Basics of Research

Methodology
UNIT 3 RESEARCH DESIGN
Structure
3.1 Introduction
3.2 Objectives
3.3 Need for Research Design
3.4 Principles of Research Design
3.5 Types of Research Designs
3.6 Developing a Research Plan- Exploration, Description, Diagnosis and
Experimentation
3.7 Sampling Techniques
3.8 Need for Sampling
3.9 Significant terms in Sampling
3.10 Types of Sampling Designs
3.11 Probability Sampling Procedures
3.12 Non-probability Sampling Procedures
3.13 Let Us Sum Up
3.14 Key Words
3.15 Suggested Further Reading/References
3.16 Answers to Check Your Progress

3.1 INTRODUCTION
The Unit ‘Research Design’ brings forth various phases of research that could
be carried out in a planned and phased manner by developing a research
design. Once the researcher finalizes the research design, the entire research
process will be under their control. The example below will make learners
realize the significance of developing a research design. The architect
prepares the design, keeping in mind that he needs to construct a building.
The plan will contain the structure of the building, number of rooms,
utilization of rooms, size of open space, size of garden space, building
material, and length and breadth of the building in square feet. The architect
will decide and prepare a blueprint or plan before starting the building. I
believe the above example will help you know the research design's
significance.

Once the design is in the hands of the researcher, they have the freedom to
make necessary changes whenever the need arises. Designing also helps the
researcher reduce wasteful expenditure on finance, time and energy. The
process of preparing a research design anticipates various difficulties, and the
same difficulties will be addressed in the research design through appropriate
strategies. Preparing the research design also helps the researcher to control
44
the unexpected situation rationally. It also facilitates the researcher to Research Design
articulate the research procedure in a better manner during the academic
presentation. It helps them to prevent the possibility of failure. With the
above introduction, the present unit facilitates you to learn the definition and
need for research design, principles of design research design, and types of
research designs. Finally, we discuss sampling techniques.

3.2 OBJECTIVES
After studying this unit, you should be able to
 define research design;
 discuss the need for the research design;
 explain various research designs;
 define sampling;
 discuss the need for sampling; and
 describe various sampling techniques.

3.3 NEED FOR RESEARCH DESIGN

Let us start with defining “what is Research Design”. We define the research
design as “such design as a symbolic construction or model”. “The research
design is the arrangement of conditions for data collection and analysis in a
manner that aims to combine relevance to the research purpose with the
economy in procedure” (Selltiz et al. 1964). If we look at the above
definition, we can see that the design talks explicitly about the research
objectives, the rationale for doing specific research, and a methodology
which includes methods and theory, sampling, data and data organization.
Besides the core procedures mentioned above, scientific research design must
explicitly describe the accepted methodology. The theory must be a core
component of the social science research design. The research design should
have the following aspects.
1. The details of the study
2. The rationale for choosing the study
3. Reasons for carrying out the study
4. Type of data required for the chosen study
5. Techniques of data collection
6. Area of Study
7. The period of study
8. Duration of the study
9. Requirements of different types of material for the research study
10. Tools and techniques required for the study
11. Selection of cases, number of cases required for the study
12. Methods of data analysis 45
Basics of Research The above-said points need to be addressed in a structured manner with
Methodology clarity in the research design. Once the researcher addresses the above points
in the research design, they could achieve the research objectives with
minimum expenditure.
1. Let us discuss the need for methodologically designed research in the
