Chapter Four:

Research Methodology

07/31/2021 Admas University 1

 Data Sources

Primary Data Secondary Data

Observation Interviewing Questionnaire Documents

Data Collection Instruments

Govt.
Non-Participant

Semi-Structured

publications
Un-structured
Participant

Structured

Collective
Earlier
Mailed

Direct
or
Researches
Census
Personal Records
07/31/2021 Admas University 2
Observation
 Observation involves the collection of information by
a way of investigators’ own observation, without
interviewing the respondents.

 Sometimes, observation is divided into two types

◦ Participant observation and
◦ Non-participant observation

 Observation also can be of two types

◦ Structured observation
◦ Unstructured observation
07/31/2021 Admas University 3
 Structured and unstructured observation
 Structured observation is a type of observation in
which the researcher decided in advance issues
pertaining to
◦ What should be observed?
◦ How the observations should be recorded?
◦ How the accuracy of observation can be ensured?

 Unstructured observation takes place when the

abovementioned issues are not thought in advance.

 Structured observation is considered appropriate in

descriptive studies, whereas in an exploratory study the
observational procedure is most likely to be relatively
unstructured. 07/31/2021 Admas University 4
 Participant and non-participant
observation
 This distinction depends upon the observer’s sharing
or not sharing the life of the group he is observing.

 An observation is participant observation if the

observer observes by making himself a member of the
group he is observing so that he can experience what
the members of the group experience.

 An observation is non-participant if the observer

observes as a detached emissary without any attempt
on his part to experience through participation what
others feel.

07/31/2021 Admas University 5

Cont'…
 But, if the observer is observing in such a manner that
his presence may be unknown to the people he is
observing, such an observation is described as
disguised observation.

07/31/2021 Admas University 6

Cont'…
 Advantages of observation
◦ Overcomes language barrier.

◦ Can be carried out any time.

◦ Helps researcher to get in touch with the

information directly.

◦ More reliable technique.

07/31/2021 Admas University 7

Cont'…
 Disadvantages of Observation
◦ Misinterpretation of what is observed.

◦ More expensive way of collecting information than

the questionnaires.

◦ Time consuming.

◦ Cannot be applied to many aspects of social life.

For instance, one cannot observe attitude and
beliefs.

07/31/2021 Admas University 8

Interview Method
 Interview involves presentation of oral-verbal
stimuli and reply in terms of oral-verbal responses.

 An interview is a conversation between two people

(the interviewer and the interviewee) where questions
are asked by the interviewer to obtain information
from the interviewee.

 Thismethod can be used through personal interviews

and, if possible, through telephone interviews.

07/31/2021 Admas University 9

Personal interview
 Personal interview method is a type of interview
which requires a person known as the interviewer
asking questions generally in a face-to-face contact
to the other person or persons.

 This sort of interview may be in the form of direct

personal investigation or it may be indirect oral
investigation.

 In the case of direct personal investigation, the

interviewer has to collect the information
personally from the sources concerned.
07/31/2021 Admas University 10
Cont'…
 But in certain cases it may not be possible or
worthwhile to contact directly the persons concerned
or on account of the extensive scope of enquiry, the
direct personal investigation technique may not be
used.

 In such cases, an indirect oral examination can be

conducted under which the interviewer has to
cross-examine other persons who are
supposed to have knowledge about the problem
under investigation and the information obtained is
recorded.
07/31/2021 Admas University 11
Types of personal interviews
 Structured interview is the one in which there is a
set of predetermined questions.

 The interviewer in a structured interview follows a

rigid procedure laid down, asking questions in a
form and order prescribed.

 Unstructured interview is a method of interview

where there is flexibility in the approach to
questioning. It does not follow a system of
pre-determined questions and standardized
techniques of recording information.
07/31/2021 Admas University 12
Cont'…
 In a non-structured interview, the interviewer is
allowed much greater freedom to ask, in case of need,
supplementary questions or at times he may omit
certain questions if the situation so requires.

07/31/2021 Admas University 13

Telephone interview
 Telephone interview involves contacting the
respondents on telephone.
 It plays an important role when the survey has to be
accomplished in a very limited time.
Advantages of Interview
◦ Interview is quite flexible.
◦ Adaptable and can be used in many people.
◦ Information can be obtained in a detail and well-
explained.

