You are on page 1of 280

ECON-532: Advance Research Methods with

Statistical Packages,
Lecture slides
Prepared for MSc Students
Department of Economics
Wolaita Sodo University
October 2019

Compiled By Solomon Kebede, MSc, PhD Candidate, 1


2019
ECON-532: Advance Research Methods with
Statistical Packages

Understanding the Elements of Economic


Research

Compiled By Solomon Kebede, MSc, PhD Candidate, 2


2019
Main Contents of the lecture
1. Introduction – What is research?
2. What is knowledge and how it is generated
3. Classification of Economic Research
4. Research ethics
n Harm to participants
n Lack of informed consent
n Invasion of privacy
n Deception
n Protecting research subjects

Compiled By Solomon Kebede, MSc, PhD Candidate, 3


2019.
Introduction -What is Research?
o Research is about advancing the frontiers of knowledge.
n Research = creation of knowledge

o It is a scientific and systematic search for pertinent


information to answer a question, and thereby help in
solving a problem.
n Social interrelationship, opinions, customs, habits, conditions
of life can be the focus of a research.

o Empirical studies are carried out to diagnose the situation, and


to make recommendations to overcome the undesirable
situation.
Compiled By Solomon Kebede, MSc, PhD Candidate, 4
2019
Introduction -What is Research?
o Research is done in an organized manner.
n A research project has a well-known structure: a
beginning, middle, and end.
o To do research, it is necessary to know what the correct tools
are, and how to use them.
o And the techniques and range of tools you use to do research
are called Research methods.
o They provide you with the knowledge and skills you need
to collect, sort and analyze information so that you can
come to some conclusions.
Compiled By Solomon Kebede, MSc, PhD Candidate, 5
2019
Introduction -What is Research?
o For instance, research methods provide the needed
information that guides policy makers to make informed
decisions by reducing uncertainty.
o With the right sort of methods you should be able to
convince other people that your conclusions have
some validity

o So, identifying a subject to research, finding and collecting


information and analyzing it are the tasks of research.
o the task of generating new knowledge.

Compiled By Solomon Kebede, MSc, PhD Candidate, 6


2019
Introduction – Objectives of Research
o The research objectives may be broadly grouped as follows:
n To gain familiarity with a phenomenon or to achieve new
insights into it - exploratory or formulative research studies
n To portray accurately the characteristics of a particular
individual, situation or a group- descriptive research studies
n To determine the frequency with which something occurs or
with which it is associated with something else --diagnostic
research studies
n To test a hypothesis of a causal relationship between
variables- hypothesis-testing research studies

Compiled By Solomon Kebede, MSc, PhD Candidate, 7


2019
Introduction – What You Can do With Research
o Research can be used to:
n Categorize - involves forming a typology of objects,
events or concepts and can be useful in explaining which
‘things’ belong together and how.

n Describe - attempts to examine situations in order to


establish what is the norm.

n Explain - aims to move beyond ‘just getting the facts’ in


order to make sense of the myriad other elements
involved.
Compiled By Solomon Kebede, MSc, PhD Candidate, 8
2019
Introduction – What You Can do With Research
o Evaluate - involves making judgements about the quality of
objects or events.

o Compare - Two or more contrasting cases can be examined


to highlight differences and similarities between them,
leading to a better understanding of phenomena.

o Correlate - relationships between two phenomena are


investigated to see whether and how they influence each
other.

Compiled By Solomon Kebede, MSc, PhD Candidate, 9


2019
Introduction – What You Can do With Research
o Predictions- if there has been a strong relationship between
two or more characteristics or events in the past, then similar
circumstances should exist in the future, leading to
predictable outcomes.

o Control- Once you understand an event or situation, you may


be able to find ways to control it.
n you need to know what the cause and effect relationships are
and that you are capable of exerting control over the vital
ingredients.

Compiled By Solomon Kebede, MSc, PhD Candidate, 10


2019
Introduction – Motivation in Research
o The motivation to do research may be due to the
n Desire to get a research degree along with its
consequential benefits
n Desire to face the challenge in solving the unsolved
problems, i.e., concern over practical problems initiates
research
n Desire to get intellectual joy of doing some creative
work
n Desire to be of service to society
n Desire to get respectability, etc.

Compiled By Solomon Kebede, MSc, PhD Candidate, 11


2019
How is knowledge generated
o Research is about generating knowledge.
o Knowledge is awareness of something, an issue, a fact or an
understanding of a matter.
o Knowledge of our social world comes from a range of
different sources:
n through reasoning
n through intuition, or
n through the use of appropriate methods.

o But, not all roads to knowledge are equally


useful/reliable.
Compiled By Solomon Kebede, MSc, PhD Candidate, 12
2019
How is knowledge generated
o There are two methods of obtaining knowledge: scientific
and non-scientific methods.
n Non-Scientific Method: gaining knowledge through
senses, experience, intuition, and revelation.

o Senses and Experiences (Experiential knowing). Some gain


knowledge through physical senses - sight, sound, touch, taste,
and smell -and experience.
n When one gets too close to a fire and gets burned once, he
or she gains the knowledge that it is dangerous to be too
close to the fire.

Compiled By Solomon Kebede, MSc, PhD Candidate, 13


2019
How is knowledge generated
o Intuition (Tenacity): knowledge is obtained if one strongly
believes what one perceives is real and true.
n Intuition is the strong feeling that what one perceives to be
the case is indeed true.

o Believing something because, based on your view of the


world and your assumptions, you don’t want to give up your
belief.
n But it is subjective.

o B/C: If two people hold mutually contradictory beliefs, both


cannot be true.

Compiled By Solomon Kebede, MSc, PhD Candidate, 14


2019
How is knowledge generated
o Revelation (Authoritative knowledge): An alternative to an
individual’s belief in what is true, could reside in what
authorities say is true.
n Revelation is the presentation of the truth from a
supernatural source.

o We may feel convinced that something is true because an


authority (the Bible, Koran, a leader or teacher) tells us it is
true.
n Authorities would force beliefs under threat of some kind of
penalty.
o But, experts with different perspectives will hold different
beliefs. How is one to know which expert / authority is actually
right? Compiled By Solomon Kebede, MSc, PhD Candidate, 15
2019
How is knowledge generated
o Knowledge acquired via experience, intuition and revelation
remain private knowledge.
n knowledge obtained through nonscientific methods cannot
be subjected to objective testing.

o Scientific Method: The most valid approach is through


science, which is objective and self-correcting
o This method of gaining knowledge is obtained by reasoning
and is considered to be the most reliable method of gaining
knowledge.
n knowledge obtained by scientific method can be subject to
testing.
Compiled By Solomon Kebede, MSc, PhD Candidate, 16
2019
The Scientific Method of Research
o The difference between scientific knowledge and other kinds
of knowledge is that scientific work is much more
systematic than casual observation.
o Scientific method consist of systematic observation,
classification and interpretation of data.
o Our scientific knowledge relies on the fact that our
observations are objective, data-driven, public, and potentially
replicable.

n The process is incremental, with a series of small steps


rather than a giant leap.
Compiled By Solomon Kebede, MSc, PhD Candidate, 17
2019
The Scientific Method of Research
o What constitutes a ‘science’, the nature of its methods and
the types of data which it should collect are open to dispute.

n Science may be defined as an objective, logical, and


systematic method of analysis of phenomena devised to
permit the accumulation of reliable knowledge.

o Science is interested to discover “ laws“ or regularities of


both the physical and social worlds.
o The law of gravity
o demand and supply

Compiled By Solomon Kebede, MSc, PhD Candidate, 18


2019
The Scientific Method of Research
o Science insists on following systematic, methodical “rules“ for
gathering empirical evidence.

n Because, evidence that is obtained in an unsystematic way


is regarded as tainted or problematic;
o i.e. it is seen as less trust worthy.

o The actual practice of science also shows that there are


alternative methods of gathering information and of analyzing
the data.

Compiled By Solomon Kebede, MSc, PhD Candidate, 19


2019
The Scientific Method of Research
o Some sciences, such as the natural sciences deal with the
physical and material world- natural science.
o The subject matter of the natural sciences is the physical world,
and is independent of the researcher and the way it is
understood.
o For example, a tree has a physical structure of its own,
regardless of the scientist who is studying the way it grows.
o But other sciences involve the study of people – their beliefs,
behavior, interactions, attitudes, institutions, and so forth-
social sciences.
o They are sometimes called soft sciences.
Compiled By Solomon Kebede, MSc, PhD Candidate, 20
2019
The Scientific Method of Research
o A misconception about research is that research is related to
laboratory research.

n But research is not limited to certain fields of study: it is


characterized by the methods used.

o This is to say that social science is not sloppy or lack rigor.


n It means the subject matter in social science (human
social life) is fluid, formidable to observe, and hard to
measure precisely with laboratory instruments.

Compiled By Solomon Kebede, MSc, PhD Candidate, 21


2019
The Scientific Method of Research
o Social research is open to a lot of debate as the focus of the
investigations is humanity.
n The problems are compounded by the researchers being
human too

o Social research is more than a reflection of our opinions and


prejudices.
o Rather:
n Social research substantiates, refutes, organizes or
generates theories and evidence which may challenge not
only our own beliefs, but also those of the society in general.

Compiled By Solomon Kebede, MSc, PhD Candidate, 22


2019
The Scientific Method of Research
o A simple summary of the steps in the scientific method could
go like this:
n Identification or clarification of the problems.
n Formulation of tentative solutions or hypotheses.
n Practical or theoretical testing of solutions or hypotheses.
n Elimination or adjustment of unsuccessful solutions.

Compiled By Solomon Kebede, MSc, PhD Candidate, 23


2019
Characteristics of the Scientific Method
o A set of overriding principles govern the scientific
investigation.
o It is based on empirical evidence
o The scientific method is based on empirical evidence
and utilizes relevant concepts
o It is concerned with the realities that are observable
through the use of verifiable experience or
observation.
o Some of the realities could be directly observed, like the
number of students present in the class or the number of
people employed in a factory.
Compiled By Solomon Kebede, MSc, PhD Candidate, 24
2019
Characteristics of the Scientific Method
o But, there are also realities which cannot be observed
directly.
n Realities that cannot be put to “sensory experience”
directly or indirectly
Ø existence of heaven,
Ø the Day of Judgment,
Ø life hereafter,
Ø God’s rewards for good deeds, etc.
n does not fall within the domain of scientific method.

Compiled By Solomon Kebede, 25


MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o it is based on logical reasoning (Critical thinking)
n The scientific method practices logical reasoning which
allows determination of the truth through steps
different from emotional and hopeful thinking

n Science is fundamentally a rational activity, and the


scientific explanation must make sense.
o Religion may rest on revelations, custom, or
traditions, gambling on faith, but science must rest
on logical reason.

Compiled By Solomon Kebede, 26


MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
q It possesses skeptical attitude
q Scientific attitude (SA) is often equated with skeptical
attitude

n This is the constant questioning of your beliefs and


conclusions.
n It requires the possession of skeptical attitudes.

n A skeptic holds beliefs tentatively, and is open to new


evidence and rational argument

Compiled By Solomon Kebede, 27


MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o It is Verifiable: Observations made through SM are to be
verified by using the senses to confirm or refute the
previous findings.
n The more the results are replicated or repeated, the
more we will gain confidence in the scientific nature of
our research.
o It is Cumulative: Prior to the start of any study researchers
try to scan through the literature and see that their study is
not a repetition.
n Instead of reinventing the wheel the researchers take
stock of the existing body of knowledge and try to build
on it.
Compiled By Solomon Kebede, 28
MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o A linkage between the present and the previous body of
knowledge has to be established, and that is how the
knowledge accumulates.
n Every new idea does not have to start from a scratch;

o The existing body of knowledge provides a huge foundation


on which researchers build on and hence the knowledge
keeps on growing.

Compiled By Solomon Kebede, 29


MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o It is Deterministic: Science is based on the assumption that
all events have preceding causes that are subject to
identification and logical understanding.
n For the scientist, nothing “just happens” – it happens for
a reason.

o The scientific researchers try to explain the emerging


phenomenon by identifying its causes.
n The researcher tries to narrow down the number of
reasons in such a way that some action could be taken.

Compiled By Solomon Kebede, 30


MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o It possess ethical and ideological neutrality: The
conclusions drawn through interpretation of the results of
data analysis should be objective;
n i.e., they should be based on the facts of the findings
derived from actual data, and not on our own subjective
or emotional values.
o Researchers are human beings, having individual ideologies,
religious affiliations, cultural differences which can influence
the research findings.
o Therefore, follow the principle of objectivity, uphold
neutrality, and present the results in an unbiased manner.
Compiled By Solomon Kebede, 31
MSc, PhD Candidate, 2019
Characteristics of the Scientific Method
o It lends to statistical generalization: The scope of the
research findings in one organizational setting applicable to
other settings.
n The more generalizable the research, the greater its
usefulness and value.

o Science Is Public: Scientists only recognize research as valid


or useful when they can scrutinize it.
n Generally, we accept research as valid if it has
undergone a peer review process.

Compiled By Solomon Kebede, 32


MSc, PhD Candidate, 2019
Ways of Reasoning
o The modern method of science is broadly of two broad
methods of reasoning: Induction and Deduction.
n However, it is often not possible practically to apply
either extreme in a pure fashion.
o The shortcomings of each can be mitigated by using a
combination that is formulated as the hypothetico-deductive
method.
n The scientific method uses the features of each approach in a
pragmatic way.