following paragraph. We will draw the need for research design through
an example. Once the researcher formulates research questions and
objectives, the researcher has no idea how accurate the results of his study
are in many cases. Of course, the researchers may have many questions in
their minds and try to inquire. One takes up research to make a valuable
contribution to the community and further scientific inquiry. In such a
situation, the researcher has to find out how much percentage of
inaccuracy can be accepted. Secondly, researchers need to know how
much percentage inaccuracy will be produced by/her methods or
instruments. In both situations, if a researcher gets clarity, there is no
need to trouble themselves by wasting time and energy to get more
incredible accuracy.
2. In every research, the collected data must be in a tabular format that
needs to be analysed. The collected data should be meaningfully
presented to interpret and draw conclusions. The researchers may not get
good results if they take lots of time to collect more unwanted data. Most
of the time, the researcher intends to collect new data. If researchers have
lots of data without understanding how they will give meaning to that
data after spending so much time collecting it, the data does not give any
meaning to their research. It is too late to give meaning to them at the
final stage of the research. The above-mentioned hypothetical situation
can be avoided by taking the time to prepare the research design. In
laboratory research, the researcher may wish to do impressive research.
At the same time, if they develop inferior quality instruments to
experiment, they may get inaccurate results. At the same, if the researcher
devotes his energy and time to developing quality instruments to do
testing, he may get accurate results.
3. As a scientist, it is their responsibility to build scientific institutions with
their capability for a better way of inquiry. They may apply various
methodological inquiries to build theories to solve societal problems and
inquire about social relations. A scientist cannot afford to satisfy with his
methods. As a scientist, it is their responsibility to question every stage of
their research by questioning their methods with an open mind. This
paves the way for improving their methods continuously while doing the
research. Once the researcher decides on the research problem, they must
develop an appropriate methodology to solve the problem. In this process,
developing a research design is very important. According to Ackoff, “the
idealized research design is concerned with specifying the optimum
research procedure that could be followed where there are no practical
restrictions”.
46
 If you look at the above definition of research design, it states that there is Research Design
no need to develop an idealized research design. One can give a choice to
improve the research design in the process of conducting the research.
Once the researcher finalized the research design, it must be converted
into the working procedure. Before starting the research, the researcher
must consider various factors, including time, money and energy.
4. If research is part of the master's or pre-PhD programme, the researcher
may face time constraints. In such a situation, the researcher may limit to
observing one or two variables rather than finalizing multiple variables to
observe. The observation method may bring subjective error in both
science and social science research. In such a situation, the researcher
may consider using triangulation as a method where the same phenomena
can be tested using more than one method. The researcher needs to collect
data using more than one method as this may also, in some situation,
helps in avoiding errors in collecting the data. They may draw a sampling
population that they need to observe multiple times. Sometimes this may
take time and energy, which is not practical. Thus, the research design
helps the researcher to organize his ideas in a form whereby it will be
possible for them to look for flaws and inadequacies. The research design
also allows researchers to present their study to the experts for critical
comments and evaluation. Without such a course of action, it will be
difficult for the critic to provide a comprehensive review of the proposed
study.

