RESEARCH METHODS

NOTES
DEFINITION OF RESEARCH

1 Research is a systematic process of gathering, analysing and interpreting data for the
purpose of providing information for decision making.

2 It is a logical system and scientific way of solving problems.

3 It is a systematic investigation of a problem/question or phenomenon using scientific

methods of acquiring knowledge.

 Research is a process – That means it takes some time to undertake.

 Research is systematic - It follows a particular procedure of gathering and analysing

data. This procedure can be followed by anyone doing research.

 It is a way of gathering/collecting data ie research results in the collection of data

therefore there must be ways of collecting the data.

 When collected, the data must be analysed in order to find answers to the problem.
A research must have a problem, that is, something of concern you find answers to.

THE PURPOSE OF RESEARCH

 Research is directed towards the solution of a problem,

 It aims at developing generalisations (ie explanations), principles or theories that will

help in predicting future occurrences.

 It describes natural or human-made phenomena, their form, structure, activity,

change over time and relationships.

 It provides statistical information about different aspects which may be useful for
policy makers.

 Research may be focused on improvement of systems, products, inventions etc.

 It results in the production of new knowledge.

 Research can upgrade/update already existing knowledge.

CHARACTERISTICS OF A GOOD RESEARCH

 It is directed towards solving a problem.

 It uses evidence that can be observed and verified.

1
  It depends on valid and reliable data gathering procedures.

 It generates and adds new knowledge.

 Research is systematic and follows particular procedures and steps.

 It requires expertise, the researcher should know what is known and seek to fill the
gap.

 It provides original and innovative answers to questions.

 It is replicable – The research process is detailed to allow other researchers to repeat

the study if there is need.

 The purpose of the research is clearly defined.

 It applies high ethical standards.

 Limitations are frankly revealed.

 Conclusions are drawn from the findings of the research.

 Research is objective; different researchers using the same procedures come up to

the same conclusions.

BROAD GROUPS OF RESEARCH

SCIENTIFIC RESEARCH

 A scientific enquiry is also known as an empirical enquiry. It emphasises direct

observation and experimentation as ways of answering questions.

 It involves collection, organisation and analysis of data,

 It draws conclusions which entail the acceptance or rejection of hypotheses.

NON-SCIENTIFIC ENQUIRY (EVERYDAY APPROACH TO KNOWLEDGE)

 It is also known as an unempirical method of research.

 It relies on logical reasoning and common sense as ways of finding knowledge.

 It also relies on personal experience and other unreliable methods of discovering the
truth.

CHARACTERISTICS OF NON-SCIENTIFIC RESEARCH

2
  Tenacity – clinging to certain beliefs regardless of lack of supporting evidence.
Statements are presented over and over again and accepted as true without
examining or evaluating them.

Once a person accepts a certain belief without data to support it, the person will
often create a reason for its truth; the person often refuses to accept new
information that contradicts it.

 Intuition – what is considered to be common sense or is self - evidence knowledge.

People act on the basis of what feels right or what feels reasonable. It is a product of
insight and what is perceived to be the truth. However, many of self-evidence truths
are found to be false.

 Authority – It relies heavily on authority. This means reference to some authority has
long been used as a source of knowledge. People accept a new idea because some
authority figure tells them it is so, and so they repeat or live by what they are told.

Authority figures include; famous educators, physicians, traditional healers,

prophets, etc. The bases of their knowledge are one or more of; opinion, tradition,
revelation or direct experience. Authorities however, often contradict each other.

Parents and teachers are authorities for their children.

 Rationalistic Method – Non-scientific enquiry involves a rationalistic (reasoning)

method of collecting data. People use reason every day as they try to solve problems
and understand relationships. Reason alone may not be accurate if there is no way
to determine the accuracy of the assumption on which it is based.

COMPARATIVE CHARACTERISTICS OF SCIENTIFIC AND NON-SCIENTIFIC APPROACHES TO

KNOWLEDGE

NON-SCIENTIFIC

(EVERYDAY) SCIENTIFIC

General Approach Intuitive Empirical

Observation Casual, Uncontrolled Systematic, Controlled

Reporting Biased, subjective Unbiased, Objective

Concepts Ambiguous with surplus Clear definitions, operation

3
 meanings specific

Instruments Inaccurate, Imprecise Accurate, Precise

Measurement Not valid, Unreliable Valid, Reliable

Hypothesis Untestable Testable

Attitude Uncritical, Acceptable Critical, Sceptical

TYPES OF RESEARCH

BASIC/FUNDAMENTAL/PURE RESEARCH

 Basic research involves development of an idea or theory (idea)/theory building.

 It deals with theoretical problems.

 It uses the laboratory setting (experimental).

 It frequently uses animals as subjects

 Conditions are carefully controlled within the laboratory setting so that the results
are not affected by other influences.

 Basic research produces results that have direct applications eg research on human
development which includes learning theories, moral development theories, social
development theories etc.

APPLIED/ACTION RESEARCH

 It tends to address problems of immediate concern.

 It uses real world settings ie it uses human subjects

 It gives results that are of direct value to the practitioner.

 It is concerned with applying or testing theory, and evaluating its usefulness in

solving problems ie it is concerned with improvement of actual practice.

 Action research is a unique form of applied research.

4
  It is a reflective process of problem solving.

 It is called ‘’ practitioner research” because of the involvement of the practitioner in

real life.

 Action means that the practitioner is involved in the collection of data, analysis and
interpretation of the results.

 She/he is involved in the implementation of the results of the research and is thus
well placed to judge the effectiveness of the interventions.

EALUATION/EVALUATIVE RESEARCH

 It is a systematic method of collecting and analysing data in order to make decisions.

 Its purpose is to establish the worthiness of a programme against a course.

 It involves value judgement.

 Sometimes it involves comparison of two similar programmes or projects eg

comparison of two teaching methods or products.

RESEARCH AND DEVELOPMENT

 It aims at developing effective products eg training materials, management systems

 This type of research is expensive (demanding in terms of objectives and personnel

as well as time to complete it.

 Products are developed to meet specific needs. The finished products are tested and
revised in the field until the final level of effectiveness is achieved.

 Research and development is expensive but results in quality products to meet

consumer needs.

HISTORICAL RESEARCH

 It is research that attempts to make accurate and authentic reconstruction of the

past through a systematic analysis and synthesis of verifiable evidence.

 It seeks to arrive at conclusions concerning causes, effects and trends of past

occurrences.

 Historical research seeks to understand present events and to anticipate future

events. Emphasis is on objectivity and authenticity of evidence because there is
reliance on other people (the past experiences they report or remember),objects

5
 which bear attributes of the past (such as social and cultural artefacts), or records
which purport to point out aspects of a past reality.

 Historical research is guided by hypotheses.

 It uses already available data and these data are referred to as primary and
secondary data.

(PRIMARY DATA

 This refers to first-hand knowledge which can be obtained from eye witness reports
and original documents. These are hard to come by but they are more accurate than
secondary data.

SECONDARY DATA

 It refers to second hand information eg description of an event by a person other

than the eye-witness.

Secondary sources are subject to criticism. These are internal and

external criticism.
INTERNAL CRITICISM

 This evaluates the worthiness of data.

 It seeks to find out the degree to which data are reliable.

 It also seeks to establish if the data support the hypothesis.

EXTERNAL CRITICISM

 It assesses the authenticity/validity of data.

STEPS IN HISTORICAL RESEARCH

1 Location of evidence

2 Gathering and evaluation of evidence

3 Interpretation

4 Drawing conclusions

CORRELATIONAL RESEARCH

 It aims at determining whether there is a relationship and to what degree a

relationship exists between two or more variables.

6
  It seeks to use relationships in making predictions eg. the relationship between age
and performance at work or the relationship between qualification and competence
at work.

 In this kind of research changes in one variable can be associated with changes in
another.

EXPERIMENTAL RESEARCH

 It tries to establish cause and effect.

 It involves group comparison. The alleged cause (activity believed to make a

difference) is referred to as the treatment.

 The cause is also known as the independent variable.

 The difference/effect is the dependent variable.

 Experimental research seeks to investigate eg the effect of rewards on attitude

towards work.

 For this, two groups are used in the experiment. Members are randomly assigned to
the groups.

 The group that is given rewards is the treatment group and the group that is not
given rewards is the control group.

CAUSAL COMPARATIVE STUDY

 This type of research is similar to the correlational one, but it goes beyond
establishing mere associations.

 It seeks to establish causality as well, ie the extent to which changes in one variable
can be attributed to changes in another or others. The way it does so is to start off
with the observable effect and work backwards ie. Regression.

 Regression ascertains the ``existence or non-existence of a causal relationship but

only when there is an established underlying theory relating to the variables being
studied.

 A study example can be, the effect of sex on job performance. In the study, the cause
(independent variable) is not manipulated. One cannot manipulate sex (male or
female).Some causes can be manipulated eg a management system.

 Groups are compared on some dependent variables eg job performance.

7
  Causal and effect relationships are of low expenses because they do not take much
time to conduct.

 It can lead to acceptance or rejection of hypothesis.

RESEARCH DESIGNS

Kothari and Garg (2014) define a research design as a laid down plan of how a research is
going to be conducted. It portrays how the study will be structured that is ,it contains
certain steps of how the study will be carried out.

SURVEY RESEARCH DESIGN

To survey is to see over and beyond.

 A survey is a systematic collection and presentation of data.

 Surveys give a clear picture of a particular situation (descriptive surveys).

 They aim at quantifying distribution of certain variables in a study population eg

attitudes, values, thoughts, feelings, opinions etc.

 They are used in a wide range of areas eg in marketing research, public opinion
polling, voting intentions surveys and media research (rating surveys).

