CHAPTER TWO

RESEARCH DESIGNS &

SAMPLING PROCEDURES

Abdella Kosa
Assistant Professor, Sheger College &
Visiting Scholar, University of Johannesburg
This chapter is mainly concerned with giving answers
to the following questions,
What are the philosophical approaches in research?
1
What are the three types of research approaches?
What is research design? Why we need research
2
design? What are the basic features of good research
design?
3 What are the different types of research design?

4 What is the meaning of sampling design? What are

the basic steps in sampling design?
5 What are the characteristics of good sample design?

6 What are the types of sample designs?

 What are the philosophical approaches in research?
• Philosophically, researchers make claims about :
– What is knowledge (ontology)? how we know it? (Epistemology)? And
how we write about it (rhetoric).
• Ontology: concerned with nature of reality. What assumptions
do we make about the way in which the world works?
• Epistemology: General parameters and assumptions associated
with an excellent way to explore the real world nature.
• Paradigm choice is a reflection of how the researcher views the
world (ontology) and believes that knowledge is created
(epistemology).
• Axiology: What roles do our values play in our research choices?
• Methodology: Combination of different techniques used by the
scientists to explore different situations.
Philosophical approaches in research . . .
• These beliefs, along with the researcher’s disciplinary
focus and past experiences, influence his or her
philosophical approach to research;
• Different worldviews/schools of thought --knowledge
claims are
1. Positivism and Postpositivism,
2. Interpretivism,
3. Pragmatism
4. Realism
 Post Positivism Interpretivism
Positivism: claims that the social • Provides an understanding of
world can be understood in an social reality that is based on the
objective way & scientist works subjective interpretation
independently. depends on researcher’s
Post-positivism: reflect the need to
interests.
identify and assess causes that
• Meanings is constructed through
influence outcomes based on careful
observation and measurement of discussions or interactions with
the objective reality. Test hypothesis other persons.
1 2 3 4
The pragmatic worldview Realistic view
• Emphasize the research problem • Based on the
principles of positivist
and use all approaches available
& interpretivist
to understand the problem. • Based on assumptions
• Argues that the most important that are necessary for
determinant of the epistemology, the perception of
ontology and axiology you adopt subjective nature of
is the RQ the human
 Research paradigms vs research philosophies
Positivism Realism Interpretivism Pragmatism
Objective, ind’t of Socially
Ontology

External, objectve External, multiple

human thought, constructed,
and ind’t of social views to answer
interpreted through subjective, may
actors RQs
social conditioning change, multiple
Epistemolog

Only observable Observable Subjective Either/both

phenomena can phenomena provide meaning, focus on observable
provide credible credible data, details of situtions phenomena and
y

data, focus on explaining within a and reality behind subjective meaning

casuality context it. based on RQs

Research is value Research is value Value plays role in

Research is value
Axiology

free, research is bound, researcher interpreting results,

laden, the
ind’t of data & is part of what is researcher adopt
researcher is biased
maintains being researched, both objective &
by world views
objective stance subjective subjective view
Methodolo

Highly structured, Methods chosen Small samples, in- Mixed or multiple

large samples, must fit the subject depth method designs,
gy

measurement, matter, quantitative investigations, quantitative &

mainly quantitative or qualitiative qualitative qualitative
Philosophical approaches in research . . .
Methodology is the bridge that brings our philosophical
standpoint (on ontology and epistemology) and method
(perspective and tool) together.
To summarize see the research ‘Onion’
(Saunders et al., 2009)

Fig2.1: research “Onion’