Compiled By Solomon Kebede, 33


MSc, PhD Candidate, 2019
Ways of Reasoning
o They represent the two contrasting approaches to acquiring
knowledge and understanding the world.

n Plato argued for deductive thinking (starting with theory


to make sense of what we observe)

n While Aristotle for inductive thinking (starting with


observations in order to build theories).

Compiled By Solomon Kebede, 34


MSc, PhD Candidate, 2019
Ways of Reasoning
o Deductive reasoning (the rationalist’s approach)
n first developed by the Ancient Greeks.

o An argument based on deduction begins with general


statements and, through logical argument, comes to a specific
conclusion.
n Sometimes this is called a "top-down" approach.

o It consists of a major general premise (statement), followed by a


minor, more specific premise, and a conclusion which follows
logically.

Compiled By Solomon Kebede, 35


MSc, PhD Candidate, 2019
Ways of Reasoning
o The research in this case is guided by the theory.
n begin with thinking up a theory – to specific hypotheses – into
observations.
n test the hypotheses with specific data- a confirmation (or
not) of our original theories.

o Theories are speculative answers to perceived problems, and


are tested by observation and experiment.
n While it is possible to confirm a theory through observations, it
can be falsified and totally rejected - inconsistent observations.

Compiled By Solomon Kebede, 36


MSc, PhD Candidate, 2019
Ways of Reasoning
o A simple example of deductive logic:
n All live mammals breathe (General statement – first
premise);
n This cow is a live mammal (Inference – second premise);
n Therefore, this cow breathes (Conclusion).

o Economic theory rests largely on deductive logic.


n Examples: utility maximization in the consumer behavior; profit
maximization in the producer behavior.

o when one theory is rejected, another is proposed and tested,


and thus the fittest theory survives.

Compiled By Solomon Kebede, 37


MSc, PhD Candidate, 2019
Ways of Reasoning

Deductive ReasoningCompiled By Solomon Kebede,


MSc, PhD Candidate, 2019
38
Ways of Reasoning
o Inductive Reasoning (the empiricist’s approach)
n Inductive reasoning starts from specific observations and
then develops a general conclusion from them.
o It is the earliest and, even now, the commonest popular
form of scientific activity – We use it daily.

o In induction one starts from observed data and develops a


generalization which explains the relationships between the
objects observed - this a "bottom up" approach.

n The inductive logic of reasoning is followed in most


empirical economic research.

Compiled By Solomon Kebede, 39


MSc, PhD Candidate, 2019
Ways of Reasoning
o Three conditions must be satisfied for generalizations to be
considered legitimate by inductivists:
1. There must be a large number of observation statements.
2. The observations must be repeated under a large range of
circumstances and conditions- how large ???
3. No observation statement must contradict the derived
generalization

o A simple example will demonstrate the line of reasoning:


n All the giraffes that I have seen (Repeated observations)have
very long necks.
n Therefore, all giraffes have long necks (Conclusion) .
Compiled By Solomon Kebede, 40
MSc, PhD Candidate, 2019
Ways of Reasoning
o The scientific revolution in the 17th century was based on this
approach, led by such scientists as Galileo and Newton
n The fall of an apple on Newton’s head from the tree led to
the theory of gravity

o According to the philosopher Francis Bacon in order to


understand nature, one should consult nature, and not the
writings of ancient philosophers such as Aristotle, or the Bible.

o Mendel’s discovery of genetics and Darwin’s theory of


evolution - developed through inductive reasoning.

Compiled By Solomon Kebede, 41


MSc, PhD Candidate, 2019
Ways of Reasoning

Inductive Reasoning42
Compiled By Solomon Kebede,
MSc, PhD Candidate, 2019
Hypothetico-deductive reasoning or scientific method
o But, choosing between one position and the other is somewhat
unrealistic in practice.

o The hypothetico-deductive method combines inductive and


deductive reasoning.
n inductively developing hypotheses (testable theories) from
observations, charting their implications by deduction, and testing
them to refine or reject them in light of the results.

o This combination of experience with deductive and inductive


reasoning is the foundation of modern scientific research, and
is commonly referred to as scientific method.

Compiled By Solomon Kebede, 43


MSc, PhD Candidate, 2019
In other words, most social scientific research involves both inductive
and deductive reasoning as the scientist shifts endlessly back and
forth between theory and empirical observations – research cycle.

Compiled By Solomon Kebede, 44


MSc, PhD Candidate, 2019
Research methodology and Research methods
o Research methodology and Research methods are not
necessarily the same.

o Research methodology is the science and philosophy behind all


research.
n It may be understood as a science of studying how
research is done scientifically.
o The scope of research methodology is wider than that of
research methods.
o It includes:
n The theoretical perspective
n The tools and techniques of data collection
n The methods of data analysis.
Compiled By Solomon Kebede, 45
MSc, PhD Candidate, 2019
Research methodology and Research methods
o Research methodology provides the principles for
organizing, planning, designing, and conducting research.

o Research methodology has many dimensions and research


methods do constitute a part of the research methodology.

Compiled By Solomon Kebede, 46


MSc, PhD Candidate, 2019
Research methodology and Research methods
o Research methods may be understood as all those
techniques that are used for conduction of research.
o Research methods comprises of two things
n Research techniques – methods or techniques of data
collection– survey methods, or other methods
n Research tools- instruments for collection of the data –
questionnaire schedule, historical record, etc.
o Research methods provide the specific details of how one
accomplishes a research task.

Compiled By Solomon Kebede, 47


MSc, PhD Candidate, 2019
The Values of Research?
q The role of research in several fields of economics, whether
related to business or to the economy as a whole, has greatly
increased in modern times.
n Research provides the basis for nearly all government
policies in our economic system.
n Research has its special significance in solving various
operational and planning problems of business and industry.

q Solving societal problem: Research reduces uncertainty for policy


makers, planners, business managers, and other stakeholders by
providing information that improves the decision-making process

Compiled By Solomon Kebede, 48


MSc, PhD Candidate, 2019
Main Attributes of Economic Research
o Economic research requires a specific plan or procedure
n It has a beginning, a middle and an end.
n We need to know whether the research project is feasible in
terms of time and money – the research proposal

o Economic research accepts certain critical assumptions


n If you do not make assumptions, you cannot make logical
conclusions

Compiled By Solomon Kebede, 49


MSc, PhD Candidate, 2019
Main Attributes of Economic Research
o Economic research is, by its nature, cyclical
n Every research project brings answers but also new
questions
n Those questions, in turn, bring new research projects

o Economic research begins with a question that the researcher is


trying to answer
n There should be a compelling question
n It is not just filling gaps in knowledge

o So, economic research has clearly articulated goals and is not


open-ended.

Compiled By Solomon Kebede, 50


MSc, PhD Candidate, 2019
Evaluation of Economic Research
o How can you tell whether a piece of research is any good?
n When doing background reading, we should be able to
assess the quality of the research projects we read about.

o Good research is systematic, logical, empirical, and is


replicable.
o In general an evaluation of an economic research involves.
n Validity
n Reliability
n Replicability
n Generalizability
Compiled By Solomon Kebede, 51
MSc, PhD Candidate, 2019
Evaluation of Economic Research
o Validity of research is about the degree to which the research
findings are true.
o Different types of validity could be considered:
n Measurement validity: The degree to which measures (e.g.
questions on a questionnaire) successfully indicate concepts.
n Internal validity: The extent to which causal statements are
supported by the study.
n External validity: The extent to which findings can be
generalized to populations or to other settings.
n Ecological validity: The extent to which the findings are
applicable to people’s everyday, natural social settings
(Bryman, 2004).
Compiled By Solomon Kebede, 52
MSc, PhD Candidate, 2019
Evaluation of Economic Research
o Reliability is about the degree to which the results of the
research are reputable.
o Three prominent factors that are involved (Bryman (2004):
n Stability: The degree to which a measure is stable over
time.
n Internal reliability: The degree to which the indicators
that make up the scale or index are consistent.
n Inter-observer consistency: The degree to which there is
consistency in the decisions of several ‘observers’ in their
recording of observations or translation of data into
categories.
Compiled By Solomon Kebede, 53
MSc, PhD Candidate, 2019
Evaluation of Economic Research
o Replicability is about whether the research can be repeated
and whether similar results are obtained.

n This is a check on the objectivity and lack of bias of the


research findings.

n It requires a detailed account of the concepts used in the


research, the measurements applied and methods
employed.

Compiled By Solomon Kebede, 54


MSc, PhD Candidate, 2019
Evaluation of Economic Research
o Generalizability refers to the results of the research and how
far they are applicable to locations and situations beyond the
scope of the study.

n There is little point in doing research if it cannot be


applied in a wider context.

n On the other hand, especially in qualitative research, there


may well be limits to the generalizability of the findings,
and these should be pointed out.

Compiled By Solomon Kebede, 55


MSc, PhD Candidate, 2019
Classifications of Economic Research
o Research comes in many shapes and sizes.

o There are different criteria on which to base the classification


of research activities.

n The classification may be in terms of:


o The research (data collection)techniques used in it,
o the time dimension,
o research environment
o data used.
o The intended uses of research;

Compiled By Solomon Kebede, 56


MSc, PhD Candidate, 2019
Classifications of Economic Research
Descriptive versus Analytical Research
o The purpose of descriptive research is description of the state
of affairs as it exists at present.

o The main characteristics of this method are that the


researcher has no control over the variables.

o The researcher can only report what has happened or what is


happening.
o Example; the frequency of shopping by people, the
preference of people, the number of employed workers
in a factory, etc.
Compiled By Solomon Kebede, 57
MSc, PhD Candidate, 2019
Classifications of Economic Research
o Analytical studies go beyond simple description in their
attempt to model empirically the social phenomena under
investigation.

n In analytical research the researcher has to use facts or


information and analyze these to make a critical evaluation
of the material.

o It asks “why” or “how” and tries to find the answer to a


problem.

Compiled By Solomon Kebede, 58


MSc, PhD Candidate, 2019
Classifications of Economic Research
Applied versus Basic Research:
o Research may be undertaken either to understand the
fundamental nature of a social reality (basic research) or to
apply knowledge to address specific practical issues (applied
research).

o Applied research aims at finding solution for an


immediate pressing problem facing a society or an
industrial unit or business organization.

o Theory is less central in applied research.

Compiled By Solomon Kebede, 59


MSc, PhD Candidate, 2019
Classifications of Economic Research
o Basic research is mainly concerned with generalizations and
with the formulation of a theory.

o It is primarily concerned with the understanding of the


fundamental nature of social reality.

o It is the source of most scientific ideas and ways of


thinking about the world.

o It is mostly exploratory in nature.

Compiled By Solomon Kebede, 60


MSc, PhD Candidate, 2019
Classifications of Economic Research
Quantitative vs Qualitative Research
q Quantitative research is based on the measurement of
quantitative figure or quantity or amount.
q applicable to phenomenon that can be expressed in
terms of quantity.
q Most often we collect data and see whether a hypothesis is
consistent with the data

Compiled By Solomon Kebede, 61


MSc, PhD Candidate, 2019
Classifications of Economic Research
o Qualitative Research
n Qualitative research is concerned with subjective
assessment of attitudes, opinions, and behavior.
n It generates results, which are not subjected to rigorous
quantitative analysis.
n Generally group interviews, projective techniques and in
depth interviews are used.
o Qualitative research is particularly important in the behavioral
sciences.

Compiled By Solomon Kebede, 62


MSc, PhD Candidate, 2019
Classifications of Economic Research h
o BUT, social research is often pluralistic:
q researchers often combine quantitative and qualitative
research methods within the same study.

q Mixed-method research strategies are particularly effective


in policy-oriented research and the contribution that
qualitative research can make to policy evaluation is
increasingly being recognized.

Compiled By Solomon Kebede, 63


MSc, PhD Candidate, 2019
Classifications of Economic Research
Other types of researches
n Research
o Can be one time or longitudinal research,
o can be field setting or laboratory based or simulation
research,
o Can be inferential, experimental or simulation studies,
o can also be clinical or diagnostic research, etc.

Compiled By Solomon Kebede, 64


MSc, PhD Candidate, 2019
Types of Research Methods

Compiled By Solomon Kebede, 65


MSc, PhD Candidate, 2019
Types of Research Methods

Compiled By Solomon Kebede, 66


MSc, PhD Candidate, 2019
Time Dimension in Economic Research
q Time is an important element of any research design.

q Quantitative research may be divided into two groups in terms


of the time dimension
q A single point in time (cross sectional)
q Multiple points research (longitudinal research)

q Cross –sectional research takes a snapshot approach to social


world.
q This is the simplest and less costly research approach.
q Limitation – cannot capture social processes or changes.

Compiled By Solomon Kebede, 67


MSc, PhD Candidate, 2019
Time Dimension in Economic Research
o Longitudinal research examines features of people or other
units more than one time.
n It is usually more complex and costly than cross sectional
research but is also more powerful especially with respect
to social changes.

o Time series research – this is longitudinal study on a group of


people or other units across multiple periods (e.g. time series
data on exports of coffee).