 They are designed to measure what occurred rather than Why.

 They are used to measure characteristics of a particular population either at a fixed

point in time(cross-section) or comparatively over time(longitudinal).

 They are normally used to collect data from a large sample of people.

 Data is collected using structured questionnaires although observation and

interviews are also used.

 Sampling techniques used often include: simple random, stratified random and
cluster sampling.

 Results are highly generalizable.

EXPERIMENTAL RESEARCH DESIGN

 Individuals are randomly assigned to at least two groups.

 One group is a control group and the other is the experimental or treatment group.

 The outcome of the intervention (treatment) is measured comparing the two groups.

8
  Randomisation is done to avoid confounding issues.

 Pre-testing is done.

 It makes use of small populations and random sampling techniques.

 The order is as follows: Pre-test – Treatment (experiment) – Post-test.

CASE STUDIES

 A case study seeks to study the characteristics of an individual unity.

 It involves a single institution, a group or an individual, a class or a community.

 It provides useful insight to the problem.

 It is mainly useful for studying behaviour, beliefs and attitudes.

 Data collection instruments are: open interviews (because there are few people
involved), direct observation or participant observation.

 A sample can be studied for a period more than a year.

 Generalizability of results/findings is limited.

EVALUATIVE DESIGN

 The design is used to collect data about a product while it is being developed or
programmed or after its completion.

 It aims at making judgements about merit, value, strengths of a system, procedure,

training programme, management etc.

 It is usually initiated by someone in need of a decision to be made concerning policy

management.

FORMATIVE EVALUATION

 It is done during the implementation of a programme or during production.

 It is used to guide the development of new products.

SUMMATIVE EVALUATION

 It is done at the end of a programme or completion of a product.

9
  It is conducted to determine how worthwhile the final product is in comparison
with other competing products.

 It is carried out to establish how effective a new system or programme is.

CONCLUSIVE RESEARCH

 This is confirmation research. The researcher already has a hypothesis.

 She/he collects data or evidence related to the hypothesis.

 The researcher examines the evidence to either accept or reject hypothesis.

 The design ends with conclusions recommendations and compilation of reports.

Findings are used as input in decision making.

 Information needed is clearly defined.

 The research process is formal and structured.

 It requires a large and representative sample.

 Data analysis is quantitative.

EXPLORATORY RESEARCH

 Research is aimed at providing insights and understanding of phenomena on which

there is little or no knowledge. These explore what is happening and try to ask
about it.

 Exploratory studies are normally conducted during the initial stages of the research
process. Studies are conducted to clarify the problem.

 In most cases exploratory research is conducted by;

 Secondary data collection

 Talking to experts in the field

 Conducting focus group discussions

 Normally exploratory research is conducted with the expectation that subsequent

research will be required to provide conclusive evidence.

 Results are generally followed by further exploratory or conclusive research.

CLASSIFICATION OF RESEARCH

[Link] by Method
10
  Historical research

 Case Study

2 Classification by Purpose

 Basic Research

 Applied Research

 Evaluative

 Research and Development

3 Classification by Design

 Experimental

 Survey

4 Classification by Type of Data

 Qualitative Research – It involves narrative descriptions (data collection – interviews

and observation).

 Quantitative – It is a research that describes using figures.

OVERVIEW OF THE RESEARCH PROCESS

-Problem identification

-Hypothesis Formulation

-Literature Review

-Writing Research Proposal

-research design

-Data collection

-Data Analysis

-Presentation of Results

-Conclusions

-Recommendation

-Report Writing
11
-Communication of Results

1 Problem Identification

 “A problem is the axial center around which the whole research effort turns,”(Leedy,
P.D. 1993:59).

 It is an issue that merits attention or requires a solution. It is the core of the study.

 It is the issue forming the central concern of the study.

CONSIDERATIONS

Tuckman (1978) advances considerations in problem identification as follows;

(a) Workability

 Is the study within limits and range of your resources and time constraints?

 Access – Will you have access to the sample in numbers required?

 Answer – Can you come out with an answer to the problem?

 Manageable and Understandable – Is the required methodology manageable

and understandable?

 Are there enough sources of literature?

(b) Critical mass

 Are there enough variables?

 Variables are characteristics that take different values for different

individuals/people at different times and/or places.

 A variable is the characteristic being measured. Are there enough potential

results?

 There should be enough to write about.

(c) Interest

 Are you interested in the area, in the specific problem and potential
results/solution?

 Does the problem relate to your background?

12
  Does the problem relate to your career interest?

 Does it turn you on/Does it enthuse you?

THEORETICAL VALUE

 Does the problem fill a gap in the literature?

 Will others recognise its importance?

 Will it contribute to advancement in your field?

Does it help to build up knowledge?

 Is it publishable?

PRACTICAL VALUE

 Will the solution improve the system?

 Are practitioners likely to be interested in the results?

 Will the system be changed by the outcome?

SOURCES OF RESEARCH PROBLEMS

 Theses and dissertations – All theses and many dissertations contain suggestions for
further research.

 Articles in academic and professional journals – Journal articles sometimes include

recommendations for further work, and they are reasonably up to date and
therefore should be given appropriate attention by researchers.

 Reports – Particularly of Government sponsored bodies are usually up to date and

these often contain recommendations on which research can be based.

 Books and book reviews – Though books in general are not as up to date as other
documentary sources, they often contain detailed accounts of particular fields. Book
reviews contain evaluations of the extent to which certain books contribute to
knowledge and therefore provide useful guide to researchers seeking for topics.

 Communication with experts in the field – Experts in the field in which one wants to
research on have distinct knowledge of what has been studied and what still needs
to be studied.

13
  Conversations with potential users of the research findings – In these conversations,
consumers of research findings point out areas they need research to be carried out.

 Discussion with colleagues – Much may be gained when a few ideas have been
formulated by discussing them informally with colleagues.

 Abstracts – These contain summaries of various research studies. Having gone

through the summaries, a potential researcher proceeds to look for the full research
report for a study whose summary he/she might have read and got interested.

 The media – It is a potential source of research topics because newspapers, radios,

and televisions report on research progress which is felt to be of general interest.
Additionally major findings are reported by the media earlier than scholarly accounts
appear in academic journals.

IMPOTANCE OF THE RESEARCH PROBLEM

 It lays the foundation for the development of the research proposal.

 It is important for the formulation of the research objectives

 It is also from the research problem that the research hypothesis is

formulated.

 It guides the selection of the literature review.

STATEMENT OF THE PROBLEM

 It is a concise description of the nature of the problem, its sites (where we

get the problem), its distribution and severity, who is affected, since when
and consequences for the affected.

 It is an analysis of the factors that may influence the problem.

 It is a brief description of the solutions that have been tried, how well they
have worked and why further research is required.

SUB-PROBLEMS/ RESEARCH QUESTIONS

 Each sub-problem is a complete researchable unit.

 They must add up to the totality of the main problem. Solving the sub-
problems results in solving the main problem.

 With sub-problems, interpretation of the results would be easier and clearer.

EXAMPLES OF TERMS USED IN THE STATEMENT OF THE PROBLEM

14
 An evaluation, an assessment, an examination, an analysis, an investigation,
an enquiry, a review

 The statement should not be too long. It should be brief, precise and clear to
understand. A problem can be stated either as a question or declarative
statement.

STATEMENT OF OBJECTIVES

 Objectives are the immediate or short term statements of what is intended

to be achieved.

 Aims are the major goals of the study.

PURPOSE OF OBJECTIVES

 They assist the research to focus on the essentials (what is important).

 They help the researcher avoid collection of irrelevant data.

 They help the researcher organise data into clearly defined paths.

 They facilitate the development of the methodology.

 They help orient the collection, analysis, interpretation and utilisation of the
research results.

 Objectives should be clearly related to the statement of the problem.

HOW TO STATE THE RESEARCH OBJECTIVES

 NB. Use action verbs.

 They should be specific enough to be evaluated.

 Avoid non-action words like ‘’understand, appreciate’’, these should be

avoided because they are difficult to measure and they subjective.

 Use the following when stating objectives;

to classify, to compare, to evaluate, to verify, to assess, to investigate, to

analyse etc.

HYPOTHESIS

It is an intelligent or educated guess about a relationship between two or

more variables.

15
 It is a tentative explanation/proposition about a relationship between two or
more variables.

 It provides direction to the organisation of the study.

 It provides a framework for the collection, analysis and interpretation of data.

 It can be tested resulting in the acceptance or rejection of the hypothesis.

 It allows escape from personal bias.

 Hypotheses can be drawn from the researcher’s theoretical knowledge. They

are powerful in advancing knowledge.

 There are two types of hypotheses;

 Null hypothesis – H0

 Alternative hypothesis/Research hypothesis – H1

 The null hypothesis states that there is no association between / among

variables.

 It states absence of a relationship or difference.

 The alternative hypothesis indicates /states the presence of a relationship

between or among variables.

 When a hypothesis is to be tested, it is appropriate to state both the null and

alternative hypotheses. Begin with the null hypothesis then the alternative
hypothesis. eg ,,

 H0 – There is no relationship between age and cancer.

 H1 – There is a relationship between age and cancer.

 There are directional and non-directional alternative hypotheses;

 Non-directional – There is a relationship between smoking and cancer.

 Directional – Smoking increases chances of contracting cancer. It shows the

direction or inclination of the relationship.

 There are errors that can be made in accepting or rejecting the null
hypothesis.

 Rejecting a true null hypothesis is called Type 1 Error.

16
  Accepting a false null hypothesis is called a Type 11 Error.

 Hypothesizing is part of everyday life eg a car breaks down, the breakdown is

followed by a series of guesses. The guesses are actually hypotheses.