• Source: Saunders et al., 2009
 What are the three types of research
approaches?
• Therefore, researchers not only select the assumptions
about the nature of reality but also the strategies of
inquiry;
• The philosophical assumptions lead to the strategies;
• Strategies of inquiry are :
– Qualitative,
– Quantitative and
– Mixed methods designs or models that provide
specific direction for procedures in a research
design.
 Qualitative Research Approach
• Explores attitudes, behaviour and experiences through
such methods as interviews or focus groups.
• Attempts to get an in-depth opinion from participants.
• Interpret phenomenon in terms of the meanings people
bring to them. Is one in which the inquirer often makes
knowledge claims based primarily on
– Constructivist perspectives (i.e. the multiple meanings of
individual experiences, meanings socially and historically
constructed, with an intent of developing a theory or pattern)
or
– Advocacy/participatory perspectives (i.e. political, issue-
oriented, collaborative or change oriented) or both.
 Characteristics of Qualitative Approach
• Tend to collect data at the site. They do not bring individuals
into lab or they do not send instruments for individuals.
• Data collected through interviews, observations, and
documents, rather than relying on a single data source.
• The researcher keeps a focus on learning the meaning that
the participants hold.
• Not intended to test a theory or hypothesis; instead, it is
exploratory and through induction it builds theories or
hypotheses.
• No fixed rules for sample size in qualitative research.
• It is flexible and emergent without being constrained by
standardized procedures.
 QUANTITATIVE APPROACH
• A one in which the investigator primarily uses post-positivist
claims for developing knowledge (i.e. cause and effect thinking,
reduction to specific variables and hypotheses and questions,
use of measurement and observation, and the test of theories)
• Employs a review of the existing literature to deductively develop
theories and hypotheses to be tested.
• Employs strategies of inquiry such as surveys and experiments &
collects information using questionnaires/structured interviews.
• Researchers engaged in quantitative strategy of inquiry have
assumptions:
– Testing theories deductively,
– Building in protections against bias,
– Controlling for alternative explanations; and
– Being able to generalize and replicate the findings
Qualitative Vs Quantitative Research
Qualitative Vs Quantitative research
 Qualitative Vs Quantitative research
• We may have written a combination of these words
which could mean two things.
– Firstly, you might want to think about combining
both qualitative and quantitative research and this
called triangulation.
– Triangulation is good way of approaching research as
it enables you to counteract the weaknesses in both
qualitative and quantitative research.
– However, it could mean that your ideas are still
unclear and that you need to focus a little more.
 How do you decide which is the best
methodology for your research?
• First, decide whether you should consider qualitative or
quantitative research;
• Secondly, looking at the five Ws (what, why, who,--)
• Have a look at the words you have used: Certain words
help to suggest a leaning towards qualitative or
quantitative research .
– For example , If you have written words such as ‘discover’,
‘motivation’, ‘experiences’, ‘think/thoughts’, ‘problems’, or
‘behave/ behaviour’, this suggests a leaning towards
qualitative research.
– If you have used “ how many', test, verify, how often, they
suggest quantitative research
 Mixed Methodology/approach
• An approach to inquiry that combines qualitative and
quantitative approaches.
• Researchers tend to base knowledge claims on pragmatic
grounds employing both quantitative and qualitative data
either concurrently or sequentially to best understand
research problems.
Strategies of mixing
• Sequential mixed methods approach - The researcher
starts with gathering qualitative data and then gathers
quantitative data or vice versa in two different phases;
• Concurrent mixed methods:Where the quantitative and
qualitative phases occur simultaneously in order to provide
a comprehensive analysis of the research problem.
What is research design?
 Tasks Involved In a Research Design
Define the Information Needed

Design the Exploratory, Descriptive, and/or Causal

Phases of the Research

Today’s
Specify the Sampling Process and the Sample Size
Topic

Specify the Measurement and Scaling Procedures

Construct a Questionnaire

Develop a Plan of Data Analysis

 Good research design is the
“first rule of good research.”

Why we need Knowledge of the needed

research design? research design allows
advance planning so that
the project may be
conducted in less time and
typically at a cost savings
due to efficiencies gained
in preplanning.
What are the basic features of a good
research design?

