Module 1: Understanding the Nature of

Educational Research
Meaning and Characteristics of Educational Research
The term „Research‟ consists of two words: Re + Search.

 „Re‟ means again and again and „Search‟ means to find out something.

 Therefore, to research means to observe the phenomena (matukio) again and again from
different dimensions.

 It is a scientific/systematic investigation/inquiry or search of knowledge. It is a critical

examination of phenomena.

 Research is the process of arriving at dependable solution to a problem through the

planned and systematic collection, analysis and interpretation of data.

Why scientific

- It is deliberate and planned (inafanywa kwa uangalifu na imepangiliwa)

- It is organized, defined and principled

- It has specific processes and procedures (uses scientific methods)

- It has some norms (e.g. ethics), and practices

- It is controlled/regulated

- It can be empirically verified (imefanyiwa majaribio yanayoweza kusibitika)

Research being that scientific requires;

• A research paradigm (model); is an overview of the whole project that uses “the bigger
picture” as the basis for deciding how to approach the research.

1
 • A research design; is a carefully constructed plan of action that is rationally designed
and likely (enye uwezekano) to offer the best prospects/possibility of success.

• Research design refers to a plan which shows the strategy of an inquiry thought
appropriate to a research

• A research problem; is a specific goal that can be achieved and which is clearly
identified.

• Methods of data collection – instruments of measurements

Roles of Research

• It addresses some questions that have no answers

• It answers certain questions which have not been answered so far and the answers depend
upon human efforts (when the answers are not available)

• Research is the most important process for advancing knowledge for promoting progress
and to enable man to relate more effectively to his environment.

• Although it is not the only way, it is one of the more effective ways of solving scientific
problems.

• To discover the relationship between seemingly unrelated events, and ultimately improve
the way we live.

• Research is frequently used for describing a thing or event, discovering the relationship
between phenomena, or making predictions about future events.

• In short, research can be used for the purposes of description, explanation, and prediction,
all of which make important and valuable contributions to the expansion of what we
know and how we live our lives.

Aims and Significance of Educational Research

• To solve educational problems (our knowledge is incomplete/has gaps)

2
 • The role of research is to provide methods for obtaining answers in education

• To generate new knowledge in education (e.g. teaching methods)

• To test educational hypotheses/fact

• To formulate educational theories

• For planning in education - prediction

Aspects of Education that one may Research

1. Curriculum issues

• Development, Implementation, Evaluation

• Relevance, review, change

2. Policy issues

• Formulation, Pilot testing, Implementation

• Evaluation,

• Review, Amendments

3. Resources (human and material) issues

• Allocation, supervision, sufficiency, recruitment, orientation

• Administration, leadership and management, financing, services, environment

4. Teaching and learning

• Methods of teaching and learning

• Assessment and evaluation of teaching and learning

• Classroom management

• Classroom interaction

3
 • Classroom attendance

• Relevance of instructional practices

• Use of text books

5. Parents and community issues

• Parental involvement in schools

• Parents‟ education level in relation to students‟ support

Sources of Acquiring Knowledge

1. Common/everyday methods or ways of knowing/knowledge

• Authority – knowledge from persons believed to have knowledge. Examples are elders
(wisdom), experts, leaders, etc. E.g. If our doctor tells us that we need to stop smoking or
we shall die we almost certainly do so without question as the person is knowledgeable in
that area.

• Mystical Method – the truth of it relies in the supernatural source (divine, ancestral
spirits)

• Tenacity “ujasiri” (customs and Tradition). …this is how we do things. It is claimed

that some things we believe simply because we have always believed them. We know
these things to be true because we know it. We take them for granted.

• Personal experience (what people encounter or what is in their mind is taken to be true)

Limitations of the common ways/methods of knowing/knowledge

- Lack of control of external influence when explaining the causes of events

- No questioning or testing the information (unverifiable)

- They are individualistic and subjective

- They make knowledge static

4
 - They do not comply with scientific methods

- They are susceptible to deceit and fallacious

2. Reason

 Ogula (2009) defines reasoning as the ability to expound/explain one‟s thoughts logically
and to make conclusions.
 Reason involves thinking about something and developing an understanding of it (Jonson
& Christensen (2016).
 Is the action of thinking about something in a logical and sensible way

There are Three (03) Types of Reasoning.

• Deductive reasoning

• Inductive reasoning

• Deductive – Inductive reasoning (combination of the two)

2.1 Deductive Reasoning

 This type of reasoning involves arriving at specific conclusion based on a priori or self
evident proposition. It is a formal logic that is based on syllogism.

Examples of deductive reasoning or syllogism

• All human beings are rational. Fundikila is a human being. Therefore Fundikila is
rational.

• Every mammal has lungs. All rabbits are mammals. Therefore, every rabbit has lungs.

• All professors are highly learned. Mwendosaa is a professor. Therefore Mwendosaa is

highly learned.

• All dictators have a tragic end. Adolf Hitler was a dictator. Therefore, Adolf Hitler had a
tragic end.

• NB: The middle term MUST be extended (applied to all members)

5
2.2 Inductive Reasoning

 Inductive reasoning involves formulating generalizations based on observation of a

number of individual cases. Reasoning from particular to a hypothesis and eventually to a
generalization

Examples:

• Every university in Tanzania is headed by a professor. Therefore all universities in

Tanzania are headed by professors.

• Every rabbit that has ever been observed has lungs. Therefore, every rabbit has lungs.

• Every professional teacher is trained. Therefore all professional teachers are trained.

2.3 Deductive-Inductive Reasoning

 This involves in a back-and-forth process of induction (from observation to hypothesis or

from the specific to the general) and deduction (from hypothesis to implications).
 Hypotheses are tested rigorously and, if necessary, revised. It means the investigator first
operates inductively from observations to hypothesis and then deductively from these
hypotheses to their implications to establish the validity of knowledge.

3. Scientific Method

Meaning of science

• The root of the word science is the Latin scientia, which simply means “knowledge.” In
simple terms, science can be defined as a methodological and systematic approach to the
acquisition of new knowledge.

• For example, research is used in such diverse scientific fields as psychology, education,
biology, medicine, physics, and botany, and other areas in which research makes valuable
contributions to what we know and how we think about things.

Characteristics of Scientific Method of Research

• Empirical Approach – knowledge obtained through experience or senses

6
 • Careful Observations - done logically and systematically

• Questions - translating research idea into answerable question that guides it

• Hypotheses - an educated and testable guess as an answer to the question.

• Experiments – manipulations and tests to determine behaviour/effects

• Measurements – specifications and quantifications of sizes/amounts

• Analyses and evaluation of data

• Valid conclusions – objective inferences from findings

• Replication – similar findings can be obtained if the same design is employed

• Therefore, scientific knowledge is based on objective data that were reliably obtained in
the context of a carefully designed research study.

• In short, scientific knowledge is based on the accumulation of empirical evidence

• The biggest benefit of the scientific method is that it provides a set of clear and agreed-
upon guidelines for gathering, evaluating, and reporting information in the context of a
research study.

• Among such benefits is the development and formulating theories

Scientific Research and Theories

• Theories refer to systematically organized set of facts and their logical relationships such
that they are able to explain and predict new facts and events.

• A theory implies that something has been proven and held to be true.

• Theories are based upon proven hypotheses and verified multiple times by researchers or
experts

Sample development of a simple theory

i. Observation: Many university students are highly knowledgeable

7
 ii. Research Question: Are all university students highly knowledgeable?
iii. Hypothesis: All university students are highly knowledgeable.
iv. Test: Random sample university students and test them.
v. Findings: All university students are highly knowledgeable
vi. Verification: Other researchers report similar findings
vii. Theory: All university students are highly knowledgeable
viii. Prediction: Any university students I meet must be highly knowledgeable

Types of Educational Research

Types of Research

- Basic/fundamental research; this done over a long period of time and aims to generate
theories
- Applied/immediate research; this meant to solve practical problems that occur in the
society and take a short period of time.

