set
+ = summation mark
Grouped data = +[f(x-X)
+f f = frequency x = class mid point
Standard deviation
Is the square root of average of the differences of the individual scores from the
aarithematic mean. Is the mostly used used measure of dispersion about the mean. The
larger the standard deviation in comparison with data size, the wider the spread of data
around the mean.

For ungrouped data = +(x – X)2

N

Grouped data - Population = +f(x – X)2

-Sample +f(x – X)

+f – 1

95
 Lorenz Curve
This curve is graphic method of showing to what extent different variables deviate from
uniformity or to show degree of deviation in order to make comparisons. If two variables
are constantly relative to each other, a percentage changes for the other variable. This
situation may be represented on a graph by a straight line. This line is always 45 degree
to the origin and is known as the line of uniformity or equal distribution.

Line of unequal distribution

Line of equal distribution

Coefficient of variation (Cv) is used to compare the variability of the data when two sets of
data are given. Is given by the formula
Cv = x 100
x(mean)

3 TIME SERIES ANALYSIS

 It shows how data will fluctuate. A set of observations of a random variable are arranged
in chronologically i.e. according to time. Represents data measured regularly over a
period The utility of time series analysis can be summarized as follows:-
 It helps to understand past behavior
 It helps in planning future operations
 It helps in knowing current accomplishment.

4 MEASURE OF RELATIONSHIP
The measures of relationship or association are correlation and regression analysis.

Correlation analysis indicates the extent to which a relationship exists between at least two
sets of data under consideration and closeness between two variables. The correlation is a
value -1 < r < 1 inclusive

When r = 1 indicates a perfect positive correlation

r = -1 indicates a perfect negative correlation
r = 0.5 indicates a positive correlation

For 0.5 < r > 1, the relationship is said to be strong correlation

0 < r > 0.5, represents a weak positive correlation

96
 -0.5 < r > 0 , is a weak negative relationship

When r = 0 there is no relationship

Regression analysis is a quantitative method which attempts to set a forecasting model

to estimate the value of dependant variables under circumstances in which an acceptable
correlation exists between the dependant variables and one or more independent variables.
The general equation for regression analysis is a straight line y = mx + c where m is the
gradient and c is the y intersection

Errors associated with data analysis

i. Absolute error:- is an error that gives the greatest the difference between the true value
and the approximate value. Is an error that gives the greatest difference that exists
between the approximate value and the true actual value. In measurement, the absolute
error is half the smallest of measurement e.g for centimeters is 0.5 cm. If a number is
rounded to 10 or 100, the absolute error is 5 and 50 respectively. For one decimal places,
the smallest unit of measurement is 0,1 therefore the absolute error is 0.05

ii. Relative error:- is the absolute error expressed as a proportion of the true value. Is given
by the formula = absolute error x 100
Actual error

iii. Standard error:- is the standard deviation of the sampling distribution.

DATA PRESENTATION
The results of the data analysis should be presented in a form suitable effective for
communication of the results.
The form of presentation will depend on :
-The purpose for which the information is desired.
-Type of results and findings, that is, qualitative or quantitative

Some of the common methods of data presentation are discussed below:-

a) NARRATIVE
 This is the textual presentation of research results i.e. reporting the results in words.
The use of words allows explanation emphasis and interpretation to be made.
However, personal bias can distort the data and data which does not fit the
researcher’s pre-concerned ideas maybe omitted or not dealt with in sufficient depth.
Suitable only for qualitative data

Disadvantages
-Personal bias can distort data
-Ambiguity is at extreme
-Language difference and vocabulary may be a problem
-Interpretation of data is not at a glance

97
 b) TABLES
 Tabulation is common method of presenting statistical data. A table displays detailed
statistical information precisely so that interpretation of results can be at a glance.
However, no interpretation of results can be given in the table. It is also difficult to
emphasize any particularly interesting reading unless heavier type, underlining or a foot
notes are used to enhance understanding of tables. The principles of good tabulation
have already been discussed.

c) DIAGRAMS AND CHARTS

 There are various diagrams and charts that can be used to give a pictorial presentation
of data. A chart or diagram cannot present information precisely, but it can give a
quick overall impression of the findings.
 Can give a quick overall impression of findings by a mere glance and then encourages
reading text to have a more complex account of findings.