Flexible, appropriate, Gives the smallest

efficient, economical experimental
and so on error
Assumes the availability
Minimizes bias and
of skills, time and money
maximizes the reliability
for the research work
of the data collected
and analyzed
Related to the purpose of
Yields relevant information
the research problem & al
& opportunity for
with the nature of the
considering many different
problem to be studied
aspects of a problem.
 What are the different types of research design?
Sample

Cross Research Nature of

Text

Sectiona No. of Design investigati

l contact on Sample Observasi
Text
onal
Before
This is a Sample
& After sample Degree of Text
text.
Problem
Longit formulation Quasi-
experiment
udinal al
Experme
Sample
Text ntal
Explorat Descript
ory ive
Explanato
ry
Summary of Study designs
Research
Design

Exploratory Conclusive
Design Research

Descriptive Causal
Secondary Experience Pilot Case
Data Surveys Studies Studies Design /Experimental
Design

Cross-sectional Longitudinal

Experiment

Secondary
Survey Observation
Data Study
 Exploratory research
Provides insights into and Techniques
comprehension of the ●Secondary Data Analysis
problem situation ●Pilot Studies: uses sampling but does not
confronting the apply rigorous standards. Includes
researcher. ○ Focus Group Interviews: Unstructured, free-
flowing interview with a small group.
It could be used for any of
○ Projective Techniques: Indirect means of
the following purposes:
questioning that enables a respondent to
●Gain background information. project beliefs and feelings onto a third party
●Formulate/clarify
problems and or an inanimate object.
Develop hypothesis. Case Studies: Intensively investigate one or
●

●Establish research priorities. a few situations similar to the problem situation.

●Identify alternative sources of ●Experience Surveys: Individuals who are

action. knowledge about a particular research problem

are questioned.
 Descriptive Research
◉ Description of something-usually market characteristics
or functions.
◉ Provides answers to questions such as who, what,
where, when and how.
◉ Important because, not always easy to isolate variable
that will explain those causes.
◉ By gathering descriptions of “what is” and comparing
them to “what we would like” educators can see the
area that needs to be addressed.
26
Descriptive Research: Survey Forms
Factors to be considered
• Written questionnaires • Sampling
• Type of population
• Personal interviews
• Question Form
• Telephone interviews
• Question Content
• Observational
• Response rates
Research
• Costs
• Portfolios
• Available facilities
• Length of data collection
• Computer assisted
techniques for data
collection
 Frequency of contact
• Best suited to studies aimed at finding out the prevalence
Cross of a phenomenon, situation, problem, ...
sectional • Obtaining an overall picture as it stands at the time of the
study about the study population.
• Eg., consumer satisfaction with a product.

before-&- • Measures changes. It is the most appropriate design for

after measuring the impact or effectiveness of a program.
design • Two sets of one shot data collection from same population
using same process as a cross-sectional study
• e.g. The impact of incentives on the productivity of employees

• A fixed sample (or samples) of population elements

Longitu
dinal
is measured repeatedly on the same variables
• A longitudinal design differs from a cross- sectional
design in that the sample or samples remain the
same over time
 Cross Sectional vs. Longitudinal Designs
Sample
Cross Sectional Surveyed
at T1
Design

Sample Same
Before and Sample also
Surveyed at Surveyed at
after T1 T2

Time
T1 T2
Same Same
Sample Sample also
Sample also
Surveyed at Surveyed at Surveyed at
Longitudinal T1 T2 T3
Design

Time T2 T3
T1
 Causal Research
◉ The major objective is to obtain evidence regarding cause –
and effect (causal) relationships.
◉ Causality may be thought of as understanding a
phenomenon in terms of conditional statements in the form
of “if X, then Y.”
◉ The “best” Research Design to determine CAUSAL RELATIONS
◉ Example: If I spend more on advertising, then sales will rise.
◉ Will replacing TV commercial A with commercial B lead to
increase in consumer preference on our brand?
◉ Can we improve profitability of our fashion clothing line by
increasing its price by 10%?
30
 Type of Experimental design
True Experimental Quasi-experimental
◎ Researchers plan to measure the ◎ Although the researcher
response variable depending on plans to measure the
the explanatory variable. response variable
◎ The most important factor is depending on the
randomization. explanatory variable,
there is a lack of
◎ The randomization check, is the
randomisation in the
process of examining the overall
quasi-experimental
effectiveness of random
assignment design
◎ It is a mixed design
◎ In performing randomization
where random and non- 31
checks, researchers compare
random experiments are
study groups or conditions on a
number of pretest variables. employed together
 Important Concepts Relating
to experimental design