Types of Educational Research basing on purpose

- Basic research
- Applied research
- Evaluation research
- Orientation research

There are three broad categories of research that stem from the three paradigms. Hence, three
types of research approaches (Basing on Method)

• Qualitative

• Quantitative

• Mixed methods

Qualitative Research
- Is the type of educational research done to solve educational problems where conclusion
is drawn through explanations.

8
 - In the course of conducting researches, researchers align problems that can be solved or
the results that can be explained within the context of the best fit paradigm. Hence each
problem fits in a specific paradigm.

Features of Qualitative Researches

 It is conducted in a natural but particular setting because they are concerned with
content

 Naturalistic enquiry is responsive; start where the person is constantly finding out
what the concerns of the people involve are.

 Use broad questions designed to explore, interpret, or understand the social context.

 It provides in depth investigation of individual groups or institutions as they naturally

occur.

 They feel that people and their actions are best understood when they are observed in
social or cultural context.

 A relatively small sample is studied

 Use human instrument (observer) to collect, filter and organize data.

 Use purposive sampling designs are used to select the sample

 It uses emergent design. In emerging design, you constantly change it

 They are more interested in observing process rather than outcomes.

 The focus is on perceptions and meanings

 The researcher plays an interactive role

 Triangulation- use a variety of instruments and methods until nothing new turns up

 Inquiry is value bound

 Aims at generating theories and hypotheses

9
  Data are collected in words or pictures – not quantifiable in statistics

 Data are analyzed inductively, attention is given to particulars

 Findings are reported in narrative form

 Findings are not generalized but transferred to similar context

Types of Qualitative Researches

Biography

 This is the study of an individual and his or her experiences as told to the researcher or
found in documents or archival materials. Biographical research in educational studies
may be conceived in many different ways; however typically the topic connotes a study
of a single life focusing primarily upon an individual who in some way is affiliated with
the professional field of education.

Case study

 Is an exploration of a single unit over time through a detailed, in-depth data collection
involving multiple sources/methods of information? It endeavours to discover meaning,
to investigate processes, and to gain insight or understanding of an individual, group, or
situation. It can be used to address exploratory, descriptive, and explanatory research
questions.

Ethnography

 The word ethnography is derived from the Greek words ethos (“tribe”) and graphos
(“something that is written”). It literally means “writing about people.” The interest is in
describing the culture of a group of people and learning. E.g. shared attitudes, values,
norms, practices, and patterns of interaction, perspectives, language, produce & use, e.g.
clothing, ethnic foods or architectural styles.

Grounded theory

10
  Focuses on developing and generating an explanatory theory in education. It uses the
inductive approach and collects data using multiple techniques over a long period of time.
However, in grounded-theory research, the data collected are continually reviewed to
build a theory that is “grounded” in the data.

Historical research

 Which is about people, places, and events in the past? Is also called narrative research
because it studies “the text of history” and it often presents its results through stories or
narratives.

 Historical research is under qualitative research because, generally the data tend to be
qualitative and use of evidence and the forming of arguments is closer to that of
qualitative research than to quantitative research. Historical research analyses documents,
and artefacts and/or uses interviews with eyewitnesses to gain insight into past events

 Educational phenomena operated in the past like educational events that, trends in
education, teaching practices and the outcomes, the origin of current practices and
changes over time. Historiography is the word historians sometimes use to mean
“research methods.”

Phenomenology

 Which focuses on individuals‟ experience of some phenomenon? The researcher attempts

to understand how one or more individuals experience a particular phenomenon.

Basic Qualitative Study

Quantitative Research
- Is the type of educational research done to solve educational problems where conclusion
is drawn through presenting numerical data.
- Base on empirical data and use hypotheses

Features of Quantitative Studies

11
 • Yields data that are quantifiable. Normally wants to know the frequency of occurrence of
an event and percentage of people involved

• The social world exists externally and is viewed objectively

• Requires large samples

• Data is analyzed deductively

• Concerned with theory testing as the theories and hypotheses are stated before the
commencement of the study.

• Results can be generalized to the population from which the sample is drawn

Researches under Quantitative Paradigm

1. Experimental designs

2. Non experimental designs

1: Experimental Design

 This consist of
i. Experimental: True experiment – Pre test – Post test control group random control
design, Post Test only Random control group design, Solomon four group design
ii. Quasi Experimental design

I: Experimental Designs

• These are used to study cause and effect relationship amongst two or more variables.

• The three essential elements of an experimental design are:

i. Randomization - The researcher assigns the participants to different groups randomly

ii. Manipulation - the researcher does something at least to some of the participants in the
research i.e. treatment

iii. Control - the researcher introduces one or more controls over the experimental situation

12
True experiment is further divided into three:

• Pretest-posttest control group design

• Post test only control group design

• Solomon four group design

1. Pretest – Post-test control group design

RE – X – Post-test

CR ------ Post-test

R= Randomization E= Experimental group, C=Control Group

• X= Treatment

Treatment

• In this design, the researcher randomly assigns participants to both the experimental and
control groups.

• Once the participants are assigned to their appropriate groups, they are then pretested.

• After that participants in experimental group are subjected to treatment for example a
new training program.

• At the end of the experiment the two groups are post tested

• This helps to assess the effectiveness of the treatment

2. Post test –Only control group design

13
 Treatment

3. Solomon four group design

• RE 01 X 03

• RC 02 - 04

• RC - X 05

• RC - - 06

• Four groups are randomly selected from a population. Two groups (control) receive
treatment and two (Control groups) do not. One group receiving treatment and one group
not receiving treatment are pretested. After the experiment all groups are post tested.
This design provides good control over threats to internal validity. The pretest post test
provides control over extraneous factors such as history and maturation.

This design is used to achieve three objectives/purposes

• To assess the effect of the treatment, the effect of a pre-test and the interaction between
pre-test and treatment conditions.

• To determine if the pre-test had any effect, compare group 1 & 3 or group 2 & 4.

• If the pre-test had an effect on experiment group, this should result in a pre-test treatment
interaction. To determine this difference in the results of the post-test b/n groups 1 & 3
should be more than that of groups 2 & 4. This is because the pre-test has effect on the
experimental group and not on the control group.

14
II: Quasi Experimental Design

• These have been defined as experiments that have treatments, outcome measures, and
experimental units but do not use random assignment of subjects to various groups in the
experiment.

• However, the independent variable(s) can be manipulated.

• The purpose of a quasi – experimental design is to approximate a true experiment.

• It is not usually possible in social science research to apply true experimental designs
because of the difficulty of obtaining equivalent groups or achieving random assignment
of the two groups.

2: Non Experimental Designs/Survey design

The survey design is used when one wants to study large populations by studying samples drawn
from it. Usually describes characteristics of participants for the purpose of building theories or
generalizations about the population. It involves random selection of participants. It is divided
into:

 Longitudinal
 Cross Sectional,
 Correlation

A: Longitudinal Survey Design

• This involves collection of data from the sample at several points in time. It is
appropriate when the researcher wants to observe changes over a long period of time on
the subjects. Data about a research subject is collected at several points with the aim of
understanding changes.

B: Crossectional Survey/Design

• Used when the researcher needs to get information at one point in time to describe the
current characteristics of a sample.

15
 • The researcher identifies the essential components of the population and randomly
samples the research subjects from each of the components at the same time.

• For instance, in a simple cross-sectional study researcher might compare the reading
skills of the sample of children from standard one, three, five and seven in one year e.g.
2021.