Compiled By Solomon Kebede, 68


MSc, PhD Candidate, 2019
Time Dimension in Economic Research
o The panel study – consists of a sample of people, often
randomly selected, who are questioned on two or more
occasions.

o the researcher observes exactly the same people group or


organization across time periods, each time using the snapshot
approach.

o In panel study the focus is on individuals or households.


n Example: interviewing the same people in 1991, 1993,
1995, etc, and observing the change is an example of panel
data set.
Compiled By Solomon Kebede, 69
MSc, PhD Candidate, 2019
Time Dimension in Economic Research
o A cohort Analysis – is similar to the panel study, but rather
than observing the exact same people, a category of people
who share similar life experience in a specified period is
studied.

n Hence the focus is on group of individuals not on specific


individuals or households.
o Example: students from a particular year of
matriculation or people on strike at a certain time.

Compiled By Solomon Kebede, 70


MSc, PhD Candidate, 2019
Research Ethics
o The value of research depends as much on its ethical veracity as
on the novelty of its discoveries - Ethics are the rules of conduct
in research.

o How can we believe the results of a research project if we doubt


the honesty of the researchers and the integrity of the research
methods used?

o Researchers, subjects, funding bodies and society may have


conflicting incentives.
n As a result of its focus on people, ethical issues are centrally
important in research - Ethical responsibility is required to
do the work honestly and with integrity.
Compiled By Solomon Kebede, 71
MSc, PhD Candidate, 2019
Research Ethics
o Researchers must ensure the accuracy, appropriateness, of
data and research results.
o Hence, when working with human participants, it is often
necessary to obtain some kind of ethical approval –
n from ethical board.

o To treat participants in your research with respect and due


consideration is a basic tenet of civilized behaviour.

Compiled By Solomon Kebede, 72


MSc, PhD Candidate, 2019
Research Ethics
o Honesty in your work
n It is easy to cheat, but the penalties resulting from
discovery are stiff and humiliating.

n Honesty is essential to engender a level of trust and


credibility that promotes debate and the development of
knowledge.

n Do not fabricate, falsify, or misrepresent data.


n Do not deceive colleagues, research sponsors, or the
public.

Compiled By Solomon Kebede, 73


MSc, PhD Candidate, 2019
Research Ethics
o Researchers-researchers relation (Intellectual ownership)
n Unless otherwise stated, what you write and the ideas will be
regarded as your own work.
o Directly copying someone else’s work into your report, thesis etc.
and letting it be assumed that it is your own – plagiarism.
o Plagiarism is the worst offence in academics

q Plagiarism is the unauthorized use of someone else's thoughts or


wording either by
q incorrect documentation, failing to cite your sources
altogether, or
q simply by relying too heavily on external resources.
Compiled By Solomon Kebede, 74
MSc, PhD Candidate, 2019
Research Ethics
q Whether intentional or unintentionally some or all of another
author's ideas become represented as your own.
q Plagiarizing undermines your academic integrity and
betrays your own responsibilities.

o It also includes informal published material such as the


"buying" of a paper from another student.

o Because it is intellectual theft, plagiarism is considered as an


academic crime with punishment
o Can easily be detected through peer reviews
o Note: Plagiarism includes copied data/information,
falsification, fabrication or omission of significant results.
Compiled By Solomon Kebede, 75
MSc, PhD Candidate, 2019
Research Ethics
o Citation and acknowledgement
n In no field of research can one rely entirely on own ideas,
concepts and theories alone.

n Standard practices have been developed to permit the


originators of the work and ideas to be acknowledged within
your own text
o This is called citation.

o You should acknowledge the assistance of others and any


collaboration with others.
Compiled By Solomon Kebede, 76
MSc, PhD Candidate, 2019
Research Ethics
o Responsibility and accountability of the researcher
n You do have responsibilities to fellow researchers,
respondents, the public and the academic community.

n Accurate descriptions are required of


o what you have done,
o how you have done it,
o the information you obtained,
o the techniques you used,
o the analysis you carried out, and
o the results of experiments , etc.
Compiled By Solomon Kebede, 77
MSc, PhD Candidate, 2019
Research Ethics
o Falsifying results—to make them fit your conclusion.
n Trimming—removing data that does not fit in with your
analysis.
n Biased or inappropriate analysis.
q Fabrication and falsification of research results are serious
forms of misconduct.
n Researcher should avoid either a false statement or an
omission that distorts the truth - Objectivity.
q In order to preserve accurate documentation, every researcher
has an obligation to maintain a clear and complete record of
data acquired.

Compiled By Solomon Kebede, 78


MSc, PhD Candidate, 2019
Research Ethics

o Fraud must be avoided in research and this can come in


several forms:
n Being selective in sampling.
n Not reporting survey response/participation rates.
n Deliberately biasing the data collection instruments—for
example, asking leading questions in surveys.
n Making up data—because you can’t be bothered doing
the data collection.

Compiled By Solomon Kebede, 79


MSc, PhD Candidate, 2019
Research Ethics
o Data and Interpretations
n There is often a temptation to be too selective in the data
used and in presenting the results of the analysis carried
out.
o Irresponsible policy advise
o Waste of research fund

o Silently rejecting or ignoring evidence which happens to be


contrary to one’s beliefs constitutes a breach of integrity.
n What could be of vital importance in developing a theory could
be lost.
o Scientific objectivity should be maintained (or attained as
closely as is practical).
Compiled By Solomon Kebede, 80
MSc, PhD Candidate, 2019
Research Ethics
o It is considered a breach of research integrity to fail to report
data that contradict or merely fail to support the conclusions,
including the purposeful withholding of information.
n Negative (unexpected) results must also be reported.

q Records should include sufficient detail to permit examination


for the purpose of
ü replicating the research,
ü responding to questions that may result from unintentional
error or misinterpretation,
ü establishing authenticity of the records, and
ü confirming the validity of the conclusions.

Compiled By Solomon Kebede, 81


MSc, PhD Candidate, 2019
Research Ethics
o Funding bodies and researchers

o Pressure and sponsorship from sources which might


influence the impartiality of the research outcomes should be
avoided.
n Fund allocations directing research
o ‘money buying research results’?

Compiled By Solomon Kebede, 82


MSc, PhD Candidate, 2019
Research Ethics
o Researchers had to deal with the ethical issue of a person's
right to service.

n Good research practice often requires the use of a no-


treatment control group.

n But, persons assigned to the no-treatment control may feel


their rights to equal access to services are being curtailed.

n Even when clear ethical standards and principles exist,


there will be times when the need to do accurate research
runs up against the rights of potential participants.

Compiled By Solomon Kebede, 83


MSc, PhD Candidate, 2019
Research Ethics
o So, there needs to be a procedure that assures that
researchers will consider relevant ethical issues in
formulating research plans.

o To address such needs institutions and organizations


formulate an Institutional Review Board (IRB)

o A panel of persons will review proposals with respect to


ethical implications and decide whether additional actions
need to be taken to assure the safety and rights of participants.

Compiled By Solomon Kebede, 84


MSc, PhD Candidate, 2019
Research ethics
o In addition, by reviewing proposals for research, IRBs help to
protect both
n the organization, and
n the researcher against potential legal implications of
neglecting to address important ethical issues of
participants.

Compiled By Solomon Kebede, 85


MSc, PhD Candidate, 2019
Research ethics
Researcher and research subjects
Some ethical principles governing data collection include: harm to
respondents, informed consent, respect for privacy and
safeguarding the confidentiality of data.

o So in collecting data researchers need to be guided by


principles of respect for persons and obtaining informed
consent.

o There are two standards that are applied in order to help


protect the privacy of research participants.

Compiled By Solomon Kebede, 86


MSc, PhD Candidate, 2019
Research ethics
o Confidentiality – participants are assured that identifying
information will not be made available to anyone who is not
directly involved in the study.

o Anonymity - means that the participant will remain


anonymous throughout the study -- even to the researchers
themselves.

o Clearly, the anonymity standard is a stronger guarantee of


privacy, but it is sometimes difficult to accomplish, especially
in situations where participants have to be measured at
multiple time points (e.g., a pre-post study).

Compiled By Solomon Kebede, 87


MSc, PhD Candidate, 2019
Research ethics
1. Harm to participants
o Physical harm, humiliation, embarrassment, loss of trust, harm
to participant’s development, loss of self-esteem, stress, etc.

o Ethical standards require that researchers should not put


participants in a situation where they might be at risk of being
harmed as a result of their participation.

o Harm can be defined as both physical and psychological.


n Minimize harms and risks and maximize benefits; respect
human dignity, privacy, and autonomy; take special
precautions with vulnerable populations.
Compiled By Solomon Kebede, 88
MSc, PhD Candidate, 2019
Research ethics
2. Lack of informed consent
o The principle of voluntary participation requires that people
should not be coerced into participating in research.

o Closely related to the notion of voluntary participation is the


requirement of informed consent.
n Essentially, this means that prospective research
participants must be fully informed about the procedures
and risks involved in research and must give their consent
to participate.

Compiled By Solomon Kebede, 89


MSc, PhD Candidate, 2019
Research ethics
o Informed consent implies that persons who are invited to
participate in the research activities should be free to choose to
take part or refuse.

o Participants should be given as much information as needed to


make decision on participation

o They are free to decide after having been given the fullest
information concerning the nature and purpose of the research.

Compiled By Solomon Kebede, 90


MSc, PhD Candidate, 2019
Research ethics
3. Invasion of privacy
o Privacy of participants should be protected
o Participants may refuse to address specific questions even
though they have agreed to participate
o Covert/secret research usually violates privacy

Protecting research subjects


o Legal protection
o Professional associations and universities (ethics committees)

Compiled By Solomon Kebede, 91


MSc, PhD Candidate, 2019
Research ethics
5. Deception
o When researchers present their research as something other
than what it is – deception.

o Researchers usually want to limit participants understanding so


that they act naturally
n Debriefing: to get subjects interpretation of the research but
handle debriefing with sensitivity.

o Thus, collection of data illegally, under false pretences, from


minors, etc is unethical.
o Getting access and consent to do research is therefore, essential.

Compiled By Solomon Kebede, 92


MSc, PhD Candidate, 2019
Research ethics
o In summary: Some important shared values for the
responsible conduct of research include:
ü HONESTY — conveying information truthfully and
honoring commitments,
ü ACCURACY — reporting findings precisely and taking
care to avoid errors,
ü EFFICIENCY — using resources wisely and avoiding
waste, and
ü OBJECTIVITY — letting the facts speak for themselves
and avoiding improper bias.
ü Social Responsibility-strive to promote social good
and prevent or mitigate social harms through research.

Compiled By Solomon Kebede, 93


MSc, PhD Candidate, 2019
• Identifying Your Area/topic

Proposal
• Developing the research problem with
focusing
on main research questions
• Working on Literature
• Working on conceptual framework
• Working on Your Methodology
• Managing Your Field Work

Proposal
Post
• Working on the Data
• Writing Your Thesis
• Presenting Your Results
94
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019
o Formulating and clarifying the research topic is
the starting point of your research project
o This is probably the most difficult, and yet
the most important, part of your research
project.
o Mainly driven by curiosity/interest
n Your past experiences
n Future goals and aspirations
n Existing problems/challenges
o Guides you in searching for researchable
problems
Compiled By Solomon Kebede, MSc, 96
PhD Candidate, 2019
A) Generated by the researcher:
q Own professional experience is the most important source
of a research problem.
q Choose a topic in which if possible, already have some
academic knowledge.
q For instance, some researchers are directly engaged in
program implementation and come up with a topic based
on what they see is happening.
q Attending conferences, seminars, and listening to
professional speakers, etc. are all helpful in identifying
research problems.
B) Examining the literature :
q Examining the theoretical or empirical literature in your
specific field.
q Published articles are excellent sources of ideas. They help to…
q Explore findings discovered in previous research.
q Identify suggestions an author gives for further research at
the end of an article.
q Extend an existing explanation or theory to a new topic or
setting.
q Challenge findings or attempt to refute a relationship.
C) Provided by a client :
Ø Requests For Proposals (RFPs) are published by government
agencies, NGOs and some companies.
Ø They describes the problem that needs to be addressed,
q the contexts in which it operates,
q the approach they would like you to take, and
q the amount they would be willing to pay for such research
- they are virtually handing the researcher an idea.
D) Technological and Social Changes:

E) Scanning the Media

F) Discussion and Brainstorming

Compiled By Solomon Kebede, MSc, PhD Candidate, 100


2019
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 101
o Without interest to sustain motivation, it is very
easy for a researcher to get tired or bored and give
up before the research is completed.

o Subject which is overdone, should be avoided


by the average researcher since it will be difficult to
throw any new light in such cases for the average
researcher.

Compiled By Solomon Kebede, 102


MSc, PhD Candidate, 2019
Con't...

Compiled By Solomon Kebede, 103


MSc, PhD Candidate, 2019
o After a topic has been selected the next step is to
define the research problem more precisely so
that it becomes a specific research question.

o You must know exactly what the problem is


before you begin work on it.
Ø Problem definition or problem statement is
a clear and precise statement of the
question or issue to be investigated.
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 104
Con't...

o It is a clear statement about an area of concern, a


condition to be improved upon, a difficulty to be
eliminated, or a troubling question that exists in
scholarly literature, in theory, or within existing
practice.
o Defining a problem involves the task of laying
down the boundaries within which a researcher
shall study the problem.