Dependent and Extraneous & control

independent variables variable Experiment

Variable is a concept which Controls: Used to Process of examining the

can take on different minimise the influence truth of a statistical
quantitative values of extraneous variable hypothesis

Confounded Experimental and

relationship control groups Treatments

When the dependent Control group: When a The different conditions

variable is not free from group is exposed to usual
under which
conditions.
the influence of extraneous experimental and control
Experimental: under
variable(s special situation groups are treated
 Comparison
Factor Exploratory Descriptive Causal
Objective Discover ideas and Describe market Determine cause and
insights characteristics/functions effect relationships
Characteri Flexible & Versatile Marked by prior formation Manipulating of one/more
stics The front end of total of hypothesis, preplanned ind’t variables, Control of
research design & structured design mediating variables
Methods expert surveys Secondary data experiments
pilot surveys Surveys
secondary data panels
qualitative research observational & others
Findings/R Tentative Conclusive Conclusive
esults
Outcomes Generally followed findings used as input into findings used as input into
by further research. decision making decision making
 Which is the “Best” Research Design &
Method?

• “You cannot put the same shoe on every foot.”

– Publilius Syrus
• It depends on the
– Problem of interest,
– Level of information needed,
– Resources,
– Researcher’s experience, etc.
 General guidelines for choosing research
designs include the following

◉ When little is known about the problem situation, it is

desirable to begin with exploratory research.
◉ Exploratory research is the initial step in the overall
research design framework.
◉ It is not necessary to begin the survey research design with
exploratory research. A research design could well begin
with descriptive or causal research. To illustrate, a
consumer satisfaction survey is conducted annually need
not begin with or include an exploratory phase each year.
◉ Exploratory research may follow descriptive or causal
research.
35
 What is the meaning of sampling?
• Sampling is the process involving the selection of a
finite number of elements from a given population of
interest, for purposes of inquiry.
• Other related terminologies;
– Target population: the population of interest to whom the
researchers would like to make generalizations.
– Study population: the actual group in which the study is.
conducted.
– A small fraction of the population from which conclusions can be
drawn about the whole population.
– Sampling frame: List of the population
– Study unit/Sample: A subset of a study population, about
which information is actually obtained : persons, housing
units, etc.
– A sample is a representative part of a population.
 Why sampling?
The reasons for sampling are;

To get a general impression For obtaining estimates on

of the total population of certain characteristics of
interest. the population.
The sampling process is undertaken
In this case the selection of
through a set of rigorous &
individuals to be included in the
objective procedures to avoid
sample can be quite subjective.
subjective bias.

Other reasons
Time, Cost and available resources, and practicability
Sampling makes possible the study of a large, heterogeneous population.
Sampling is for economy and for speed.
What are the steps in the
sampling design?

01 02 03 04

Defining the Listing the Selecting a Obtaining an

population population representativ adequate
e sample sample size
Target population Sampling frame Probability or non-
For statistical
probability
measurement > 30
sample size (Stutely
2003) or use formulas
What are the characteristics of good
sample design?

Truly representative & Systematic bias is

Similar to population controlled

Results can be applied to the

Having small sampling population in general with a
error reasonable level of confidence
⇒ reliability

Economically viable Optimum size

(adequately large)
What are the types of sample designs?

