RESEARCH DESIGN AND METHODOLOGY (RESEARCH DESIGN)

Research designs
A research design is the overall strategy used by the researcher as a means of collecting
and analyzing data in order to test research hypotheses or to answer research questions.
The following section explains types and qualities of a good research design.

Types of research designs

1. Experimental designs
Experimental research designs use the symbols discussed herein:

R = Random selection of subjects and assignment to

experimental and control groups.
X = Experimental treatment (variable) applied.
C = Control condition.
O = Observation or test.
- = A line between levels indicates equivalence of groups.

a) True experimental designs have the following characteristics

 They employ randomization to provide for equivalence between

experimental and control groups.
 They are the strongest type though difficult to arrange especially in
social sciences, but used whenever possible.

i) The post – test only, equivalent groups

R X O1

R C O2

 The experimental and control groups era equated by randomization.

b) The Solomon four – groups design, for example,

R O1 X O2
R O2 C O4
R X O5
R X O6

i) Subjects randomly assigned to four groups

ii) Two groups receive treatment (x)
iii) One experimental group receives a protest (01).
iv) Two control groups receive no treatment.
v) One control group receives a pro-test.
2. Survey designs (survey research)

These are research designs which study large and small populations (or universe) by
selecting and studying samples chosen from the population to discover the relative
incidence, distribution and interactions of variables. They employ a flow plan or chart to
outline the design and subsequent implementation of a survey. The flow plan starts with
the objectives of the survey, lists each step to be taken and ends with a final report as
illustrated here below:

Formulating objectives of the study

Sampling and Sampling Design

Designing methods of data collection and constructing

instrument

Collection of data, time and location

Processing, coding, tabulation and analysis of

data

Interpretation and reporting of findings

3. Exploratory design (field studies)

The exploratory design is mainly used in qualitative studies. It is a non-experimental

scientific investigation aimed at discovering the relation and interaction between
variables in real social structures. The investigation in a field study first looks at a social
or institutional situation and then studies the relation among the attitude, values,
perceptions and behaviours of individuals and groups in the situation. Data is mainly
collected through observation and interviews.

Purpose of exploratory study is to discover significant variables in the field situation; to

discover relations among variables; and to lay groundwork for later, more systematic and
vigorous testing of hypothesis.

Exploratory research primarily uses qualitative research.

Validity and reliability of a research design

The two essential characteristics of research designs in terms of the arrangement of

conditions that ensures relevance to the research purpose and accuracy of results are:
i) Reliability
ii) Validity

1. Reliability
It means the consistency or accuracy of the research instrument, in measuring whatever
it measures. It is the degree to which an instrument will give similar results for the same
individuals at different times.

Statistically, or in any measurement process, theory assumes that there is an ‘error’

contained in all forms of measurements. An observation score can be seen as consisting
of two parts, namely:

 The individual’s ‘true score’

 An ‘error score’, which is due to the inaccuracy in measurement

‘True score’ + ‘error’ = ‘observation score’

Reliability is related to these scores. If scores have large error components, then
reliability is low, but if there is little error in the scores, then reliability is high. Reliability
is thus a statistical concept based on the association between two sets of scores
representing the measurement of individuals on two different occasions.

2. Validity

Validity refers to the extent to which a measurement does what it is supposed to do.
Validity may be determined by use of well-devised research designs which:

 Provides a good strategy for the hypothesis (es) or answering research questions
 This means an adequate plan of procedures for data collection and analysis that
should be undertaken to evaluate a particular theoretical perspective: accurately
and purposefully.

Internal validity: refers to the extent to which relationships between independent and
dependent variables in a research study can be said to be genuine. It implies the minimum
control arrangement of conditions necessary to interpret the results. It is concerned with
the adequacy of procedures for collection and analysis of data and interpretation of results
in a manner that is relevant to the research purpose. Specifically, it refers to the extent to
which the independent variable (s) can be said to have a genuine effect on the dependent
variable(s).

Internal validity, therefore deals with the degree of control of the influence of extraneous
and other variables factors.
Activity
Find out the meaning of extraneous variables
External validity: refers to the, extent to which the research findings / results can be
generalized to other populations, other than the research setting. It is concerned with such
questions as:

i) To what group of people can the findings be generalized?

ii) In which type of settings and under what conditions can the results be
generalized
iii) To what other variables, situations and so forth do the results generalize?