CHARACTERISTICS OF HYPOTHESES

1. They are reasonable.

2. They are testable.

3. They are brief and understandable.

4. They specify the variables studied.

5. They are assertive (not suggestive)

6. They conform to existing theory.

THEORY; a logically organised set of propositions (claims, statements, assertions)

which serves to define events (concepts), describe relationships among these events,
and explain their occurrence.

The major functions of theory are;

 To guide research

 To organise empirical knowledge

 To describe and explain phenomena (events)

IMPORTANCE OF HYPOTHESES

1. They are very useful in organising the research effort.

2. They specifically limit the enquiry to the interaction of certain variables.

3. They suggest appropriate methods for collecting, analysing and interpreting the
data.

4. The confirmation and rejection of hypothesis through empirical testing gives a

clear indication of the extent of knowledge gained.

SIGNIFICANCE OF THE STUDY

 The researcher must demonstrate why it is worth the effort, time and
expense required to carry out the study.

17
  The researcher should also state the benefits believed to be derived from the
study ie. the theoretical value, practical value and any other benefits.

ASSUMPTIONS

 These are conditions that are taken for granted and without which the
research effort would be impossible eg the co-operation of the respondents,
the acceptance of the problem etc.

 These are statements that the researcher believes to be facts but cannot
verify because they are not the object of the research.

LIMITATIONS

 These deal with boundaries and they are established by factors or people
other than the researcher himself/herself.

 Thomas and Nelson (1996) define limitations as ‘’possible shortcomings or

influences that cannot be controlled by the researcher,’’ eg the time factor,
financial and material resources.

DELIMITATIONS

 These are limits. The researcher sets guidelines to show how far, how wide
and how deep the research would go.

 Delimitations should be your choice but should be practical and achievable.

 Delimitations help the researcher not to choose a topic which is too wide, too
narrow or too shallow.

 It should show what will be covered and what will not be covered. In brief
Krall (1971) asserted/argues that delimitations help the researcher keep
focused on the research problem.

 They also allow the researcher to complete research in a reasonable period

therefore delimitations and limitations are concerned with research
boundaries.

OPERATIONAL DEFINITIONS

 These define a concept solely in terms of the operations used to produce or

measure it.

18
  The concepts are defined specifically according to how they are used in the
study.

LITERATURE REVIEW (CHAPTER 2)

 Literature review refers to the summary and critical analysis of what

others have written about the problem being studied.
This seeks to provide a background on the theoretical underpinnings
(foundation) of the research.

 It is the critical reading of literature.

 It can also be referred to as a concise and focused critique of a topic or issue

which lies within a well-defined field of study and a sub-area of specialisation.

 As such it must reflect the researcher’s ability to:

 Survey the content area of the study ie. What knowledge is related to the
main research questions.

 Trace the development of such knowledge over time, highlighting the major
landmarks.

 Address issues surrounding the topic. Issues here refer to those views,
theories and perspectives which are conflicting and have not been resolved,
or are parallel and so remain diffuse and unclear.

 Suggest authentic possibilities which could be exploited as a way of filling in

gaps or resolving disagreements.

 Building a case through logical and systematic analysis and synthesis of

diverse viewpoints often through the integration of logical arguments into a
coherent whole.

 The identification of research designs and methodologies which worked or

did not work, and the contexts to and within which they were applied and
conducted.

IMPORTANCE OF LITERATURE REVIEW

 Revealing similar investigations to one’s own, getting to know how others

handled their research.

 It helps to determine one’s methodology.

 It assists the researcher to contextualise his/her own research

19
  It helps to provide areas of gaps in knowledge.

 It assists the researcher to evaluate his/her own research effort.

 It helps the researcher to assess strengths and weaknesses of previous

studies.

 It introduces the researcher to significant research personalities whose work

he might not have known before.

 It enables researchers to define the frontiers of their fields.

 It assists the researchers to limit their questions, clarify and define the
concepts of the study.

 It enables the researchers to place their questions in perspective. It leads to

insight into the reasons for contradicting results in an area.

 It prevents unintentional replication of previous studies.

 It places researchers in a better position to interpret the significance of their

own results.

 It is from a good review that the researcher’s confidence is enhanced.

 After reading various sources of literature around the topic, the researcher
becomes an expert in the area studied.

 It demonstrates the key theories and controversies in the field.

STEPS IN REVIEWING LITERATURE

1 Choosing a research topic

2 Locating relevant information

3 Evaluating content

4 Synthesizing knowledge

5 Summarising insights

1 CHOOSING THE TOPIC

What is the specific problem a research question that your literature review helps to
define?

20
2 LOCATING RELEVANT INFORMATION

Find out about the location of relevant sources in the library.

Have key words related to the problem to guide your research.

Target the most significant literature in your field.

Unless used for a historical perspective, sources should be current.

Focus on sources published by experts in the field.

Review references cited by experts as possible leads.

Gather from both primary and secondary sources.

3 EVALUATING CONTENT

After selecting sources, you need to evaluate the content in greater detail.

The following are the questions to consider;

What is the ‘big picture’ of your subject area?(what are the past and present
research trends).

Does the author show any sign of bias?

What are the strengths and weaknesses of the source?

What are some of the themes common to the collection of literature?

Where does your narrowly- focused topic fit into the ‘big picture’ of your subject
area?

SYNTHESIZING KNOWLEDGE

As you review and analyse your collected literature, you will identify the following
key pieces of information and evaluate their relevance and usefulness to your topic

-hypotheses/research questions

-study location, subjects

-methods and materials

- results

-conclusions, application
21
The next step is to group related elements of the sources into themes eg identify all
the relevant results of each source as a single theme.

Organise them into a logical sequence.

The organisation depends on your specific topic and the theme/subheading or your
literature review.

The review should have a smooth flow.

SUMMARY

Conclude the review with a brief summary of the literature and its implication. The
question to ask yourself is;

‘’What is the relationship between the proposed study and the literature? ‘’

In particular this shows how the proposed study moves beyond what is known. This
might mean taking some main trends in the literature but extending them;
confirming or challenging current findings, or researching on accepted theory but in
a new field.

TYPES OF DATA

 In any research study that is undertaken, there is the collection of data.

 These data are of different types eg qualitative and quantitative data,

collected from various sources, primary and secondary and are of various
forms, overt and non-overt (covert).

SECONDARY DATA

It is data already collected and documented by others.

Sources – government documents, academic and organisational journals, historical

studies, college and school magazines, data bases and periodicals books and book
reviews.

ADVANTAGES

Sometimes resources for data collection are usually limited therefore it is cheaper to
use or buy secondary information.

It takes time to collect and analyse data or there may be no time to train
interviewers therefore secondary data serves the time.

Secondary data provides unobtrusive measures.

22
Data is collected quickly and likely to provide high quality data.

Secondary sources can provide comparable and contextual data. It is useful for
triangulation of findings.

Secondary sources can lead to unforeseen discoveries if they are reanalysed.

DISADVANTAGES

There is need to collect data from several different sources therefore chances of
making errors are high.

There is possibility of transcription or printing errors in published data eg

newspapers and magazines.

Data collected may not be up to date.

Collection and completion of data is a long process therefore it could be a problem if

the investigation item changes rapidly.

There may be biases due to poor collection techniques.

COMPARISON OF PRIMARY AND SECONDARY DATA

PRIMARY DATA SECONDARY DATA

It is first hand Information It is second hand Information

It is used by people for whom the data It is used by persons other than people
was collected for whom the data was collected

Data is relatively accurate because it is The accuracy of data is questionable

directly observed by the researcher. because usually there is no proof of the
data.

Data is usually up to date Data is usually not up to date and needs

some adjustments.

Data precisely relates to the problem Data provides a good starting point in
under investigation. problem solving before any attempt to
direct observation.

It is also used when secondary data fails It is used when it is not possible to
to provide sufficient answers to the collect primary data.
research problem.

23
 OVERT AND NON-OVERT DATA

The distinguishing feature between these two types of data is the extent
to which explicit communication takes place between the researcher
and study participants during data collection.

OVERT DATA

 These are primary data obtained by asking people questions and

obtaining their answers.

 There is either written or verbal communication between the

researcher and participants.

 Almost all overt primary data consist of data derived from or

relating to people.

 The communication is by means of telephone interviews, face to

face/personal interviews and mail.

NON-OVERT PRIMARY DATA

 This refers to all primary data other than that obtained through
asking questions to the participants.

 There are five major approaches to collect non-overt data namely,

audits, trace analysis, content analysis, mechanical observation
and personal observation.

(a) Audits – This is done when a researcher examines physical records or

performs an inventory analysis eg a sales audit.

 In a sales audit ,data from a particular shop, or geographical area

is analysed to investigate the existence of trends or relationships.

(b) Content Analysis

 It is a research technique for the objective, systematic and

quantitative description of the manifest content of
communication,.

24
  It studies content of a communication, potential causes of a
communication message and the potential effects of a
communication message.

(c) Trace Analysis

 Trace analysis is an approach to non-overt data collection in which

the data are based on physical traces of a past behaviour.

 Traces may be left intentionally or unintentionally by the

individuals being studied.

 Because these individuals are not directly studied (trace analysis

data are, in a sense, historical data), they have no knowledge of
the research, and the research has no impact on their behaviour.

 For this reason trace analysis sometimes termed unobtrusive

measures research or non-reactive research.

(d) Mechanical Observation

 Researchers employ mechanical devices to collect non-overt data.

 Such devices include the stand alone variety and the facilitating
variety: their general purpose is to continually record data for
later analysis.

(e) Personal Observation

 This is the most widely used approach to non-overt primary data

collection.