Compiled By Solomon Kebede, 105


MSc, PhD Candidate, 2019
o Defining and formulating a research problem is
more important than its solution.
ØA problem clearly defined is a problem half
solved.

o Defining the problem clearly will help to find


answers to questions like:
ØWhat data are to be collected?
ØWhat characteristics of data are relevant and
need to be studied
ØWhat relations are to be explored
ØWhat techniques are to be used for the purpose.
Compiled By Solomon Kebede, 106
MSc, PhD Candidate, 2019
Con't...
o Therefore, in the definition of the problem you
are required:
ØTo describe the background of the study,
ØIts theoretical basis and the underlying
assumption,
ØThe specific and workable questions, and
ØA n e x p l a n a t i o n o f w h y i t i s i m p o r t a n t o r
significant, and what advantage (i.e. to society as a
whole or to other researchers) the research may
deliver.
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 107
Con't...
o The problem statement should make a convincing
argument that there is insufficient knowledge
available to explain the problem or there is a need
to test what is known and taken as a fact.
o The problem definition should also provide a brief
overview of the literature and research done in the
field related to the problem and of the gaps that
the proposed research is intended to fill.

Compiled By Solomon Kebede, 108


MSc, PhD Candidate, 2019
Con't...
o The problem statement may focus on:

A research gap is an area


where no or little research has been
carried out.
The research problem is
defined by asking a question to which the
answer is unknown, and which you will
explore in your research.
Compiled By Solomon Kebede, MSc, PhD Candidate, 109
2019
ü Building on work already done, but taking it further
(by using a new sample, extending the area studied,
taking more factors into consideration, etc.).
or

here a conflicting claim, theory


or method is put forward.

Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 110


o Defining the problem involves the undertaking of
the following steps generally one after the other:
ü Statement of the problem in a general way;
ü Understanding the nature of the problem;
ü Surveying the available literature
ü Developing the ideas through discussions;
ü Rephrasing the research problem into a working
proposition.
Ø This is the step of rephrasing the problem into
operational or analytical terms, i.e. to put the
problem in a specific terms as possible.

Ø Objectives are generally taken as evidence of the


researcher’s clear sense of purpose and direction.
o The general objective provides a short statement
of the specific goals pursued by the research.
ØUsing the general focus of your research
question you will develop a set of research
objectives.
o The specific objectives are operational and
indicate the type of knowledge to be produced,
audiences to be reached, etc.

Compiled By Solomon Kebede, 113


MSc, PhD Candidate, 2019
What precisely do you hope to achieve from
undertaking the research?
W hat i n di c ato rs wi l l y o u u s e t o
determine whether you have achieved your objectives?
Are the targets you have set for yourself
achievable given all the possible constraints...inputs?
will you have the energy to complete the
research on the ground?
ØTi Will you have time to accomplish all your
objectives in the time frame you have set?
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 114
Research Question

o A research question is an aspect of your


research area that you want to explore
o It points to:
n the theory you need to explore;
n the literature you need to review;
n the data you need to gather; and in
particular,
n the methods you need to call on.

Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 115


– indicates the questions you are interested in
(whether you are interested in what, where, how,
when, and why)
– indicates whether you foresee a relationship
between concepts you are exploring (whether you
are looking for impacts, causes, comparisons, etc.)
– indicates the theory and literature you will need to
explore and review,
– indicates the data you will need to gather, and the
methods you will use to collect and analyze these
data

Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 116


o You may want to consider the
following to find your focus issue:
n Past experience
n Theory
n Observation
n Contemporary timely issues
n Gaps in literature/research

Compiled By Solomon Kebede, MSc, 117


PhD Candidate, 2019
Brainstorming
Life /work
Re

Experience
ad
ing

Research
Questions

Research Question Development


Compiled By Solomon Kebede, MSc, 118
PhD Candidate, 2019
•From Topics to Researchable Questions

Identifying Your
Research Area

Generating Questions
Finding an Angle

Researchable
Questions Narrowing
and clarifying

Compiled By Solomon Kebede, MSc, 119


PhD Candidate, 2019
Examples
How does the National Bank of Ethiopia fix
foreign exchange rate?
o Why did inflation occur during economic
boom in Ethiopia?
o What factors determine the level of food
security among farm households?
o Does inflation hamper FDI?

Compiled By Solomon Kebede, MSc, 120


PhD Candidate, 2019
Hypothesis

Do I Need Hypothesis?
 There are two schools of thoughts in this
regard
 Positivists believe that the hypothesis is a
cornerstone of scientific method and it is a
necessary component of the research
process
 Post-positivists often view the hypothesis
as a reductionist devise to constrain social
research and take all life force from it

Compiled By Solomon Kebede, MSc, 121


PhD Candidate, 2019
From Research question to hypothesis
o What factors determine the level of food security
among farm households? (Research question)
o Landholding affects the level of food security
among farm households. (hypothesis)
o Does inflation hamper FDI? (Research question)
o Inflation hampers FDI. (hypothesis)
o But not all research questions can be convertible
to a testable hypothesis.
o How does the National Bank fix foreign exchange
rate?
Compiled By Solomon Kebede, MSc, 122
PhD Candidate, 2019
o A hypothesis may not be appropriate if:
- you do not have an intuitive feeling or
educated guess about a particular situation
- you do not have a set of defined variables –
Your research may be explorative
- your question centres on phenomenological
description
- your question centres on an ethnographic
study of a cultural group

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 123
o Research may be done alone, but is never done
in isolation
o The production of new knowledge is
fundamentally dependent on past knowledge
o It is virtually impossible for researchers to add
to a body of literature without being conversant
with it
o Working with literature is an important part of
the research process
o It inspires, informs, educates, and enlightens

Compiled By Solomon Kebede, MSc, 124


PhD Candidate, 2019
o Working on literature is often seen as
a tedious task.
o Working with literature involves
n Finding it
n Managing it
n Using it, and
n Reviewing it

Compiled By Solomon Kebede, MSc, 125


PhD Candidate, 2019
What type of literature should I use?
o Working with literature starts with finding
appropriate ones
o Appropriate literature categories include:
n Subject specific books (introductory and
advanced texts, research reports, seminar
works)
n Journal articles
n Grey literature: refers to both published and
unpublished materials that do not have ISBN or
ISSN (e.g. conference papers, unpublished
research thesis, newspaper articles, brochures)

Compiled By Solomon Kebede, MSc, 126


PhD Candidate, 2019
n Official publications, statistics and
archives: can be a valuable source of
background and contextual information; can
also be used as a source of data
n Writing aids: include bibliographic reference
works, dictionaries, encyclopedia, almanacs, etc

Compiled By Solomon Kebede, MSc, 127


PhD Candidate, 2019
How do I search for literature?
Don’t do it alone!!!
Other
Researche
rs/practitio Your
Librarians ners Supervisor

s e
U ur
yo rch
e a ls
s ki l
s
Compiled By Solomon Kebede, MSc, 128
PhD Candidate, 2019
How do I manage the literature I
accessed?
 You may accumulate large number of
literature (downloading, copying, buying, or
borrowing)
 But you can’t read all of them at once
 Even there are literature on which you
shouldn’t kill your time and energy
 You should select and prioritize your literature
 How? Be systematic in this: If you are reading
a journal article first look at its abstract or
summary

Compiled By Solomon Kebede, MSc, 129


PhD Candidate, 2019
o In a book, look at the table of contents, the
back cover summary, and the introduction;
also look at the conclusion at the end of each
chapter
v Within a few minutes you can assess whether
the material is relevant or not to your work
v You may rank the work as 1 (highly relevant),
2 (somewhat relevant), 3 (minimally relevant),
or 4 (not relevant)

Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 130


A
ab ac
ABC

B bc
C
Sorting Literature

Compiled By Solomon Kebede, MSc, 131


PhD Candidate, 2019
o Keep track your references (highly
important!!)
o Record the ideas with full reference
information (including page number!)
o Once you lose a source it will be tedious to
get it again (you may not get it at all!!)
o Helpful to annotate your references (may
contain a brief summary, a critical comment,
notes on the relevance)
o Keep also good referencing
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 132
How should I use the literature?
o Literature are used at every stage of the
research process
o At every stage of the research process your
demand is different; you seek literature to
answer different questions in your mind
o Depending on your question the type of
literature you require varies in quality and
the way of your review also should vary

Compiled By Solomon Kebede, MSc, 133


PhD Candidate, 2019
 .
What do I know Background reading for
about this topic? Exploring your own learning using a
How can I find Your Topic broad array of materials
more? mainly popular texts
Need for more in-depth
What research
Developing engagement in research
questions should I Research literature (e.g. journal
answer? How shall Question articles)
I fix them?
Background/contextual
How do I develop readings that put the
a convincing
Articulating a
significance of the
rationale for my Rationale research in a broader
study? societal/scientific context

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 134
What
 . Informing Writings of contemporary
theory/theories your study and/or classic theorists
should I use in my
study? with theory
A review of past studies
can inform your choice of
How should I Designing methods. Background
design and apply Method reading of specific methods
suitable methods? is also necessary

What research Writing Need for thorough and


has already been Literature critical review of the
conducted in this literature (generally
area?
Review journal based)

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 135
A few points on the formal “Literature
Review”
o Is included as a section or chapter in your work
o Generally relies on articles published on well
established research journals
o Should be relevant, critical, and comprehensive
o Conducting and writing a good literature review,
however, is no easy
o You need to negotiate multiple purposes;
decide on appropriate content; work towards a
logical structure; make convincing arguments

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 136
o Reviewing literature is not an end but is a
means towards an end; it is purpose-based
o Within a ‘literature review’ the literature is
reviewed or explored so that researchers
can
n Inform readers of developments in the field
n Establish their own credibility(the contribution of
the researcher)
n Argue the need for and relevance of their study
& indicate research gap
o A reasonably adequate coverage is required
Compiled By Solomon Kebede, MSc, 137
PhD Candidate, 2019
o Two practical problems are observed:
n The struggle to find literature because only few
literature exist
n The struggle to limit literature because of too
many literature
o In case of too few literature
n Check for your research question
n Ask for assistance of others for searching
n If couldn’t find any, try to review related
literature
Compiled By Solomon Kebede, MSc, 138
PhD Candidate, 2019
Con't...
o If you are confused with too many literature
n Try to make your research question more
focused
n Start with more thorough review of highly
relevant literature— “ABC” in previous example
A note on the writing process
o A literature review is beyond a “he said -
she said” report; It is rather an
argumentative piece of writing

Compiled By Solomon Kebede, MSc, 139


PhD Candidate, 2019
Con't...
o In order to write a good literature review
n Read a few good, relevant reviews
n Write critical annotations(explanatory note/+ing
critical comment) as you go
n Develop a structure (work on a potential outline
for your literature early)
n Write purposefully( don’t write a single sentence
without a purpose!!)
o Literature review is an iterative and ongoing
process (be prepared to produce several
drafts)
o Get plenty of feedbacks
140
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019
o This can be part of literature review,
methodology section, or stand alone
o This forms a conceptual and theoretical basis
o Introduces key concepts
o Provides relationships among key variables in
broader sense
o Can be adopted from others but should be
adapted to your context
o Commonly presented in graphical forms
o Not all research papers put their CFs explicitly

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 141
o Some confusing terms: methodology, methods,
tools, methodological design
o Methodology: The framework associated with a
particular set of paradigmatic assumptions that
you will use to conduct your research i.e.
scientific method, ethnography, action research
o Methods: The technique you use to collect and
analyze data, i.e. interviewing, surveying,
participant observation
o Tools/instruments: The devices you use to
collect data, i.e. questionnaire, checklist,

Compiled By Solomon Kebede, MSc, 142


PhD Candidate, 2019
o Methodological design: The plan for
conducting your study that includes all of
the above
o Methodological design sketches the path
that you take to answer your research
question

Compiled By Solomon Kebede, MSc, 143


PhD Candidate, 2019
Con't...

Q A

Q A
 You may develop your own method, rely on
existing methods, or combine the two
 Both have their own pros and cons

Compiled By Solomon Kebede, MSc, 144


PhD Candidate, 2019
Con't...
Three prerequisites for methodological design.

Address the
research
question

Your
methodologic
Be within
al design Be
your
capacity practical
and & doable
interest
145
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019
Con't...
q Developing your methods and tools is about
being able to assess your answers in
relation to the three prerequisites above
q It is also about being able to answer the
basic questions of who, where, when, how,
and what
q Who
ü Who do you want to speak about?
ü Who do you plan to speak to or observe?
q Where
ü What is the physical domain of your sample?
ü Are settings relevant to the credibility of your
methods

Compiled By Solomon Kebede, MSc, 146


PhD Candidate, 2019
o How
n How will you collect your data?
n How will you conduct your methods?
o When
n How do your methods fit into your time
frame?
n Is timing relevant to the credibility of your
methods?
o What
n What will you look for/what will you ask?

Compiled By Solomon Kebede, MSc, 147


PhD Candidate, 2019
Research Design
o Once the objectives of a research project have been defined the
appropriate research design has to be identified.
n how these objectives can be met.

o The research design is the plan, structure and strategy of


investigation conceived so as to obtain answers to research
question and to control variance.

o A research design provides the framework for the collection and


analysis of data and subsequently indicates which research
methods are appropriate.
n It is a plan that directs the collection, estimation and analysis of data.