Volunteer
What are the types of sample designs?
Probability sampling Non probability sampling
Not based on known
Employs random
probability. Every item has an
procedure. Done on the unknown chance of being
basis of chance. selected

Every sampling unit has a Procedures are not valid for

obtaining truly representative
known and non-zero and
sample of a population.
equal probability
Quick, inexpensive and
More complex, time convenient
consuming & costly
There is no way to measure
the precision of the resulting
Reliable estimates can be sample. used when it is
produced and inferences unfeasible or impractical to
can be made conduct probability
sampling.
 Probability sampling
Selected at random from the sampling frame,
Simple
Random To use a SRS method:
• Make a numbered list of all the units in the
Sampling population
• Each unit should be numbered from 1 to N
• Select the required number.
The randomness of the sample is ensured by:
Use of “lottery’ methods, Table of random
numbers, & Computer programs

Advantages: No bias, Small variability, …

Limitations: Requires a sampling frame,
Difficult if the reference population is
dispersed, Minority subgroups of interest may
not be selected, …
 Probability sampling
• Selection of individuals from the sampling frame
systematically at regular intervals down the list and
the starting point is chosen at random
Systematic • Taking individuals at fixed intervals (every kth)
based on the sampling fraction.
(interval) random Steps
sampling 1. Number the units on your frame from 1 to N.
2. Determine the sampling interval (K) by dividing
the number of units in the population by the
desired sample size.
3. Select a number between one and K at
random.
4. Select every Kth unit after that first number
• Less time consuming & easier to perform as
compared to SRS.
• Systematic sampling should not be used when a
cyclic repetition is inherent in the sampling frame
 Probability sampling
• The population is divided into homogeneous,
mutually exclusive groups called strata, and a
Stratified
population can be stratified by any variable.
random • Among strata there is heterogeneity and
sampling within each strata units are homogeneous.
• A separate sample is taken independently
from each stratum.
• It can make the sampling strategy more
efficient.
• A larger sample is required to get a more
accurate estimation if a characteristic varies
greatly from one unit to the other.
• Stratified sampling ensures an adequate
sample size for sub-groups in the population
of interest.
• Equal allocation:
– Allocate equal sample size to each stratum
• Proportionate allocation:
n
nj  Nj
N
– nj is sample size of the jth stratum
– Nj is population size of the jth stratum
– n = n1 + n2 + ...+ nk is the total sample size
– N = N1 + N2 + ...+ Nk is the total population size
Village A B C Total
HHs 100 250 150 500
S. size ? ? ? 60
 Probability sampling
Sometimes it is too expensive to carry out SRS
• Population may be large and scattered.
• Complete list of the study population unavailable
Cluster • Travel costs can become expensive if
sampling interviewers have to survey people from one end
of the country to the other.
Cluster sampling is the most widely used to reduce the
cost
The clusters should be homogeneous, unlike stratified
sampling where the strata are heterogeneous
Steps
1. Cluster sampling divides the population into
groups or clusters.
2. A number of clusters are selected randomly to
represent the total population, and then all units
within selected clusters are included in the
sample.
3. No units from non-selected clusters are included
in the sample.
Advantages
• Cost reduction
• It creates 'pockets' of sampled units instead of spreading the
sample over the whole territory.
• Sometimes a list of all units in the population is not available,
while a list of all clusters is either available or easy to create.
Disadvantages
• Creates a loss of efficiency when compared with SRS.
• It is usually better to survey a large number of small clusters
instead of a small number of large clusters.
• You do not have total control over the final sample size.
 Probability sampling
Similar to the cluster sampling, except that
it involves picking a sample from within
Multi-stage each chosen cluster, rather than including
sampling all units in the cluster.
This type of sampling requires at least two
stages.
The primary sampling unit (PSU) is the
sampling unit in the first sampling stage.
The secondary sampling unit (SSU) is
the sampling unit in the second sampling
stage, etc.
If more than two stages are used, the process
of choosing population units within clusters
continues until there is a final sample.
• With multi-stage sampling, you still have the
benefit of a more concentrated sample for cost
reduction.
• However, the sample is not as concentrated as
other clusters and the sample size is still bigger
than for a simple random sample size.
• Admittedly, more information is needed in this
type of sample than what is required in cluster
sampling.