07/31/2021 Admas University 14

Cont'd…
 Disadvantages of Interview
◦ Interviewees may feel shy or become discomfort
about the question.
◦ Sometimes questions may be wrong and may
take a lot of time from the respondent to give
respond.
◦ Untrained interviewers may be one of the major
weaknesses or demerits of the interview.

07/31/2021 Admas University 15

 Cont'd…
Advantages of telephone interview:
• Relatively cheap and quick (as long as the
respondent picks and is willing to take the
interview)
• Can cover reasonably large numbers of people or
organizations.
• Wide geographic coverage
• No waiting and spontaneous response
• Help can be given to the respondent
• Can tape answers
Cont'd…
Disadvantages of telephone interview:
• Questionnaire required
• Not everyone has a telephone.
• Repeat calls are inevitable
• Straightforward questions are required to ask (canno
address complex issues)
• Respondent has little time to think on an issue
• Good telephone manner is required
Focus Group Discussion

A focus group could be defined as a group of

interacting individuals having some common
interest or characteristics, brought together by
a researcher, who uses the group and its
interaction as a way to gain information about
a specific or focused issue
Con’d…
Advantages of FGD method:
 People naturally interact and are influenced by others
( high validity).
 Provide data more quickly and at lower cost than if
individuals are interviewed on one-to-one basis.
 Generally requires less preparation and is comparatively
easy to conduct.
 Researcher can interact directly with respondents ( allows
clarification, follow-up questions, probing)
 Data uses respondents’ own words: can obtain deeper
levels of meaning, make important connections
 Very flexible; can be used with wide range or topics,
individuals, and settings
Con’d….
Disadvantages of FGD method:
 Researcher have less control over group; less able to control
what information will be produced.
 Data analysis is more difficult (narrative, textual).
 Small sample size & sampling method severely limit ability
to generalize to larger populations.
 Requires carefully trained interviewer who is knowledgeable
about group dynamics.
 Researcher Bias: He/she may provide clues about what types
of responses are desirable.
 Respondent bias: Results may be biased by presence of a
very dominant or opinionated member; more reserved
members may be hesitant to talk.
Questionnaire
Advantages of questionnaire method:
 Can be used as a method in its own right or as a basis for
interviewing or a telephone survey
 Can be posted, e-mailed or faxed
 Can cover a large number of people or organizations
 Wide geographic coverage.
 Relatively cheap and no prior arrangements are needed.
 Avoids embarrassment on the part of the respondent.
 Respondent can consider responses.
 Possible anonymity of respondent.
 No interviewer bias.
Questionnaire…
Disadvantages of questionnaire method:
Susceptible to design problems.
Questions have to be relatively simple.
Historically low response rate.
Time delay whilst waiting for responses to be returned
Require a return deadline.
Several reminders may be required
Assumes no literacy problems.
No control over who completes it
Problems with incomplete questionnaires.
Steps to design a questionnaire:
1. Write out the primary and secondary aims of
your study.
2. Write out concepts/information to be collected
that relates to these aims.
3. Review the current literature to identify
already validated questionnaires that measure
your specific area of interest.
4. Compose a draft of your questionnaire.
5. Revise the draft.
6. Assemble the final questionnaire.
07/31/2021 Admas University 23
Questionnaire General Format

Self-administered (by mail or

personal contact)
In person (face-to-face)
Telephone interviews
 Questions Context
Standardize the interviewer
Standardize the response format
Ask questions in a social, cultural,
and economic context
Keep confidentiality and voluntary
participation
Include a letter of introduction or
presentation
 Questions Format
Open ended questions - more
information but difficult to
codify, enter, and analyze
Closed ended questions - less
information but easy to codify,
enter, and analyze
 How To Obtain Valid Information

Avoid two-edged questions

Avoid negative questions
Adopt/adapt questions used
successfully in other questionnaires
Secondary Data
Sources of Secondary data:
Federal/provincial/state governments
Statistics agencies and Trade association
General business publications, Annual reports,
Magazine and newspaper articles
Academic publications, Library sources
Computerized bibliographies and syndicated services
Internal records and reports
Cont’d…
Advantages of Secondary data:
It often allows researcher to extend the scope of your
study
It saves time and cost
It usually provides a larger database.
In many small research projects, it is impossible to
consider taking a national sample because of the costs
involved.
cont’d….
Disadvantage of secondary data:
Researcher has less control over how the data was
collected.
There may be biases in the data, but not known.
Data may not exactly fit your research questions.
Data may be obsolete; can distort research results
Can also raise issues of authenticity and copyright.
Sample design
Sample defined:
A sample is a subset, or some part, of a larger
population.
A larger population could be anything out
which sample is taken.
A complete group of entities sharing some
common set of characteristics is population.