Compiled By Solomon Kebede, 148


MSc, PhD Candidate, 2019
Research Design
o A research design helps you to set out an overall framework
for your research and enables you to make decisions
regarding
n what (objectives),
n where (scope),
n when (time period),
n how much (sample size),
n by what means (methodology, techniques, etc.)
constitute the elements of a research design.

o you can combine two or more types of research designs in


the study of human behaviour.

Compiled By Solomon Kebede, 149


MSc, PhD Candidate, 2019
Research Design

o So, answering the following questions will help you determine a


specific design for the study.

n Does your research question call for a detailed examination of


one individual, or

n would you find a better answer by looking at the average


behaviour of a large group?

n Should you make a series of observations on the same


individual over a period of time, or should you compare the
behaviour of different individuals at the same time? etc.

Compiled By Solomon Kebede, 150


MSc, PhD Candidate, 2019
Types of research designs
o There are numerous types of research design that are
appropriate for the different types of research projects.
n However, some methods are more widely used than others.

o The choice of which design to apply depends on the nature of


the problems posed by the research.

n Each type of research design has a range of research


methods that are commonly used to collect and analyze the
type of data that is generated by the investigations.

Compiled By Solomon Kebede, 151


MSc, PhD Candidate, 2019
Con't...
o Here, we consider five major types of research designs:
n experimental, quasi experimental, cross-sectional,
longitudinal and case studies.

Experimental design: identifies effects by a random assignment of


a treatment between a control and a treatment group

n This design begins with the assumption that the material or


cases that are being studied can be manipulated by the
researcher so that changes or differences can be measured.

o the researcher deliberately does something to something,


in order to provoke and then measure a reaction.
Compiled By Solomon Kebede, MSc, PhD Candidate, 2019 152
The Experimental Design
o Experimental studies are done in carefully controlled and
structured environments and enable causal relationships to be
identified and analyzed.
n Changes are made to an independent variable and the effects
are observed on a dependent variable – i.e. cause and effect.
o Experiments usually require a hypothesis (prediction) to be
formulated first in order to determine what variables are to be
tested and how they can be controlled and measured.

o There are several classes of experiment – true, quasi, etc. which


are characterized by the amount of checking and control
involved in the methods.
Compiled By Solomon Kebede, 153
MSc, PhD Candidate, 2019
The Experimental design
nExperimental research differs from other approaches through its
greater control over the objects of its study.
oChemical experiments in a laboratory are examples.
n However, we cannot manipulate always all the data sources
used in social research.
nExample: we cannot deliberately deprive a child of
schooling to see how far they would develop without that
sort of education because:
§such action would be regarded as highly unethical
§It would also be impractical since we would not be able
to control all the possible influences on each child.

Compiled By Solomon Kebede, MSc, PhD Candidate, 154


2019
The Experimental design
n Nevertheless, there are cases in which the principal
characteristics of an experiment can be helpful in designing social
research that approximates to an experiment.

nAn experiment usually has two key elements:


1. Experimental group– this is the group of people or
materials that are manipulated or changed.

2. Control group– this is a group of people or materials that


are the same as the experimental group in every way except
the aspect of manipulation or change.

Compiled By Solomon Kebede, 155


MSc, PhD Candidate, 2019
The Experimental design
o Manipulation helps to isolate and control the influence of
confounding variables (i.e. variables correlated with both the
dependent and the independent variable) in a randomisation
process .

n Experiments allow identification and separation of


unwanted variables by controlling them.

o And, by controlling and manipulating variables systematically,


we can determine which variables influence the behavior that
we are studying.

Compiled By Solomon Kebede, MSc, PhD Candidate, 156


2019
The Experimental design
o The experimental and control groups need to be similar before
the experiment starts.

o Then, participants are assigned at random to experimental and


control groups in the hope that this will result in equal
assignment of people to both groups.

n So
n we will have two groups: experimental and control groups
(with and without ‘treatment’)
n changes in the experimental groups before and after
‘treatment’ compared with changes in the control group.

Compiled By Solomon Kebede, MSc, PhD Candidate, 157


2019
The Experimental design

Compiled By Solomon Kebede, MSc, PhD Candidate, 158


2019
The Experimental design
Method of Pre-test Intervention Post-test
allocation to (X)
groups

Random Measure on ‘Treatment’ Measure on


(experiment outcome outcome
al group) variable (Y) variable (Y)

Random Measure on No Measure on


(control outcome ‘treatment’ outcome
group) variable (Y) variable (Y)

Compiled By Solomon Kebede, 159


MSc, PhD Candidate, 2019
The Experimental design
o Randomized controlled trials (RCTs) are commonly held when
testing medical drugs or treatments.
n ‘Classic’ case: clinical trial for a new medicine
o Patients randomly assigned into experimental/treatment
and control groups
o Health status of participants measured
n New medicine given to experimental group and the control
group receives placebo
n Improvements in the two groups are measured after some
time
n If conditions of experimental group significantly improved
compared to control, the medicine is working

Compiled By Solomon Kebede, MSc, PhD Candidate, 160


2019
Randomised control trials – examples
o Nowadays, randomised control trials (RCTs) are being
extensively used in economics as well.

Example 1: E. Duflo, M. Kremer and J. Robinson, “How high are


rates of return to fertilizer? Evidence from field experiments
in Kenya”, 2008 American Economic Review: 98:2, 482–488

o aim: to vary fertilizer use experimentally on real-world farms


and measure the impact of potentially complementary inputs
use on output

Compiled By Solomon Kebede, MSc, PhD Candidate, 161


2019
Randomised control trials – examples
o Question: Hybrid seed and fertiliser increase yield between 40-
100% on experimental plots; but only 60% of farmers use
them

o Do fertiliser and hybrid seed increase yield in real small farms


where conditions are not ideal?

o Design:
n Farmers randomly selected from list of parents of students;
implemented with an NGO
n A series of six field trials made over three years starting
from July 2000

Compiled By Solomon Kebede, MSc, PhD Candidate, 162


2019
Randomised control trials – examples
o On each farm 3 adjacent 30 square meter plots selected
n One plot was randomly assigned to receive Calcium
Ammonium Nitrate (CAN) fertilizer to be applied as top
dressing
n On the second plot, hybrid seeds were used in place of
traditional varieties and Di-Ammonium Phosphate (DAP)
fertilizer was supplied for planting along with CAN for top
dressing; the official package recommended by the Ministry
of Agriculture
n The third plot was a comparison plot on which farmers
farmed as usual with traditional seed and without fertilizer

Compiled By Solomon Kebede, MSc, PhD Candidate, 163


2019
Randomised control trials – examples
o Additional input costs covered; field workers of the NGO
worked with farmers
n Then maize output measured
o Farmers were instructed to farm their plots just as they would
have otherwise; no additional labour on treatment plots
o Result: the full package recommended by the Ministry of
Agriculture found to be unprofitable on average for the
farmers in the sample.
o Example 2: E. Miguel & M. Kremer (2001), “Worms: Education
& health externalities in Kenya” - Econometrica, Vol 72, No. 1,
159-217.

Compiled By Solomon Kebede, MSc, PhD Candidate, 164


2019
Quasi-experimental
q uses a comparison group similar to the treatment group (but
there is no random assignment)

n It is a research design used in situations where two or more


groups of participants or data can be identified, and one is
used as the control and the other as the experimental group.

o when a group is tested for the influence of a variable and


compared with a control group which has not been subjected to
the variable.
n Example: participants and non participants in a program

Compiled By Solomon Kebede, MSc, PhD Candidate, 165


2019
Cross-sectional design
q In cross sectional designs information is gathered to look at
similarities or differences between them only at one point in
time;
qit provides a snapshot of ideas, opinions, information, etc.

quses variation between subjects; Less expensive and most


common type.
n Using cross sectional studies, we can work with data about
past and current experiences to identify possible cause-
and-effect associations.

o Cross-sectional studies are often associated with large-scale


surveys using questionnaires.
Compiled By Solomon Kebede, MSc, PhD Candidate, 166
2019
Cross-sectional
o A cross-sectional research design:
n includes more than one case;
n collects data at one particular time;
n includes within its research participants groups of people
or cases that can be compared,
o e.g. men and women, people of different ages or
ethnicities, or people from different cities.
o Cross-sectional studies often take the form of questionnaire
surveys, with participants selected using random sampling
processes.
o other ways such as observation and documents or
interviews could also be used.

Compiled By Solomon Kebede, MSc, PhD Candidate, 167


2019
Longitudinal design
q Longitudinal designs consists of repeated cross-sectional
surveys to ascertain how time influences the results.
q useful for capturing changes over time.

q If your research questions relate to changes over time you may


wish to consider designing a longitudinal study.

n This will enable you to look at the same people or situations at


key points in time and to consider the changes over time.
o i.e. data is gathered on at least two occasions separated in
time.
o Because of the repeated nature of this research design, it tends
to be expensive and time consuming
Compiled By Solomon Kebede, MSc, PhD Candidate, 168
2019
Longitudinal
o Two types of study are commonly identified:

o Panel studies– these consist of a sample of people, often


randomly selected, who are questioned more than once.

o Cohort studies– these concentrate on a group that shares


similar characteristics, such as students from a particular year
of matriculation or people on strike at a certain time.

Compiled By Solomon Kebede, MSc, PhD Candidate, 169


2019
Case study design
q This is a detailed analysis of individual cases.
n includes either a single case or a small number of cases but
each case is explored in detail and in great depth.
q The information may be both qualitative and quantitative.

n The subject of the case may be a person, an organization, a


situation or a country.
n The selection of the case is important, in terms of its potential
to produce data that will enable you to address your research
question.

n In case study design, no claim is made for generalizability.


Compiled By Solomon Kebede, MSc, PhD Candidate, 170
2019
Case study
o The selection of the case may be based on:
1. Critical case. Here the case is chosen as one that will
enable the researcher to test a theory or hypothesis.
2. Extreme or unique case. Here the focus is on a situation or
group that is perceived to be different to any other and
the focus is on its uniqueness.
3. Representative or typical case. This is a case that is chosen
because it is seen to represent many other similar cases
4. Revelatory case. This is a case that has the potential to
shed light on the research topic.
o It may be that the researcher has access to a situation that has
been hidden.
Compiled By Solomon Kebede, MSc, PhD Candidate, 171
2019
Survey research
o Surveys are the most widely used method of collecting a wide
variety of information in economics.

n Through surveys we collect structured data sets and


present them in the form of a variable by case data grid

n ‘Case’ is the unit of analysis (e.g., household in household


surveys or firms in a firm survey)

n Mainly structured questionnaires are used but other


instruments can also be used to collect survey data (e.g.,
observation)

Compiled By Solomon Kebede, 172


MSc, PhD Candidate, 2019
Survey Research
Variables
Variables
hh size income hh head
Hh 1 hh10
size income
450 hh head
F
Cases Hh 1 10 450 F
Hh 2 5 1000 F
Cases Hh 2 5 1000 F
Hh
Hh33 7
7 745
745 MM
… … … …
… … … …

Compiled By Solomon Kebede, 173


MSc, PhD Candidate, 2019
Survey Advantages and Disadvantages

o If planned correctly, a survey makes it easy for people to


participate.
o Participants can remain anonymous.
o Same questions can be asked in several ways to double-check
for accuracy and consistency.
o Survey Disadvantages
n People surveyed may not answer all questions.
n It is easy for people to give habitual responses.
n Low response rate is common.
n People can and do lie.
n Can’t test knowledge with mailed surveys.
Compiled By Solomon Kebede, 174
MSc, PhD Candidate, 2019
Survey Sampling
o Some studies involve only small number of people and thus all
of them can be included.
n But when the population is large, it is usually not possible
to undertake a census.

o Therefore, we will be forced to work with a smaller, more


manageable number of people to take part in their research.

o The selection of some cases from a larger group of potential


cases is called sampling.
n It aims at obtaining consistent and unbiased estimates of
the population parameters.

Compiled By Solomon Kebede, MSc, PhD Candidate, 175


2019
Survey Sampling
o The best sample is representative, or is a model of the
population.

o A sample is a portion of a larger whole.

o Of course, no sample is perfect, as it usually has some degree


of bias or error.
n But, if this sample is chosen carefully using the correct
procedure, it is possible to generalize the results to the
whole of the population.

Compiled By Solomon Kebede, MSc, PhD Candidate, 176


2019
Survey Sampling: Reasons for Sampling
o Reduced cost: since data are secured from a small fraction of the
population, cost will be reduced.

o Greater speed: for the same reasons as above, sample surveys can
be reported faster than that of census.

o Greater scope and accuracy: since samples deal with fewer units
than complete census, it is possible to attain greater accuracy
and cover greater scope by involving highly trained personnel,
careful supervision and specialized equipment, which otherwise
wouldn’t be possible.

Compiled By Solomon Kebede, 177


MSc, PhD Candidate, 2019
Survey Sampling: Reasons for Sampling
o Feasibility: some investigations could only be addressed by
sample surveys: for example
n when studying infinite populations,
n laboratory testing one’s blood,
n when conducting quality assurance tests (especially when
the test involves the destruction of the product), etc.