• However, multi-stage sampling still saves a great
amount of time and effort by not having to create a
list of all the units in a population.
 Non-probability random sampling
Convenience sampling
Easily and conveniently accessed
samples are selected
Easy to use, but offset by the
presence of bias. Volunteer sampling
deliver accurate results when the Occurs when people volunteer to
population is homogeneous be involved in the study.
Eg., determining whether a lake is volunteers accept the possibility of
polluted or not. a lengthy or unpleasant process in
exchange of payment..
Judgment sampling Introduce strong biases.
sample is taken based on Eg., drug testing
certain judgments
More biased than haphazard Quota sampling
sampling A no. of quotas for sub-populations
Used in exploratory studies have been selected.
Does not meet the basic requirement
Snowball sampling of randomness.
Existing study subjects recruit Effective when information is
future subjects. urgently required or no suitable
Often used in hidden frame.
populations. Less expensive than random
Eg., drug users or commercial sampling, Easy to administer
sex workers.
 Sampling Error
• It is created because of chance variation.
• It is the difference between the sample statistics
(such as the mean and standard deviation) and the
corresponding population values, known as
parameters.
• In relation to this point, one should note that the
mean of the sampling distribution and the mean of
the population are equal.
• Sampling error is unavoidable: It can be reduced by
two factors:
– A larger sample produces a smaller sampling error
– A homogenous population produces a smaller sampling
errors
 Sampling/systematic bias
• It results from some imperfect aspect of the research
design or from a mistake in the execution of the
research/administrative error (data processing error,
sample selection error, etc).
• Non-sampling Errors
– An inadequate sampling frame
– Non-response from participants
– Field errors
– Response errors, accessibility bias
– Coding and data entry errors
– Defective measure device
 Sample size determination
• The size of the sample is one of the most important
determinants of the accuracy of survey estimates.
• Samples size is estimated using formulae
• The selection of a formula depends on
Sampling strategies (techniques)
Population size & Nature of the population- Heterogeneous or
homogenous- Dispersion factor (or variability)
Nature of study (qualitative or quantitative)
Study and design type (experimental vs non-experimental )
The type of variable being studied.
Type of statistical comparison planned.
Available finance and other resources (trained investigators)
 Basic questions that should be asked when
choosing a sample
• How large a sample can you collect?
• What is the prevalence of the condition you are
studying?
• What level of budget do you have for the study?
• What staff are available to gather the sample?
• How much time do you have for the research?
• Into how many cells or categories are you going to
divide your data for analytical purposes?
Tips
 Article Review Guideline
The review should address the following vital issues:
(1) What is the major theme of the article?
(2) Did the authors provide enough background information for the
intended audience to understand it?
(3) Do the researchers clearly develop a major research question,
proposition, or hypothesis that is to be evaluated in the empirical
study and discussed in this article?
(4) What are the theories discussed in the paper and how do they
compare?
(5) Are the particulars in the research design and methods (e.g., research
design, sampling techniques, data collections and analysis methods,
and data validity and reliability) appropriately stated? Justify.
(6) If the author argues supporting/rejecting a theory, what are her/his
reasons and how valid are the reasons?
(7) In discussing the results of the empirical study, do the authors remain
true to the actual findings of the study
(8) How well are conclusions of the article aligned with the findings of the
study?
(9) What exactly does the work contribute and relate to the overall topic
of your course?
(10) In your opinion, what were the strengths and weaknesses of the
paper?
Instructions Contents
 Chose two articles and review  Preliminary part
independently.  Title of article,
 The page limit for the review is 2  Name of author/s, and year
pages (for each), written in Times of publication
New Roman 12, Line spacing 1.5, &  Title of journal, Volume
paragraphs Justified number, issue number, and
 Your work should be neat and clear. page numbers
 Submission Date: Sunday, April 21,  Main Body
2023  Summary,
 Any delay for submission will not be  Strengths and limitations,
accepted unless reasonable evidence  Significance (lessons drawn)
is presented.  Supplementary part
 Academic dishonesty (plagiarism,  References,
cheating and etc) will make your  Appendices (if another
value zero. works are referred)
 Individual Assignment

Select two of the provided articles

and review individually