Chapter 4

Research Methodology
 Learning Objectives

After completing this chapter the student will be able to :

 Understand sampling design

 Identify the data collection methods
 Know how to design a well-structured questionnaire
 Assess the appropriate use of quantitative and qualitative
data analysis methods
 Introduction
Overview of Research methodology
• The main aim of research is not merely to gather information. Instead, it
goes beyond that. The true goal of research is to seek answers to
previously unanswered questions to contribute to the body of knowledge
in a discipline (Goddard and Melville, 2001).
• To have your newly discovered knowledge recognized by the scientific
community, you need to provide evidence to demonstrate its validity and
truthfulness. The scientific community values empirical evidence and
rigorous testing, so presenting a well-designed study or experiment that
supports your findings is crucial.
• Without such evidence, your discoveries are unlikely to be accepted and
integrated into the broader body of scientific knowledge. The key is to
follow the established norms and practices of the scientific method to
build a compelling case for the legitimacy of your work.
• Research methodology: Science of understanding how research is
performed methodically.
• It provides the reasons why certain methods are used in the process.
• Simply put, the research methodology section must shed light on how
you are able to collect or generate your research data and demonstrate
how you analyze them.
 Introduction

• It is worth noting that the exact

components/contents/structure of the methodology
chapter will vary depending on the field of research as well as
the university. So, it is always a good idea to check the
guidelines provided by your institution for clarity and, if
possible, review past dissertations and theses from your
university.
 General structure for a research methodology
 Introduction:
 In this section, you should remind the readers what the
focus of your study is, especially the research aims. Your
research design needs to align with your
research aims, objectives and research questions, so it is
useful to frontload this to remind the reader (and yourself!)
what you are trying to achieve with your design and
methodology.
 In this section, you can also briefly mention how
 Introduction
 Research Methodology Design
 In this section, the key research design choices should be
explained and justified in a logical manner.
 you need to get specific.
 most common research design choices:
o Research Philosophy (positivism/interpretivism/pragmatism),
inductive versus deductive, Exploratory /Descriptive/
Correlational/ Experimental, cross-sectional/longitudinal,
population and study area, study variables, sampling strategy
(sample size determination/ sampling techniques), data
collection methods (interviews/FGD/observations/…), data
analysis methods.
 Introduction
 Methodological Limitations

• With the key research design choices outlined and justified, the next
step is to discuss the limitations of your design. No research design
or methodology is perfect – there will always be trade-offs between
the “ideal” design and what is practical and feasible, given your
constraints. Therefore, this section of your methodology chapter is
where you will discuss the trade-offs you had to make, and why
these were justified given the context.
• It is important to be critical of the shortcomings of your study.
There is no use trying to hide them (your examiner will be aware of
them regardless).
• By being critical, you will demonstrate to your examiner that you
have a strong understanding of research design.
• State the limitations, why these were justified, how
you mitigated their impacts to the best degree possible, and how
your study still provides value despite these limitations.
• Methodological limitations can vary greatly from study to study,
ranging from common issues such as time and budget constraints
to issues of sample or selection bias.
 Introduction
Illustrative Example

Let us consider a hypothetical study aimed at understanding the impact of remote work on employee
productivity.
Ideal Research Design
•Research Question: How does remote work affect employee productivity?
Ideal Methodology:
Type of research: A longitudinal study that tracks the same group of employees over several years.
Sample Size: A large, diverse sample of employees from various industries to ensure generalizability.
Data Collection methods: Use a combination of qualitative and quantitative data collection methods to
gather information from employees and employers.
Data Analysis methods: Use advanced statistical methods to analyze the data, ensuring robust
conclusions.
 Trade-offs and Practical Constraints
1. Time and Resources:
• Ideal: A longitudinal study is time-consuming and requires significant resources to track participants
over years.
• Practical: Instead, the researcher opts for a cross-sectional study that surveys employees at one point in
time. This is quicker and less expensive but may miss changes in productivity over time.
2. Sample Size and Diversity:
• Ideal: A large, diverse sample would provide more reliable and generalizable results.
• Practical: Due to budget constraints, the researcher can only survey a small group of employees and
employers from one company.
 Introduction
 Concluding Summary

 In this section, you should concisely summarize what you

have presented in the chapter.
 It can be useful to use a figure to summarize the key design
decisions(for example, Saunders’ Research Onion).
 Importantly, this section needs to be brief – no longer than a
paragraph or two.
 Include only what you have already discussed in your
chapter; don’t add any new information.
 Introduction
Basic Concepts
Census and Sample Survey
 Surveys are used as a tool to collect data from some or all units of a
population and compile the information into a useful form. There are two
different types of surveys that can be used to collect data in different
circumstances to satisfy differing needs. These are sample survey and
censuses survey.
 Population is the entire set of units/observations being studied (it
composed of all observations on the entire group under our
consideration).
Census survey: A complete enumeration of all items in the target
population.
 When all items are covered, no element of chance is left and highest
accuracy is obtained.
 This type of inquiry involves a great deal of time, money and energy.
Sample survey: In a sample survey, only part of the total population is
approached for information on the topic under study. These data are then
'expanded' or 'weighted' to make inferences about the whole population.
 Sample is the set of observations taken from the population for the
purpose of obtaining information about the population.
 Introduction
 Once a sample of the population is determined, data is collected for
the sample members.
 This data can include both quantitative and qualitative information,
which can then be analyzed to serve three key purposes: description,
inference, and interpretation.
 Descriptive analysis: Summarizes the quantitative data numerically or
graphically to describe observed patterns and trends in the
sample/population. For qualitative data, it involves identifying recurring
themes, categories, and narratives to provide a detailed understanding
of the sample/population.
 Inferential analysis: Uses patterns and relationships observed in the
sample data to make generalizations or draw conclusions about the
broader population, accounting for the variability and uncertainty
inherent in the sampling process.
 For a sample to be used as a guide to an entire population, it is important that it is truly a
representative of that overall population.
 Appropriate and scientific sampling procedures assure that the inferences and
conclusions can be safely extended from the sample to the population as a whole.
 Interpretative analysis: Focuses on deriving deeper insights from
qualitative data by interpreting meanings, contexts, and underlying patterns to
provide a richer understanding of the experiences or phenomena being
studied.
 Introduction
• The raw materials for any research analysis are the data.
• We collect relevant data using appropriate methods.
• Once the tedious task of collecting data is completed, this
collection of raw data in itself reveals very little.
• It is extremely difficult to determine the true meaning of a
bunch of numbers or texts that have simply been recorded on a
piece of paper.
• It remains for us to organize and describe these data in a
concise and meaningful manner.
 In order to determine their significance, we must display the
data in the form of tables, graphs and charts (so that we can
have a good overall picture of the data).
• Then, we have to analyze the data thruogh summary measures
such as the mean and standard deviation; assess the extent of
relationship (correlation) between two (or more) variables;
identify themes or patterns, and the like.
• Finally, based on the analysis, we have to make generalizations
and arrive at reasonable decisions.
 Introduction
 Variables
 Categorical
 Numerical
 Discrete
 Continuous