• The aim is to determine if the level makes a difference in reading skills.

C: Correlation Survey Design

• This design is used when the researcher wants to describe in quantitative terms the degree
to which two or more variables are related.

• Correlation studies involve the collection of data on two or more variables on the same
group of subjects then computing the correlation coefficient

Sample Questions of Correlation Studies

• Is there a relationship between the distance that children walk and their weight?

• Is there a relationship between the number of teacher absences and student test scores?

Causal Comparative design or Ex-Post Factor Design

• Ex-post facto means from what is done afterwards.

• In the context of social and educational research the phrase means “after the fact”.

• It refers to studies which investigate possible cause and effect relationships by observing
existing conditions or state of affairs and searching back in time for possible causal
factors.

• Ex-Post Facto Design therefore is the method of finding out possible causes of events
after they have occurred and cannot therefore be manipulated by the researcher.

• Both types involve quantifying the relationship between two or more variables.

Difference between Causal comparative and Ex-post factor

16
 • However, causal-comparative research is more often used to examine variables that
involve dichotomies (such as having a history of abuse or not) or categories (such as
single-sex versus coed school).

• Correlation research is more likely to explore relationships among variables that are
continuous.

• Causal-comparative research typically involves comparing two groups on one dependent

variable. Correlation research often measures many different variables. Although good
causal-comparative studies certainly attempt to control many different extraneous
variables, the major focus of the study is on whether the two groups differ with respect to
the dependent variable. This difference is usually reported in terms of differences
between group means rather than by using correlation coefficients.

Major Differences between Qualitative Research and Quantitative Research

Qualitative Research Quantitative Research

It requires small sample It requires large sample
Data analyzed inductively Data analyzed deductively
Finding are not generalized but transferred to Finding can be generalized to the population
similar context from which the sample is drawn
Use subjective approach Use objective approach
It is exploratory research type It is conclusive research type
Use purposive sampling Use randomly sampling
Use words, pictures and objects element of Use numerical data as element of analysis
analysis
Use non-structured techniques or method like Use structured techniques or method such as
in-depth interviews and group discussion. surveys, questionnaires and observation.

Mixed Methods Studies

17
  Mixed method or research design involves data collection (both quantitative and
qualitative), analysis and interpretation of studies that, singly or together, address a
particular phenomenon.

Things researcher need to consider while conducting mixed methods studies.

- The methodology dimension; which methodology/approaches will be used (Eg; survey,

experiment, case study, ethnography, action research, historical study) which will impact
on the research design, sampling, instrumentation, data analysis, ethics?
- The time dimension; when and where will that quantitative and qualitative elements be
present in the study, in what sequence and/or concurrence or simultaneity? Should be
quantitative and qualitative data be analyzed together or separately.
- The priority dimension; which and what have priority (if any), where and when
quantitative and qualitative (e.g. paradigms, methodologies, data types, data analysis)?
- The relationship dimension; will the research types and data types be independent,
interactive, complementary, and additional to each other? What is the relationship
between different types of data at different points in the research, both within
phase/within stage and cross phase/cross stage?
- Integration; where and when, at which stages, and why do the integration of quantitative
and qualitative methods and data occur?
- Independence, the obverse of integration; where when and why will methods and data be
kept concurrent, separate, interactive or independent.
- Differentiation; will mixed methods and data be used to address the same issue of
different issues?
- Matching; which kind of data are required for which stages of the research?
- Issue in question; around which issue do the mixed methods occur, for example at the
levels of the context of educational research, constructs, variables, research questions,
purpose of the research.
- Transformative intention; does the research have an explicitly political agenda.
- Scope; does the mixing of methods occur within a single study or across more than one
study in a set of coordinated studies within a single programs of the research
- Strands; how many different stands are mixed in the study (Green, 2008:14)?

18
There are three core designs in mixed methods research

i) Convergent mixed methods

Convergent mixed methods is a form of research design in which the researcher converges or
merges quantitative and qualitative data in order to provide a comprehensive analysis of the
research problems. In this design the investigator typically collects both forms of data at roughly
the same time (concurrently) and then integrates the information in the interpretation of the
overall results. Contradictions or incongruent findings are explained or further probed in this
design.

In this single-phase approach, a researcher collects both quantitative and qualitative data
analyses them separately, and then compares the results to see if the finding confirms or
disconfirm each other. Ideally, the key idea with this design is to collect both forms of data using
the same or parallel variables, constructs or concepts. For example, if the concept of interests in
Basic French is being measured during quantitative data collection, the same concept is asked
during the qualitative data collection process such as in an open-ended interview. Some
researcher will use this design to associate certain themes with statistical data using different
forms of data for the quantitative and qualitative data collection.

ii) Explanatory sequential mixed methods

Is one in which the researcher first conducts quantitative research analyses the results and then
builds on the result to explain them in more detail with qualitative research. It is considered
explanatory because the initial quantitative data results are explained further with the qualitative
data. It is considered sequential because the initial quantitative phase is followed by the
qualitative phase.

This mixed methods design appeals to individuals with a strong quantitative background or from
field relatively new to qualitative approaches. It involves a two-phase data collection project in
which the researcher collects quantitative data in the first phase, analyses the results, and then
uses the results to plan (or build on to) the second, qualitative phase. The quantitative results
typically inform the types of participants to be purposefully selected for the qualitative phase and
types of questions that will be asked of the participants. The overall intent of this design is to

19
have the qualitative data help explain in more detail the initial quantitative results, thus it is
important to tie together or to connect the quantitative results to the qualitative data collection.

iii) Exploratory sequential mixed methods.

Exploratory sequential mixed method is the reverse sequence from the explanatory sequential
design? In this explanatory sequential approach the researcher first being with a qualitative
research phase and explores the views of participants. The data are then analyzed, and then
information used to build into a second, quantitative phase. The qualitative phase may be used to
build an instrument that best fits the sample under study, to identify appropriate instrument to use
in the follow-up quantitative phase, to develop an intervention, for an experiment, to design an
app or website, or to specify variables needed to go into a follow-up quantitative study.

This is the reverse of explanatory sequential approach as it starts with a qualitative phase
followed by a quantitative phase. A three-phase exploratory sequential mixed methods is a
design in which a researcher first being by exploring with qualitative data and analysis, then
builds a feature to be tested and tests this feature in a quantitative third phase. Like the
explanatory sequential approach, the second feature builds on the results of the initial database.
The intent of this design is to explore with a sample first to that a later quantitative phase can be
tailored to meet the needs of the individuals being studied.

Interpretation; the researchers interpret the mixed method results in a discussion of a study. The
order of interpretation is to first report the qualitative findings, the development or design of the
feature to be tested (the development of new quantitative measures), and then the quantitative
test in the final phase of the study. It does not make sense to compare the two databases, because
they are typically drawn from different samples (as noted above in the data collection discussion)
and the intent of the strategy is to determine if the qualitative themes in the first phase can be
generalized to a larger sample.

Particularly challenges to this design reside in focusing in on the appropriate qualitative findings
to use and the sample selection for both phases of research. These basic or core designs then can
be used in more complex mixed method strategies. The core design can augment an experiment
by, for example, collecting qualitative data after the experiment to help explain the quantitative

20
outcome results. The core designs can be used within a case study framework to deductively
document cases or to generate cases for further analysis.

Educational Research Paradigms

 A paradigm is a worldview. It means “a basic set of beliefs that guide action”.

 It is a way of looking at phenomena, a view of what counts as accepted or correct way of

working or an „accepted model or pattern‟.

 It is a shared belief system or set of principles, consensus on what problems are to be

investigated and how to investigate them, typical solutions to problems, and an
understanding that is more acceptable.