Thus, external validity is a matter of sampling .The broader or wider, the sampling from
a given population, (for example, work, home school), and conditions (for example, work
demand, income levels, educational attainment), the greater the independent – dependent
variables relation.

Factors affecting internal validity

i) History
History is made up of specific external events beyond the control of the researcher that
may have stimulating or disturbing effect on the performance of subject. It is also made
up of the unanticipated events occurring while the research is in progress that can affect
the dependent variable. Examples are anxiety, emotional stress and fatigue.

ii) Maturation
Maturation refers to progress operating within the subject as a function of time. It
includes the physical and mental changes of subjects over a period of time.
Examples are: a subject may become tired, bored or wiser, etcetera leading to reduced
or increased effect on final observation

iii) Instrumentation
An instrument is valid for a particular purpose and group. Example: an interview
schedule designed for CBD employees in a service industry cannot be valid for
employees in a manufacturing concern.

Instrumentation is an effect due to inconsistent use of measuring instruments. For

instance:

o A questionnaire may be changed between pre-test and post-test. This change is

like likely to result in an effect that is independent of the effect due to the research
variables.

o In observation, observers may become tired, acquire a second dimension and

behave as if they were ‘only human’.

iv) Testing (effect of pre-testing)

When a pre-test is given, it is possible that the initial experience with the test during the
first assessment period can cause inflated (exaggerated) scores on the second measure.
Subjects become “aware” after the first test, so that improved scores on the second tests
threaten the internal validity.
 v) Statistical regression
This occurs when a sample which is extreme on some variable is selected, for example,
poor readers, and remedial groups, etcetera.

The group is usually obtained by using some measurement in the first instance, if a
second measure is administered with the purpose of variable. An extreme group, initially
below the mean will, on second testing, seem to have improved even if it has no treatment
whatsoever.

vi) Selection (maturation interaction)

A combination of differential selection and maturation produce a joint (interaction)
effect, either additional to or in the absence of any biases resulting from the two
separately.

Example:
A group of delinquents treated over a period of some years might show substantially
greater gains on behaviours rating compared to a ‘normal’ group solely because of
selection bias’ and maturation ( long period ) effects.

Factors affecting external validity

a) Interaction between testing and experimental variable

Subjects may be extremely sensitized by the pre-tests, which may have serious effects
on the influence of experimental variation.

Example:
Evaluation of fancy dietary scheme on group of fat girls

ii) Interaction between selection and experimental variable

iii) Reactive effects of the experimental arrangements (Hawthorne effect).

As soon as people realize they are guinea pigs, they change (usually, temporary) and
thus produce changed effect. Any aspect of the experiment may produce this effect.

iv) Multiple – treatment interactions

These occur when the same groups of subjects are used for various treatment involved
in the experimental design. This occurs because it is impossible to erase the effects
of prior treatment before embarking on new ones.

Activity

Give the meaning and purpose of research. Give an example to explain the difference
between experimental research and predictive inquiry.
How to design a research study
Conducting research requires a logical sequence of related steps:
1. First, you will develop an interest to study something. There will be questions to be
raised about the interest.
For example;
 Why do you want to study “students’ discipline”
 What information is missing about it that must be sought?
 What do you want to achieve at the end of the study?
 The interest is then transformed into an idea. For example, is Guidance
and Counselling related to discipline? I s A related to B?
 Theory or previous research would help to explain or answer these
questions. Theory explains why things happen and what makes them
happen.

2. Conceptualize the problem (what is to be studied). Here you must specify the meaning
of the concepts and variables to be studied. In this case;

 What is discipline, as a concept?

 Which variables represent discipline?

3. Make a choice of the research method to use in that study. Find out which method is
most appropriate. Is it descriptive research, or Correlational research etcetera?

4. Determine whom you want to draw conclusions about. Who will be observed and
measured to obtain the necessary data? Will they be teachers of G&C and students,
only?

5. Ask yourself how the variables will be measured to obtain data.

6. Observation and measurement follows, that is, collecting data for analysis and
interpretation. Will you use questionnaires or interviews?

7. Data processing must then be done in order to transform the data collected into a form
appropriate for analysis.

8. Data analysis is then conducted to help you develop interpretations and draw
conclusions.

9. The results are then reported and their implications developed.

Activity
Assuming that you are required to carry out research study relating to your school
programmes, such as the “Effects of Guidance and Counselling on students
Discipline”.
 What would be your first activity?
 What other activities would you carry out?
 Where would you go to search for information necessary to conduct the
activities?