3.4 PRINCIPLES OF RESEARCH DESIGN

The following principles must be kept in mind while designing the

research—the existing literature mentioned the following six principles for
research design.
1. Research design should be holistic.
2. It should be agnostic.
3. It should be prepared with a purpose.
4. The research design should be prepared early.
5. It should be flexible.
6. It should be shared with experts and other researchers.

3.5 TYPES OF RESEARCH DESIGNS

According to Manheim, there are three types of research designs.

1. Research design in case of exploratory research studies;
2. Research design in case of descriptive and diagnostic research studies,
and
47
Basics of Research 3. Research design in case of hypothesis-testing research studies.
Methodology
Let us discuss each type of research in detail in this section.

1. Research Design in case of Exploratory Research Studies

It is also called formulative research studies. The study based on exploratory

research design is either to formulate a research problem for further precise
investigation or to develop a working hypothesis for further research. The
following example will help you to know exploratory research with clarity. A
person has visited a doctor with a fever. The doctor did not know the reasons
for his/her fever. The doctor might treat the patient with the visible
symptoms. If the fever went within a few days, the doctor diagnosed it rightly
and treated for the disease the doctor diagnosed by giving drugs based on
visible symptoms. His exploration of particular patients for visible symptoms
might pave the way for knowing symptoms of malaria or dengue, facilitating
further research on symptoms and drugs. With the above example, the
research design emphasizes discovering ideas and insights. As I mentioned in
an example in this section, exploratory research design provides space and
flexibility to explore different aspects of a specific research problem. In the
initial stage, the research problem will be defined. It will be converted into a
specific research problem after exploratory studies. In the exploratory
research design, there will be flexibility in changing the research design. In
exploratory research design, the following research is mainly carried out.
1. Reviewing existing literature for chosen research problem;
2. The experience survey to document the experiences of individuals on
specific issues; and
3. The analysis of ‘insight-stimulating’ examples.
The survey of an existing literature review may pave the way for formulating
particular research problems or developing hypotheses. It is a simple and
fruitful method. The researchers need to visit libraries to review the existing
literature. The existing studies and the hypotheses stated in the existing
studies may be reviewed, and their usefulness is evaluated. Based on that,
further research for formulating research problems may be planned. It may
also be considered whether the already stated hypotheses suggest a new
hypothesis. In this way, the researcher reviews and builds upon their work
based on the work done by others. However, in cases where hypotheses have
not been formulated, her/his task is to review the available material for
deriving the relevant hypotheses. Besides, the bibliographical survey of
studies already made in one's area of interest may also facilitate the
researcher to formulate the problem precisely. Researchers should also
attempt to apply concepts and theories developed in different research
contexts to the area in which they are working. Sometimes the works of
creative writers also provide a fertile ground for hypothesis formulation.
Experience survey means surveying people with practical experience with a
specific problem to be studied. Such a survey aims to obtain insight into the
48
relationships between variables and new ideas relating to the research Research Design
problem. For such a survey, competent people can contribute new ideas. The
sample population to explore new ideas may be selected carefully.
Researchers must ensure that respondents represent different types of
experience. The investigator may then interview the selected respondents.
The researcher must prepare an interview schedule for the systematic
questioning of informants. However, the researcher must ensure flexibility in
conducting interviews because the respondents should be allowed to raise
issues and questions that the investigator has not previously considered.
Generally, collecting information through their experiences in an interview is
likely to be long and may last a few hours. Hence, it is often considered
desirable to send a copy of the questions to be discussed to the respondents
well in advance. This will also allow the respondents to do some advance
thinking over the various issues involved so that, at the time of the interview,
they may be able to contribute effectively. Thus, an experience survey may
enable the researcher to define the problem more precisely and help
formulate the research hypothesis. The survey may also provide information
about the practical possibilities for doing different types of research.
Analysis of ‘insight-stimulating’ examples is also a fruitful method for
suggesting hypotheses for research. It is particularly suitable in areas with
little experience as a guide. This method consists of intensively studying
selected instances of the phenomenon in which one is interested. For this
purpose, the existing records may be examined, unstructured interviewing
may take place, or any other approach may be adopted. The investigator's
attitude, the study's intensity and the researcher's ability to draw diverse
information into a unified interpretation are the main features that make this
method an appropriate procedure for evoking insights.
Now, what sort of examples are to be selected and studied? There is no clear-
cut answer to it. Experience indicates that certain types of instances are more
appropriate for particular problems than others. One can mention a few
examples of 'insight-stimulating’ cases such as the reactions of strangers, the
reactions of marginal individuals, the study of individuals who are in
transition from one stage to another, and the reactions of individuals from
different social strata and the like. Thus, in an exploratory formulative
research study which merely leads to insights or hypotheses, whatever
method or research design outlined above is adopted, the only thing essential
is that it must continue to remain flexible so that many different facets of a
problem may be considered as and when they arise and come to the notice of
the researcher.
2. Research Design in case of Descriptive and Diagnostic Research
Studies
The researcher will decide to do descriptive research studies to describe the
characteristics of a particular individual or a group. Unlike descriptive