 Therefore a paradigm is a dominant way of conceptualizing a phenomenon, of

approaching it methodologically and of looking for solutions to research problems.

 A system of understanding viewpoints that shapes the conduct of research.

 A guide that commits the researcher to particular research approaches & designs.
Paradigms provide philosophical, theoretical and methodological foundations for
conducting research.

 A research paradigm is an approach to thinking about and doing research.

 A paradigm may dominate for decades even centuries with its adherents defending it
bitterly in the face of conflicting information, until ultimately a revolution occurs and a
new paradigm emerges.

Major Paradigms in Research

1. Positivism

Worldview contending that there are universal laws and truths that can be objectively studied,
tested, and verified.

21
 • Believes in experimental designs in which quantitative variables are used to measure
relationships, explanations and predictions. Argue that research should follow the
methods of natural science

• Under this knowledge is objective and tangible (subjective knowledge here is not
accepted as realy)

• What can work is that is based on particular laws

• Positivists believe in the objective knowledge of external reality which is rational and
independent. They deny the validity of metaphysical speculation. Uses research
instruments that yield statistical data.

• The aim of the positivist researcher is to seek generalizations and hard quantitative data.

2: Interpretivism/Constructivism

 Believes in non experimental (qualitative designs) in which peoples conception can

change at any time.

 Under this knowledge is subjective

 Knowledge is constrictive than base on an individual interpretation

 Interpretivism or constructivism asserts that knowledge is subjectively constructed.
 Hence it cannot be reduced to causal relationships or universal truths because actions are
based on social and cultural beliefs, values, and experiences.
 It employs naturalistic inquiry - states that social research ought to be studied in natural
settings.
 Interpretive is phenomenological; qualitative; naturalistic; constructivist; interactionist;
phenomenological etc.
 It seeks to understand how the world is socially constructed and experienced. Naturalistic
researchers believe in multiple, situated truths and realities.

3. Pragmatism/Mixed

22
  Focuses on “what works” to answer the research question. Pragmatism shifts energies
away from philosophical underpinnings and focuses on actions. This is a mixed approach.
 Pragmatism is problem-centred; action oriented; utility oriented.
 Truth knowledge is what serve the purpose
 Believes in both qualitative and qyantitative
 It favours mixed methods in research to balance between the two – positivism and
constructivism (interpretivism).

Module 2: Basic Elements of Educational

Research
Variables in Research
A variable is a condition, property, factor or quality that can vary (e.g.in quantity, intensity or
levels) from one case to another.

It is a construct or a characteristic that can take on different values (levels or scores). It is the
opposite of a constant, which does not vary between cases.

• Height is one example of a variable; it can vary in an individual from one time to another,
among individuals at the same time, among the averages for groups etc.

• Social class, gender, education level, vocabulary level, intelligence, and spelling test
scores are other examples of variables.

• Normally, researchers study variables and the relationships that exist among variables.

Independent and dependent variables

An Independent Variable
 Is a characteristic which is not affected by another variable, it is independent of other
variables. It is a variable that the researcher can manipulate or control. It is a predictor
variable.

23
  It is often considered to be a stimulus that influences a response or an antecedent which
may be manipulated or modified (e.g. under experimental or other conditions that might
control the amount or frequency of something) to affect an outcome.

A Dependent Variable
 Is a condition whose (numerical) value depends to some degree on that of one or more
independent variables? This is an Outcome variable (also called criterion or response
variable).

Mediator/Intervening Variables
• A mediator or intervening variable is one which explains the relationship between an
independent and dependent variable, or between two other variables.

• Intervening or mediating variables stand between the independent and dependent

variables, and they transmit the effect of an independent variable on a dependent variable.
They moderate it.

Moderating Variables
 Are predictor variables that act on or intersect with the independent variables, and then
together in combination with the independent variables influence the dependent variables.

• They make a causal chain by linking a cause and effect relationship.

• It is recognized as being caused by the independent variable and as being a determinant

of the dependent variable.

• A mediator variable (B) receives the effect of one independent variable (A) and this
affects the outcome variable (C).

• Example: Higher education (independent variable) typically leads to higher income

(dependent variable).

• Occupation is an intervening variable between education and income because it is

causally affected by education and itself affects income.

24
 • School leadership might moderate the relationship between teacher commitment and
school effectiveness.

Extraneous Variables
• These are the variables that affect the research study either because the researcher is not
aware of them or if he or she is aware, he/she cannot control them.

• It is any uncontrolled variable which can influence the outcome of the study or the
effect of the independent variable on the dependent variable.

• Extraneous factors are also called confounding variables, or confounds for short.

• These are variables that influence the outcome of the experiment, though they are not the
variables of interest. They are undesirable because they add error to the experiment.

• It is obscures the effects of another variable.

• If one mathematics teacher used a different book in his class and another mathematics
teacher used another book in her class to teach the same topic and at the end of the topic
the students were given an achievement test to see how they mastered the topic.

• Teacher effectiveness and the text book would be confounded.

• There is no way to determine if the differences in the performance between the two
classes were caused by either the text book or teacher effectiveness.

• Age may change student‟s performance in exams irrespective of the teaching method.

• Hours of part-time study, might affect the interaction between stress level and
performance in the history test.

• That is, the interaction of stress level and hours of study might exert an effect on the
dependent variable (performance in the history test).

• In the test of comparing the effects of lecture with cooperative approaches to

performance, students‟ personality type (introvert and extrovert) is a confounding
variable.

25
 • Cooperative learning works better for extroverted students and lecture methods works
better for introverted students.

Identifying Research Problems

Meaning of Research Problem

A research problem is a practical issue or concern that is intellectually puzzling or challenging.

Creswell & Creswell (2018) describes it as an issue or concern that needs to be addressed (e.g,
the issue of racial discrimination). The problem comes from a void in the literature, and conflict
in research results, topics that have been neglected in the literature; a need to lift up the voice of
marginalized participants; and “real-life” problems found in the work place, the home, the
community, and so forth, P64).

Background to the problem refers to the issues that are driving forces behind the puzzle of the
research problem. They are sub problems that constitute the main problem (Research problem,
Kombo & Tromp, 2006), refer to the background to the problem as the setting or position of the
study (P. 24). As such the background to the problem is not just a summary of literature but a
setting that derives a real research problem.

It is the setting that indicates clearly the way the researcher problematizes puzzles to lead to a
real research problem. It entails identifying gaps in knowledge of the existing body of literature,
as if it were puncturing them by identification of unanswered questions.

Nduguru (2007) contends that knowledge gap arises from inadequately described statements or
explained by existing theories, contrary findings, views, disagreement between experience and
theoretical bases among others. That is why it is done by widely introducing the topic (Mugenda
& Mugenda, 2003) so as to integrate sub puzzles from global level to a particular geographical
area for deducing a research problem. One may start with general observation and then identify
something that needs explanation and define some variables that one wishes to measure (Field,
2013). This entails the justification of the selected area of study.

Condition of a Research Problem

Not every problem is researchable.

26
To problem that require research must have the following conditions:

- There is the practical concern or issue that has no answer for the time being.
- There should be a discrepancy between what is known and what should be known (the
current state differ from idea) – knowledge gap.
- The reason (s) for the difference should be unclear i.e. questionable.
- There should be more than one possible answer.

It must pass the three tests;

- Contributory
- Original and
- Researchable

To identify it: Look at existing literature, social concerns, controversies, gaps, personal
experience.

Statement of the research problem

- A statement of a research problem is a concise description of the issue or concern that a

researcher wishes to dress.
- It provides the context and the purpose of the study
- Normally ask about the relationship between two or more variables
- Topic, evidence of the issues in literature, deficiencies in the evidence, remedy of the
problem, deficiencies in remedies, what do we need to know as researchers, policy
makers, decision makers etc. knowledge gap (categorical statement).