07/31/2021 Admas University 31

Why sample?
 Saves Cost, Labor, and Time

 Quality Management/supervision

 Accurate and Reliable Results

 Sampling may be the Only Way

◦ For example, consider the case of electric bulbs. In
testing the life of bulbs, if we were to burn every
bulb produced, there would be none left to sell. This
is destructive sampling.
07/31/2021 Admas University 32
Sampling frame
A sampling frame is the list of elements from which
the sample may be drawn. A simple example could be
listing of all University students meeting the criteria
of target population.
 Example; Registrar in a university, payroll in
organizations, telephone directory,
 Sampling frame is useful in providing a listing of each
elements in the population
 A sampling frame is also called the working
population because it provides the list that can be
worked with operationally.
 This is possible only for finite population.

07/31/2021 Admas University 33

 Sampling unit
A sampling unit is that element or set of elements
considered for selection in some stage of sampling.

 Sampling may be done in single stage or in multiple

stages.
◦ Single-stage sample, the sampling units are the
same as the elements.

◦ In more complex samples, however, different

levels of sampling units may be employed – the
case of clustering or stratified techniques
07/31/2021 Admas University 34
Parameter
A parameter is the summary description of a
given variable in a population.
◦ The mean income of all families in a city and
the age distribution of the city’s population are
parameters.
 More precisely, parameters is the characteristics
of a POPULATION

07/31/2021 Admas University 35

Statistic
A statistic is the summary description of a given
variable in a survey sample.
◦ Thus the mean income computed from the
survey sample and the age distribution of that
sample are statistics.

 Like the parameters, statistic stands for

characteristics of a SAMPLE.

07/31/2021 Admas University 36

Sampling…….

STUDY POPULATION

SAMPLE

TARGET POPULATION
The Sample is the Subset of the Population

Sample
Population
Determining Sample Design
In sampling design, pay attention to:
a) Type of universe:- finite or infinite universe
b) Sampling unit:- geographical (state, district, village,
etc.), or a construction unit (house, flat, etc.), or a
social unit (family, club, school, etc.), or it may be an
individual.
c) Source list (sampling frame): is the frame from which
sample is to be drawn.
Determining Sample Design...
d) Size of sample: Sample size is the determination of the
number of elements to be selected from a population to
serve as representatives. It can be done in two ways:
Rule of Thumb and Using Formula
e) Parameters of interest: must consider the question of the
specific population parameters (e.g., mean, median,
mode) .
f) Budgetary constraint: This may even lead to the use of a
non-probability sample.
g) Sampling procedure: for a given sample size and for a
given cost, it should lead to smaller sampling error.
Determining Sample Design...
Systematic Bias & Sampling Errors:
A. Systematic bias: errors in the sampling procedures,
cannot be addressed by increasing sample size.
Sources: 
•Inappropriate sampling frame:
•Defective measuring device;
•Non-respondents:
•Indeterminancy principle: individuals act differently when
kept under observation.
•Natural bias in the reporting of data: people tend to give
what they think is the ‘correct’ answer rather than revealing
their true feelings. E.g., understate or overstate income.
 Determining Sample Design...
B. Sampling errors:
•Related to the sampling procedure and size
•Decreases with the increase in the size of the sample, & in
case of homogeneous population.
•But increasing the size increases the cost & may enhance
the systematic bias.
•Researcher must ensure that the procedure causes a
relatively small sampling error & helps to control the
systematic bias in a better way.
 Determining Sample Design
Characteristics of a good sampling design:
•Must result in a truly representative sample.
•The results of the sample study can be applied, in
general, for the universe with a reasonable level of
confidence.
•Must result in a small sampling error.
•Must be viable in the context of:
•Funds available for the research study and
•Practicality to pick the selected elements of the
sample.
•Must enable controlling systematic bias in a better way.
 Sampling Techniques