 Scales of measurement

• Nominal: The nominal scale assigns numbers as a way to label

or identify characteristics. The numbers assigned have no
quantitative meaning beyond indicating the presence or
absence of the characteristic under investigation. he numbers
we assign for the various categories are purely arbitrary, and
any arithmetic operation applied to these numbers is
meaningless.
 Introduction

• Ordinal: The ordinal scale ensures that the possible categories

can be placed in a specific order(rank) or in some ‘natural’
way; categories that have an inherent level. The numbers are
not obtained as a result of a counting or measurement
process, and consequently, arithmetic operations are not
allowed.
• Interval/ratio: the numbers/data in these scales are obtained
as a result of a measurement process; natively numeric data.

 Why is level of measurement important?

 knowing the level of measurement helps you decide on how to

interpret the data. For example, if your measure is nominal,
then you know that the numerical values are just short codes
for (qualitative) categories.
 knowing the level of measurement helps you decide what type
of statistical analysis is appropriate.
 Introduction

Variable

Categorical Quantitative/
numerical (Discrete
or Continuous)

Nominal Ordinal Ratio Scale

Interval Scale
(ordered)
(not ordered)

14
 Introduction
 Tips on the types of variables
• Researchers often manipulate or measure independent and
dependent variables in studies to test the cause-and-effect
relationships.
• Researchers who focus on causal relations usually begin with an
effect, and then search for its causes.
Dependent Variable: is a variable that is being tested and
measured in a research work (it is also referred as
response/outcome variable/effect variable).
Independent Variable: As per the name, an independent variable
(IV) stands alone. The value does not change due to the impact of
any other variable. The researcher manipulates or changes the
independent variable to measure its impact on other variables.
o It is what we expect will influence the dependent variable (the
cause variable).
• The independent and dependent variables may be viewed in terms
of cause and effect, respectively. If the independent variable is
changed, then an effect is seen in the dependent
variable. Independent variable is “independent of” prior causes that
act on it, whereas the dependent variable “depends on” the cause.
 Introduction
• Moderating Variable: is one that has a strong contingent
effect on the independent variable-dependent variable
relationship (i.e., the presence of a third variable (the
moderating variable) modifies the original relationship between
the independent and the dependent variable.
• A moderating variable (or moderator) affects the strength and
direction of relationship between the two variables.
• A Moderating variable is used to check whether the third
variable influences the strength or direction of the relationship
between an independent and dependent variable.
• An easy way to understand this is that the moderator variable
might change the strength of a relationship from strong
to moderate, to nothing at all. It is almost like a turn dial on the
relationship; as you change values of the moderator, a statistical
relationship that you observed before might dissolve away.
 Introduction
For Example,
o A strong relationship has been observed between the quality
of library facilities (X) and the performance of the
students (Y).
o Although this relationship is supposed to be true generally, it
is nevertheless contingent on the interest and inclination of
the students.
o It means that only those students who have the interest and
inclination to use the library will show improved performance
in their studies.

• Mediating variable (or Mediator): explains the process

through which two variables are related.
• explains the relationship between independent and
dependent variables.
• For instance, imagine that you find a positive association
between note-taking and performance on an exam. This
association may be explained by study hours, which would
be the mediating variable.
 Introduction

DV: Purchasing intention; IV: Perceived originality; Mediator: Attitude

towards the brand; Moderator: Resistance to advertising
Introduction
 Estimation of sample size

Anyone designing a survey must decide on:

 what amount of sampling error in the estimates is tolerable
and,
 must balance the precision of the estimates with the cost of
the survey.

 Even though many variables may be measured, an

investigator can often focus on one or two responses that
are of primary interest in the survey, and use these for
estimating the sample size.
In designing sample surveys, instead of asking about the
required precision, many people ask:
 ‘What percentage of the population should I include in the
sample?’
This is a wrong question since precision is obtained through
the absolute size of the sample – not the proportion of the
population covered.
 Estimation of sample
size
Formulas for sample size determination often involve certain
unknown properties of the population.

These must be estimated in order to give specific results.

 Some methods of obtaining such estimates are:
a) Use the results from previous sampling of the same or a
similar population.
b) A pilot sample (survey), a small sample taken to provide
information and guidance for the design of the main survey,
can be used to estimate quantities needed for setting the
sample size.
c) Apply guesswork about the structure of the population
assisted by some mathematical results.
 For example, if you believe the population to be normally
distributed and have an idea of the range of the data, then
estimate the standard deviation as: S = Range/4
 Estimation of sample size
Sample size determination for numerical data

where

For unknown population, sample size is:

n is the sample size desired

α is the level of significance
Z α/2 is the critical value that cuts off an area of α/2 in the normal curve (the
standard normal deviant)
 is the standard deviation
E is an error of estimation/margin of error/allowable error
 Estimation of sample size
Sample size determination for
proportions

Where

For unknown population, sample size is:

here, P is the proportion of category of interest in the response variable

 Estimation of sample size
Yemane’s Formula

Taro Yamane (Yamane, 1967,73) derived the following formula:

Most common in business research works

 Sampling Techniques
 Sampling: is a process of selecting representative observations
from the target population.
 a sample should have the same characteristics as the
population it is representing.
 To draw valid conclusions from your results, you have to
carefully decide how you will select a sample that is
representative of the target population. This is called
a sampling technique/method.
 There are several reasons for taking a sample instead of a
complete enumeration of the whole population or census. These
include:
 A census may be very expensive.
 A census may require too much time.
 A census can not be applied for infinite population
 A carefully obtained sample may be more accurate than a
census.
• For example, in a large inventory census or in a complete audit,
errors due to fatigue or carelessness on the part of the census
 Sampling Techniques

 Broadly speaking, there are two types of sampling techniques:

random sampling and non-random sampling.
 In random sampling, the elements to be included in the sample
entirely depend on chance.
 Random sampling techniques often yield samples that are
representative of the population from which they are drawn.
 In non-random sampling, the units in the sample are chosen by the
investigator based on his/her personal convenience and beliefs.
 Random sampling
 Simple: This is a method of sampling in which every member of the
population has the same chance of being included in the sample.
 Random number generators

 Systematic: An initial starting point is selected by a random process,

and then every nth number on the list is selected.
 In some instances, the most practical way of sampling is 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. This is called systematic sampling.
 Sampling Techniques

 Stratified: the procedure of dividing the population into

relatively homogeneous groups, called strata, and then
taking a simple random sample/systematic random
sample from each stratum.
 If the population elements are homogeneous, then there is
no need to apply this technique.
 Example of stratified random sampling: If our interest is the
income of households in a city, then our strata may be: –
low income households – middle income households – high
income households
 To obtain a sample from each stratum: – Take a sample of
size proportional to the sub-population (stratum) size, i.e.,
draw a large sample from a large stratum and a small
sample from a small stratum– This is known as proportional
allocation.
 Sampling Techniques
 Cluster:
 This is a method of sampling in which the total population is divided into
relatively small subdivisions, called clusters, and then some of these
clusters are randomly selected using simple random sampling.
 Once the clusters are selected, one possibility is to use all the elements in
the selected clusters.
 However, this seems uneconomical. Instead, we take a random sample of
elements from each of the selected clusters (called two-stage sampling).
Example: Suppose we want to make a survey of households in Addis Ababa.
 Collecting information on each and every household is impractical from the
point of view of cost and time.
 What we do is divide the city into a number of relatively small sub-divisions,
say, Kebeles. So the Kebeles are our clusters.
 Then we randomly select, say, 20 Kebeles using simple random sampling.
 To collect information about individual households, we have two options:
a) We visit all households in these 20 Kebeles
b) We randomly select households from each of these 20 selected Kebeles
using simple random sampling–This method is called two-stage sampling
since simple random sampling is applied twice (first, to select a sample of
Kebeles and second, to select a sample of households from the selected
Kebeles)
 Sampling Techniques

 Non-random Sampling Techniques

 Convenience or Accidental sampling: members of the
population are chosen based on their relative ease of access.
 Purposive or Judgment sampling: The researcher chooses the
sample based on who he/she thinks would be appropriate for the
study.
 Purposive sampling starts with a purpose in mind and the
sample is thus selected to include people or objects of interest
and exclude those who do not suit the purpose.
 Purposive sampling can be subject to bias and error.

 Quota sampling: A quota is established and researchers are

free to choose any respondent they wish as long as the quota is
met.
 Sampling Techniques
 Snowball sampling: is particularly appropriate when the
population you are interested in is hidden and/or hard-to-
reach (e.g. samples of the homeless, criminals, or users of
illegal drugs).
 Snowball sampling asks respondents to recommend other
respondents who might subsequently be invited to take the
survey; It is a technique where existing respondents recruit
future respondents from among their connections.
 NB: The actual process of sampling causes sampling
errors. For example, the sample may not be large enough
or representative of the population. Factors not related to
the sampling process cause non-sampling errors. A
defective counting device can cause a non-sampling error.
Question: Describe the advantages and disadvantages of
random and non-random sampling techniques.
 Sampling Techniques

Classification of Sampling Methods

Sampling
Methods

Probability Non-
Samples probability

Systematic Stratified Convenience Snowball

Simple
Cluster Judgment Quota
Random

31
 Sampling Techniques
 Please note that:

• Non-probability sampling techniques cannot be used to infer from the

sample to the general population.
• Does that mean non-probability samples aren't representative of the
population? Not necessarily.
• With non-probability samples, we may or may not represent the population
well, and it will often be hard for us to know how well we have done so.
• Researchers prefer probabilistic or random sampling methods over non-
probabilistic ones, and consider them to be more accurate and rigorous.
• A sampling error is an error that occurs when the sample used in the
study is not representative of the whole population. In line with this, the
results found in the sample do not represent the results that would be
obtained from the entire population.
• A sampling error is an error in a survey that occurs because a sample is
used; a sample is only an approximation of the population from which it
is drawn.
• Sampling errors can be reduced by increasing the sample size. As the
sample size increases, the sample gets closer to the actual population,
which decreases the potential for deviations from the actual population.
 Sampling Techniques

