Chapter 3

Research Design (Planning Research Project)

Meaning of Research Design
People define it in various ways as follow:
 It is plan of collecting and analyzing data in economic, efficient
and relevant manner.
 It is plan of organizing framework for doing study and
collecting necessary data.
 It is conceptual structure within which research is conducted.
 It constitutes blueprint for collection, measuring and analysis of
data.
 It includes outline of what researcher will do from writing
hypothesis and its operational implications to final analysis of
data
 Cont’d
 Whatever be nature of research design the following general
rules should be followed in planning and preparing research
design:-
i. Define nature and scope of problem
ii. Specify related variables (specify sources and types of
information needed)
iii. Exclude variable not relevant to study
iv. Start with logical hypothesis.
 Cont’d
 The design decision should be oriented in answering the
following questions:
(i) What is the study about?
(ii) Why is the study being made?
(iii) Where will the study be carried out?
(iv) What type of data is required?
(v) Where can the required data be found?
(vi) What periods of time will the study include?
(vii)What will be the sample design?
(viii)What techniques of data collection will be used?
(ix) How will the data be analyzed?
(x) In what style will the report be prepared?
 Cont’d
 Keeping in view the above stated designing decisions, one may
split overall research design into the following parts:
(a) sampling design which deals with method of selecting items to
be observed for given study
(b) observational design which relates to conditions under which
observations are to be made;
(c) statistical design which concerns with question of how many
items are to be observed and how information and data
gathered are to be analyzed;
(d) operational design which deals with techniques by which
procedures specified in sampling, statistical and observational
designs can be carried out.
 Cont’d
 In brief, research design must, at least, contain:-
(a) clear statement of research problem;
(b) Procedures and techniques to be used for gathering information;
(c) population to be studied
(d) methods to be used in processing and analyzing data.
 Need for Research Design
 Why we need to have research design? Research design is
necessary because:
a) It facilitates smooth sailing of research operation
b) It makes research project as efficient as possible and help to
yield maximum information with minimum expenditure,
time and effort.
c) It helps researcher to organize his ideas in a form where by it
will be possible for him to look for flaws and inadequacies
d) Design will be given to others for their comment and critical
evaluation. In absence of such course of action, it will be
difficult for critics to provide comprehensive review of proposed
study.
 Features of a Good Design
 Important features of a good research design can be summarized
as follow:
a) It is a plan that contain clear statement of research problem
and specifies source and types of information relevant to
research problem
b) It is strategy specifying which approach will be used for
gathering data or relevant information
c) It indicates population to be studied and methods to be used
in processing and analyzing data
d) It also tentatively includes time and cost budgets, since most
studies are done under these two constraints.
 Important Concepts Relating to Research Design
 the various concepts relating to designs are:
1. Dependent and independent variables: A concept which can
take on different quantitative values is called a variable.
 As such concepts like weight, height, income are all examples of
variables.
 Qualitative phenomena are also quantified on the basis of
presence or absence of concerning attribute(s).
 Phenomena which can take on quantitatively different values
even in decimal points are called ‘continuous variables’.
 But all variables are not continuous.
 If they can only be expressed in integer values, they are non-
continuous variables or in statistical language ‘discrete
variables’.
 Cont’d
 Age is example of continuous variable, but number of children
is example of non-continuous variable.
 If one variable depends upon consequence of other variable, it is
termed as dependent variable, and the variable that is antecedent
to dependent variable is termed as independent variable.
 For instance, if we say that height depends upon age, then height is
dependent variable and age is independent variable.
 Cont’d
2. Extraneous variable: Independent variables that are not related
to purpose of study, but may affect dependent variable are termed
as extraneous variable.
 E.g., if some one wants to test the relationship between intensity of
light on the level of productivity, other variables like age of
workers, heat in the working place or personal problem of worker
may as well affect the level of productivity. Since they are not
related to the purpose of a study, they are called extraneous
variable.
3. Control: A good research design has to minimize the influences
of extraneous variable. To do so researcher uses control as a
remedy to minimize the effects of extraneous variable. In
experimental research control refers to restrain to experimental
condition.
 Cont’d
4. Confounded relationship: When dependent variable is not free
from influence of extraneous variable, then the relationship
between dependent and independent variables is said to be
confounded extraneous variable.
5. Experimental and control groups: In experimental research
when a group is exposed to usual condition, it is called control
group, but when a group is exposed to special condition it is an
experimental group.