Stages/Processes of Educational Research and Format

i. Identification of research problem

ii. Literature review
iii. Research purpose
iv. Data collection (method of data collection i.e. questionnaire, interview, observation,
focus group discussion)
v. Data analysis

27
 vi. Report writing (is the final stage, where a researcher write or present data obtained
from the field, which involve presentation of data from chapter 1 to 5)

Research Objectives, Questions and Hypotheses

Research Problem or Objectives

- There are sub problems break the research problems into sub problems questions or
objectives.

Research Question

Each type of research by purpose has specific research questions. The research questions in
comparative and experimental studies are quantitative questions that can be used for assessment
studies i.e. determining relationship or differences in variables. Descriptive studies can use
descriptive or theoretical question like:

 What are the most important factor determining the students‟ choice of BEDSN at
AMUCTA?
 To what extent students‟ at AMUCTA participate spiritual activities?
 What is the status quo of academic performance in Basic English and Basic French at
AMUCTA?

Comparative studies can have research questions as: what is the difference in weekly library use
between BAED and BEDSN students? What are the differences in perceptions towards
undergraduate research defense between lecture in education foundation and those in other
departments? What are the differences in attitudes towards sports and games between AMUCTA
students and those from other universities?

Experimental and quasi experimental designs employ relationship research question focuses on
causal relationship, associations, trends and interactions among two or more variables or one or
more groups. Examples;

 What is the relationship between gender and attitude towards spiritual services at
AMUCTA?

28
  What is the relationship between study time and learners‟ academic performance in
Tanzania private schools?
 What is the relationship between performance in linguistics and literature? (Ary; Jcobs;
and Sorensen, 2010; Gay, Mills, and Airasian, 2012).

Research question are used to address the research problem in a more specific manner.

The first factory you need to consider is whether your study is a quantitative, qualitative or
mixed method (designs)

Quantitative Research Question

There are number of factors to consider

a) The types of quantitative research questions you‟re trying to answer.

b) The variables you want to measure manipulate and/or control.
c) How you should structure your research questions opposed to research hypothesis

What type of quantitative research questions are you trying to answer?

i. Descriptive research questions

Describe the variables you‟re measuring (mainly quantification of the variables in percentages,
proportions, frequencies). Description of individuals or groups etc. example

What are the most important factors determining the choice of BEDSN at AMUCTA?

To what extent students at AMUCTA participate in extracurricular activities?

What is the status quo of performance in secondary schools in Tabora Municipality?

ii. Comparative research questions

Aims to examine the differences between two or more groups, on one or more dependent
variables. These questions normally start by asking, “What is the difference in…performance
between students”. Example;

 What is the difference in weekly TV watching between students and lectures?

29
  What are the differences in perceptions towards BEDSN program between women and
men?
 What are the differences in attitudes towards sports between boys and girls?
iii. Relationship research questions

Focus on causal relationship, associations, trends and interactions among two or more variables
or one or more groups. These research questions are popular in experimental and quasi
experimental design.

These research question may generate hypothesis testing based on correlations (r, and rho), chi-
squared for association of categorical variables, regression analysis. Example;

 What is the relationship between gender and attitude towards inclusive classes?
 What is the relationship between study time and performance?
 What is the relationship between performance in geography and biology?

The variables you intended to measure

 Categorical data: nominal and ordinal levels of data

 Continuous data (numerical data-interval/ratio data
 Independent (cause) and dependent (effect) variables

How should structure your research questions

i. Choose the type of quantitative research question your trying to create (descriptive,
comparative and/or relationship)
ii. Set out the first words that start the research question like
- “What are”
- “How do
- “Does”
- “How often”

Determine the correct order for the variables your investigating

- (i.e., independent, dependent and/or control variables).

30
 - As a general rule, independent variables are set out first, followed by dependent
variables, and then control variables (if there are any).

Qualitative Research Question

Qualitative researchers pose research questions

- Not objective
- Not hypothesis

Two types of qualitative research questions to focus a study‟s purpose;

i. Central question – broad question that asks for exploration of the central phenomenon
- Sub questions
ii. Questions that narrow the focus of the study

Writing qualitative research question

- Ask 1-2 central questions and no more than 5-7 sub questions
- In these questions should: Relate the central question to the strategy of inquiry.
- Begin with “What” or “How”
- Focus on a single phenomenon or concept
- Use exploratory verbs like discover or describe
- Avoid directional words such as “Affect” or “Impact”
- Evolve during the study
- Be open-ended without references to the literature
- Specify the participants and research site (unless stated previously)

A script of writing a Qualitative Central Question

(How or what) is the (“story for” narrative research; “meaning of” the phenomenon for
phenomenology; “culture-sharing pattern” for ethnography; “issue” is in the “case” for case
study) of (central phenomenon) for (participants) at (research site).

Mixed Design Research Questions

31
In mixed research designs, questions are set to link qualitative and quantitative data in one
question. Alternatively the researcher can divide the questions into two types of data, starting
with either of the two. For example, if the study has five questions, it can start with two
quantitative questions and then three qualitative questions etc. examples;

 How to students at AMUCTA rate and explain the effectiveness of research committee at
their college?
 Do participants‟ qualitative answers sufficiently explain quantitative responses on the
relevance of research methods for bachelor students?
 Do teachers‟ explanations match with the issue of massive failures of students at Gongoni
primary school?
 To what extent does teacher‟s evaluation report of performance reflect the students‟
grades at Gongoni primary school?
 Is there any congruence between qualitative and quantitative responses?

Conceptual Framework (For Quantitative Studies)

This is the researcher‟s conceptualization of the study based on relationships of concepts or

variables that is presented in diagram. It is demonstration of the relationships of the measured
variables.

Punch (2009) articulates “A conceptual framework is a representation, either graphically or in

narrative form, of the main concepts or variables, and their presumed relationship with each
other. It is usually best shown as a diagram” (P.83).

Nachmias and Nachmias (1996) describe it as a way, in which descriptive categories are
systematically placed in a broad structure of explicit propositions and stating the relationship
between two or more empirical properties (P.38)

Dooley (2001) refers to the framework as diagramming theories, which state separated relations
among concepts (P.59) for instance, if a researcher seeks to assess the implementation of the
national strategy for inclusive education at AMUCTA, one can measure the following variables:
the extent of compliance with the strategy guidelines; staff and students‟ knowledge of the

32
guidelines, relevance of implementation strategies in departments etc. The following can be
conceptual framework.

Independent Variables Intervening Variables Dependent Variables

Extent of
compliance with Clarity of
the guidelines national Effectiveness of strategy
inclusive implementation
strategy

Availability of
Relevance of
resources
inclusive
strategies
The diagramed variables must be related with the research questions or hypotheses of the study.
They must be covered in either or both of the two (questions or hypotheses).

Hypotheses

- Is the statement that confirm availability of something

Developing a Research Proposal (Purpose and Components)

Research Proposal Format

KC + V + P

Key Concept – main theme or issue under study

Verb – main verb in the study

P – Population or subjects or individuals of the study i.e. affected population.

Title

 Refer to the heading or label of the study.

 Describes what the research is all about.

33
  Means it is the summary of the whole research because covers all that are inside the
study.
 Is a quick summary of a key idea of the study? (Means when you look the title you know
or get what are inside.

Characteristics of a good Research Title

i. Should be brief but clear and including the main idea and geographical location.
Example assessment on implementation of learning centred a method in secondary
school at Tabora Municipal.
ii. It is informative & catchy (it is attractive to the mind & eyes of the leader).
iii. It must be in line with the research problem, research questions & objectives.
iv. Should be indicating the issue that is researchable.
v. Normally it reflects the relationship between the independent & dependent variables.