 This is where a researcher personally observes behaviour as it

occurs.

 It is also used in conjunction with overt data collection eg during a

personal interview, a researcher can collect non-verbal behaviour.

 Personal observation can take place in a natural setting as when

an individual makes a shopping trip in a supermarket.

25
  It can take place in a controlled environment eg children brought
together to assemble a particular machine and they are observed
as they do so.

QUANTITATIVE DATA

 This refers to data in numerical form or data that can be

quantified.

 Quantitative data can be both primary and secondary.

 Quantitative data in raw form convey very little meaning unless

processed and analysed.

 Processing raw data is turning them into information.

 Quantitative data can be obtained using structured instruments eg

questionnaires, tests, interviews etc.

 Quantitative data are obtained by measuring quantifiable

variables such as test scores, prices of goods, rental costs etc.

 Quantitative data are analysed using quantitative techniques eg

tables, graphs/charts and statistics.

 When quantitative data are analysed, they bring out relationships

and trends.

 Quantitative data analysis involves calculations and chart drawing,

manually or using computer software packages such as spread
sheets, Excel, and SPSS etc.

 Quantitative data can be obtained when research designs such as

experimental, surveys etc. are used.

 Quantitative data can either be discrete or continuous.

 Quantitative data can be obtained from samples selected using

probability sampling techniques.

 The collection of quantitative data is guided by hypotheses.

26
QUALITATIVE DATA

 Qualitative data refers to data based on meaning expressed

through words, images, objects etc.

 This is data in narrative form.

 Qualitative data are collected by studying non-quantifiable

variables such as depression, honest, beliefs, colour etc.

 Data are collected using semi-structured or unstructured

instruments such as in open interviews.

 Data are collected from sample selected using non probability

sampling techniques.

 Collection results in non-standardised data requiring classification

into categories.

 Analysis of qualitative data is by use of conceptualisation.

 Data collection is not guided by hypotheses but hypotheses evolve

with the development of the study.

 Data can be analysed manually or using computer assisted data

analysis software.

METHODOLOGY (CHAPTER 3)
This is the section of the study where the researcher lays down procedures used to
carry out the study.

It answers to the following questions:

 What approach will be used?

 How will the data be collected?

 How will the data be analysed?

27
1 It is likely this section that receives the most attention from those scrutinizing and
evaluating the research proposal.

2 It is therefore this section that requires greatest attention to detail and accuracy.

3 The methodology section requires a description of:

(a) The research design

(b) The population and sampling methods chosen.

(c) The data collection instruments, techniques and procedures for collecting and
recording data.

(d) The data analysis procedures to be used.

(e) How ethical issues are to be dealt with.

(i) THE RESEARCH DESIGN

1 The research design is the blueprint for the collection, measurement and analysis of data.

2 It includes an outline of what the researcher will do from writing the hypothesis and their
operational implication to the final analysis of data.

3 There are various definitions of the research design but they provide the following
essentials as stated by Cooper and Schindler,( 2003:146):

 The design is an activity and time based plan.

 The design is always based on the research problem.

 The design guides the selection of sources and type of information.

 The design is a framework for specifying the relationships among the study’s
variables.

 The design outlines procedures for every research activity.

The research design provides guidelines to the selection of the research approach/selection
of the type of research eg quantitative, qualitative or mixed methods approach.

The various research designs that can be used include case studies, surveys, experiments,
exploratory, conclusive, evaluative quasi experiments etc.

28
Justifications are made to the choice of the design to be used. On this point it is important
to do the following:

(i) Outline the strengths and weaknesses (merits and demerits) in general.

(ii) Explain how the design suits your particular circumstances and study (as a
design, given its merits and demerits, how can its merits make your study
strong)?

This section is very important and nothing should be left to the reader to guess.

NB. It is on the basis of this description and justification that the validity of your research
findings can be assessed.

Any possible sources of invalidity should be acknowledged.

POPULATION AND SAMPLING

Population is the totality of all subjects that conform to a set of specifications, comprising
the entire group of persons that is of interest to the researcher and to whom the research
results can be generalised.

 Describe the population.

 Specify the characteristics that the population must have in order to be included in
the study.

SAMPLING

 A sample is a proportion or subset of the research population selected to participate in a

study representing the research population.
 Sampling is the process of selecting the subset of the population to participate in the
study.

 Identify the sampling technique that you will use to come up with the sample.

 Give a justification of the type of sampling that you select.

 Provide a thorough discussion of sampling procedures employed.

RESEARCH INSTRUMENTS

These are the tools you would use to obtain the required data.

These include questionnaires, interview guides, observation guides, tests etc.

29
Research instruments need to be clearly described so as to bring out their strengths and
weaknesses as a way of justifying their selection and suitability to the research study.

Measures taken to control the weakness identified should be spelt out so as to ascertain the
validity and reliability of these instruments and the data to be collected.

DATA COLLECTION PROCEDURES

These are steps taken to administer instruments and collect data from the study
participants. Such steps include;

Making appointments with research subjects through phones, letters, fax etc

Distribution and administering of instruments on the sample, for example, by hand mail or
through research assistants

Retrieval of instruments

NB. These steps need to be clear, orderly and justified. They are a deliberate and well
thought out strategy meant to enhance timely collection of comprehensive data.

WHY NEED TO KNOW HOW DATA WAS COLLECTED

 Methods used to collect data affect the results.

 Knowing how data were collected help the reader to evaluate the validity and
reliability of the results and conclusions drawn from the results.

 It makes clear reasons why a particular method or procedure has been chosen. The
reader wants to know whether data were collected or generated in a way that is
consistent with acceptable practice in the field of study.

 It shows whether there is link between methods and objectives.

 It explains problems anticipated and steps taken to prevent them from occurring,
and problems that occurred and the way their impact was minimized.

 Often recording how you overcame obstacles is an interesting part of the

methodology, and means you can give rationale for certain decisions. It also gives a
realistic view of using the methods you have chosen.

 It is useful to those who want to replicate your methodology so sufficient

information should be given to allow others to use the work.

DATA PRESENTATION AND ANALYSIS

30
This section spells out the overall procedures to be used to organise, describe and analyse
the collected data.

The process entails suggesting the manner in which findings are to be

(a) Presented eg using tables, graphs and diagrams.

(b) Described to reveal their meaning

DATA COLLECTION TOOLS/INSTRUMENTS

1. QUESTIONNAIRES

A questionnaire is a data collection method in which a list of questions is directed to

respondents to solicit responses.

ADVANTAGES OF QUESTIONNAIRES

 They are preferable when a given answer is required or there is need for
consultation.

 They permit contact at minimum cost.

 They are a very efficient way of collecting data.

 They are less expensive than interviews because there is no need to train staff
to administer questionnaires.

 They record data which can be easily referred back to.

 People can be asked the same question throughout therefore there is consistence in
questioning.

 Questionnaires reach anyone in different geographical locations.

 People are more likely to say the truth because questionnaires are anonymous and
confidential.

 They reduce errors caused by bias which results from using interviews.

 They provide reliable data as facts can be checked by other researchers using the
same instrument.

 They provide quick and efficient data analysis both manually and using a computer.

DISADVANTAGES

31
  They have a low return rate i.e. not all questionnaires are returned. One should have
70% of return for data to be
valid and reliable.

 Some questionnaires are not answered by the person we expect to answer.

 In some cases people answer what they think the researcher wants to hear.

 They are answered by the literate people only therefore they are biased.

 They are expensive to reproduce.

 There is no face to face interaction.

 They do not provide opportunity for probing to clarify questions.

 Questionnaires are less flexible than interviews.

 Respondents may interpret questions differently.

 Validity of data may be reduced by unwillingness, inability to give full and accurate
responses to questions.

QUESTIONS

The foundation of all questionnaires is the question.

The questionnaire must translate research objectives into specific questions.

Answers to such questions will provide data for hypothesis testing.

NB. The question must motivate the respondent to provide the information being
sought.

VALIDITY IN RESEARCH

This refers to the extent to which research instruments measure what they purport to
measure.

It refers to the appropriateness, meaningfulness, and usefulness of the specific inferences

made from the measurement.

RELIABILITY

It refers to the consistence of a measuring instrument each time it is used to measure the
same thing.

It can also be defined as the degree to which observed scores are free from errors of
measurement. We can gauge
reliability by the consistence of scores.

32
Reliability refers to the stability, consistency, dependability and accuracy of an instrument.

ACCURACY

This refers to the difference between what an instrument says is true and what is known to
be true.

CONTENT OF QUESTIONS

Survey Questions – These are concerned with facts, opinions, attitudes, respondents’
motivation and their level of familiarity with certain subjects.

These can be classified into (a) factual questions and (b) questions about subjective
experiences.

FACTUAL QUESTIONS

 They are designed to elicit objective information from the respondents regarding
their background, environment,
habits [Link] gender, age, marital status, education or income.
 Such classification may in turn aid in explaining differences in behaviour and
attitude.
 Some questions are designed to provide information on the social environment of
the respondents.
 How accurately the people report depends on what and how they are asked.

QUESTIONS ABOUT SUBJECTIVE EXPERIENCES

 These involve the respondents’ beliefs, attitudes, feelings and opinions.
 Such questions are included in surveys conducted in the social sciences.
 Attitudes are general orientations that can incline a person to act or react in a
certain manner when confronted
with certain stimuli eg on the issue of abortion would you describe your personal
position.
1. Favourable 2. Neutral 3 opposed

CHARACTERISTICS OF A GOOD QUESTIONNAIRE

 It should be brief, simple, clear and straight forward.