Probability Non-probability
Sampling Sampling

Simple Systematic Stratified

Cluster
Random Random Random
Sampling
Sampling Sampling Sampling

Convenience Snowball Quota

sampling sampling sampling

07/31/2021 Admas University 44

Determining Sample Design...
Random/Probability Sampling Techniques
1. Simple Random Sampling:
• Every member of the population has the same chance o
being included in the sample.
• Techniques: Lottery method, Random numbers.
2. Systematic Random Sampling:
• Used to select, say, every 20th name on a list, every 12th
house on one side of a street, every 50th piece of item
coming off a production line, and so on.
•Requires a source list
• The first item can be selected using simple Random
Sampling
Determining Sample Design...
3. Stratified Random Sampling:
• Involves dividing the population into relatively
homogeneous groups, called strata,
• Example:
• Stratifying households by income level - low
income HHs; middle income HHs; high income
HHs
• Stratifying by gender, education level, etc
• Then, to select individuals from each strata, we can
use simple or systematic random sampling techniques.
Determining Sample Design...
4. Cluster Sampling (Multi-Stage Sampling)
The total population is divided into relatively small
subdivisions, called clusters,
Some of these clusters are randomly selected using
simple or systematic random sampling.
One possibility is to use all the elements in the
selected (sample) clusters.
If elements within the sample clusters give similar
results, we take a random sample of elements from
each of the selected clusters using simple or systematic
random sampling techniques.
 Cluster sampling
Section 1 Section 2

Section 3

Section 5

Section 4
 Determining Sample Design...
Non-random Sampling Techniques
1. Convenience, Haphazard or Accidental sampling
•members of the population are chosen based on their
relative ease of access) – selection could be biased.
2. Judgmental sampling or Purposive sampling
•The researcher chooses the sample based on who
he/she thinks would be appropriate for the study
•This is used primarily when there is a limited
number of people that have expertise in the area
being researched.
•can be subject to bias and error.
Determining Sample Design...

3. Case study (selection of a case or

multiple cases) - The research is limited
to one group, often with a similar
characteristic or of small size.
4. Deviant Case - Get cases that
substantially differ from the dominant
pattern (a special type of purposive
sample).
Determining Sample Design...

5. Ad hoc quotas - A quota is established and

researchers are free to choose any respondent
they wish
6. Snow-ball sampling: one respondent leads to
others; useful when the researcher does not
know where the potential respondents could be
found; their number is unknown and no source
list.
 Census and sampling survey
A complete enumeration of all items in the ‘population’ is
known as a Census.
 All items in any field of inquiry is a population or universe.
 Under census, the assumption is that no element is left and highest
accuracy is obtained.
 Alternatively,sample respondents can be selected from
the population.
 The respondents selected should be as representative of
the total population as possible.
 The selected respondents constitute what is technically
called a ‘sample’ and the selection process is called
‘sampling technique.’ The survey so conducted is known as
‘sample survey’.
07/31/2021 Admas University 52
 The Survey Method
A Survey:
Is detailed and quantified description of a
population (i.e., a set of people, organizations,
institutions, countries, etc)
Involves systematic collection of data by
interview, questionnaire or observation
methods,
Requires precise samples (if it is not a census)
and standardization of instruments.
Survey Method…
Characteristics of Survey Methods
Involves a team of enumerators + supervisors
Large number of observation/data can be collected
within a certain period of time from a relatively large
number of respondents.
The sample can be spread over a wide area - making
the study some what more generalizable
the principal researcher need not spend too much
time in the field
 Survey Method…
Surveys require respondents to answer questions
about their:
◦ Opinions, attitudes, or preferences, and
◦ Socio demographic characteristics.
It is essentially cross-sectional – conducted during a
particular point in time.
It is not concerned with the characteristics of the
individual respondents per se but with that of the
target group (population) as a whole.
 Survey Method…
It is conducted based on:
◦ Clearly defined research problem/questions
◦ Definite research objectives
◦ Well defined concepts, factors, variables, etc
It requires:
◦ Careful design of sampling methods, data collection
instruments;
◦ Careful planning of the data collection process/procedure
before hand.
◦ Careful analysis and interpretation of the data gathered.
◦ Logical and skilful reporting of the findings.
 Survey Method…
Types of Survey Studies based on 3 criteria:
(a) Nature of variables:
i) Status survey – simply reports status, or
ii) Survey research – involves more research objective
than simply reporting status .
(b) Group Measured:
(i) Sample (sample survey), or
(ii) Population (census)
(c) Based on the nature enquiry:
(iii)Descriptive survey, or
(iv)Explanatory/analytical survey
(v) Mixed – descriptive and analytical
Case Studies
Tend to be much more specific in focus than surveys
Can explore many themes & subjects, from a much more
focused range of people, organizations or contexts
Can explore subjects & issues where relationships may
be ambiguous or uncertain (i.e., exploratory)
Can also investigate causal relationships (i.e.,
explanatory)
Often (but not always) tend to be inductive rather than
deductive in character.
Not longitudinal study; focuses on contemporary events
 Case Studies…
Itgives a detailed knowledge about the phenomena
and not able to generalize beyond the knowledge.
Does not require control of behavioral events
Prediction cannot be made; no statistical inferences
can be drawn from the exploration of a phenomenon
Should be used when addressing ‘how’ and/or ‘why’
questions
Can be single case study (one case) or multiple case
study (two or more cases).
◦ One case- one person/household/school/event, etc
◦ Multiple case- multiple persons/households/schools/ events
 Case Studies…
Sources of data in case-study:
Documentation – policy/proclamation, specific
report of organizations, etc
Archival records – e.g., from national archives
Interviews – face-to-face, telephone, etc
Direct observations – participant/non-participant
Focus group discussions
Physical artifacts – e.g., costumes, household
utensils, hunting/production tools, etc
 Case Studies…
The multiple case-study process: Steps
1. Develop a theoretical stance - A provisional
hypothesis or set of questions is developed.
2. Select multiple cases – depending on the unit of
analysis (e.g., persons, organizations, schools)
3. Design and pilot research tools, protocols and field
procedures
4. Conduct the multiple case studies (i.e., case 1, case 2,
case 3, …)
5. Draw cross-case conclusions (synthesize/combine)
and develop theories/hypothesis (hypotheses)
6. Write the case study report
 Case Studies…
Guidelines for Case Studies: the case study must
Be significant: There must be a convincing reason for selecting
the case/cases from among similar other cases.
Be complete: the one case/cases must provide a
complete (not partial) view of the phenomenon
Display sufficient evidence: all evidence required to
describe/analyze the phenomenon must be obtained
from the selected case/cases. Otherwise:
◦ Collect additional evidence from the selected case/cases
◦ Add more cases into the study to provide the missing evidence
Be written in an engaging way (e.g. for policy
development or for further research such as surveys)
Measurement and scaling concepts
 Measurement can be qualitative or quantitative.
◦ Qualitative measurement focuses on
interpretation and analysis findings textually.
◦ Quantitative measurement is numerical.
Analysing quantitative data
Whilst measuring quantitative data, measurements
can be nominal, ordinal and interval and ratio
levels of measurement.