• Random sampling is an additional way to minimize the occurrence

of sampling errors. Random sampling establishes a systematic
approach to selecting a sample.
• Sampling errors is unavoidable but knowledge helps minimize
it.
 Causes of sampling error:
• The study did not use a representative sampling method. Hence,
the study obtains a sample that misrepresents the population.
• Non-response bias occurs when the observations with particular
characteristics do not respond. The sample might reflect the
population, but the responses do not.
• The study sampled the wrong population. The study might have
used a representative sampling method but somehow drew a
sample from the incorrect population.
 Using a perfect or precise sample from the wrong population is a
potential cause of sampling error. This can happen in several
ways:
 Sampling Techniques
1. Incorrect Definition of the Target Population
•If the target population is not well-defined or misunderstood, the sample may be
drawn from a group that does not represent the intended population.
Example: A study aiming to assess the preferences of urban youth may mistakenly
include suburban or rural youth, leading to a mismatch between the study’s goals
and the sample collected.
2. Inaccurate Sampling Frame
•A sampling frame is the list or source from which a sample is drawn. If the frame
is not aligned with the target population, the sample will not represent the intended
group.
Example: Using a database of private school students to study educational
attainment in all schools (public and private).
3. Misinterpretation of Study Scope
•Researchers might select a sample that fits a specific group without realizing that it
excludes other critical segments of the intended population.
Example: Studying consumer preferences for a product but sampling only from
premium stores, neglecting the general consumer base.
 Sampling Techniques
4. Operational Errors
Field staff might misunderstand instructions and collect data from the wrong
geographic location, age group, or demographic segment.
Example: A survey designed for 18-25-year-olds being conducted with respondents
aged 30+ due to unclear guidelines.
Implications and Prevention
Implications: Results derived from such samples may not generalize to the
intended population, leading to flawed conclusions or recommendations.
Prevention:
Clearly define the target population.
Use a valid and reliable sampling frame.
Train field staff to avoid operational errors.
Review the sampling process rigorously before data collection.
 Questionnaire Design Process
 Questionnaire: is a systematic data collection tool that consists of a series of
questions which help in achieving research objectives.
 Information which can be obtained from questionnaire:

o Demographic and Personal Information: age, gender, educational level,

occupation, income, and other biographical details.
o Knowledge and Awareness: Understanding of specific topics, events, or
concepts.
o Behaviors and Actions: Past activities, habits, or routines; Current behaviors and
practices.
o Opinions and Perceptions: Preferences, judgments, and views on particular
topics or issues.
o Attitudes and Sentiments: Emotional responses, levels of agreement,
satisfaction, or approval.
o Motivations and Reasons: Factors influencing decisions, actions, or preferences.
o Intentions and Future Plans: Likely future behaviors, goals, or aspirations.
o Experiences: Past encounters, challenges, or achievements in various contexts.
 Questionnaire Design Process

 Assignment for presentation: Discuss briefly on the

Questionnaire Design Process.

 Steps involved in the development of a questionnaire:

1. Decide on the information required

2. Identify the target respondents.
3. Specify the type of interview method
4. Determine the content of individual questions
5. Develop the question wording.
6. Put questions into a meaningful order and format.
7. Check the length of the questionnaire.
8. Pre-test the questionnaire.
9. Develop the final survey form.
 Questionnaire Design Process
General advice about questionnaire design process
1. Your questionnaire must tightly align with your golden thread (research
questions, and aim and objectives)
o Keep your study’s research questions and objectives front of mind during
questionnaire design process
o Each question should help achieving your objectives
2. Your questionnaire should contain a clear introduction on the cover page
(opening page); the introduction should be informative and stimulate
respondents’ interest:
o interviewers give the respondent their name and provide identification
o explain that a survey is being conducted
o describe the survey's purpose
o give the respondent time to read or be informed about confidentiality
issues, etc
 Inform the respondent of the estimated time needed (10-20 minutes)
 Missing data may occur if the questionnaire requires more time
3. Be careful not to get distracted by irrelevant variables when designing your
questionnaire
 Including irrelevant variables can lead to confusion and may result in misleading or
unreliable data, thus impacting the overall accuracy and validity of your research findings.
4. The midsection of your questionnaire should focus on the core
variables/constructs of your research work.
 Questionnaire Design Process
5. Your questionnaire design must adhere to all ethics policies and
data protection laws.
o Seriousness on ethical considerations and all data collected should be
handled and stored securely
6. Avoid lengthy questions
o try to make the questions as concise as possible.
o The questions should be read well and have a good flow.
7. Stick to plain language that all respondents can understand.
o provide simple and clear definitions of any jargon/complex language
o The words should be simple, direct and familiar to all respondents.
o be sure to carefully edit and proof-read your questionnaire before
launching it (grammar, spelling, …)
8. Create logical flow of questions
o The questions should read well and have a good flow.
o keep your Likert scales ordered in the same fashion and a consistent
format (coding…)
o Be aware that Likert scales can be problematic if they have too few
or too many options.
o If the questions are close-ended, the response categories should be
mutually exclusive and exhaustive?
 Questionnaire Design Process
9. Provide a progress indicator and use headings to help orient your respondents.

 A progress indicator serves as a visual cue that informs respondents about their
advancement within the questionnaire. This feature not only alleviates anxiety
regarding the length of the survey but also encourages completion by providing a
sense of accomplishment as they move through the questions.
 By categorizing questions under clear and descriptive headings, researchers can
guide respondents through different sections, making it easier for them to
understand the context of each question. This structured approach minimizes
confusion and helps respondents focus on the specific topic at hand, thereby
improving the quality of the data collected.

10. Avoid using loaded questions (complex questions)

o A question that has particular words chosen to suggest the answer that is wanted.
o A loaded question contains assumptions that might force a respondent to agree or
answer in a particular way. These questions can introduce bias and affect the
quality of data collected.
Examples of Leading Questions and Improvements:

Loaded Question: Why do you think our product is the best on the market?
Problem: It assumes the respondent agrees that the product is the best.
Improved Question: How does our product compare to others on the market?
(Options: Much Worse, Worse, About the Same, Better, Much Better)
 Questionnaire Design Process
11. Avoid using leading questions.
oPushes the respondent to answer a certain way.
oLeading questions can result in skewed responses that do not truly reflect
the respondent’s opinion.
Examples of Leading Questions and Improvements:
Leading Question: Most people say this product is affordable. Do you agree?
Problem: It biases the respondent by mentioning others' opinions.
Improved Question: How would you rate the affordability of this product?
(Options: Very Expensive, Expensive, Neutral, Affordable, Very Affordable)
12.Avoid using double-barreled questions.
oA question may contain two (or more) variables which require independent
response
oExample: Does your company provide training for new employees and
retraining for existing staff? This example is double-barreled as it asks
two questions rather than one.
13. Avoid vague questions.
ounclear what is being asked or very open-ended (although open-ended
questions are useful for understanding nuance and finer details)
oonly use open-ended questions if they are central to your research
objectives
14. Be careful when using questions that allow respondents to select
multiples choices
15. Avoid sticking only on 1 or 2 response types that you are comfortable
 Questionnaire Design Process
16. You need to ensure that your questionnaire produces valid and
reliable data (demonstrate the validity and reliability of your
questionnaire).
o Validity: Does the questionnaire actually measure what you are
trying to measure?
o Reliability: Do questions that are measuring the same construct
get the same response?
o You can often modify existing scales to suit your specific research
needs.
o It is completely acceptable to make use of existing scales in your
research (you can use a “tried & tested” scale).
17. Consider your data analysis methods at the questionnaire design
stage
o Questions should be appropriate for the respondent you have in
mind and data analysis methods.
o Keep your analysis methods in mind when designing each
question in a questionnaire (questions should be meaningful and
quantifiable).
 Questionnaire Design Process
o It is a good idea to start with the end (data analysis) when
designing questionnaire
o You should have a clear idea of what statistical methods you
plan to use (think carefully about which descriptive/inferential
methods you will use for each of the constructs)
o If you are not sure about the appropriate methods you will
use, seek out statistical help before launching the your
questionnaire.
18. Make sure you conduct a pilot study to test your survey
questions and responses
o Help purify your questions (you need to make sure that each
question should have a clear focus and a clear point).
o A pilot study will provide you with some data to do preliminary
analysis.
o Use your pilot study data to assess the validity and reliability
o Confirmation for moving forward
 Data Collection Methods
• Data collection is the process of gathering and/or measuring
information on variables of interest, in an established systematic
fashion that enables one to answer stated research questions,
test hypotheses, and evaluate outcomes.
• The collection of data that are relevant to the problem being studied
is commonly the most difficult, expensive, and time-consuming
part of the entire research project.
• The data collection component of research is common to all fields of
study including physical and social sciences, humanities, business,
etc.
• While methods vary by discipline, the emphasis on ensuring
accurate and honest collection of data remains the same.
• Regardless of the field of study or preference for defining data
(quantitative, qualitative), accurate data collection is essential to
maintaining the integrity of research.
• The selection of appropriate data collection instruments (existing,
modified, or newly developed) and clearly explained instructions for
their correct use reduce the likelihood of occurrence of errors.
• Some of the consequences from improperly collected data include:
o inability to answer research questions accurately
o distorted findings resulting in wasted resources
 Data Collection Methods
There are two sources of data:
 Primary data: are collected specifically for the analysis desired
 Secondary data: Previously collected by someone else for a different purpose
(have already been compiled and are available for statistical analysis)

Types of data collection methods

 Data collection methods are fundamental to research and are broadly classified into
quantitative and qualitative approaches. Each approach serves different research purposes
and employs unique methods and tools. Below is a comprehensive overview of data
collection methods:

I) Qualitative Data Collection Methods

 Qualitative data collection focuses on exploring subjective experiences, perceptions, and

meanings. This method provides rich, detailed, and nuanced insights into phenomena.

1)Personal Interviews: One-to-one, back and forth, discussion between the researcher and
the study participant
• Data collection tools are structured forms with closed-ended questions.
 Distribution of questionnaires can be done via: Face-to-face or Telephone.
Face-to-face: This involves trained interviewers visiting the desired people (respondents) in
person to collect data.
o To determine whether workers in a given factory are satisfied with their salary/work or not;
an investigator may contact each worker and ask his/her opinion.
• The advantage of this method is that it ensures a high response rate, and researchers
 Data Collection Methods
Telephone: This involves trained interviewers phoning people to collect
data. This method is quicker and less expensive than face-to-face
interviewing.
oTo determine whether a new product is favored by the public or not, a
company may randomly pick telephone numbers from the telephone
directory, call each of these numbers and then ask the preference or
opinion of the respondents.
2) Focus group discussion: Participants interact with the researcher
(trained interviewer) and each other.
oGroup discussions guided by a facilitator to gather collective
perspectives.
oThe participants can be influenced each other.
3) Document Review: The researcher analyzes existing text (for
example: quarterly and annual reports, historical records, …).
oAnalyzing existing texts, images, videos, or physical objects for patterns
or meanings.

II) Quantitative Data Collection methods

 Quantitative data collection focuses on numerical data that can be

measured and analyzed statistically. These methods are typically
used in research requiring objectivity, generalization, and
reproducibility.
 Data Collection Methods

I) Self-completed (written questionnaire): In this

method, written questions
are mailed or hand-delivered to participants.
 Mail survey: Here questionnaires are mailed to participants and
mailed back to the researcher after completion.
 It is a relatively inexpensive method of collecting data: one can
distribute a large number of questionnaires in a short time.
o This requires the questionnaire to be simple and straightforward
– a major disadvantage of a mail survey is that it usually has lower
response rates than other data collection methods.
o Also, people with limited ability to read or write may experience
problems.

 Hand-delivered questionnaire: This is a self-enumerated survey

where questionnaires are hand-delivered to participants and filled out
or mailed back to the researcher after completion.
o This method usually results in better response rates than a mail
survey, and is particularly suitable when information is needed from
several household members.
 Data Collection Methods

II) Online: An online questionnaire distribution refers to the process of

distributing and collecting forms/questionnaire or assessments electronically
through platforms such as websites, emails, social media, or mobile apps.
This method offers several advantages over traditional methods like Self-
completed questionnaire :
• Cost-effective: Online distribution reduces the need for printing, mailing, and
other physical resources, saving both time and money.

• Wider reach: People from all around the world can participate in online
surveys, increasing the sample size and providing a more diverse range of
opinions.