 It should have a suitable title.
 It should not be double barrelled (avoid asking two questions in one).
 It should not have ambiguous questions (having more than one meaning).
 It should not have leading questions.
 It should not ask sensitive or too personal questions.
 It should be specific and related to the research questions and objectives.
 Questions relating to the same idea should be addressed together

33
  Questions related to the same idea should start from the simple to the complex.
 It should have exhaustive choices to questions with pre-determined answers.
 Each question contributes towards an answer to the problem.

Cover Letter/Transmittal Letter

1. It introduces your questionnaire on your behalf, as well as your role in the study.
2. It motivates the respondent to answer the question.
 It should be brief, about one page.
 It must include salutation by name to provide a more personal touch.
 It should also highlight the purpose of the study and importance of the study, time limit and
importance of the respondents.
 Enclose a stamped envelope.
 Request for immediate return of filled in questionnaires.
 Express confidentiality assurance.
 Offer access to results as incentive
 Show appreciation; thank the respondents in advance for their co-operation.
 Give your name and signature.
 Provide your contact number for further questions.

INTERVIEWS

TYPES OF INTERVIEWS

Face to Face Interviews, Telephone Interviews etc

ADVANTAGES

 Interviews have a high response rate at 80%

 They are appropriate where a large population is illiterate.

 One can build rapport with an interviewee.

 The researcher can probe inadequate responses.

 One can monitor non-verbal cues eg facial message/body language.

 Interview is flexible in terms of conduct of the interview eg venue and time.

 They have immediate response.

 They are a primary source of data.

 They are useful when probing complex information

34
  They are a primary source of data.

 They are useful when probing complex information –it is possible to change the
question.

DISADVANTAGES OF INTERVIEWS

 They are time consuming’

 They are labour intensive.

 They are expensive if travelling is involved.

 They lack anonymity and confidentiality.

 They are high in bias due to presence of the interviewer.

 The responses are varied therefore difficult to classify and analyse.

 The social influence of the interviewer/interviewee relationship may result in

withholding some important information.

 They are affected by the tendency of respondents to give what they think is
expected.

 Respondents may adopt avoidance tactics if questioning is too personal or too deep.

 Respondents may not trust the researcher and therefore respond in suspicion.

OBSERVATION

 It involves studying or watching physically the subject of the research.

 It helps to describe activities.

 Observation helps to refine and reorganise one’s own thinking as one observes.

 It is planned systematically.

 It is subject to checks and controls through observation characteristics.

ADVANTAGES

 It is objective

 It is useful where it is not possible to give verbal responses eg among the deaf.
35
  Behaviour is recorded as it occurs (the researcher gets first-hand information)

 Data is collected in its natural setting.

 Verbal and non-verbal behaviour can be collected.

 Observations can collect more information than anticipated.

 It can yield more accurate data than given by the person himself or herself (people
do not normally do what they say they can do).

DISADVANTAGES OF OBSERVATION

It is time consuming (sitting and observing).

It demands planning and being systematic. There is no room for planning on reaching the
observation point.

The observer should have good memory to make mental notes.

Some situations in human experience are taboo to be observed.

There is possible bias due to differing perceptions, values, beliefs, ethics and norms.

Information concerning the past cannot be observed.

People change behaviour when they know they are being observed.

TYPES OF OBSERVATION

Participant and Non-participant Observation

In participant observation the researcher joins the respondents and pretends to be one of
them.

In non-participant observation the participants are aware that they are being observed.

PARTICIPANT OBSERVATION

Definition: - The process of learning through exposure to or involvement in the day to day or
routine activities of participants in the researcher setting (Schensul and Lecompte,1999;91)

Participant observation requires a certain amount of deception and impression

management.

The process of establishing rapport within a community and learning to act in such a way as
to blend into the community so that the members will act naturally.

36
The researcher should then remove himself or herself from the setting or community to
immerse oneself in the data to understand what is going on and be able to write about it.

It includes observation, natural conversations, and interviews of various sorts, checklists,

questions and unobtrusive methods.

It is characterised by actions such as having an open non-judgemental attitude, being

interested in learning more about others.

The researcher should be aware of the propensity for feeling culture shock and for making
mistakes, the majority of which are overcome by;-

(i) Being a careful observer

(ii) Being a good listener

(iii) Being open to the unexpected in what is learnt.

FOCUS GROUP DICUSSIONS

Definition: -Focus groups are small homogeneous groups representative of the target
population and made up of key informants brought together to discuss pertinent issues.

A focus group discussion is a group interview or discussion consisting of a small group of

individuals who meet together to express their views about a particular topic defined by the
researcher.

Membership ranges from 10 – 15 and in some groups 6 – 10.

People in the focus group are often chosen because of their interest, involvement or
knowledge in relation to the research issue.

The group process during the discussion allows people to develop their ideas with each
other and to brainstorm different options/opinions.

A facilitator/moderator leads the group and guides the discussion.

The discussion usually lasts 1 -2 hours and is tape-recorded.

SKILLS FOR SUCCESSFUL FOCUS GROUP MODERATION

Show genuine interest in all the participants.

Be a moderator not a Participant

Be prepared to hear unpleasant views/views you do not agree with.

37
Make a decision about the appropriate level of moderation.

ADVANTAGES OF FOCUS GROUP DISCUSSIONS

With skilled facilitation, focus groups can come up with ideas and solutions which none of
the individual members had thought about beforehand.

Focus groups can be economical means of gathering data.

They can produce creative data (They generate fresh ideas)

They give immediate response.

Response rate is relatively high.

There is opportunity to probe for clarity or more information.

They can collect both, verbal and non- verbal responses.

They are generally versatile and more natural.

They allow clients to observe.

DISADVANTAGES OF FOCUS GROUP DISCUSSIONS

‘Group think’ can operate so that individual members feel afraid to voice their true opinions
and diversity is lost.

Much depends on the skill of the facilitator.

They can be expensive if some members come from faraway places.

The talkative members dominate the discussion and the opinions of the quieter ones cannot
be recorded.

Time can be wasted on arguments therefore decision making on the side of the facilitator
becomes difficult.

It needs probing skills.

It needs careful planning and being systematic.

Data obtained is difficult to analyse

SAMPLING

Sampling is a process of selecting a number of subjects for a study in such a way that the
subjects represent the larger population from which it was selected
38
A sample is a part or subset of a population taken to be representative of the population as a
whole for the investigative purpose of research.

The aim of sampling is to gain information about a population ie. it is used to make an
estimate of the characteristics of the population as a whole.

A population is the group to which a researcher would like the results of a study to be
generalizable.

A population can also be defined as the total number of individuals/ elements in the group in
which we are interested.

Sampling is normally done instead of using a census because;

1. It is cheap.

2. it is efficient.

3. It is easier to manage.

4. It is subject to statistical verification.

5. It allows for a high degree of precision.

LIMITATIONS OF A SAMPLE

1. Low accuracy; In comparison to a census technique, the conclusions based on sample

results are more liable to error.

2. Changeability of units; It is unethical to extend the conclusions derived from one set of
sample to other sets which are not similar or are prone to change.

3. Misleading conclusions; If the sample is not selected carefully, the conclusions derived
from them will become misleading if extended to all units.

4. Need for specialised knowledge; Sampling can be successful only if a competent

researcher makes the selection.

5. There are some cases in which sampling is not possible; Sampling cannot be used if the
material is of a heterogeneous nature or where there is need for 100% accuracy.

CHARACTERISTICS OF AN IDEAL SAMPLE

1. Representativeness .The sample should represent adequately the whole population. Units
selected should have the same set of qualities and features as found in the whole
population.

2. Independence; This has to do with interchangeability of units, ie. every unit should be
available to be included in the sample.

39
 3. Adequacy; The number of units included in a sample should be sufficient to enable
derivation of conclusions applicable to the whole population. A sample having 10% of the
whole population is adequate.

4. Homogeneity: The units included in the sample must bear resemblance with other units
for the sample to be scientific.

DETERMINING THE SAMPLE SIZE

A sample should be as large as possible.

The larger the size the more representative and the more generalizable the results are.

MINIMUM ACCEPTABLE SIZE;

These depend on the size of the research.

If it is descriptive research 10% of the population should form the sample.

For correlational research about 30% of the subjects should form the sample.

For causal comparative, 30 subjects should form the sample.

In experimental research about 15 subjects per group should form the sample.

NB. Even large samples can lead to wrong conclusions if not well selected.

STEPS INVOLVED IN SAMPLING

1 Define the population of interest.

2 Determine whether to sample or census.

3 Select the sampling frame.

4 Choose a sampling method/technique

5 Determine the sample size

6 Specify the sampling plan/ procedure.

7 Select the sample.

TYPES OF SAMPLING PROCEDURES

Probability sampling

Non-probability sampling

PROBABILITY BASED SAMPLES

40
 These are procedures in which every member of the sampling frame has an equal chance of
being selected. The chance of every unit in the population to be selected is known. ‘

Selection of specific units for the sample depends entirely on chance.

(a) RANDOM SAMPLING

It is a probability based sample selection procedure.

All individuals in the defined population have an equal chance of being selected.

It is the best single way of obtaining a representative sample.

The sample can be selected using;

Computer Selection

Random Number Tables

Lottery Methods

RANDOM NUMBER TABLES

The numbers move vertically or horizontally.

A random sequence of numbers is produced by selecting from the random lists those participants
who fall in the ranges required eg. when selecting 5 people from 50. Give each person a number 1-
50.

Select those who hold the first 5 numbers which occur as you move through the table.

THE LOTTERY METHOD

When selecting a group of 30 from college students. Give every student a number, draw thirty
numbers from a well shuffled barrel.