6. Treatments: The different condition under which experimental
and controlled groups are put are referred to us treatment. The
usual study program and the special study program are an
example of two treatments in studying the effects of new or
special study program on performance of students.
 Cont’d
7. Experiment: is process of examining truth of statistical
hypothesis, relating to some problem. E.g., examining the
usefulness of a newly developed drug. Experiment can be
comparative or absolute experiment. If we want to determine
impact of newly developed drug against existing drug it is
example of comparative experiment. But the previous example is
an example of absolute experiment.
8. Experimental unit: the pre-determined plots (or blocks or group)
where different treatments are used.
 Forms of research Design
 Research design is unique to methodology.
 Two broad methodology groups can be used to answer any research
problem:- These are experimental and non-experimental.
 The main difference between these two methodologies lies in the
control of extraneous variables and manipulation of at least
one independent variable by intervention of the investigator in
experimental research.
 Take for instance previous example “The impact of intensity of
light in the working places on workers productivity”.
 Here researcher can manipulate independent variable, i.e., intensity
of light.
 Cont’d
 The researcher by changing light intensity in working place can
observe level of productivity and compare result recorded.
 But in non-experimental research, there is no intervention beyond
that needed for purpose of measurement of variable in question.
 Research design for non-experimental research
A) Design for Exploratory research study:
 Exploratory research is a research aimed at identifying and
formulating a research problem or developing working hypothesis.
 The main purpose of such study is only formulating problem for
further, precise investigation or of developing working hypothesis
from operational point of view.
 The major concentration in such study is on discovery of ideas
and insights.
 An exploratory study is often used as introductory phase of a
large study and results are used in developing specific
technique for larger study.
 The design for such study is characterized by great deal of
flexibility.
 Cont’d
 Since no clear hypothesis has been developed about problem, the
following forms of research design can be used:
i. The survey of concerning literature. This is most simple and
fruitful method of formulating precisely research problem or
developing hypothesis. Hypothesis stated by earlier works may
be reviewed and their usefulness be evaluated as basis for further
research.
ii. Experience survey: This implies survey of people who have
practical experience with problem to be studied. The objective of
such survey is to obtain insight into relationship between
variables and new ideas relating to problem. For such purpose
people who are competent and can contribute new ideas are
carefully selected as respondent and investigator then interview
them.
 Cont’d
 The researcher must prepare interview schedule, which ensures
flexibility in a sense that respondent should be allowed to raise
issues and questions, which researcher has not previously
considered.
 It is advisable to send copy of interview schedule to be discussed
to the respondent well in advance.
iii. Analysis of insight (Examining analogous situation): is also
important method for suggesting hypothesis for research.
 It is particularly suitable in areas where there is little experience to
serve as guide.
 This method consists of intensive study of selected instances of
phenomenon. For that purpose the existing records may be
examined (if any).
 Cont’d
 The unstructured interview may be conducted.
 The question however is that what type of examples (instances
should be selected and studied?
 The instances commonly chosen in social science are like
 Cases that provide sharp contrasts or
 Cases having striking features are considered to be relatively more
useful to identify problems
Example
 Reaction of stranger
 Reaction of marginalized people
 Study of individuals who are in transition
 Reaction of people from different social strata
 Cont’d
iv. Examination of case histories: is one of older methods of
research in any branch of science.
 Case histories that are similar in content are generally available
and provide very important area of investigation for exploratory
study.
 However, the result of investigation of case histories is always to
be considered as suggestive rather than conclusive.
 B) Research design for descriptive study
 Descriptive researches are concerned with describing
characteristics or function of particular individual, or of group
or phenomenon.
 Most of social researches come under this category.
 This study involves description of extent of association between
two or more variables.
 Descriptive research in contrast to exploratory research is marked
by prior formulation of specific research problem (question).
 Investigator already knows substantial amount about research
problem, may be as result of exploratory study, before project is
initiated.
 Thus, researcher should be able to define clearly, what he wants to
measure and setup appropriate and specific means for measuring it.
 Cont’d
 In descriptive research researcher needs to define not only what he
wanted to measure but also must find adequate method for measuring
it along with clear-cut definition of population he wants to study.
 Since purpose is to obtain complete and adequate information,
procedure to be used must be carefully planned.