studies, diagnostic research studies determine the frequency with which
something occurs or its association with something else—the studies 49
Basics of Research concerning whether certain variables are associated are examples of
Methodology diagnostic research studies. For example, the world has faced the Covid 19
pandemic, which made it standstill. The diagnostic research may determine
the frequency and interval of a pandemic from the 19th century. As against
this, studies concerned with specific predictions, with the narration of facts
and characteristics concerning individuals, groups, or situations are all
examples of descriptive research studies. Most of the social research comes
under this category. From the point of view of the research design, the
descriptive and diagnostic studies share common requirements. We may
group these two types of research studies. In descriptive and diagnostic
studies, the researcher must clearly define what he wants to measure and find
adequate methods for measuring it along with a clear-cut definition of the
'population' he wants to study. Since the aim is to obtain complete and
accurate information in the said studies, the procedure must be carefully
planned. The research design must protect against bias and maximize
reliability, with due concern for the economical completion of the research
study. The design in such studies must be rigid and not flexible. It must focus
attention on the following: (a) Formulation of the objective of the study, (b)
finalization of methods of data collection, (c) Selecting the sample, (d)
Collection of data, (e) Processing and analysing the data. (f) Reporting the
findings. The first step in a descriptive/diagnostic study is to specify the
objectives after stating the research question(s).
The researcher may encourage using “structured instruments” to reduce
biases. The sample design is significant in this type of research design. We
have discussed different sample designs in this unit. The researcher must
ensure that the selection of the sample must be error-free. It is suggested to
design sampling through probability or random sampling. To obtain data free
from errors introduced by those responsible for collecting them, it is
necessary to closely supervise the field workers' staff as they collect and
record information. Checks may be set up to ensure that the data collecting
staff perform their duty honestly and without prejudice. “As data are
collected, they should be examined for completeness, comprehensibility,
consistency and reliability. The data collected must be processed and
analysed. This includes coding the interview replies and observations,
tabulating the data, and performing several statistical computations. To the
extent possible, the processing and analysing procedure should be planned in
detail before actual work is started. This will prove economical because the
researcher may avoid unnecessary labour. Coding should be done carefully to
avoid coding errors.
Similarly, the accuracy of tabulation may be checked by having a sample of
the tables re-done. In the case of mechanical tabulation, the material (i.e., the
collected data or information) must be entered on appropriate cards, which is
usually done by punching holes corresponding to a given code. The accuracy
of punching is to be checked and ensured. Finally, statistical computations
are needed, so averages, percentages, and various coefficients must be
50 worked out. Probability and sampling analysis may as well be used. The
appropriate statistical operations and tests of significance should be carried Research Design
out to safeguard the drawing of conclusions concerning the study. Lastly, the
completed research must be appropriately communicated through well-
written reports. The report's structure needs to be well planned so that all
things relating to the research study may be well presented in a simple and
practical style. Thus, the research design in the case of descriptive/diagnostic
studies is a comparative design throwing light on all points narrated above
and must be prepared to keep in view the objective(s) of the study and the
resources available. However, it must ensure the collected evidence's
minimization of bias and reliability. The said design can be appropriately
referred to as a survey design since it considers all the steps involved in a
survey concerning a phenomenon to be studied.
3. Research Design in case of Hypothesis-testing Research Studies
It is generally called experimental studies. This research study tests the
relationship between two or more variables. In this experimental design, the
researcher first finalizes the experimental procedure to reduce bias and ensure
reliability and validity. After completing the experiment, the researcher could
infer the results about causality.
Experimental design is conducting research in an objective and controlled
fashion so that precision is maximized and specific conclusions can be drawn
regarding a hypothesis statement. Generally, the purpose is to establish a
factor or independent variable's effect on a dependent variable. Thus, in an
experiment, we observe and measure the effect of treatment given to a few
variables by controlling other variables affecting our observations. The term
“treatment” refers to a particular experimental condition. The material to
which the treatment is applied and on which the variable under study is
measured is known as the experimental unit. Since all variables cannot be
controlled, it may cause an error in our observations. It is an experimental
error. The whole experiment is conducted according to some plan called the
experiment or experimental design. Without a design, a research study is like
a construction without any plan or map. The design enables us to answer
research questions as validly, objectively, precisely and economically as
possible.
The principles of experimental design play an essential role in research.
Professor R.A. Fisher’s name is associated with experimental designs. He
made the beginning of such designs in agricultural research. As such, the
study of experimental designs has its origin in agricultural research. Professor
Fisher found that by dividing agricultural fields or plots into different blocks
and conducting experiments in each block, whatever information is collected
and inferences are drawn from them is more reliable. This fact inspired him
to develop specific experimental designs for testing hypotheses concerning
scientific investigations. Today, experimental designs are being used in
research relating to phenomena of several disciplines. Since experimental
designs originated in the context of agricultural operations, we still use
51
Basics of Research several technical terms of agriculture (such as treatment, yield, plot, block) in
Methodology experimental designs.
Basic Principles of Experimental Designs