• Faster response rate: Respondents can complete online questionnaires at

their convenience, leading to quicker turnaround times for data collection
and analysis.
• 5. Enhanced data quality: Digital surveys can include features such as skip
logic, branching, and conditional questions that streamline the process for
respondents, reducing the risk of errors or inconsistencies in responses.
• Environmentally friendly: Online distribution eliminates the need for paper,
ink, and other physical resources, making it an eco-friendly option.
 Other data Collection methods such as experiments can also be used to collect
 Data Preparation and Analysis
Data
Data preparation is crucialPreparation
steps in the research process, ensuring data quality
before analysis. These steps help researchers understand the dataset's structure,
detect errors, and prepare it for statistical modeling.
Poor data preparation can lead to biased results, misleading conclusions, and
invalid research findings.
Data preparation involves: Handling Missing Data, Duplicates, Outliers,
inconsistencies, etc
1) Handling Missing Data
Deletion: Remove rows (listwise deletion) or columns with excessive missing
values.
Imputation: Replace missing values using mean, median, mode, or multiple
imputation.
2) Handling Duplicates
Detecting Duplicates: Check for identical rows using unique identifiers.
Removing Duplicates: Eliminate unnecessary repeated entries while retaining
relevant ones.
3) Handling Outliers
Winsorization: Replace extreme values with some statistical measures such as
percentiles.
Transformation: Use link functions such as Log, square root, or Box-Cox
transformation to reduce skewness.
4)Handling Inconsistent and Anomalous Data
Cross-Checking Values: Verify responses with reference data.
 Data Analysis Methods
Quantitative Data Analysis Methods
• Data Analysis can be explained as the process of discovering useful
information by evaluating data.
• Data Analysis is an important part of research as a weak analysis will
produce an inaccurate report that will cause the findings to be faulty,
invariably leading to wrong and poor decision-making.
• It is, therefore, necessary to choose an appropriate data analysis method
that will ensure you obtain reliable and actionable insights from your data.
• Finding differences, patterns, connections, and relationships in your data
can be a difficult task, but with the correct tools and data analysis
techniques, you can go through the amount of data you have and
generate knowledge about it.
• Quantitative data analysis :Quantitative Data Analysis is a technique
by which an analyst relies on mathematical and statistical calculations,
figures, and models to garner specific data.
o Quantitative analysts aim to use mathematics to represent a given reality
or predict an outcome.
o The purpose of quantitative analysis is to provide an objective prediction
of reality.
o It is based on describing and inferring observations statistically.
o Quantitative data analysis techniques typically work with algorithms,
mathematical analysis tools, and software to gain insights from the data.
 Quantitative Data Analysis Methods

• Types of quantitative data analysis:

 Descriptive data analysis
 Inferential data analysis
Descriptive data analysis
It is used to describe data through tables, graphs, and summary calculations.
 focus on describing the visible characteristics of a dataset (a population or
sample).
 Tables, Chart, Bar chart, Line graphs, Mean, Standard deviation, Skewness.
 Indicate whether your data are appropriate for your research
questions.
(For example: examine whether or not sample size is enough for the groups
to be compared)
 Depending on the research questions, descriptive statistics may be all you
need.
 Know what type of data you are working with
 Know how to describe and explore it as appropriately as possible.
 Help you identify which inferential method is suitable for your data.
 highlight potential relationships between variables
 Use visualization to explore data
 Help you understand the data and identify trends far quicker…histogram,
 Quantitative Data Analysis Methods

Inferential data analysis

 In quantitative data analysis, the expectation is to turn raw
numbers into meaningful insight using numerical values.
 Descriptive statistics is all about explaining details of a
specific dataset using numbers, but, it does not explain the
motives behind the numbers.
 Hence, the need for further analysis using inferential
statistics.
 Inferential data analysis: aims to make predictions about
the target population from the analyzed sample data
obtained from descriptive statistics. The focus is on
making predictions about the target population based on a
representative sample of the population.
 It is used to generalize results and compare several groups,
show relationships that exist between multiple variables,
and are used for hypothesis testing that predicts changes
or differences.
 Quantitative Data Analysis Methods
Overview of Hypothesis
 A hypothesis is a tentative
Testing
assumption about a population
parameter.
o Two types of hypotheses:
a) The null hypothesis () is often established as:
 No significant association between two or several items
 No significant difference between two or several items
 No significant influence of one item on another
o Its mathematical presentation always includes the equality sign.
b) The alternative hypothesis ( or Ha) is the alternative available
when the null hypothesis has to be rejected.
―The sign used in formulating the alternative hypothesis is always
strict inequality.

53
 Quantitative Data Analysis Methods
Overview of Hypothesis
Testing
 Hypothesis testing is a procedure for checking the
validity of a statistical hypothesis.
―It is the process by which we decide whether the null
hypothesis should be rejected or not.
―A value computed from a sample that is used to
determine whether the null hypothesis has to be
rejected or not is called a test statistic.

54
Quantitative Data Analysis Methods
Overview of Hypothesis
Testing

55
 Quantitative Data Analysis Methods

Overview of Hypothesis
Testing

•
 Quantitative Data Analysis Methods

Parametric vs. Non-parametric tests

 Parametric tests:
 Normally involve data expressed in absolute
numbers (interval or ratio scale) rather than
ranks and categories (nominal or ordinal).
 Always assume some distribution about the data
(e.g., a normal distribution).
 If assumptions are correct, these tests can
produce more accurate and precise estimates.
 Example: t-tests, one-way ANOVA, Pearson's
Correlation, Regression, etc.
 Quantitative Data Analysis Methods

Parametric vs. Non-parametric tests

 The t-test (One-sample t-test, Independent sample t-test, and

Paired-sample t-test) and One-way ANOVA examine whether
group means differ from one another.
 The t-test compares two groups, while ANOVA can do more
than two groups.
 The t-test and ANOVA have three assumptions:
1) Independence assumption: the elements of one
sample are not related to those of the other sample.
2) Normality assumption: samples are randomly drawn
from the normally distributed populations with unknown
population means.
3) Equal of variance assumption: the population
variances of the groups are equal.
 Quantitative Data Analysis Methods

Parametric vs. Non-parametric tests

 The Regression model helps the analyst to understand the

relationship of independent variables to a dependent
variable.

o More specifically it focuses on how the dependent

variable changes in relation to changes in independent
variables.