 Thus design in such study must be rigid and must focus on:
i. Formulating objective of study
ii. Design method of data collection
iii. Select sample of study (sampling)
iv. Collect data relevant for study
v. Processing and analysis of data collected
vi. Reporting findings
 Cont’d
i. Formulating objective of study (what is study about? why is it
being made?): The first step in descriptive study is to specify
objective with sufficient precision to ensure that data collected
are relevant.
ii. Design method of data collection: After formulating of objective
of study next comes planning of method of data collection. First
of all the researcher has to determine relevant and type of data
needed to study. After defining relevant information needed for
designed study, method of data and technique for collecting them
must be devised. Method of data gathering (like, observation,
questionnaires, and etc.) with their merit and limitation are
considered and best one is selected for particular study.
 Cont’d
iii. Select sample of study (sampling): In most studies, the researcher
takes out sample(s) and then wishes to make statements about
population on the basis of sample analysis. In short sampling has
to be designed. There are different sample techniques among
which appropriate sampling technique would be selected. The
sample expected to be representative and yielding accurate
information with minimum amount of research effort.
iv. Collect data relevant for study (Plan of data collection): To
obtain data free from errors introduced by those responsible for
collecting them, it is necessary to supervise closely staff or field
workers as they collect and record information. As data are
collected, they should be examined for completeness,
comprehensibility, consistency and reliability.
 Cont’d
v. Processing and analysis of data collected: The data collected
must be processed and analyzed. Data processing design includes
steps like coding interview replies, observation, tabulating data;
and performing several statistical computations. To the extent
possible, processing and analyzing procedure should be planned
in detail before actual work started.
vi. Reporting findings: This is task of communicating findings to
others and researcher needs to be well planned so that all things
relating to research study may be well presented in simple and
effective style.
 Such design appropriately referred to as survey design since it
takes into account all steps involved in survey research.
 Research design for experimental research
 Experimental studies are those in which researcher can control
and manipulate at least one of independent variable and test
hypothesis of causal relationship between variables.
 Experimental design refers to framework or structure of
experiment.
 The study of experimental design has its origin in agricultural
research.
 Planning of such design was made first by Fisher (1960) when he
was working at center of Agricultural Research in England.
 Today, experimental design being used in researches relating to
phenomenon of several disciplines.
 Basic principle of experimental design
 Fisher has enumerated three principles of experimental design.
1. The Principle of Replication: the experiment should be repeated
more than once. That is, each treatment is applied in many
experimental units instead once. By doing so statistical accuracy
of experiment is increased. Replication is introduced in order to
increase precision of study.
 Suppose the researcher want to examine effect of new teaching
method on students' performance. The researcher can divide
students' population into two groups, and one treatment (old
teaching method) will be given to one group and new teaching
method to another group. After treatment the result is compared in
both groups. Applying this principle implies students population is
divided into more than one group and each treatment is given more
than once.
 Cont’d
2. The Principle of Randomization: This principle of experiment is
applied in order to reduce the effect of extraneous factor by
randomizing application of treatments. In other words, this
principle indicates that we should design or plan experiment in
such a way that variation caused by extraneous factor can all be
combined under general headings of “chance“. We can apply
principle of randomization to above example.
 To reduce effect of extraneous variable, which is intelligence, a
researcher can randomly assign students in different groups or we
can apply different treatments randomly.
 Cont’d
3. The Principle of Local Control: This is other important principle
of experimental design. This is used to measure and eliminate
variability caused by extraneous variable. The extraneous factor,
known sources of variability, is made to vary deliberately over as
wide range as necessary and this need to be done in such a way
that variability it causes can be measured and hence eliminated
from the experimental error.
 Here the researcher will try to fix extraneous variable. The students
are divided on basis of their IQ. Students in each IQ level are
divided into two groups and each group is given different treatment
and results are compared. By doing so effect of extraneous variable
(intelligence) is controlled.
 LITERATURE REVIEW
INTRODUCTION
 Literature review is a critical look at existing sources that are
significant to your topic.
 The purpose of literature review is to “look again” (re + view) at
what other researchers have done regarding specific topic
 literature review is means to end, namely to provide background
to and serve as motivation for objectives and hypotheses that guide
your own research
 In writing literature review, you are not expected to merely
summarize available previous researches. Rather, you should be
able to critically evaluate, re-organize and synthesize work of
others.
 Cont’d
 Literature means also works you consulted in order to
understand and investigate your research problem.
 It is description of your topic area, supported by references
 It is summary, discussion and critical analysis of academic
work related to your research question
 The Purpose is to convey to reader what knowledge or ideas
have been established on topic, and what their strengths and
weaknesses are.
 The literature review must be defined by guiding concept (e.g.
your research objective, problem or issue you are discussing).
 Cont’d
 It is not just a descriptive list of material available or set of
summaries.