Compiled By Solomon Kebede, MSc, PhD Candidate, 178


2019
Survey Sampling
o Representativeness
n There are two principles in representativeness:
o The need to avoid bias and the need to gain maximum
precision.

n Bias can arise:


q if the selection of the sample is done by some non-
random method
q if the sampling frame (i.e. list, index, population record)
does not adequately cover the target population.
q if some sections of the population are impossible to find
or refuse to co-operate.

Compiled By Solomon Kebede, MSc, PhD Candidate, 179


2019
Survey Sampling
q Representativeness is important particularly if you want to
make generalization about the population.

q So, for Quantitative Studies:


q Samples should be drawn in such a way that it is
representative of the population.

q For Qualitative Studies:


q representativeness of the sample is NOT a primary concern.
q In qualitative studies we select study units which give you
the richest possible information.
q you go for INFORMATION-RICH cases!

Compiled By Solomon Kebede, MSc, PhD Candidate, 180


2019
Steps in Sampling Design
a) Identifying the relevant population:
n Determine the relevant population from which the sample
is going to be drawn.
n Example: if the study concerns income, then the definition
of the population as individuals or households can make a
difference.
n Apart from persons, a study population could also be
villages, institutions, records, etc.
b) Determining the method of sampling:
o Whether a probability sampling procedure or a non-probability
sampling procedure has to be used.

Compiled By Solomon Kebede, 181


MSc, PhD Candidate, 2019
Steps in Sampling Design
c) Securing a sampling frame:
n A list of elements from which the sample is actually drawn
is important and necessary.

d) Identifying parameters of interest:


n what specific population characteristics (variables and
attributes) may be of interest.

e) Determining the sample size


o The determination of the sample size deepens on several
factors.
q Obviously, the bigger, the better

Compiled By Solomon Kebede, 182


MSc, PhD Candidate, 2019
Determining the sample size
o One of the questions researchers tend to ask is ‘how many
people should I speak to?’
n This is not an easy question, as a number of different
aspects need to be taken into account.
o The decision on the sample size hinges on how large an error
one is willing to tolerate in estimating population parameters
or,
n Put differently, what effect size will be required for the
result to be considered significant.
o This obviously depends on the type of research.
n But in the final decision, statistical precision must be
balanced against time, cost, and other practical
considerations.
Compiled By Solomon Kebede, 183
MSc, PhD Candidate, 2019
Determining the sample size
o Research designs with too small of a sample size are unethical
because they waste resources since they can only provide
anecdotal evidence.

q If the sample size is too small, the data will be unusable and
the experiment must be repeated, with a larger sample to
test the hypothesis with greater rigor.
q Not possible to make valid generalizations.

q Thus, the issue sample size is a very important thing in


quantitative research.

Compiled By Solomon Kebede, 184


MSc, PhD Candidate, 2019
Determining the sample size
o Research studies that use too large sample i.e., larger than
needed also are unethical because
n time and financial resources wasted,
n human subjects undergo experimental procedures that
could be distressful, painful.

o It is unethical to start work on a research project, which


burdens research subjects and consumes resources, without
sufficient methodological skills to ensure a useful product.

Compiled By Solomon Kebede, 185


MSc, PhD Candidate, 2019
Determining the sample size
o The sample size will also depend on what you want to do with
your results.
o For large scale, quantitative surveys you will need to contact
many more people than you would for a small, qualitative piece
of research.
n For instance, if you intend to produce large amounts of cross
tabulations, then you need to contact more people.
o The general rule in quantitative research is that the larger the
sample the better it is.
o However, the additional benefits derived from increasing
the sample size become smaller as the sample size
increases.

Compiled By Solomon Kebede, 186


MSc, PhD Candidate, 2019
Determining the sample size
o However, you have to remember that you are probably
restricted by time and money
n Therefore, you have to make sure that you construct a
sample which will be manageable.

o Also, you have to account for non-responses and you may need
to choose a larger sample size to overcome this problem.
n There will be some non-responders – people who do not
agree to take part in your research – and some whom you
may not be able to contact.
o select at least 10% more than you hope to gain
responses from.

Compiled By Solomon Kebede, 187


MSc, PhD Candidate, 2019
Determining the sample size
In summary, the sample size depends on several factors.
i) Degree of homogeneity: The size of the population variance is the
most important parameter.
n The greater the dispersion in the population the larger the
sample must be to provide a given estimation precession.

ii) Degree of confidence required: Since a sample can never reflect


its population for certain, the researcher must determine how
much precision you need.
o Precision is measured in terms of
n An interval range.
n The degree of confidence
o The sample size needed in a research study is determined by
the level accuracy required in the study.
Compiled By Solomon Kebede, 188
MSc, PhD Candidate, 2019
Determining the sample size

iii) Number of sub groups to be studied:


q When the researcher is interested in making estimates
concerning various subgroups of the population then the
sample must be large enough for each of these subgroups
to meet the desired quality level.

iv) Cost: cost considerations have major impact on decisions about


the size and type of sample, as well as the data collection
methods.
q All studies have some budgetary constraint and hence cost
dictates the size of the sample.
o The level of precision obviously increases with increase in the
size of the sample thus researchers are usually challenged by the
interest of balancing accuracy and cost.
Compiled By Solomon Kebede, 189
MSc, PhD Candidate, 2019
Determining the sample size
v) Practicality: Of course the sample size you select must make
sense.
o Therefore the sample size is usually a compromise between
what is DESIRABLE and what is FEASIBLE.

o For researchers with limited time and resources, the sample


size is more likely to be influenced by the resources available
and the ease of access to the sampled cases.

n It is important, though, to be aware of the limitations of a


smaller sample and to discuss these in your research
report or dissertation.

Compiled By Solomon Kebede, 190


MSc, PhD Candidate, 2019
Determining the sample size
VI) Other Considerations:
o (i)Prior information: If our process has been studied before, we
can use that prior information to determine our sample size.
q This can be done by using prior mean and variance
estimates and by stratifying the population to reduce
variation within groups.

o Note: If you know the mean and variance statistical methods


can be used to determine the size of the sample required for a
given level of accuracy.

Compiled By Solomon Kebede, 191


MSc, PhD Candidate, 2019
Determining the sample size
q (VII) Rule of Thumb: is based on past experience with samples
that have met the requirements of the statistical methods.
q For small populations (under 1000 a large sampling ratio
(about 30%). Hence, a sample size of about 300 is required.

q For moderately large population (10,000), a smaller sampling


ratio (about 10%) is needed – a sample size around 1,000.
q To sample from very large population (over 10 million), one
can achieve accuracy using tiny sampling ratios (.025%) or
samples of about 2,500.

Compiled By Solomon Kebede, MSc, PhD Candidate, 192


2019
Determining the sample size
o (VIII) Using Cochran’s Formula: You need to determine a few
things about the sample you need.

o Margin of Error (Confidence Interval) — No sample will be


perfect, so you need to decide how much error to allow.
n The confidence interval determines how much higher or
lower than the population mean you are willing to let your
sample mean fall.

o It will look something like this: “68% of voters said yes to


Proposition Z, with a margin of error of +/- 5%.”

Compiled By Solomon Kebede, 193


MSc, PhD Candidate, 2019
Determining the sample size
o Confidence Level — How confident do you want to be that the
actual mean falls within your confidence interval?
n The most common confidence intervals are 90% confident,
95% confident, and 99% confident.

o Standard of Deviation — How much variance do you expect in


your responses?

o The safe decision is to use .5 – this is the most forgiving number


and ensures that your sample will be large enough.

Compiled By Solomon Kebede, 194


MSc, PhD Candidate, 2019
Determining the sample size
o When the population of the study is unknown or very large,
Cochran (1963) has developed the following formula to
determined sample size.

o where n represents sample size, Z is z-score which is confidence


level of 95%, p represents standard deviation which commonly
takes a value of 0.5, q= 1-p, e represents the level of precision
(margin of error commonly used as +/- 5% with confidence
interval of 95%).
Compiled By Solomon Kebede, 195
o MSc, PhD Candidate, 2019
Determining the sample size
Sample Size in Qualitative Studies
q The size of a sample is not big issue in qualitative research since
the emphasis is obtaining new information to have deep
understanding of a phenomenon.

q There are no fixed rules for sample size in qualitative


research.

q The size of the sample depends on WHAT you try to find out,
and from what different informants or perspectives you try
to find that out.

q the sample size is therefore estimated, but not


determined.
Compiled By Solomon Kebede, 196
MSc, PhD Candidate, 2019
Probability and non-probability sampling
o We can arrange the different approaches to the sampling
process in a spectrum.

o At one end of the spectrum are the sampling approaches


that are based on statistical theory.

n They aim to produce a sample that can be highly


representative of the whole population
n These are called probability samples.

Compiled By Solomon Kebede, 197


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o At the other end of the spectrum are approaches to sampling
that are concerned with selecting cases that will enable the
researcher to explore the research questions in depth,

n These are called purposive or non probability samples

Compiled By Solomon Kebede, 198


MSc, PhD Candidate, 2019
Probability and non-probability sampling
q Probability sampling is based on the concept of random
selection that assures that each population element is given a
known non zero chance of being selected.
q A scientific sample drawn in such a way that the probability of
being chosen is equal or is known – for example, a random
sample.

q A randomization process is used in order to reduce or


eliminate sampling bias so that the sample is representative
of the population from which it is drawn.

q Probability sampling requires a sampling frame (a listing of


all study units).

Compiled By Solomon Kebede, 199


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o Only probability sampling provides a statistical basis for saying
that a sample is representative of the target population.

o And a sample will be representative of the population from


which it is drawn if all members of the population have an
equal chance of being included in the sample.
o random selection minimizes human bias

Compiled By Solomon Kebede, 200


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o In summary:
n probability samples are more representative than any other
type of sample.
q sampling errors can be calculated only for probability
samples.
q probability samples rely on random process, i.e. the selection
process operates in a truly random method.
q since each element has an equal chance or probability of
being selected it is possible to get consistent and unbiased
estimate of the population parameter.

Compiled By Solomon Kebede, 201


MSc, PhD Candidate, 2019
Probability and non-probability sampling
q Several types of probability sampling methods could be
identified:
q Simple Random Sampling Technique
q Systematic sampling Technique
q Stratified Sampling Technique
q Cluster Sampling Technique.
q Hybrid Sampling

Compiled By Solomon Kebede, 202


MSc, PhD Candidate, 2019
Sampling Techniques

Probability Non-probability
Sampling Sampling

Simple Systematic Stratified Multistage


Cluster
Random Random Random Sampling
Sampling
Sampling Sampling Sampling

Convenience Volunteer Snowball Quota


sampling sampling sampling sampling

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 203
Probability and non-probability sampling
1. Simple Random Sampling (SRS): It is a sampling method that
specifies that each person in a population has the same chance
of being included in a sample as every other person.
n each element of the population has an equal chance of being
selected into the sample.

o This is the simplest and easiest method of probability sampling.


n It assumes that an accurate sampling frame exists.

o Selection could be done either by using table of random


numbers or by the lottery method.

Compiled By Solomon Kebede, 204


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o E.g., simple random sampling for household surveys
Ø Population = all households in the country
Ø Sampling frame = the list of all households (20 million in
Ethiopia?)
Ø Sample size = say we have resources to cover only 20,000
households
Ø Sampling fraction 20,000/20,000,000 or 0.1%
Ø Select randomly 20,000 households from the long list of
20,000,000 households

Compiled By Solomon Kebede, 205


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o Merits of SRS
n No investigator bias or discretion
n Help us to obtain a more representative sample
n Can produce better estimates for the population
o Limitations
n It needs up-to date list of the population units
n Units selected might be scattered geographically, hence,
high cost of data collection

Compiled By Solomon Kebede, 206


MSc, PhD Candidate, 2019
Probability and non-probability sampling
2. Systematic Sampling Technique
o In systematic sampling individuals are chosen at regular
intervals (for example every kth) from the sampling frame.

n The first item is selected randomly and then the remaining


units will automatically be selected with some
predetermined pattern.

n A sampling interval (the standard distance between the


elements selected in the sample) identified.

o Simplicity and flexibility are its major advantages.

Compiled By Solomon Kebede, 207


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o Steps to draw a systematic sample:
n Calculate K (the sampling interval).
n Select a number between 1 and K at random, say that
number is r.
n Then this means we select the rth element at random.
n Then the rth, (r+K) th, (r+2K) th, …, [r+(n-1)K] th elements of
the population will be selected in our sample.

Compiled By Solomon Kebede, 208


MSc, PhD Candidate, 2019
Probability and non-probability sampling
q E.g., a systematic sample is to be selected from 1200 students
of a school.
ü The sample size to be selected is 100.
o The sampling fraction is: 100/1200= sample size/study
population = 1/12
o The sampling interval is therefore 12.
o The first student in the sample is chosen randomly, for
example by blindly picking one out of twelve pieces of paper,
numbered 1 to 12.
q If number 6 is picked - every twelfth student will be
included –i.e. 6, 18, 30, 42, etc.
Compiled By Solomon Kebede, 209
MSc, PhD Candidate, 2019
Probability and non-probability sampling
3. Stratified Sampling
q Useful when we have heterogeneous populations.

q One disadvantage of the SRS is that small groups in which the


researcher is interested may not appear in the sample
q Stratified sampling divides a population into the
appropriate strata and a simple random sample taken using
either SRS or SS techniques from each stratum.

o The elements in a stratum are supposed to be homogeneous


with respect to the given characteristics.