LIMITATIONS OF RANDOM SAMPLING

1. It is expensive; the cost of pulling together a large sampling frame is prohibitive.

2. Respondents selected must be interviewed to ensure the integrity of the process. This means up
to three call backs to individuals may be made before classifying a non-response.

STRATIFIED SAMPLING

It is a process of selecting a sample in such a way that identified subgroups in a population are
represented in a sample in the same proportion that they exist in the population.

Steps are similar to those in random sampling except that one is selecting from sub-groups in a
population rather than from the whole population.

SYSTEMATIC SAMPLING
41
It is done by taking every nth person in a list eg. selecting 20 students from 200.

Make an alphabetic register then select every 10th name. Random select a starting number eg 4.

CLUSTER SAMPLING

This is sampling in groups not individuals.

Groups are randomly selected.

Any intact group of similar characteristics is a cluster.

Steps in cluster sampling are similar to those in random sampling except that the random selection
of groups is involved not individuals.

MULTISTAGE SAMPLING

It is a modification to cluster sampling.

The population would be found in the first stage or primary units. Each of the primary units may be
made up of second stage units and so forth.

A sample is taken of the first stage units and then of the second stage units eg. for national surveys,
the administrative districts for voting are usually rural and urban districts (PSU) then parliamentary
constituencies (SSU) and lastly polling stations (TSU).

NON- PROBABILITY SAMPLING TECHNIQUES

These are sampling techniques in which there is no way to estimate the probability of each element
to be included in the sample. There is no guarantee that each element has some equal chance of
being included.

Non-probability sampling techniques are used when it is not possible to obtain an absolutely
representative sample.

ADVANTAGES

1. Lower cost

2. Faster

3. Smaller sample sizes.

4. Important respondents can be targeted.

DISADVANTAGES

1. Results are purely indicative

2. Sampling error cannot be computed.

3. The degree of representativeness of the sample to the population is not known.

42
4. Assumptions need to be made about the groupings with the population of interest.

QUOTA SAMPLING

It is a non-probability sampling technique in which a sample is selected based on known proportions

in the populations.

In quota sampling, the population is divided into groups according to selected characteristics eg.
gender, age, economic status or religious groupings etc. One may use census data to find out the
proportion of the adult population who fall under various age or sex classes.

The investigator makes a final selection from the groupings eg. 5% aged, 40% adults and 55% young.

The quota in this sense is a percentage of respondents selected from each given category.

It is one type of purposive sampling in which the investigator selects respondents within the quota
according to their judgement.

The quota sampling technique is very suitable for attitude and opinion surveys.

ADVANTAGES

It has a high speed of data collection.

It is cheap to undertake.

I t allows sampling to take place where a sampling frame may not be available but key characteristics
of the population are known’

Investigators may not need to interview named individuals; they are screened in or out via a small
number of classification questions

The data, when compared to random methods which are perhaps double the cost, has been proved
to be acceptable provided that the research is properly managed.

Cost savings may be used to improve the quality of research through increasing sample sizes or using
a different method in support of the survey.

It is a popular technique thus its popularity shows that it works.

DISADVANTAGES

Whilst known characteristics may be distributed in correct proportions, unknown characteristics that
may be relevant to the survey may not be. Hidden bias may exist that is not discovered.

Researchers may be biased as to the types of respondents they choose to interview or the location
where they choose to carry out the interview. A quota of the young people may be filled at one
youth club but will not be truly representative of the population as a whole.

JUDGEMENTAL OR PURPOSIVE SAMPLING

43
It is a nonprobability sampling technique in which the interviewer handpicks the respondents that
are to be included in the sample on the basis of expert judgement. This means the investigator
exercises his/ her own discretion in selecting sample observation from the population.

It is often used when working with small samples.

The key assumption underlying this type of sampling is that, with sound judgement or expertise and
an appropriate strategy, one can carefully and consciously select the sample.

The investigator in this case should know the types of people who are typical of the population in
question.

The individuals selected are those judged to have special characteristics or those who are likely to
provide the most useful information for the study.

MERITS/ADVANTAGES

It is low cost sampling technique.

It is convenient to use.

I t saves time.

It is almost as good as probability sampling.

It can give adequate representation of all the characters of the population if it is used on a small
population.

DEMERITS/DISADVANTAGES

It allows for bias or personal prejudice.

The results of the sample cannot be extended /generalised to the population because various
samples drawn from the same population using this technique may give different results due to
absence of scientific application of statistical techniques.

The sample estimates vary widely from the population parameters leaving room for large sampling
errors.

In the absence of scientific application of statistical techniques, it becomes difficult to establish the
significance of the statistical error.

CONVENIENCE/ OPPORTUNITY SAMPLING

It is a non-probability sampling technique where sampling units are selected because they are
conveniently available, accessible and easy to measure eg a class of students, a congregation (people
at a church), people visiting a supermarket and can be stopped for questioning.

ADVANTAGES OF CONVENIENCE SAMPLING

It is cheap to conduct.
44
It saves time since it studies those who are available.

It does not require a sampling frame so it is easy to conduct.

DISADVANTAGES

The sample may not be representative of the population so it is not possible to generalise results.

It is not possible to identify the sampling error.

SNOWBALL SAMPLING

This is a non-probability sampling technique in which initial respondents are selected at random and
then subsequent respondents are selected by referrals, ie. key individuals are found from initial
interviews. One person’s interview can produce a snowball effect as each new interviewee leads in
turn to several more interviewees. Those who are selected can either report on the feelings of the
community or especially when trying to understand a particular phenomenon under research.

This is very useful where there is a low incidence of the population eg where unusual behaviour is
under consideration.

SELF SELECTING/ VOLUNTARY SAMPLING

This is a technique where volunteers offer to be participants therefore volunteers are self-selecting
eg those who answer advertisements requiring participants.

Field studies may recruit participants in this manner. This is also suitable for studies which really are
sensitive’

ADVANTAGES

Volunteers are likely to cooperate, be open and provide accurate information.

Volunteers are not likely to withdraw before the end of the study.

DISADVANTAGES

Volunteers may not make up a representative sample.

It is not always easy to come up with volunteers’

It is not possible to come up with the sampling error so it is not safe to generalise the results where
such samples have been used.

5 Common Errors in the Research Process

1. Population Specification
This type of error occurs when the researcher selects an inappropriate population or universe from
which to obtain data.

45
2. Sampling
Sampling error occurs when a probability sampling method is used to select a sample, but the
resulting sample is not representative of the population concern. Unfortunately, some element of
sampling error is unavoidable. This is accounted for in confidence intervals, assuming a probability
sampling method is used.

3. Selection
Selection error is the sampling error for a sample selected by a nonprobability method.

4. Non-responsive
Nonresponse error can exist when an obtained sample differs from the original selected sample.

5. Measurement
Measurement error is generated by the measurement process itself, and represents the difference
between the information generated and the information wanted by the researcher.

VALIDITY

Validity is an important concept in the field of research.

It refers to the appropriateness of, meaningfulness and usefulness of the specific inferences made
from the scores or results. It also refers to how well an instrument measures what it is supposed to
measure.

TYPES OF VALIDITY

1 CONTENT VALIDITY

It refers to the accuracy with which the instrument measures the factors or situations under study
eg. Content under study – how accurately do questions asked elicit the information required.

2 CONSTRUCT VALIDITY

It refers to the degree to which construct is measured.

A construct is any idea or concept eg honest, anxiety, etc.

It is difficult to measure. Different methods of measuring construct should focus on the results.

3 CRITERION VALIDITY

It refers to the standard against which to measure the results of the instrument that is doing the
measurement.

Data from the measuring instrument should co-relate highly with the results of the criterion.

4 FACE VALIDITY
46
It is based on the subjective judgement of the researcher.

It should measure what it is designed to measure, and the sample should adequately represent the
population.

5 INTERNAL VALIDITY

It is where there is freedom from bias in forming conclusions.

It is the extent to which other variables which affect the research are controlled.

A good research must endeavour to minimise other extraneous effects.

The researcher must ensure that he/she has isolated the cause because she/ he might think that the
cause lies with one variable yet there may be other hidden variables which may affect.

To ensure validity, the researcher must ensure that the research is carefully designed to control
possible sources of errors which may come out of choice of subjects, instruments and procedures
used.

6 EXTERNAL VALIDITY

It is concerned with the generalizability of the results.

The question to ask is: Can the conclusions drawn from the sample be generalised to the whole
population or to other cases?

THREATS TO INTERNAL VALIDITY

1 HISTORY

During the time that an experiment is taking place, some events may occur that confuse the
relationship being studied.

2 MATURATION

Changes may occur within the subject due to passage of time but not specific to any particular event.

This usually occurs when the study takes a long time; the subject can become hungry, bored or tired.
This can affect response results.

3 TESTING

The process of taking a test can affect the results of the next test.

Taking a pre-test may influence the results of a post-test.

4 INSTRUMENTATION

Changes in the measuring instrument or the observer threatens validity

47
Even with the same observer used for all measurements, boredom, fatigue and anticipation of
results can distort results of separate observations.

5 SELECTION

Differential selection of subjects for experimental and control groups

Groups should be equivalent in every respect.

6 STATISTICAL REGRESSIONS

The tendency of scores to concentrate around the mean

7 MORTALITY

Loss of subjects due to death, withdrawal etc

With each dropout, the composition of the group changes and this affects the results.

NB. Control for almost all these threats is randomization.

COMPENSATORY THREATS TO VALIDITY

1 DIFFUSION OR IMITATION OF TREATMENT

Exchange of information between members of the control group and those of the treatment group
eliminates the difference in treatment thus affect the results.

2 COMPANSATORY EQUALISATION

Compensatory actions by the control group where the treatment is desirable can influence results.