 It is well-written analytical narrative that brings reader up-to-
date on what is known on given topic, but also provide fresh
insights that advance knowledge, Resolve conflicts between
studies, Identify new ways to interpret research results and
Creating path for future research.
 Cont’d
 New research grows out of old

 fig 2.1 and others.docx page 1

 Review of Key Elements of Previous Definition
 Literature Review is summary of researches: It is not a “list” of
found research but coherent and articulate account of past and
current research findings
 Suggestion: read 2 or 3 Literatures in order to become familiar
with Literature Reviews
 What to review?
 Original research articles, review articles, opinion pieces, short
communications, conference proceedings, edited books, editorials,
databases etc.
 Text-Books are not good source of materials for review-as they
usually contain knowledge that is too established.
 Literature Review should be exhaustive and as current as
possible.
 How many articles?
 fig 2.1 and others.docx page 5
 There is no set number. As long as the search is exhaustive and
focused on research topic, the review will be acceptable.
 Planning a Search Strategy
 fig 2.1 and others.docx page 2
 How to review?
 First establish logical sub-topics or categories of topics
 Read papers, paraphrase their key findings and put them in pre-
established categories
 Do not cite papers for non-content, therefore, always focus on
abstract as important source of information
 New sub-topics or categories may emerge.
 It is okay to modify sub-topics as one goes into details of available
knowledge
 Review of Key Elements of Previous Definition
 How far back should one search? A reasonable and widely
accepted timeframe includes research conducted during past 10
years.
 Important studies (i.e., studies that had significant impact on the
field of study) should also be mentioned even if these go beyond
mentioned timeframe.
 Common connection phrases used in Literature
review
1. Agreement or addition
i. Besides
ii. In addition to
iii. Similarly
iv. moreover
v. like
2. Comparison
i. While
ii. Unlike
iii. As compared to
iv. In contrast
 Cont’d
3. Disagreement
i. Nevertheless
ii. While
iii. Unlike
iv. Yet
4. Connection
v. Because –therefore
vi. as,
vii. Since, --
viii. Neither nor
ix. Either or
 Bad reviewing
 Making a list of unrelated or related but fragmented research
articles,.
• Jackson (2007) said…..
• Betty (2009) said….
• Thomas (1998) conducted a study about….
Paraphrase, Summary, and Quotation
 Paraphrase: Para means by the side of. Paraphrase restates all
original material in different words and is approximately the
same length as the original.
 Summary: Gives an overview of general ideas and is shorter
than the original. It avoids specific examples and is more concise
than the original.
 Quotation: Contains exact words of your source and includes
quotation marks.
 Plagiarism
 Plagiarism is passing off somebody else’s work as your own
 In literature review you are discussing other people’s work so you
must clearly reference it
 If you cannot describe something in your own words then you
must not include it in your literature review.
 If you paraphrase other people’s ideas you still need to reference
them
 Referencing your work
 You MUST reference your work- means giving the author name
and publication date for all ideas and facts you have used
 List the sources at the end (Harvard format)
 Failure to reference your work is Plagiarism and is an automatic
fail.
 References
 In text citation
• You must follow the format indicated by the
– Source of fund
– Journal
– Supervisor etc.
 Smith (1999) says that C++ is faster than Java. Part of the
sentence
 C++ is faster than Java (Smith 1999). Not part of the sentence
 Common Problems in writing literatures
 Trying to read everything: if you try to be comprehensive you
will never be able to finish the reading!
 The literature review should not provide a summary of all the
published work that relates to your research, but a survey of the
most relevant and significant work.
 Reading but not writing: Writing takes much more effort than
reading- don't put writing off until you've "finished" reading.
 Not keeping bibliographic information: When preparing your
reference you might notice that you have forgotten to keep the
information you need.
 To avoid this nightmare always put references into your writing.
 Theoretical Literature Review
 A theoretical framework is collection of interrelated ideas based
on theories.
 It is reasoned set of prepositions, which are derived from and
supported by data or evidence.
 A theoretical framework accounts for or explains phenomena.
 It attempts to clarify why things are the way they are based on
theories.
 A theoretical framework is general set of assumptions about
nature of phenomena.
 Importance of Theoretical Literature Review
 Theoretical framework plays major role in research which include
the following:
 It introduces researcher to new view of research problem. This
enables researcher to understand total realm of problem.
 It enables researcher to conceptualize topic in its entirety as an
outgrowth of larger society. This helps researcher to acknowledge
the problem from wider perspective and not from a narrow
personalized self-interest approach. This enhances the
researcher's objectivity.