07/31/2021 Admas University 63

A) Nominal scales
 Nominal scales classifies elements into two or more
categories.
 In simple terms, it is a type of data which reflects
classification characteristics, but do not indicate any
mathematical or qualitative differences.
 When the data is nominal, it is meaningless to find
mean, standard deviations, correlation coefficients,
etc.
 Example: Gender, Political Party, Nationality,
Department, e.t.c

07/31/2021 Admas University 64

Example
 Grouping into categories based upon political party
preference (EPRDF, PP, Others) or upon sex (Male or
Female) or upon religion (Muslim, Orthodox,
Protestant, Catholic, etc).

 Inthe political party preference system EPRDF might

be assigned the number "1", PP "2", and Others "3",
while in the latter females might be assigned the
number "1" and males "2".

07/31/2021 Admas University 65

Cont'…

07/31/2021 Admas University 66

b) Ordinal Scales
 Ordinal scale possesses the property of magnitude.
 It classifies scores in the algebra of inequalities (a <
b < c) (I.e a is not equal to “b” and b is not equal
to “a” etc)
 Ordering, ranking, or rank ordering is involved.
Examples: the ranking of people for height, weight,
education, job category, etc.
 In ordinal scales, the numbers attached to values
might indicate a ranking or ordering of the values.
 Can apply the median, rank order, correlations and
percentile.
07/31/2021 Admas University 67
c) Interval Scales
 Establishes an equal unit in the scale. The difference
between any two scores is equal.
 Suitable to calculate arithmetic mean and standard
deviations.
 The classic example of an interval scale
is Celsius temperature because the difference
between each value is the same.  For example, the
difference between 60 and 50 degrees is a measurable
10 degrees, as is the difference between 80 and 70
degrees.  
 Other examples: five liker scale, exam results,
attitudinal results and most behavioral concepts
07/31/2021 Admas University 68
Examples

07/31/2021 Admas University 69

D. Ratio Scales
The highest level of measurement scales.
It is the most powerful of the other
measurement scales
Example
◦ Salary, financial performance, annual budget,

07/31/2021 Admas University 70