3 COMPENSATORY RIVALRY

Competitive pressure – members of the control group try harder

4 RESENTFUL DEMORALISATION OF THE DISADVANTAGED

The control group becomes resentful of their deprivation and thus lower their co-operation and
output

TRIANGULATION

It is a technic that facilitates validation of data through cross verification from two or more sources.

In particular it refers to the application and combination of several research methodologies in the
study of the same phenomenon.

It can be employed in both validation of quantitative and qualitative (enquiries) studies

It is a method – appropriate strategy of founding the credibility of qualitative analysis.

48
It becomes an alternative to traditional criteria like reliability and validity.

PURPOSE OF TRIANGULATION

To map out, explain more fully, the richness and complexity of human behaviour by studying it from
more than one standpoint.

To give a more detailed and balanced picture of the situation

To crosscheck information to produce accurate results for certainty in data collection

To crosscheck data from multiple sources to search for regularities in the research data

TYPES OF TRIANGULATION

Data triangulation; involves space, time and research persons.

Investigator Triangulation – involves multiple researchers in an investigation.

Theory Triangulation – involves using more than one theoretical scheme in interpretation of the
phenomenon.

Methodological Triangulation – The strategy of using more than one method to gather data, such as
interviews, observations, questionnaires and documents.

NB By combining multiple observer, theories, methods and empirical materials, researchers can
hope to overcome the weaknesses or intrinsic biases and the problems that come from single
methods, single observer and single theory studies.

.Triangulation can also be achieved by collecting essentially the same data from different samples, at
different times and in different places.

RESEARCH ETHICS/ ETHICAL CONSIDERATIONS IN RESEARCH

DEFINITIONS

1 research ethics are precautions to be taken, to do with respect for human and other rights.

2 Ethics can refer to the appropriateness of your behaviour in relation to the rights of those who
become the subjects of your work or are affected by it.

3 These are concerned with the extent to which the researcher is socially and morally responsible to
his/her participants, sponsors, general public and his/ her beliefs.

(i) ANONYMITY AND CONFIDENTIALITY

The identity of participants should remain anonymous.

All information from respondents should be held in confidence.

49
Any information should be released with the participant’s permission.

Individual scores should not be reported or made public.

Present data in group statistics

Individual scores if presented should be coded and not associated in real.

Observing this ethical consideration enables the researcher to gain trust from the respondents
therefore obtain more accurate data.

(ii) INFORMED CONSENT

It is a mechanism of ensuring that people understand what it means to participate in a particular

study, so they decide in a conscious, deliberate way whether to participate.

Participants should be informed of the purpose of the study.

Participants should be aware of all the procedures to be followed.

They should be informed of what is expected of them including the amount of time likely to be
required for participation.

They should be aware of the likely risks and benefits, including psychological and social.

They should be informed of how confidentiality will be protected.

They should be aware of the fact that participation is voluntary and that they have freedom to
withdraw any time with no negatives repercussions.

Participants should be given contact information of the researcher in case of questions or problems
related to the research study.

1. It is important to observe this consideration because it reveals how much respect the researcher
gives to potential participants thus gain their acceptance.

2. Research participants are likely to have more confidence in the research and so be more open and
frank in their responses.

3. Greater confidence in the research should improve participation rates.

4. Researchers are forced to sharpen and clarify the purpose of their research which helps not only
participants but also researchers themselves.

INVASION OF PRIVACY – The right to privacy is one of the basic tenets of living in a democratic
society.

Every individual has a right to privacy.

Researchers do not have any right to intrude in a respondent’ personal affairs

50
Participants should volunteer to be included in the study not by coercion

Intimate behaviour should not be recorded unless the subjects give their explicit consent.

Participants have the right to withdraw or not to answer certain questions that they feel are too
personal/ sensitive

The nature and any timing of approach to the intended respondents needs to be agreed upon, and
has to be during sociable hours if possible.

It is important to observe these considerations because:

1. Voluntary research participants are likely to be more open and give accurate information.

2. Voluntary research participants are not likely to withdraw before the end of the study or data
collection process.

3. The relationship between the researcher and the respondent is kept warm if the researcher does
not seem to obtain data on private life.

FREEDOM FROM HARM – Respondents should not be harmed in any way, physical, mental and
emotional.

Research is considered harmful if it causes a participant to be embarrassed, ridiculed, belittled or

generally subject to mental distress, or if it produces anxiety or stress to participants.

The researcher should schedule interviews to avoid disruption of the respondent’s important
programmes.

The study should be so designed that it has a high chance of yielding significant results to respect the
dignity of the respondents.

Experiments should be designed in such a way that they do not cause death, disability or any
physical or mental harm to participants.

The participants should be assured of their safety in participation.

1. Failure to observe this consideration may result in the researcher facing legal repercussions in
case any participant is negatively affected by it.

2. A high response rate is likely to be achieved if potential respondents are assured of safety in
participation.

(V) KNOWLEDGE OF OUTCOME OF RESULTS

Participants have the right to access the final research report.

Participants should be assured that results respect their dignity.

They have the right to receive explanation of any change of procedures.

51
Observing this consideration increase response rate and compels the researcher to treat the data
with the strictest confidentiality.

(vi) PERSONAL INTEGRITY

The researcher should not falsify data.

The researcher should have personal integrity.

Data or findings should not be manipulated to agree or disagree with hypothesis. In research this is
unethical, unforgivable and unprofessional.

Observing this consideration enables the researcher to maintain rapport with participants. This
increases future response rate.

DATA PRESENTATION, ANALYSIS AND INTERPRETATION (Chapter 4)

Once you have collected the data you require for your study, the next thing is to process it.

Processing the data helps in communicating something about the problem under study.

The data should ‘speak’ to you. You should be able to derive some meaning from the data.

Data will not communicate anything unless they have been presented in a systematic way.

To present data in a systematic way you need to use presentation techniques:

(a) Tables

(b) Graphs eg. Histograms, pie-charts etc

(c) Any quantitative technique that would render the data analysable

(d) Rich verbal description of data that do not need statistical processing (qualitative). The narrative
should be precise and not flowerly.

DATA ANALYSIS

Analysing data requires a very sharp and alert mind.

When analysing data the researcher should be able to pick up; similarities, differences, agreements,
disagreements, uniqueness, diversions, conversions, trends, tendencies and any patterns.

Unless the data is conscientiously analysed, it will be difficult to interpret.

DATA INTERPRETATION

After analysing the data, you now require an appropriate interpretation of them. For good
interpretation, ask yourself the following questions:

52
(i) What do these patterns / tendencies mean?

(ii) What do these differences/similarities mean?

(iii) What do these relationships mean?

(iv) What meaning do I get from the data analysis as a whole.

The data interpretation process is a reflective process. It should also relate to your research
questions (sub-problems) in Chapter 1 of your research.

It is after the interpretation process that you will be able to formulate your own theories, reject or
accept the hypothesis of the problem under investigation.

CONCLUSIONS AND RECOMMENDATIONS

This is the final part of your research. The stages are;

(i) Write a summary ie. a brief overview of what the research was about.

In the summary, salient features of the study are highlighted eg the problem under investigation, the
methodology, the major findings and their significance, as well as any new theories built from the
findings.

(ii) Conclusions and recommendations

Conclusions are the major findings drawn from the study.

To recommend is to make suggestions for implementation, and for further research work based on
your new insights.

Link your recommendations to the research work you have done (especially your research findings).

Point out questions still to be done.

MEASUREMENT SCALES
Measurement means assignment of numbers to variables.

In any research we measure the variables involved. However the way we measure varies eg. when
we arrange people based on salary, we measure in a different way from when we give percentage
scores to students.

Each method of measurement has its merits and demerits. It is based on how the measurement is
done.

The scales of measurement can be grouped into 4 as follows;

1 Nominal scale

2 Ordinal scale
53
3 Interval scale

4 Ratio scale

1NOMINAL SCALE
This is the categorisation of individual characteristics. It involves frequency counts of individuals.

This is the least refined/ precise scale of measurement. It describes differences between things by
assigning them to categories eg race, nationality, gender (male/ female) religion, occupation etc.

2 ORDINAL SCALE
When the measurement indicates order then it falls under the ordinal scale.

When arrangement of qualities with the smallest to the highest or vice versa is done, the scale of
measurement is ordinal eg. rank ordering of people using age, height, weight etc.

It shows that things differ in amount or degree. Arranging of students in class from best to least
based on overall performance is ordinal.

The ordinal scale is more refined (accurate) than the nominal scale.

With the ordinal scale, the real differences between adjacent ranks may not be equal. Ranking
spaces them equally though they may not be equal.

3 INTERVAL SCALE
This is more refined than the other two. It has order and equal intervals eg. a score of 90 is 10 units
above the score of 80.

The interval scale is very useful when measuring variables such as temperature, percentage scores of
students etc.

On this scale there is no true (absolute) zero. Zero degrees Celcius on a thermometer does not mean
absence of temperature.

4 RATIO SCALE
Of the four scales of measurement, this is the most refined scale. There is order, there are equal
intervals and the scale has an absolute or true zero eg. if a person weighs 60 kg. and another weighs
90 kg, it is correct to say that the second person is 2 11/2 times as heavy as the first one.

The variables which are measured on the ratio scale include height, length and weight. These have a
true zero.

It is considered as the most superior of all the scales in terms of accuracy.

STATISTICS

54
It is a body of mathematical techniques/ processes for the collection, analysis and interpretation of
numerical data for use in decision making.

The term statistics is commonly used to mean figures.

FUNCTIONS OF STATISTICS
(a) To enhance knowledge about the environment we live in by providing numerical information or
facts about various natural processes which could be physical, biological, socio economic etc and
laws governing them.