Chapter one

1. Background of the problem (establishment of a study)

2. Statement of the problem (Research problem)
3. Research objectives and question
4. Research question/Hypothesis
5. Significance of the study
6. Scope and delimitation
7. Conceptual/theoretical framework or both
8. Operational definition of key terms

Chapter two

2.1 Literature Review

2.2 Theories of Model
2.3 Empirical Studies
2.4 Gap Identification

Chapter three

34
 3.1 Approach/approaches – paradigms

3.2 Design

3.3 Target population

3.4 Sample size and sampling procedures

3.5 Description of data collection instruments

33.6 Validation of research instruments

3.7 Description of data collection procedures

3.8 Data analysis technique

3.9 Research ethical consideration

Module 3: Literature Review

Is a critique of other of theories, students or other existing documents it involves locating,
reading and evaluating previous studies or documents.

Characteristics of a good Literature Review

- A logical flow of idea

- Correct and relevant references with consistent reference style
- Proper use of terminology
- Unbiased and comprehensive
- It is detailed or exhaustive in the area

Importance of Literature Review

- Provide the researcher with handy guide to the particular topic

- Provide a solid foundation of the study – consolidating the problem
- Providing an overview and a stepping stone to a problem
- It ensure credibility and legitimacy to the researcher and a study

35
 - Confirms the existence of knowledge gap in the area, by raising issues, showing where
there are gaps in the research field, and providing a partial justification or need for the
research.
- It indicates the researcher‟s own critical judgment on prior research or theoretical matters
in the field and indeed provide new theoretical, conceptual, methodological and
substantive insights and issues for the research.
- Helps in designing of the study i.e. strategies, procedures etc.
- Helps in demonstrating familiarity and increases confidence in the researcher‟s
professional ability.
- Helps in demonstration of understanding of the problem. It sets the context for the
research and establishes key issues to be addressed.
- Helps in narrowing down the topic for clarity.
- Helps to avoid replication/duplication of works or studies.
- It enables the researcher‟s to rise questions that still need to be answered in the field, how
to move the whole field forward, and how to look differently at the field.
- Helps to clarify the key concepts, issues, terms and the meaning of these for the research.
- Forms the framework within which the findings are interpreted.

Source of Literature Review (Primary and Secondary)

Primary source of literature review

- Are the sources that contain data from the field. Can be obtained through questionnaire,
interviews etc.

Secondary source of literature review

- Are the sources that contain data from the primary source. Example, books, dictionary,
dissertation, research report and thesis

Tertially source of literature review

- This combine primary and secondary literature review to form huge books like
encyclopedia

36
 Review Process

This is the review of other people‟s idea about a topic; here you pass through books, internet and
articles

Identification of Research Gap

Gape of knowledge left by other researchers where the researcher needs to undergo or conduct a
research in order to come out with the knowledge which will fill that gap.

Module 4: Research Methodology

Research designs, Methods and Approaches in
Educational Research
Research Design

- Refers to a plan which shows the strategy of an inquiry thought appropriate to a research,
these include Qualitative, Quantitative and Mixed design.

Issue to consider when choosing the research design

i. What data one need to address the research problem?

ii. What/who will be the source of the data (target population)?
iii. Are the needed information accessible?
iv. What are the possible obstacles a researcher is likely to encounter?
v. How will the data be collected (instruments)?
vi. When will the data be collected? For how long?
vii. How will the data be analyzed?
viii. Will one need research assistance?
ix. Will the study results be controversial?
x. Will the design match with the ethical requirements/standards?

Population and Sampling (Target, Size, Techniques )

37
Population

- Is a group of individuals, objects or items from which samples are taken for measurement

A sample

- Refers to a group of respondents or elements drawn from a population with individuals

who have one thing in common that provides representative characteristics of the entire
population from which a researcher interested to gather data and draw conclusions

Sampling Technique/procedures

- Is the scientific procedure of selecting sampling units which provide the required
estimates of results with associated margins of uncertainty arising from examining only a
part and not the whole. Under this we have;
i. Purposive Sampling Technique
- Is a technique used in selecting categories with special attributes
ii. Random Sampling Technique
- Is a technique employed to select categories with common attributes
iii. Cluster Sampling (makundi yana sifa zinalingana ila hayapo karibu, can be Nzega to
Kahama)
iv. Snow Ball Sampling
v. Convenient Sampling

Sample Size

Is the number of participants needed to be included

Data Collection Procedures and Tools/Instruments

Data collection is a process whereby a researcher collects data from various sources of
information in accordance with the research plan. Data can be obtained through;

- Personal interview
- Telephone calls
- Documentary review

38
 - Questionnaires
- Observation
- E-mails and internet search. Depending on the survey design (i.e. longitudinal,
crossectional and correlation)

Validity and Reliability of Research Instruments

Validation is the way of making sure that the research instruments bring intended outcomes that
can apply to other situations. The reason for testing the instruments is to ensure that the
instruments are appropriate to tape intended data

Module 5: Data Collection

Research instruments or tools

Characteristics of close ended question

i. They are in restricted form; they present the range or responses which participants
may choose e.g. yes or no question
ii. Provide greater uniformity of responses
iii. They involve alternative answers
iv. Choices are normally ranked in order of importance
v. Categories are mutually exclusive, they do not allow more than one answer
vi. They keep the respondents on subject
vii. They are relatively objective

Strength (advantages) of questionnaires

i. Can reach the large group of participants within a short time with little cost
ii. The bias that must result from personal characteristics of the researcher is avoided or
reduced
iii. Provide opportunity to participants to reflect and answer the way they like
iv. They encourage objective answers due to the absence of the researcher

Weaknesses (disadvantages) of questionnaires

39
 i. Researcher have no control over the participants
ii. No opportunity for participants to get clarifications of the questions
iii. Contaminate information for participants who are not independent in answering
iv. No assurance that all participants understand the questions
v. Illiterate people or those persons with impairment in writing skills cannot participate

Strength (advantage) of an interview schedule

i. Collect in-depth data that cannot be collected through questionnaires

ii. Allow for collecting all but relevant data via clarifications and probing questions
iii. They are flexible as they allow adapting to situation
iv. Even critical information can be obtained from the participants. E.g. feeling, gestures
v. They assure high response return rate
vi. Permit establishing rapport between interviewer and the interviewee
vii. Allow to collect information from the illiterate or from participant who are not ready
to commit themselves in writing
viii. Permit interviewee to clarify his/her thinking and the responses

Weakness (disadvantage) of interview schedule

i. It is expensive in term of time and personnel

ii. It risk bias as the interviewee is likely to influence the participants responses
iii. Can collect data from small sample or fewer sample
iv. The data are difficult to categories and to analyze
v. Some respondent may not be able to provide honest answers in the presence of
interviewer, especially for sensitive issues
vi. There is a challenge for uniformity in questioning when there are more than one
interviewer. Consider using the research assistance due to distance or other factors

Different Methods of Collecting Data in Qualitative

Research
- Interview guides
- Observational forms

40
 - Open ended questionnaire items
- Focus group discussion

Different Methods of Collecting Data in a Quantitative

Research
- Test/experiment guides
- Check lists
- Structured questionnaires (closed ended items on questionnaires)

Research Ethics
Ethical implication is the codified principles that are intended to ensure that the researchers
consider all potential risks and ethical conflicts when designing and conducting a research. The
researcher informed the informants in advance on where and when to meet. Also the respondents
were free to withdrawn from the study area.

Research Misconduct

 Is defined as a fabrication, falsification, or plagiarism in proposing or forming or

reviewing research or in reporting research results.