Compiled By Solomon Kebede, 210


MSc, PhD Candidate, 2019
211
Probability and non-probability sampling
The reasons for stratifying
1. To increase a sample’s statistical efficiency (smaller
standard errors-less variation).
2. To provide adequate data for analyzing the various
subpopulation.
3. To enable different research methods and procedures to be
used in different strata.
4. The absence or poor quality of a sampling frame makes it
necessary to first select a sample of geographical units, and
then to construct a sampling frame only within those
selected units.

Compiled By Solomon Kebede, 212


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o multiple stage stratified random sampling could also be
considered.
n E.g., in the household survey we may be interested to have
sufficient number of households from each region of
Ethiopia;
n So stratify by region!
How to Stratify
n Three major decisions must be made in order to stratify the
given population into some mutually exclusive groups.

Compiled By Solomon Kebede, 213


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o (1) What stratification base to use: stratification would be based
on the principal variable under study such as income, age,
education, sex, location, religion, etc.

o (2) How many strata to use: there is no precise answer as to how


many strata to use.
n The more strata the closer one would be to come to
maximizing inter-strata differences and minimizing intra-
strata variances.

Compiled By Solomon Kebede, 214


MSc, PhD Candidate, 2019
Probability and non-probability sampling
(3) What strata sample size to draw: different approaches could be
used:
n One could adopt a proportionate sampling procedure.

o If the number of units selected from the different strata


are proportional to the total number of units of the strata
then we have proportionate sampling.

o Or non-proportional sampling where the number of items


studied in each stratum is disproportionate to the
respective proportion of the stratum in the population.

Compiled By Solomon Kebede, 215


MSc, PhD Candidate, 2019
Probability and non-probability sampling
4. Cluster Sampling:
o Similar to stratified sampling as you need to divide the
population into discrete groups prior to sampling.
o It may be difficult or impossible to take a simple random sample
because a complete sampling frame does not exist, or
n Logistical difficulties may also discourage random sampling
techniques
n E.G: interviewing people who are scattered over a
large area may be too time-consuming.
o The selection of groups of study units (clusters) instead of the
selection of study units individually is called CLUSTER
SAMPLING.
o It is cost effective (High economic efficiency)
Compiled By Solomon Kebede, 216
MSc, PhD Candidate, 2019
Probability and non-probability sampling
q For instance, if the total area of interest is a big one it can be
divided into a number of smaller non –overlapping areas
(clusters) and some of the clusters are selected randomly.

q Clusters are often geographic units (e.g., districts, villages)


or organizational units (e.g., clinics, etc.).

o The primary sampling unit is not units of the population but


groups within the population (clusters)

Compiled By Solomon Kebede, 217


MSc, PhD Candidate, 2019
Probability and non-probability sampling
o E.g., sampling for household survey in Addis Ababa
n Probably no complete sampling frame and costly to cover
through simple random sampling Procedures

n Randomly select sub-cities (clusters)


n Randomly select kebeles from selected sub-cities (clusters)
n Then randomly select households from the selected
kebeles

Compiled By Solomon Kebede, 218


MSc, PhD Candidate, 2019
Probability and non-probability sampling
Non-Probability Sampling: selection is non random i.e., sampling
units/elements have unequal chance of being selected

o Non-probability samples are chosen based on judgment


regarding the characteristics of the target population and the
needs of the survey.

o Sometimes a probability sample is infeasible.


q Example: If we want to conduct a lengthy experiment using
human subjects we use whoever is willing to participate.
q E.g. a study of drug users (hearsay, criminal records)
Compiled By Solomon Kebede, 219
MSc, PhD Candidate, 2019
Probability and non-probability sampling
o Three conditions to use non-probability sampling.
n First, if there is no desire to generalize to a population
parameter.

n Secondly, because of cost and time requirements.


o probability sampling could be prohibitively expensive
since it requires more planning and repeated callbacks.

o Thirdly, probability sampling may break down in its


applications.
n The total population may not be available.

Compiled By Solomon Kebede, 220


MSc, PhD Candidate, 2019
Probability and non-probability sampling
(1) Convenience or accidental sampling: The method selects anyone
who is convenient.
n Units that are convenient for the investigator are selected
(e.g. volunteers)
n It can produce highly un-representative samples.

o Such samples are cheap, however, biased and full of systematic


errors.
n E.G: the person on the street interview conducted by
television programs is an example of a convenient sample.
n Drop by the cafeteria and ask questions of whoever is
there.
Compiled By Solomon Kebede, 221
MSc, PhD Candidate, 2019
Probability and non-probability sampling
(2) Quota Sampling: subgroups are identified and a specified
number of individuals from each group are included in the
research – based on certain criteria.

n Identify categories of people (e.g., male, female) then


decides how many to get from each category.
o In a school, find 10 elementary teachers, 10 middle
school and 10 high school teachers.
n is used in opinion pollsters, marketing research and other
similar research areas.

o No randomization – difficult to know the sampling error.


Compiled By Solomon Kebede, 222
MSc, PhD Candidate, 2019
Probability and non-probability sampling
(3) Purposive or Judgment sampling
n When we select a limited number of informants,
strategically so that their in-depth information will give
optimal insight into an issue is known as purposeful
sampling.

o It uses the judgment of the expert in selecting cases.


n participants are selected because of some desirable
characteristics, like expertise in the area.

o Could be useful when used by skilled investigator

Compiled By Solomon Kebede, 223


MSc, PhD Candidate, 2019
Probability and non-probability sampling
(4) Snowball (Network) Sampling – chain sampling
n This is a method for identifying and selecting the cases in a
network.
o You start with one or two information-rich key
informants and ask them if they know persons who know
a lot about your topic of interest.
o You may know only a few users of a certain product, but they
know other users. Contact the first few and ask them for names of
others, and so on.

o Useful when there is no sampling frame


n E.g., Becker’s (1963) study of marijuana users
n Illegal migrants, sex workers, drug users, etc.
Compiled By Solomon Kebede, 224
MSc, PhD Candidate, 2019
Problems in Sampling
o Survey errors: The discrepancy between statements from
survey estimates and the reality (the true value) is called
survey errors.

o There are two types of survey errors.


n sampling errors and non-sampling errors.
o and survey error is the sum of the two.

Compiled By Solomon Kebede, 225


MSc, PhD Candidate, 2019
Problems in Sampling
o Sampling errors are errors which are attributable to sampling,
and which therefore, are not present in information gathered in
a census.
o It is not a mistake
o Can be controlled by well developed sampling theory

n Sampling errors can be calculated only for probability


samples.
n Increasing the sample size is one of the major instruments
to reduce the extent of the sampling error.

Compiled By Solomon Kebede, 226


MSc, PhD Candidate, 2019
Problems in Sampling
o Sampling error is related to confidence intervals.
o A narrower confidence interval means more precise estimates
of the population for a given level of confidence.
o The confidence interval for the true population mean is given
by: 
Mean  z
n
o The sampling error is given by:


z
n

Compiled By Solomon Kebede, 227


MSc, PhD Candidate, 2019
Problems in Sampling
o Non-sampling error: Such errors are present whether it is
sampling or census survey that we are dealing with.
o these include all errors apart from sampling error and are
mostly mistakes by one party or another.

Non-Sampling Error includes:


n Non-coverage error
n Wrong population is being sampled
n Non response error
n Instrument error
n Interviewer’s error

Compiled By Solomon Kebede, 228


MSc, PhD Candidate, 2019
Problems in Sampling
Non-Coverage sampling error: This refers to sample frame defect.
n Omission of part of the target population (e.g., soldiers,
students living on campus, people in hospitals, prisoners,
households without a telephone in telephone surveys, etc.).

The wrong population is sampled


n Researchers must always be sure that the group being
sampled is drawn from the population they want to
generalize about or the intended population.

Compiled By Solomon Kebede, 229


MSc, PhD Candidate, 2019
Problems in Sampling
Non response error – Common in self-administered surveys
n This error occurs when you are not able to find those whom
you were supposed to study.

n Some people refuse to be interviewed because they are ill,


are too busy, or simply do not trust the interviewer.
n When one is forced to interview substitutes, an unknown
bias is introduced.

Compiled By Solomon Kebede, 230


MSc, PhD Candidate, 2019
Problems in Sampling
Instrument error
n The instrument in sampling survey is the device in which
we collect data- usually a questionnaire.
n When a question is badly asked or worded, the resulting
error is called instrument error.

o Example: leading questions or carelessly worded


questions may be misinterpreted by some researchers.

Compiled By Solomon Kebede, 231


MSc, PhD Candidate, 2019
Problems in Sampling
Interviewer Error :
n Enumerators can distort the results of a survey by in-
appropriate suggestions, word emphasis, tone of voice and
question rephrasing.
n Cheating by enumerators -with only limited training and
under little direct supervision.
n Perceived social distance between enumerator and
respondent also has a distorting effect.

o E.G: questions about sexual behavior might be


differently answered depending on the gender of the
interviewer.
Compiled By Solomon Kebede, 232
MSc, PhD Candidate, 2019
o The lessen up to this stage helps us to develop our
research proposal
o Purpose is to get support either financial/material
o Proposal should be written in a convincing way
o Both theoretical and practical considerations are
needed
o Proposals fail if they don’t fulfil basic requirements
of critical stakeholders (e.g. donor)
o Donors use different parameters to evaluate
proposals depending of their objective and
mandates
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 233
 This is a critical step at which you collect your data
 Data sources may vary but your main objective is
to collect credible and detailed data
 Collecting credible data is not an easy task but
requires both mental and physical strengths
 How should I behave during data collection?
 Do your homework
◦ Be prepared to talk about your research: the ability to
clearly articulate the rationale, aims, and objective, and
methods of your project can be instrumental in getting
the right doors opened

Compiled By Solomon Kebede, MSc, 234


PhD Candidate, 2019
n Prepare a brief outline of your project: certain
individuals or organizations may want to have
certain document they consider to and/or present
to ‘gatekeepers’
n Have a letter of introduction/support: a letter of
introduction or support from the college/your
organization can help you by answering questions
like, ‘so who are you and where are you from?’
n Find out about appropriate approaches: sometimes
formal approaches may not help to get access to
data; you may try informal ways( like
brother/sisterhood approach).
Compiled By Solomon Kebede, MSc, 235
PhD Candidate, 2019
o Be Professional
ü Go through the appropriate channels: going through a ‘backdoor’
is both risky (it is more likely to be shut prematurely) and
unethical( contact the desired person not 3rd except the
delegation by z desired person 2 3rd person).
ü Be respectful: choose the right time to approach, be punctual,
dress appropriately, and be modest in your initial request(try 2
indicate the contribution of the respondent 2 this paper)
ü Plan for the unexpected: very rarely does the research process
run smoothly, especially when you are dealing with individuals( u
may not get z respondent @ desired time)
ü Leave doors open: many researchers swear they have collected
all the data they need, but latter wish they could go back and ask
just a few more questions ( by giving tx indicate that u may
contact them in z future if zere is missed data)

Compiled By Solomon Kebede, MSc, 236


PhD Candidate, 2019
o Offer Something Back
n Don’t disappear: let your contacts know how
things are progressing and/or send a note of
thanks ( via email/ other means if the
technology permits you)
n Make results available: it is quite natural to have
a sense of curiosity ( eagerness 2 know
something) about studies of which you are a
part; results of your study can be quite valued
by those who have facilitated your research

Compiled By Solomon Kebede, MSc, 237


PhD Candidate, 2019
o Data can be collected in different ways:
◦ Survey
◦ Focused group discussions
◦ Key informant interviews
◦ Observation
o Survey
◦ Is a common method of collecting data because it is so
versatile (2 be used in many d/t ways)
◦ Any one of the several goals of research (exploration,
description, explanation, or evaluation) can readily be
pursued via survey research
◦ Entails using a questionnaire

Compiled By Solomon Kebede, MSc, 238


PhD Candidate, 2019
n In advanced countries, surveys can be
conducted via mails, telephone, or emails.
n Hence a large survey can be conducted with
relatively small cost
n In developing countries, however, surveys are
conducted via personal interviews
n Thus, they are expensive
n Require extensive time in the field, energy, and
mental capabilities

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 239
o Step to conduct a good survey
o Plan
n Consider what aspects of your research
questions can be answered through a
questionnaire
n Plan how to address questions that cannot be
answered through a questionnaire (e.g. through
FGD, KII)
n Plan for the appropriate time to undertake the
survey (from your angle and respondents angle)