(b) To give precision to ideas, accurate and precise measurements and or projection of levels of
important variables such as population figures, agricultural outputs, industrial production, state of
economy, and quality of goods and services.

BRANCHES OF STATISTICS
There are two branches of statistics

(a) Descriptive Statistics

(b) Inferential Statistics

(a) DESCRIPTIVE STATISTICS – (Deductive Statistics)

This branch utilises numerical and graphical methods to look for patterns as a way of summarising
and presenting the information in a set of data. It involves collecting and manipulating data.

It includes summary calculation of summary measures such as the mean and variance, drawing
graphs such as the bar graphs, histograms, frequency distributions and presentation tables.

Descriptive statistics summarises large amounts of data to manageable form that can be presented
and studied.

(b)INFERENTIAL STATISTICS – (Inductive Statistics)

The branch utilises sample data to make estimates, predictions or other generalisations about a
particular case or study or portion of data to the entire target population.

It is concerned with making inferences about a population.

IMPORTANCE OF STATISTICS
It enables one to interpret information contained in numbers in order to;

(a) Make intelligent decisions

(b) Make inferences

55
(c) Make generalisations

TYPES OF DATA
Both descriptive and inferential statistics are concerned with measurement of one or more variables
of a sample unit drawn from a population.

These measurements are called data.

The nature of data helps the user to assess the quality and select the right statistical method for
analysing the data.

Data quality is influenced by the following;

(a) Data type

(b) Source of the data

(c) Methods of data collection

CLASSIFICATION OF DATA
Qualitative and Quantitative Data

(a) QUALITATIVE DATA

This data cannot be measured numerically but can fall into one or more non-numerical categories

A variable which cannot assume a numerical value is called a qualitative or categorical variable.

Examples of qualitative data are;

(a) Data regarding blood groups of individuals eg A, B or C

(b) Sex or gender of individuals ie. male or female.

QUANTITATIVE DATA
The data can be measured numerically eg. number of students passing through the college gate.

Quantitative data has two classes;

1 DISCRETE DATA
Quantitative data is referred to as discrete if it can only take specific fixed values eg. numbers of cars
arriving at a garage at a given time, insurance policies taken in a given month, number of garments
produced in a week etc.

2 CONTINUOUS DATA

56
Data is continuous if it has no specific fixed values eg. age of a person, height of a tree, time taken to
complete a project etc.

TYPES OF DESCRIPTIVE STATISTICS

(a) Measures of central tendency

(b) Measures of variability

(c) Measures of relationships

(d) Measures of relative position

(a) MEASURES OF CENTRAL TENDENCY/LOCATION

They are used to determine the typical/average score of a group of scores. It includes the following;

(a) Arithmetic mean

(b) Median (middle - despite value)

(c) Mode (most frequent number)

a) THE MEAN

The mean is the ratio between the sum of the observations and the number of observations.

The mean or arithmetic average is expressed in a formula as follows; =

Where

The mean has a precise calculation.

It is the most sensitive measure

It represents the exact centre point of all data points.

The mean can easily be distorted by a few rogue score in one direction eg:
57
1; 3; 4; 5; 7; 64

-In the above set of data the mean is 14. The mean is therefore greater than 5 out of 6 values in the
set. The median will be more representative and in this case the median is 4.5.

-The mean can also be funny where the original figures can only be whole numbers eg an average of
2.4 children in a family.

(b)The Median

The median is the middle score or central value in a given data set.

It is much easier to identify the median when the number of scores is an odd number eg. 2, 7, 9, 8,
10, 6, 4

The first step is to arrange the numbers in order beginning with the smallest.

Pick up the score that leaves the same number of scores below and above it

In the above data set the median is 7.

When the number of scores is an even number one cannot simply identify the median without
calculating.

To calculate the median one should add the 2 middle scores and divide the total by two as in the
following example,

120, 118, 115, 114,114, 112

Median = 114+115 = 114,5

2
The median is an ordinal measure

The median is a lot more representative of the central values than the mode when dealing with
small data sets. It is not distorted by extreme values in one direction as the case with the mean.

It can be affected by skewed data as in the following example ; 1, 3, 5, 7, 23, 35, 46.

The median in this case 7 but it is misleading in really describing the data.

The Mode
The mode is the score that occurs most frequently as in the following example.

6, 5, 4, 7, 6, 8, 9, 10, 6, 11, 15, 3, 2, 6

In the above example, the mode is 6.

Example Two: 6, 5, 4, 7, 6, 8, 9, 10, 6, 11, 15, 3, 2, 6, 4, 5, 4, 4

58
This is an example of a bimodal – with two modes 4 and 6.

The mode is unreliable as a measure in a small data set eg 2, 3, 3, 3, 3, 5, 6, 8, 8, 8, 9, 9,

There is need for only 2 more 8 for the modal value to change to 8, again there is need for only one
8 for the set to have two modal values.

NB The mode is useful when dealing with large data sets as in a survey especially with large data sets
especially when data are gathered in categories.

Graphical Data
Graphical display is a way of summarising data

Graphs are referred to as figures in a research report - when they are used, they are numbered
consecutively eg Figure 4.3. This means Figure number 3 of chapter 4.

Examples of graphical displays are bar graphs, histograms, frequency polygon, stem and leaf charts,
pie charts etc.

Before calculating any measures it is useful to present data in classic form.

Data is recorded in summary sheets in columns in ascending order.

Graphing data enables the researcher to see the pattern of the distribution of scores. The common
method of graphing data is to construct a frequency polygon.

Steps in Constructing A Frequency Polygon.

1. List all scores

2. Tabulate how many received each score.

3. Place all scores on horizontal axis at equal intervals from lowest to highest.

4. Place frequencies of scores at equal intervals on vertical axis starting from zero.

5. Where the score intersects with the frequency of occurrence

SCORE Frequency (f) make a dot.

85 12 6. Connect all the dots with straight lines and the result will be the
frequency polygon.
86 10

87 7
Frequency Distribution

88 6 This is the distribution that shows the number of people who got a
particular score
89 3

90 4

91 2 59
Frequency Polygon: Construct the frequency polygon

Stem and Leaf Display

It is an economic way of graphing data. It presents only the first digits in the stem column.
The trailing digits are in the leaf column eg scores ranging between 20 and 70 below.

62, 40, 37, 68, 36, 65, 42, 28, 31, 22, 61, 28

Stem Leaf

2 2, 2,8

3 1, 6, 7

4 0, 2

6 1, 2, 5

Measures of Variability
They have to do with how spread scores are from the mean in a given set of data.

Measures of variability are useful with nominal and ordinal data eg.

Set A: 59, 59, 59 60 61, 61, 61, Set B: 30, 40, 50, 60, 70, 80, 90,

These are two sets of data, they are different but they have the same means and medians.

In Set A, scores are close to the mean.

There isn’t much variability (spread of scores) in Set A.

In Set B, scores are spread out therefore one can conclude that there is more variability in
data Set B than in data Set A.

There are two measures of variability which are the range and the standard deviation.

The Range
The range is the simpler of the two measures of variability.

60
The range is the difference between the highest and lowest scores in a data set. It is a
measure of the width of the total distribution.

The range is given as a single value.

Finding the Range:

R = H – L for the ordinary range.

In data Set A above, the range is small (61 – 59 = 2) therefore the scores are close together.

In Set B, the range is great (90 – 30 = 60) therefore the scores are spread out.

The range gives a quick rough estimate of variability. It is used extensively in quality control.

The Standard Deviation

It is the standard measure of variability in a statistical operation.

It is the most frequently used indicator (index) of variability.

It is the most stable measure of variability.

It takes into account every score in the data set.

Computing the Standard Deviation

Score

60 66 -6 36

50 66 -16 256

40 66 -26 676

80 66 +14 196

100 66 +34 1156

= 2320

Variance = = 2320 = 580

N-1 4

61
S=

N-1

S=
4

S =

S = 24.8

1. Calculate the mean.

2. Find out how far a score is from the mean by subtracting the mean from each score.

3. After squaring the differences, add all the squared differences. With the total (2320), you should
now divide by the number of scores (5) to find out the variance.

4. After finding the variance you should now find the square root of the variance. The answer you
get is the standard deviation.

A small standard deviation means the scores are close together

A large standard deviation means the scores are spread out.

Regression
Relationships serve as a basis for prediction

When observed values of × are taken to estimate corresponding y values, the process is called
simple prediction.

When more than one x variable is used, the outcome is a function of multiple predictors.

The technique that is used to make simple and multiple predictions is called Regression Analysis.

Regression is the dependence of a variable on one or more variables.

Regression analysis is used when the values are based on interval data.

The variable whose values are being predicted is called the dependent variable. It is measured on
the y axis of the graph.

The variable whose values are used to predict the dependent variable is called the independent
variable. Its outcome is independent of the other variable.
62
It is measured on the x axis of the graph (scatter diagram).

The line drawn through the points of the scatter diagram is called a regression line/line of best fit.

On the line the sum of the deviations of all the co-ordinate points has the smallest possible value.

The slope of the regression line is not determined by guess or estimation but by a geometric process.

Predicting unknown y values from known x values can be done using a regression equation.

The formula for prediction is: y = a + bx

Where y = the predicted score

X = the predictor score

b = slope

a = constant, y intercept

Measures of Relationships/Correlation

.Correlation is the degree of association between two variables and it is represented in terms of a
coefficient known as the correlation coefficient.

.The relationship can be either positive or negative

.The relationship can either be weak or strong.

.The important quality of correction is not its sign but its absolute value.

.A correlation coefficient of -0.72 is stronger than that of +0.46 even though with the former the
relationship is strong.

63
64
65
-

66
67