Ethical Issues in Educational Research

i. Authorization by different authorities (letter of introduction)

ii. Informed consent
iii. Confidentiality
iv. Safety of participants
v. Honest (uaminifu)
vi. Integrity (uadilifu)
vii. Cultural sensitivity
viii. Anonymity (don‟t expose names of those give you information)

41
 Module 6: Data Analysis
We conduct education research to get the information for filling in the knowledge gaps in
education. Not every material is information for knowledge.

The aim of data analysis is ensuring that we get and or construct new knowledge to fill in the
respective gap.

We may collect varied data from participants that may not fit in our purpose or objective of the
study of either providing facts of refusing existing claims to be facts. Data analysis seeks to align
relevance information with the research problem or questions.

The responses and observations collected are raw data (unanalyzed data), not information. Data
are raw information. The raw data become information when analyzed and become relevant to
the research problem under investigation.

Information becomes facts when there is data support (it). Facts are revelations by data

Facts leads to knowledge and understanding = new information, views and new experience

The aim of data analysis is to arrive at this stage of knowledge and understanding of
phenomenon under investigation and in so doing the knowledge gap is filled in, that is, the
research problem is solved or decisively addressed.

Data analysis includes: Data organization = elimination of unusable data; interpreting

ambiguous answer; verifying and or ignoring wrong answers; excluding participants who failed
to answer the questions or the questions that was significantly not answered.

Need for Data Analysis

 To summarize the data collection from participants or observations for ease of use
 To determine the extent to which the data answer the research questions or research
problem
 To interpret the data collection and to derive some meaningful conclusions about the
population

42
  To present the data in a way which communicates the results of the study successfully i.e.
to communicate or share knowledge to other readers or stakeholders in understandable
manner.
 To gauge the data against other facts; theories or normal practice – to see whether we are
in line with i.e. do support or we can modify them etc. this means checking whether we
contribute to Theory and Normal Practices or Not.

Qualitative Data Analysis

Qualitative data analysis typically involves categorization, synthesizing, searching for significant
patterns or themes, and discovering what is important, and then interpreted by the researcher.

Analysis involves reducing and organizing the data. The researcher must organize what he or she
has seen, heard, and read and try to make sense of it in order to create explanations, develop
theories, or pose new questions. Data analysis in qualitative research is often done concurrently
or simultaneously with data collection through an iterative, recursive, and dynamic process. Data
collection, analysis and report writing do not occur in distinct steps as is typical in quantitative
studies.

All qualitative analysis involves attempt to comprehend the phenomenon under study, synthesize
information and explain relationships, theories about how and why the relationships appear as
they do, and reconnect the new knowledge with what is already known.

Data analysis in qualitative research is a time-consuming and difficult process because typically
the researcher faces many amount of field notes, interview transcripts, audio recording, video
data, reflections, or information from documents, all of which must be examined and interpreted.

Transcription; is a transformation of audio/voice record data into text data. For example putting
down the TAP recorded data in interview and then identification of themes and categorization of
the themes. It is applied if the involved recording.

In qualitative research the researcher gets a huge amount of data from the field in terms of words,
picture and voices. Unless care is taken the researcher could find himself or herself overwhelmed
by the data. So it suggested that the data analysis process follows the following steps:

43
 - Data Organization – field data from the field interviews, observation and recordings are
cleaned up. This means relating the data into categories that correspond to research
questions or themes
- Creating Categories – the researcher must be able to detect various categories in the
data which should be distinct from each other. Relationships between categories should
be established. These categories should respond to research questions or themes.
- Develop codes for materials that fall under each category or themes. These can be
done manually or through SPSS package.
- Analyzing and Interpreting – the identification of themes, categories and patterns is
followed by evaluation and analysis of data to determine the adequacy of information.
- Reviewing – appraising or critiquing. Indicate what is reported and evaluate the findings
and or methodology where possible.
- Then a Report is Written by using APA style and in Past Form

Data Analysis in Quantitative Research

 There are two main categories of quantitative data analysis, namely; descriptive
methods or statistics and inferential methods or statistics.

Descriptive Method/Statistics

Descriptive statistics do exactly what they say: they describe and present data, for example, in
terms of summary frequencies. No attempt is made to infer or predict population parameters, and
they are concerned simply with enumeration and organization.

 They include those with central tendency: the Mode (the score obtained by the greatest
number of people); the Mean (the average score); the Median (the score obtained by the
middle person in a ranked group of people, i.e. it has an equal number of scores above it
and below it); minimum and maximum scores.
 Those in dispersion (the action/process of distributing things or people over a wide
area): the Range (the distance between the highest and the lowest scores); the Variance
(a measure of how far scores are from the mean, calculated as the average of the squared
deviations of individual scores from the mean); the Standard Deviation (a measure of

44
 the dispersal or range of scores, calculated as the square root of the variance, yielding the
average of all the individual deviations of scores from the mean).
 Other are: the Standard Error (the standard deviation of sample means); the Skewness
(how far the data are systematical in relation to a “normal” curve of distribution); the
Kurtosis (how steep or flat is the shape of a graph or distribution of data; a measure of
how peaked a distribution is and how steep is the slope or spread of data around the
peak).

Such statistics make no inferences or predictions; they simply report what has been found, in a
variety of ways.

Inferential Method/Statistics

Analysis of variance (ANOVA), analysis of covariance (ANCOVA), Chi – square test, factor
analysis, multivariate analysis of variance (MANOVA), Regression analysis, T-Test. These
allow generalizations of findings to the population.

Hypothesis Tests

There are two ways to hypothesis; that is either by observing what naturally happens or by
manipulating some aspects of the environment and observing the effect it has on the variable
that interests us (Field, 2013:13). There are several inferential statistics tests. They include.
among others, correlation; chi-square; regression; t-tests; analysis of variance (ANOVA);
multivariate analysis of variance (MANOVA); analysis of covariance (ANCOVA); factor
analysis, etc.

i. All the hypothesis will be tested at 0.05 significant level

ii. Decision rules

If p – value is ≤ 0.05, we reject H0

If p - value is ≥ 0.05, we fail to reject H0

Correlation Analysis

45
In correlation or cross-sectional research, who observe what automatically goes on the world
(field) without directly interfering with it, whereas in experimental research we manipulate one
variable to see (to grasp or realize) its effect on another. By using this model (tool), the
researcher indicates the degree of relationship between two variables. Data is analyzed, mainly,
using the correlation coefficient (r).

In correlation research, the researcher is only looking at the degree of the relationship between
variables and not the effect of one variable on another variable. Correlation value (coefficient)
lies between -1 and 1. Correlation coefficient is the number ranging from -1 to 1. If the value is
less than -1 or more than 1, something must has gone wrong. [-1 = negative correlation; 0 = no
correlation; 1 = positive correlation]. High correlation = r = 0.75 and above, r = 0.50 – 0.74
acceptable and uncertain r = below 0.50 items with coefficients of 0.5 and below can be either
rewarded to reverse the validity or substituted.

The value of -1 signifies perfect negative correlations; so as one variable increases, the other
decreases by proportionate amount. The coefficient 0 (zero) signifies no relationship between the
variables; that is if one variable changes, the other remains (stays) the same. Coefficient 1
indicates that the two variables are perfectly positively correlated.

There are two Types of Correlation Coefficient

- Karl Pearson correlation coefficient and

- Spearman‟s rho (r) coefficient correlation. For our purpose we shall use the most
accepted model of Pearson.