Compiled By Solomon Kebede, MSc, 240


PhD Candidate, 2019
o Construct the questionnaire
n Much thought must be given to the exact wording
of our questions, the structure of our questions,
and the way we sequence and format our
questions
n If relevant questionnaires exist, adopt or adapt
them
n If your questionnaire requires the construction of
any new questions, have a shot at drafting them.
n Check whether the questions cover each aspects of
your research stated in your objectives.
n Decide on the response categories of each question
n Decide on whether to use an open-ended or a
close ended ones forCompiled
each question
By Solomon Kebede, MSc,
241
PhD Candidate, 2019
n Carefully read each question and response choices,
and avoid questions that are ambiguous, leading,
offensive, or double-barrelled, complex and full
of jargons (example)
n Put your questions in a logical order that facilitates
uninterrupted flow of responses
n Start with simple, non-sensitive questions
n Write instructions for your enumerators(data
collector) Construct a clear and aesthetically
attractive layout
n Insert page number
n Ask your peers/supervisor for comment on the
questionnaire and respond to constructive
comments Compiled By Solomon Kebede, MSc, 242
PhD Candidate, 2019
n Pretest your questionnaire with a few people ( 5
sample at the study area)
o helps to know whether you and the respondents are on
the same wave length(their angle of view same/not with
u’rs)
o allows you to assess the impact of word selection,
question sequencing, and various formatting and
layout issues.
n Modify (if necessary) based on the feedback from
the respondents
o Recruit your enumerators
n Experience in fieldwork matters
n Knowledge on the subject matter also matters
n Devotion/commitment and trustworthiness matter
Compiled By Solomon Kebede, MSc, 243
PhD Candidate, 2019
 Train your enumerators
◦ Clarify each aspect of your research to the
enumerators
◦ Check their performance by asking them to try
the questionnaire via asking one another
 Execute the survey
◦ Get approval from appropriate people
◦ Ask for some facilitation
◦ Assign your enumerators
◦ Replace samples which are missing because of
different reasons immediately (You may prepare a
reserve list to facilitate fast replacement)

Compiled By Solomon Kebede, MSc, 244


PhD Candidate, 2019
n Monitor your enumerators
n Check for inconsistencies and other problems
immediately and seek for corrections
n Provide vote of thanks to local personalities
who helped you in your fieldwork

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 245
o Major activities at this stage include:
n Data coding and entry
n Data cleaning and processing
n Data analysis
Data coding and Entry
o Data coding is a process of transforming a
human response into a suitable way for
computer-assisted analysis
o If you use a pre-coded questionnaire coding
is very simple
Compiled By Solomon Kebede, MSc, 246
PhD Candidate, 2019
o However, if the data collection instruments
are not structured or semi-structured coding
takes significant amount of time
o In such a case, responses are put under
various categories based on their similarity
and the categories are coded
o After coding data will be entered into a
computer using a data sheet of an
appropriate statistical software; SPSS, Stata,
Limdep, Excel data sheets can be used
Compiled By Solomon Kebede, MSc, 247
PhD Candidate, 2019
 Take adequate time to enter data
 Use a well trained data entry operator if you
don’t do the entry yourself
Data cleaning and processing
 Before analysis, data have to be checked for
possible errors during data entry
 Cleaning requires a close case-by-case and
variable-by-variable checking of figures in the
data sheet
 Erroneous figures are checked against the
figures recorded in the questionnaire
Compiled By Solomon Kebede, MSc, 248
PhD Candidate, 2019
o But you need to save the cleaned data using a
different file name
o Data processing is an activity to be done
before and during data analysis
o This is to produce variables which can be
directly interpreted and explained in your
report
o Target variables can be produced through
recoding a variable, computing or generating
a variable, and aggregation of values at
predefined level
Compiled By Solomon Kebede, MSc, 249
PhD Candidate, 2019
Data analysis
 Data analysis is a fundamental step in your
research process
 This is a step where you show the relationship
among the variables of interest using an
appropriate method
 Basically data analysis produces two types of
results
◦ Descriptive results
◦ Inferential results
 Descriptive analysis produces output to describe
a phenomena

Compiled By Solomon Kebede, MSc, 250


PhD Candidate, 2019
o Shows preliminary relationships
o Common statistical analysis used include:
means, frequencies, mean comparisons,
cross tabulations, and correlations
o Associated test statistics include: t tests, F
tests, and Chi-square tests
o Descriptive test helps as a screening devise
for important variables
o Helps to check inferential results at the next
step
Compiled By Solomon Kebede, MSc, 251
PhD Candidate, 2019
q Inferential statistics produce outputs to
explain a phenomenon (what causes
things happen in the way they happen!)
q Results from inferential statistics can be
inferred to other similar situations
q Relationships can be of three types
◦ Association (implies sign of relationship)
◦ Correlation (implies sign and magnitude of
relationship)
◦ Causality (implies sign, magnitude, & direction
of relationship)

Compiled By Solomon Kebede, MSc, 252


PhD Candidate, 2019
o Research is often judged not based on what
you did, but based on your ability to report
what you did
o Writing is a process; consider it as a part and
parcel of the research journey rather than as
a single account of the research journey
o The standard procedure

Conduct
Prepare a Writeup the
the
proposal Report
research
writing writing writing writing writing
writing
Compiled By Solomon Kebede, MSc, 253
PhD Candidate, 2019
o Be aware the purpose of your report is to
convince people that you have produced a
good and sound research finding
o Be aware that audience vary in terms of
expectations, intellectual capabilities, and
experience
o Be aware that you should produce several
drafts before your arrive at the final one
o Be aware that ideas evolve after you start
writing (Don’t afraid to start writing!!)

Compiled By Solomon Kebede, MSc, 254


PhD Candidate, 2019
o May be you are required to use a standard
format but try to be creative in defining
topics and subtopics within the given format
o It is possible that your findings may
contradict current thinking.
o If your findings “surprised” you in some
way, and this is one important
phenomenon inherent in the research
process, then your findings are likely to
surprise others.

Compiled By Solomon Kebede, MSc, 255


PhD Candidate, 2019
o It is your responsibility to convince your
readers at least to consider the accuracy of
your analysis by providing sufficient
information for them to make independent
judgments
o There are several criteria which are useful
in drafting a compelling argument:
o 1) A thorough interpretation: For an
interpretation to be considered “thorough,”
it needs to account for all the major
dimensions which you studied, as identified
in your research question
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 256
o A thorough interpretation of your data
synthesizes the individual units of analysis
(e.g. words, sentences, responses, test
scores, weight, etc.) into a more
comprehensive context.
o Your readers should have a sense that your
interpretations represent an exhaustive search
for meaning from your data
o Your honesty as a researcher is attested to by
your comprehensive inclusion of all your data,
even data which may challenge your
inferences.
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 257
2) A clear interpretation: is easy to follow.
o Information is presented systematically and
predictably.
o Symbols in graphs and within the text are
explained.
o Details are provided to enable an intelligent
reader to understand the issues.
o The information presented in a table should
be consistent with the information presented
in the narrative.
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 258
o Your careful choice of words will reflect your
clear understanding of your findings.
o 3) A Logical Interpretation: Your reader
will consider your interpretation “logical” if
you organize your presentation in a
systematic and predictable fashion.
o From your understanding of your data, you
decide which issue needs to be addressed
first, and which follows naturally from that
first issue.

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 259
o You will also provide a sketch of the order in
which you will discuss your findings, enabling
your readers to create a mental map of your
presentation, and then to find that
information as they peruse your text
o 4) A Relevant Interpretation: Your
findings must be directly related to three
components of your study:
n your research question and the theories
n the data which you collected; and
n your data analysis.
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 260
o Information from other sources is irrelevant
at this point.
o Your interpretation must be focused on
important issues.
o You want to highlight the most prominent,
and the strongest findings
5) A Cautious Interpretation: Some
researchers make major pronouncements from
their research, implying that they have solved
all the questions in the world with their research

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 261
o Others are timid about making any
statements beyond the data which they
collected.
o A wise researcher takes care not to over-
generalize from limited data.
o Being cautious, while making a case for what
you did find, you will find an appropriate
balance.
o Seeking critical readers’ reactions to your
analysis is important
o Language matters!!!
Compiled By Solomon Kebede, MSc,
PhD Candidate, 2019 262
o Ten important points why theses fail
n There is no logical structure (structure is not
attractive & the flow of ideas is not smooth)
n Ideas are not well thought out (odd ideas come
in between)
n Thesis is disorganized
n Assumptions are made which cannot be
justified by evidence

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 263
n There are too many grammatical and spelling
mistakes
n Sentences and paragraphs are too long or too
obscure
n Ideas are directly taken from other sources
without acknowledgement
n There is too much repetition
n There is too much irrelevant information
n Summary and conclusion are weak/irrelevant

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 264
EXAMPLES
1) Simple and direct language

 It is interesting to note that the effectiveness of the overall


government rules, regulations, and guidelines was undeniable.

 The government policy was effective.

Use simple verbs: “use” instead of “utilize”

Concentrate on what you want to say .....


and say it in the simplest, most direct way!

Compiled By Solomon Kebede, MSc, 265


PhD Candidate, 2019
2)Double negatives

 It is not unlikely …
 It is likely …
 The effect of the new investment policy is not unimpressive.
 The effect of the new investment policy is impressive
 At no time was investment response absent in unemployment
 Unemployment always responds to investment.

Compiled By Solomon Kebede, MSc, 266


PhD Candidate, 2019
3) Abstract nouns made from verbs
 Assessments were carried out on households’ quality of life.
 Households’ quality of life was assessed.

Other examples of such nouns: measurement, observation


They make sentence long… avoid them.
 Observations were made on …….
 ….was observed

 It is possible to say that the growth of agricultural output in


Ethiopia is a reflection of effective utilization of modern inputs.
 The growth of agricultural output in Ethiopia may reflect effective
use of modern inputs.

Compiled By Solomon Kebede, MSc, 267


PhD Candidate, 2019
4) Noun clusters
 Research results dissemination improvement methods

 Methods of improving the dissemination of research results

 income
 annual income
 households‘ annual income
 Urban households‘ annual income

 The mean urban households‘ annual income was computed.


 The mean annual income of urban households was computed.

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 268
6) Sentence structure
Avoid long sentences! Any sentence of more than two
typewritten lines may be too long.

Use subject and verb at the beginning and follow with a


short list….

 In view of increasing the reliability of results and capturing


several dimentions of it, poverty was measured in three
different ways.

 Povery was measured in three different ways in view of


increasing the reliability of results and capturing several
dimensions of it.

Compiled By Solomon Kebede, MSc, 269


PhD Candidate, 2019
7) Personal pronouns

If you did the work and you think something is right, then
you should say it….take responsibility for your work:

 It is felt by the author that ….

 I believe that ……

Compiled By Solomon Kebede, MSc, 270


PhD Candidate, 2019
8) Unbiased language
The African farmer and his wife...
But many African farmers are women!!
“Man” as a verb

 The company is overmanned.


 The company is overstaffed.

“Man” as a suffix
 A spokesman of MIDROC company …..

 A representative of MIDROC company…..

Compiled By Solomon Kebede, MSc, 271


PhD Candidate, 2019
Gender sensitivity

 Each shareholder received his dividend …

 All shareholders received their dividends...

 The farmer has to do all the weeding himself …


 The farmer has to do all the weeding alone…

 Scientists often neglect their wives and children …


 Scientists often neglect their family…

Compiled By Solomon Kebede, MSc, 272


PhD Candidate, 2019
9) Non Preferable Words

As far as .... is concerned As for ....

At an early date... Soon...

At the present time At present or now

By means of .... by....

In the case of ....


In the latter case .... Remove them !
In this instance ....

In order to ..... to .....

Compiled By Solomon Kebede, MSc, 273


PhD Candidate, 2019
Non Preferable Words
Needless to say

For your information

It goes without saying

It is important to know

It is of interest to note

It should be noted

It stands to reason

It was demonstrated that...

Fortunately / Unfortunately
Compiled By Solomon Kebede, MSc, 274
PhD Candidate, 2019
Do’s Don’ts
o Arrive early and o Rush in late, find
make sure the room that the overhead
is set out in the way projector doesn’t
that you want. Make work and that you
sure that all the have no pen for the
equipment is whiteboard.
available and that o Worry about showing
you know how to your nerves.
work it. Everybody gets
o Try to relax and nervous when they
breathe deeply. first start giving
Acknowledge that presentations and
this is your firstCompiled By Solomon your audience should
Kebede, MSc, 275
presentation and PhD Candidate, 2019 know this.
Do’s Don’ts
o Use slides only to o Read everything
remind yourself from your slides
what to talk o Look at your notes,
o Look around the one direction,
room while you are never raising your
speaking – if it’s a head.
small group, make
eye-contact with as
many people as
possible.

Compiled By Solomon Kebede, MSc, 276


PhD Candidate, 2019
Do’s Don’ts
o Present interesting o Produce visual
visual information information which
such as graphs, people can’t see,
charts and tables either due to its
in a format which size or print quality
can be viewed by o Be drowsy, shy and
everyone. slow while
o Be alert and speak speaking.
laud enough to
your audience

Compiled By Solomon Kebede, MSc, 277


PhD Candidate, 2019
Do’s
o Alter the tone and Don’ts
pitch of your voice, o Present in a
length of sentence monotone voice
and facial/hand with no facial/hand
gestures to gestures.
maintain audience o Present everything
interest. in your thesis
o Be selective in your
presentation
o Stick to time
allowed to your
presentation.
Compiled By Solomon Kebede, MSc, 278
PhD Candidate, 2019
Presenting Your Thesis

...do not distract the audience

with silly animations

or with unrelated information

Drink more beer

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 279
Good Luck In Your Research!!

Compiled By Solomon Kebede, MSc,


PhD Candidate, 2019 280

You might also like