Professor Fisher has enumerated three principles of experimental designs:

(1) Principle of Replication;

(2) Principle of Randomization; and

(3) Principle of Local Control.

1. Principle of Replication
The experiment should be repeated more than once to increase the statistical
accuracy. Each treatment is applied in many experimental units instead of
one. For example, if we need to compare the yield of two varieties of paddy,
then each variety is applied to more than one experimental unit. The number
of times these are applied to experimental units is called their number of
replications.
2. Principle of Randomization
The principle of Randomization involves allocating experimental units at
random to avoid any bias resulting from the influence of some unknown
extraneous factor that may affect the experiment. In the analysis of variance,
we assume that the errors are random and independent. In turn, the
observations also become random. The principle of randomization ensures
this. It eliminates the systematic bias. “Randomization is analogous to
insurance in that it is a precaution against disturbances that may or may not
occur, and that may or may not be serious.
3. Principle of Local Control
The third principle is called local or error control. As already mentioned,
replication with local control reduces the experimental error. In the simplest
case, experimental units are divided into homogenous groups. The variation
among these groups is eliminated from the error, thereby increasing
efficiency. The choice of size and shape of experimental units, and
environmental conditions, also affect the experiment, causing errors. It may
be controlled by using an analysis of co-variance also.
Local control helps in controlling the systematic error or general factors or
say "G" error; Randomization controls the sampling error or “S” error and
also a little “G” error. Replication mainly controls the variations in treatment
effects. Let us call it an "R" error. Measurement error is controlled by
replication. Replication with randomization would be able to control all three
errors and the "G" error and measurement error with local control. So,
replication is a fundamental principle. Experimental design refers to the
framework or structure of an experiment, and as such, there are several
experimental designs. We can classify experimental designs into two broad
52
categories: informal experimental and formal experimental designs. Informal Research Design
experimental designs are those designs that generally use a less sophisticated
form of analysis based on differences in magnitudes. In contrast, formal
experimental designs offer relatively more control and use precise statistical
procedures for analysis. Important experiment designs are as follows:
(a) Informal Experimental Designs:
(i) Before-and-after without control design
(ii) After-only with control design
(iii) Before-and-after with control design
(b) Formal Experimental Designs:
(i) Completely Randomized Design (CRD)
(ii) Randomized Block Design (RBD)
(iii) Latin Square Designs (LSD)
(iv) Factorial Designs
Before-and-after without control design: In such a design, a single test
group or area is selected, and the dependent variable is measured before the
introduction of the treatment. The treatment is then introduced, and the
dependent variable is measured again. The effect of the treatment would be
equal to the level of the phenomenon after the treatment minus the level of
the phenomenon before the treatment.
After only with control design: In this design, two groups or areas (test
area and control area) are selected, and the treatment is introduced into the
test area only. The dependent variable is then measured in both areas at the
same time. Treatment impact is assessed by subtracting the value of the
dependent variable in the control area from its value in the test area.
Before-and-after with control design: In this design, two areas are selected,
and the dependent variable is measured in both areas for an equal period
before the treatment. The treatment is then introduced into the test area only,
and the dependent variable is measured in both for an identical period after
the introduction of the treatment. The treatment effect is determined by
subtracting the change in the dependent variable in the control area from the
change in the dependent variables in the test area.
Completely Randomized Design (CRD): CRD involves only two
principles, viz., the principle of replication and the principle of randomization
of experimental designs. It is the most uncomplicated possible design, and its
analysis procedure is also more effortless. The essential characteristic of the
design is that subjects are randomly assigned to experimental treatments or
vice-versa.
Randomized Block Design (RBD) is an improvement over the CRD. In the
RBD, the principle of local control can be applied along with the other two
principles of experimental designs. The variable selected for grouping the
53
Basics of Research subjects is believed to be related to the measures obtained in respect of the
Methodology dependent variable.
Latin Square Design (LSD) is an experimental design very frequently used
in agricultural research. The conditions under which agricultural
investigations are carried out differ from those in other studies, for nature
plays an essential role in agriculture.
Factorial Designs: Factorial designs are used in experiments where the
effects of varying more than one factor are to be determined. They are
significant in several economic and social phenomena where many factors
usually affect a particular problem. Factorial designs can be of two types:
(i) simple factorial designs and (ii) complex factorial designs.

3.6 DEVELOPING A RESEARCH PLAN-

EXPLORATION, DESCRIPTION, DIAGNOSIS
AND EXPERIMENTATION
The entire research process starts with identifying the research problem. Once
the researcher identifies the research problem, the researcher must organize
their thoughts in a specific order, and he needs to define the terms related to
the research. They may use existing definitions for the terms or define them.
The researchers thought need to be converted as a 'Research Plan’.
Converting the researcher's ideas into a specific research plan will help the
researcher
structure in the following manner.
1. Experts, in particular, may provide suggestions to do research after
looking at the proposal. Thus, the researcher could be able to rectify
flaws in the research design.
2. It facilitates the researcher to focus on a review of the literature.
3. The research plan starts with a short introduction, research question,
objectives, definitions for the concepts, methodology, experimental
procedure (if it is experimental research), and theory ( if it is social
science research). The methodology includes sampling methods, data
collection tools and technology, statistical procedure to analyse data
and the of the research report.
4. The research plan also contains the time and cost of the study.