 Assumptions of the Multiple Linear Regression model :

1) Linearity
2) Linearity
3) Multicollinearity
4) Homoscedasticity
5) Autocorrelation
 Quantitative Data Analysis Methods
Parametric vs. Non-parametric tests
 Non-parametric tests:

 Can be used when the test variable is nominal or ordinal data

 Can also be used when the test variable is continuous and if the
assumption of normality of the distribution of the data fails and
sample size is small.
 Example: Chi-square test, Spearman rank Correlation, Mann-
Whitney U test, Wilcoxson signed-rank test,, Kruskal-Wallis test, etc.
 Chi-square test: It is a widely utilized non-parametric statistical
method designed for categorical data analysis.
 Mainly used in testing for independence between variables (Chi-
square test of independence: it is used to determine whether there
is an association between two categorical variables) and goodness-
of-fit to a distribution (Chi-square goodness-of-fit test: it is a
statistical method used to determine whether the distribution of
observed categorical data matches an expected distribution. This
test is particularly useful when you need to verify if your data
follows a specific distribution such as normal distribution).
 Quantitative Data Analysis Methods

Spearman rank Correlation: measures the strength

and direction of association between two ranked variables.

Mann-Whitney U test: Compare the median scores of

two independent groups

Wilcoxson signed-rank test: Compare the median

scores of two related or paired groups.

Kruskal-Wallis test: Compare medians across three or

more independent groups
 Quantitative Data Analysis Methods
 General advice about quantitative data analysis

 Align your analysis with the golden thread

 Your golden thread should be highly visible in all chapters of
the research including the sample, research design,
analysis, findings, and conclusions sections.
 The nature of your research questions influence how you
analyze your data.
 Quantitative research should logically follow to answer the
research questions.
 Once you have finished analysis, revisit the golden thread to
ensure the alignment
 An outsider’s view is necessary for a researcher to verify the
alignment.
 Example: Poor alignment
o Analysis method: Descriptive statistics
o Research question: Assessing the relationship between
variables/constructs
o Appropriate method: Inferential statistics
 Quantitative Data Analysis Methods

 Use a suitable quantitative data analysis method/statistical

test for the data type at hand.
 Don’t overcomplicate the analysis.
 What is most important is that your quantitative data analysis
method aligns with the golden thread. Aim for a relatively
simple but well-aligned approach and execute it well.
 Choose a statistical technique that actually fits what you need
to do.
 Markers want to see that your design choices are suitable and
applicable (well executed).
 Be cautious of the temptation to keep adding additional
variables unless they are well-justified and relevant to the
study.
 Quantitative Data Analysis Methods

 Clean your data and check that it meets the assumptions.

 It is essential to clean your data before you run all your
statistical tests.
 Data cleaning helps you resolve/clear out any incorrect,
corrupt, or suspicious data.
 Not cleaning and preparing your data will lead to incorrect
statistical results. A set of fresh eyes can help you identify blind
spots in relation to your data.
 Don’t chase a statistically significant results
 It is very tempting to get fixated on identifying statistically
significant results
 How you deal with the data that you collected is the most
important factor.
 A study without statistically significant results can still be very
valuable and can lead to a very important research.
 Qualitative Data Analysis Methods
 Qualitative data: are data that cannot be counted, measured or easily
expressed using numbers. It is collected through texts, audios and
images.
 Qualitative data: limited to text-based data.
 Qualitative research focuses on words, subjects, descriptions,
concepts, or ideas.
 Qualitative data analysis: is the process of organizing, analyzing,
and interpreting qualitative data to capture themes and patterns, and
answer research questions.
 Qualitative data analysis is heavily dependent on the researcher’s
analytic and integrative skills and personal knowledge of the social
context where the data is collected.
 Qualitative data analysis can be a great way to explore a topic and come
up with a new idea or support your own hypothesis. You need to go
through large amount of unorganized qualitative data to find trends
(uncover hidden trends) and observations to support your hypothesis.
 Qualitative data analysis generally involves transcription, translation,
transformation, organization, searching for relevant information,
coding/labeling, and interpretation.
 A code is a short word or phrase that symbolically assigns a
summative, salient, essence-capturing, and/or evocative attribute for a
portion of language-based or visual data (Saldana, 2012).
 Qualitative Data Analysis Methods
 Qualitative research is fundamentally about discovering the
nuances.

 Qualitative research. Why is it needed?

 Helps to understand the underlying reasons, opinions and
motivations
 Provides insights into a problem
 Helps identify a hypothesis or idea for potential
quantitative research
 Sources of qualitative data are:
o Interview transcripts
o Focus group discussion
o Documents
o Open-ended survey responses
 Analyzing qualitative data allows us to explore ideas and
further explain quantitative results.
 Qualitative data Analysis Methods

Types of qualitative data analysis

 Thematic analysis: Searching and evaluating the
themes/patterns within qualitative data.
 Content analysis: quantifies and analyzes the presence,
meanings, and relationships of words or concepts.
 Narrative analysis: Analysis of stories. A common
assumption of narrative analysis method is that people tell stories to
help organize and make sense of their lives and their storied accounts
are functional, and purposeful.
 Discourse analysis: analysis of language (written-verbal)
with in its social context; draw meaning from language in its
context.
 Grounded theory: Development of theory based on the
collection and analysis of real world data.
 Interpretive phenomenological analysis (IPA):
Interpretation of personal lived experience.
 Qualitative data Analysis Methods
 Thematic analysis: is a method for analyzing qualitative
data that involves reading through a set of data and looking
for patterns in the meaning of the data to find themes.
 Thematic analysis is an iterative process.
 An iterative review approach will help you identify the
nuance and unexpected insights; requires multiple rounds
to identify the themes and sub-themes.
 You need to demonstrate to a marker that you have used
systematic iterative approach.
 Thematic analysis is one of the most frequently used
qualitative analysis approaches.
 Qualitative data Analysis Methods
Steps:
1. Familiarization
 Read the data several times until you become familiarize with the data.
As you read, you write down everything that comes in your mind.
2. Coding data
 Identify the main features of data and make notes
 These notes can be in the form of codes, labels, or simply highlighting
the text with different colors. Each color should correspond to a
particular research question to which the text is relevant
 Be thorough while coding data and leave no stone unturned.
3. Generating initial themes:
 Idetify initial patterns or themes in the data
 These themes will form the basis of your thematic analysis
4. Reviewing the themes