 Qualities of Effective Theoretical Literature
 An effective theoretical Literature should:
 Account for and explain a phenomenon,
 Be specific and well-articulated.
 Reflect research problem being addressed.
 Be measured in practical situation.
 Provide tentative answers to questions, issues and problems
addressed in research problem.
 Should systematically address various aspects of problem,
particularly key factors that are assumed to influence or cause
problem.
 Guidelines in Formulating Theoretical Literature
 In formulating effective theoretical framework, the researcher
should adhere to the following:
 Reflect on existing theories for purpose of identifying fitting
context.
 Analyze research title to identify independent and dependent
variables. The researcher should then reflect on relationship
between these variables.
 Find out which theories best explain relationship between
variables. This can be achieved by using library and reading books
and articles related to topic selected.
 Cont’d
 Formulation - The researcher should then write down theories
applicable, link ideas and identify relationship. After this the
researcher should formulate theoretical framework. This will
involve discussing selected theories in an attempt to answer the
research question.
 Evaluation - After formulating theoretical framework, the
researcher should evaluate it to find out if it addresses all sections
of research problem.
 Empirical Literature Review
 It consists of review of empirical data on subject matter.
 It includes comparing and contrasting of different
interpretation on the same subject by different scholars.
 While doing empirical literature review, researcher should relate
literature to each objective of research.
 Moreover empirical literatures that are used by given researcher
should not be outdated and related with study that researcher
going to investigate.
 Cont’d
 The researcher should demonstrate thoroughness in field being
investigated by critically reviewing empirical studies that have
been done in same or related study.
 This should be done at global, African and Ethiopian context.
 This analysis should be critical clearly identifying where studies
were conducted, sampling issues, key findings and observed
weaknesses in studies.
Summary of Related Literature and Research Gaps
 After reviewing literature researcher should summarize shortly
both theoretical and empirical literatures and identify gaps in it.
 Conceptual/Theoretical Framework
 Research is based on conceptualization of pertinent problem and
relationships between problem variables and how they affect one
another.
 The conceptual framework is diagrammatic model or
representation of relationship between variables and how they
are operationalized for purpose of research.
 It should indicate independent and dependent variables (cause/
effect relationship) or vice versa, moderating and intervening
variables (where applicable).
 It is not mandatory that every student must have moderator or
intervening variable but it depends on facts in real world and logic
as well as student’s own conceptualization.
 Cont’d
 Inclusion of such variables demands that candidates must test their
effects in research report.
 The source of variables / framework and explanation of
diagrammatic representation should be made.
 The framework must be linked to theory of study.
 Variables and dimensions presented in conceptual framework must
be derived from appropriate theoretical framework or known
authorities who must be cited and acknowledged.
 Cont’d
 Conceptual framework is used in research to outline possible
courses of action or to present preferred approach to idea or
thought
 Conceptual frameworks (sometimes called theoretical frameworks
by some authors) are type of intermediate theory that attempt to
connect to all aspects of inquiry (e.g., problem definition, purpose,
literature review, methodology, data collection and analysis).
 Conceptual frameworks can act like maps that give coherence to
empirical inquiry.
 Because conceptual frameworks are potentially so close to
empirical inquiry, they take different forms depending upon
research question or problem.
 Cont’d
 Proponents claim that when purpose and framework are aligned,
other aspects of empirical research such as methodological choices
and statistical techniques become simpler to identify.
 A conceptual framework is researcher’s own position on problem
and gives direction to study.
 It may be adaptation of model used in previous study, with
modifications to suit inquiry.
 Aside from showing direction of study, through conceptual
framework, researcher can be able to show relationships of
different constructs that he wants to investigate.
 Cont’d
 Once conceptual framework has been determined, next for
researcher is to determine what research methods to employ to best
answer research problem through proposed framework.
 The conceptual framework must clearly identify key variables of
study, dimensions and measurable indicators for each of variables
as well as relationships that exist among variables.
 The source of variables should be clearly indicated by student
through appropriate citations.
 The candidate must coherently describe conceptual framework to
enable reader understand what study will cover and student must
clearly operationalize all study variables.
Example of conceptual framework, determinants of
economic growth

Human
Government
Capital Financial
spending Developm
ent
gross
social
capital
welfare
formation /
expenditur Economic gross
e Growth investment
population
Export
growth
External Foreign
debt Aid
 R 3
P T E
H A
F C
D O
EN