General rules to be followed when drawing diagrams or charts etc.

 Give clear and brief tittles.
 Label the axis clearly
 Scales of values and their origins should be clearly checked.
 Excessive details should be avoided.
 Reference to the source of data should be made or an accompanying table provided.

The popular types of diagrams and charts are bar charts, pie carts and pictograms.

d) GRAPHS
 Diagrams and charts are primarily intended to present data whereas graphs go further
than this and help in the analysis and interpretation of data.
 In some cases, graphical methods are usually not as precise as calculation but their
speed and overall insight they give them evaluate tools of the researchers.

e) EQUATIONS AND SYMBOLS

 The symbols used in equations should be adequately explained.

4) INTERPRETATION OF DATA
-Is done after data presentation
-The product of data analysis is research results or findings.
-Presenting a mere summary of statistical results is not enough.
-The researcher should interpret the results in light of the research questions, objectives and
hypothesis being addressed.

The process of interpretation is essentially one of the stating what the results of the findings
show. Among others it shows or answers the following questions;-
 What do the research results show?
 What is their significance?
 What is the solution to the original problem?

98
 The interpretation of the results leads to conclusions and recommendations as discussed
below

a) Developing Conclusions from findings and hypothesis

In interpreting results, conclusions are drawn with direct reference to the research objectives
and hypothesis of the study. There should be a conclusion for each objective and for the
acceptance or rejection of hypothesis.

The researcher should compares the objective with the findings and draws conclusions on
how much and in what manner the objectives have been achieved to enable the readers to
turn to the conclusions section, and find specific conclusion relative to each objective.

In the same way the researcher must also be specific with reference to the hypothesis of the
study. Researcher should be specific to the hypothesis of study by indicating whether it has
been accepted or rejected. In doing so, should carefully review the evidence for and against
each hypothesis.

The element of comparison is fundamental to all research interpretations. Comparisons of

one’s investigational findings, with a criterion, with results of other comparable
investigations with normal or ideal conditions or with opinions of expert forms an important
aspects of:-
 Appropriate depth and breath so as to meet caliber of research.
 Feasibility of research problem.
 Time frame.
 Availability of resources.
 Justification and purpose of the research.
 Usefulness of the aims and objectives.

INTERPRETATIVE EFFORTS OF THE RESEARCHER

Once conclusions have been drawn reflecting some properties of the population, one can
estimate how far these characteristics or conclusions can be generalized to a large population
and predict modification which would have to be taken into consideration when undertaking
such a generalization.

Each generalization must agree with facts revealed by the investigation.

The researcher should also check each generalization against the facts and experience of
other investigations and previously established generalization.

Developing recommendations from findings and conclusions.

- Based on the research results and conclusions the researcher makes recommendations to the
decision maker.
- Recommendations are suggestions for application and implementation to solve the
organizational problem drawn by the researcher basing on findings and conclusions
- Recommendations should be feasible, practical, actionable and directly usable as inputs into
managerial decision making.

99
COMMUNICATION OF RESEARCH RESULTS

Items to include when reporting survey research

a) The sampling frame used e.g. telephone directories

b) Dates on which the survey was conducted
c) The population represented by the sample
d) The size of the sample from which the information was collected
e) Sampling methods and steps
f) Methods of survey used e.g. face to face, telephone interviews
g) The exact wording of the question asked
h) Organizations that sponsored the survey
i) The response rate or percentage of those contacted who actually completed the
questionnaire
j) Any missing information or Do not know responses when results in specific questions are
reported.

100