 To make your analysis more efficient, you have to make sure that the
themes you have identified are appropriate and sufficient for the
research objectives.
 This will give you an opportunity to ensure that the thematic analysis is
going tin the right direction and you can make final changes in the
theme and criteria as needed.
 Qualitative data Analysis Methods
5. Naming and defining the themes
 It is an important part of thematic research , as it enables
you to set boundaries and make distinction between data.
 It helps in organizing your analysis and ensures that you do
not mix up or get confused between different themes and
data.

6. Writing
 After you have identified, named, and defined the themes
carefully, the next step is to write up your report for your
thematic analysis.
 Your report should be formal and should include all
sections of a official report, such as introduction,
literature review, methodology, results and
observations, and final conclusion.
 Qualitative data Analysis Methods
 Content analysis: is a research method that examines and
quantifies the presence of certain words, subjects/themes, and
concepts in text, image, video, or audio messages. The
method transforms qualitative input into quantitative data to help
you make reliable conclusions about topic of study.
 The researcher uses objective and systematic counting and recording
procedures to produce a quantitative description of the symbolic
content in a text.
 Researchers can quantify and analyze the presence, meanings, and
relationships of certain words, subjects, or concepts.
 One of the most common qualitative data analysis method
 Content analysis can bring a slight quantitative angle into a
qualitative analysis
NB: While content analysis and thematic analysis seem similar, they are
different in concept:
• Content analysis can be applied to both
qualitative and quantitative data, and focuses on identifying
frequencies and recurring words and subjects.
• Thematic analysis can only be applied to qualitative data, and focuses
on identifying patterns and themes.
 Qualitative data Analysis Methods
Narrative analysis: involves the analysis of people’s stories to
derive meaning (i.e., listening to people telling story and
analyzing what that mean).
 The revision of primary qualitative data by the researcher.
 It provides insight into people’s worldviews and perspectives.
 How the story is said is very important.

 It is difficult to reproduce in subsequent researches (i.e., it is

difficult to test the findings obtained through narrative
analysis).
 Be very cautious of your own biases if you undertake narrative
analysis (potential biases can have a strong influence on the
results of the research).
 Do not draw broad conclusions.

Reading Assignment: Read discourse, grounded theory, and

Interpretive phenomenological analysis (IPA) analysis methods.
 Qualitative data Analysis Methods
 General advice about qualitative data analysis

 Expect the unexpected

 Unexpected findings are good things as long as they are supported
by data.
 Real-world findings do not always align with the literature.
 Regular short breaks will help you keep your mind fresh and avoid
tunnel vision. Actually, this applies to all chapters in a research.
 Detailed notes can help remind you of small forgotten details or
insights down the line.
 Visualizing your codes can help you identify the categories or groups
of related codes.
 You may use Excel to organize your data visually using color coding.
 Consider using word clouds to generate ideas and help you identify
common phrases.
 It is always a good idea to get outsider input regarding your analysis.
 It is sometimes useful to ask your respondents whether you captured
their thoughts accurately
 Qualitative data Analysis Methods
Relationships of
codes
Table: Relationships of codes
S. No. Title Form of relationship
1 Strict inclusion x is a kind of y
2 Spatial x is a place in y; x is part of y
3 Cause-effect x is a cause of y; x is a result of y
4 Rationale x is a reason for doing y
5 Location for action x is a place for doing y
6 Function x is used for y
7 Means-end x is a way to do y
8 Sequence x is a step in y
9 Attribution x is an attribute/character of y

Adapted from J. P. Spradley, 1979.

 Qualitative data Analysis Methods

 The reliability and validity of qualitative data have to be checked

properly.
 Reliability:
 Inter-coder
 Intra-coder

 Validity:

 trustworthiness,
 credibility,
 neutrality,
 applicability/transferability

Question: What are the strategies used to promote qualitative

research validity?
 Discussion
 In the results chapter you present and describe the analysis
results
 The discussion chapter is a section where you interpret and
explain the meaning of the analysis.
 You interpret the meaning in relation to your research questions.
 You will explain how your findings help answer your research
questions and aims.
 You will compare your findings to the existing studies/literature
 The purpose of the discussion section is to explore the
meaning, identify the importance/significance of the
information you presented in the results section.
 The discussion section must focus on the findings relevant to
the research questions.
 Avoid spending too much time on data and analysis that are not
directly related to your research questions.
 Do not present any claims that are not rooted in data in the
results section
 Discussion
Steps to write the discussion section
1. Restate your golden thread
 Researchers often read only the introduction and
discussion sections in a research work
2. Summarize your key findings
 Explain how your findings help answer the research
questions
 Typical language…
o “The findings suggest that…”
o “The data/information support/oppose the theory that….”
3. Interpret your results
 In this section unpack your findings in more detail and dig
into the meaning of each finding.
 How does your study contribute to the field?
 Don’t be afraid to discuss findings that contrast existing
literature
o How do your results compare to existing literature?
 Discussion
4. Acknowledge the limitations of your study

 Good research involves clearly identifying the limitations

and shortcomings of the study.
 Provide a balanced discussion of both the limitations and
strengths of your study.
5. Craft a concluding summary

 The concluding summary should briefly highlight the key

findings of the study
 Discussion
 Discussion tips

 Aim for strong linkage between the introduction section

and the discussion section.
 Spell out your findings and their meaning clearly
 Take inspiration from other researches
 Avoid absolute terms or definitive language
 !!
!
o u
Y
n k
h a
T

Research Methods [Muluken D. (PhD 80

)]