Two methods are normally used to compute correlation coefficient. These are;

i. Spearman‟s Rank Correlation

ii. Pearson Product Moment Correlation

Correlation Coefficient (r) tells us at a glance the strength and direction of the relationship
between distributions. Correlation coefficient range from -1 to +1. The close a coefficient gets to
-1 or +1, the stronger the relationship. The sign of the coefficient tells us whether the relationship
is negative or positive. The following examples are illustrative of correlation confidents:

46
Coefficient Strength Direction
r = -0.85 Strong Negative
r = +0.85 Strong Positive
r = +0.22 Weak Positive
r = +0.03 Very weak Positive
r = -0.42 Moderate Negative

Example: Correlation of single mothers and students‟ dropout from school

Based on these findings we conclude that there is significant relationship between single mother‟
family background and secondary school dropout in Moshi Rural district. It implies that when
students from single mothers‟ family background increase in secondary schools, their dropout
rates from school increase.

An example of correlation can be determining whether teachers‟ work experience is correlated

with their job satisfaction can be demonstrated as here under. Hypothesis: “there is a significant
relationship between teachers‟ work experience and their job satisfaction”. Decision rule: if the P
value (.05) is less than the observed correlation coefficient, we reject the null hypothesis.
Suppose the data collection from 15 teachers were as follows;

47
The following assumptions were made

i. That the subject were randomly and independently selected from same population
ii. The score are normally distributed
iii. The population variances of the two groups (variables) are homogeneous
iv. There are no significant outliers
v. There is linear relationship

T-Test

Is an inferential statistic used to analyze data in both causal comparative and experimental
research designs. The t-test is used to measure (test) the significance in the differences between
means of two groups such as the experimental and the control groups in the study. The two
groups are compared on the mean scores of the dependent (outcome) variable.

Example: Comparing learners‟ score on a test to determine the effects of the use of projector on
effectiveness in students‟ notes making (in terms of: grammar/structure, spelling mistakes,
content of the notes and the correct use of words. The researcher may select two groups from
BAD II – Linguistics student at AMUCTA. The researcher can compare the group scores of
student, who attended the lectures that employed computer projection and the other group scores
that attended the lectures without using projectors. A researcher can compare the two groups of
their final linguistics grades‟ scores. This is an independent sample t-test.

Paired T-Test (Correlated Groups) – e.g. pre test and post test. A pair after with before test

One Sample T-Test

One sample t-test is conducted to compare individuals‟ mean with certain fixed value or known
population mean (standard score). Example, the post coordinator sought to determine whether
mean scores of coursework for PGDE class is different from the university standard measure of
40 marks.

Procedure of Testing Hypothesis

i. Formulating Hypothesis

48
Null hypothesis (H0 ): there is no significant difference between students‟ mean scores and their
gender.

Alternative hypothesis (H1 ): There is a significant difference between students‟ mean score and
their gender

ii. Selection of Significant Level

This hypothesis will be tested at 0.05 significant levels

iii. Decision Rules

If the p - value ≤ 0.05 significant level, reject H0

If the p – value ≥ 0.05 significant level, fail to reject H0

Table of independent t-test

Gender N Mean Std. deviation T df Sig. (2-tailed)

Males 10 56.60 12.295 2.623 18 .o17

Females 10 39.40 16.701

Independent-samples t-test was conducted to compare the mean scores of students in arts
subjects and their gender. There was a significant different significance in mean scores between
the two groups, Males (M = 56.6, SD = 12.3) and females (M = 39.4, SD = 16.7; t (18) = 2.623,
p = .017). The magnitude of the differences in the mean score was very large (eta square = 0.28).
These results imply that there is a difference in understanding in arts subjects between males and
females students.

NB. The eta square is the total amount of variation in the data.

When there are three or more groups or variables for comparison, analysis of variance (ANOVA)
is the appropriate modal. AMOVA model is used to determine the significance of differences in
multiples means.

49
Basic statistical techniques of data analysis
 Summarizing data: measures of central tendency (mode, mean and median)
 Measures of relationships (correlation)
 Tests of significances (T-Test)
 Computer application in data analysis

Scales of Measurement
Measurement can be categorized in terms of the type of information communicated by the
symbols or numbers that are assigned

Stevens (1946, 1951) developed 4 levels or “scales” of measurement: Nominal, Ordinal, Interval,
and Ratio scales. Each of these levels conveys a different kind of information. Each level
suggests the type of statistical manipulations of the data that are appropriate and identifies the
type of information being communicated.

The following are the scales and its characteristics

 Nominal; names, labels, categorizes, classifies, types, identifies unquantifiable things, no

magnitude
 Ordinal; provides rank order of things, magnitude e.g. first to last, best to worst, but no
distance between values.
 Interval; includes rank orders, magnitude plus equal intervals/distances between values
or numbers.
 Ratio; have all the features of the first three plus absolute zero point that permits forming
ratio statements and multiplications or divisions

A Nominal Scale uses symbols, such as word or numbers, to label, classify, or identify people or
objects. Variables measured at this level are called categorical variables. Example nominal
variables: gender, organization type, state of residence, teaching method, types of brand,
leadership style, counseling method, and personality type. These can be labeled by numbers. The
symbols that are used to mark the nominal variables cannot be added, subtracted, ranked, or
averaged. However, you can count the frequency within each category, and you can relate a
nominal variable to other variables.

50
This scale is applied in areas as registration number, sex, list of students, place or birth, school
location, type of the school and ethnic group, subject combinations etc. Used in coding for data
analysis like 1 = females, 2 = males, 3 = rural school, 4 = urban school, or 1 = pass, 2 = fail etc.
but these numbers do not have the properties of the real number system.

Ordinal Scale; characterized by to ability to measure the variable in terms of both identity and
magnitude. Variables are ranked based on amount of the attribute possessed. Represents an
ordering of variables, with some number representing more than another. The scale is used to
rank students‟ achievement like division 1, 2 etc., GPD from the minimum to the maximum,
entry qualification into some education levels like first, second and third class or determining
who is higher or lower than the other in some performance etc.

Interval Scale; builds on ordinal scale and adds equal distances between values. Example inches
on a ruler. To say that data are interval, we must be certain that equal intervals on the scale
represent equal distances in the property being measured. It has no absolute zero point. Zero
point is arbitrary. Hence addition and subtraction are possible, but no division and multiplication.
Classifies based on level, age groups, grading performance and arithmetic operations such as
means, standard deviations results.

Ratio Scales; the ration of value along the scale are meaningful. Most used in inferential
statistics (sophisticated statistical techniques) that allows quantification and generalizations in
researches. Zero represents absence of the characteristic being measured.

Ratios like student-teacher, books and pupils etc. determine the number of test items a student
got correct or the amount of time taken to complete an assignment – time in seconds, weight in
pounds, and temperature in degrees. Most physical measurements such as height, weight, age,
distance, are done at the ratio level. Zero formal education, not married is possible, not
attempting the question etc.

Module 7: Research Report Writing

Presenting, Interpreting, Discussions and Conclusions of
Research Findings

51
Three Sections or Stages of Research Report

1: Preliminary stage

i. Cover page
ii. Title page
iii. Certification
iv. Declaration and copyright
v. Acknowledgement
vi. Dedication
vii. Abstract
viii. Table of content
ix. List of tables
x. List of figures
xi. List of abbreviation

2: Main Body

- Chapter one up to chapter five

3: Appendix

Writing the Preliminary Research Information (Table of

Content, Abstract, Acknowledgement, Figures,
Abbreviations etc.)
There are seven major sections in the research report:

i. Title page (involve title, author and institution)

ii. Abstract (summary of content research report)
iii. Introduction (presents the specific problem being investigated, describes the research
strategy, statement of the problem, literature review and tells what you did in the
study)
iv. Method (participant/subjects, instruments, and procedure)
v. Results (summarize the collected data)

52
 vi. Discussion (interprets and evaluates the study result)
vii. References (all citation in the text of the research report)

Bibliography and Referencing Styles (APA 6 th Edition

Format)

53