3.7 SAMPLING TECHNIQUES

As we mentioned earlier, sampling is an effective research technique as part
of the research process. There are various terms related to sampling. We will
explain those terms in this section, along with types.
First, we will learn the term ‘Universe/Population’. All items in any field of
inquiry constitute a ‘Universe’ or ‘Population.’ A complete enumeration of
all items in the ‘population’ is known as a census inquiry. You all must be
54
heard about the census survey of India. The government of India conducted a Research Design
census survey once in 10 years to generate complete data on population and
related indicators. The collected data from all populations in India will be
tabulated under various heads, and the researcher may find these data useful
for their research to bring certain inferences. However, it needs to be
emphasized that when the Universe is small, there is no use in adopting a
sample survey. At the same time, when field studies are undertaken in
practical life, considering time and cost, we have to select respondents, i.e.,
selecting only a few items. The selected respondents should be as
representative of the total population as possible to produce a miniature
cross-section. The selected respondents are called a ‘sample', and the
selection process is called the 'sampling technique.' The survey so conducted
is known as a ‘sample survey’. The researcher must prepare a sample design
for his study, i.e., plan how a sample should be selected and what size such a
sample would be.
Sample Designs
The researcher needs to finalize the size of the sample for the study. The
finalization of several items to be included in the sample, i.e., the sample
size. The selection of the size of the sample is called sample design. The
researcher needs to finalize the sample size before proceeding with data
collection.
There are two ways of selecting the sample: probability and non-probability.
The learners need to understand the fundamental difference between
probability and non-probability sampling before we proceed further. First, let
us learn what probability is. You all must have learned the definition of
probability in the mathematics and statistics courses. If one is unsure about
something, does not have enough information, and can only guess in the
process of guessing, there is uncertainty about the future course of action.
In daily life, we may face lots of uncertain situations. We always try to think
that a particular situation may come or may not. In this situation, the concept
of probability gives us a hand to answer a question like what is the likelihood
that X will win the election or “A” team will win the match? It is illustrative
of the concept of probability. If the chance of winning is one in five, the
probability is 1/5 =0.2. If the chances are 1 in 100, the probability is 0.01.

3.8 NEED FOR SAMPLING

Due to various reasons (including limited period and limited resources), it
may not be possible for the researcher to include the entire population in
survey research. Thus, they select a part of the population to understand
the characteristics of a population.
The process of systematically selecting cases from the population or Universe is
called sampling. Thus, a researcher gets a set of cases (or a sample) that
is more manageable and cost-effective to work with those cases to bring
scientific inferences. 55
Basics of Research The next question is, how we can generalize from a handful of cases?
Methodology Here we need to research logically with statistical reasoning, testing
repeatedly with empirical evidence. Moreover, a researcher cannot use
just any sample for generalization. Well-laid down sampling procedures
require rigorous exercises to increase a study's precision level.

3.9 SIGNIFICANT TERMS IN SAMPLING

Let us learn about essential terms in sampling
1. Sampling Ratio
The ratio of the size of the sample to the size of the target population is
the ‘sampling ratio’. For example, a college has 2000 students, and a
researcher draws a sample of 200 from it. Thus, the sampling ratio is
200/2000 = 0.1, or 10 per cent.
2. Sampling Frame
A population is an abstract concept. Except for specific small populations, one
can never indeed freeze a population to measure it. For example, in a city,
at any given moment, some are born, some may leave the city, some may
join the city as new entrants, and some may lose their life. The researcher
must decide exactly whom to count. Should she or he count a city resident on
holiday or outside the city when the time is fixed for the study? Therefore, the
notion of ‘population’ is abstract and exists in the mind but is impossible
to pinpoint concretely. Since it is an abstract concept, except for small
populations, a researcher needs to estimate the population. Thus, it
requires an operational definition.

A researcher operationalizes a population by developing a specific list

that closely approximates all the elements in the population. This list is a
‘sampling frame’. The researcher may choose from many sampling frames:
telephone directories, driving licences, ration cards, membership in a club, and
students' registration in a university. A reasonable sampling frame is crucial to
good sampling. A mismatch between the sampling frame and the conceptually
defined population can be a significant source of error. For instance, if
you select the telephone directory as a sampling frame, it constitutes
only 5-10 per cent of the population in a city, say, Delhi. The directory
does not list those who do not have a telephone connection. It also does not
update the frequent shifts of residences and changes in telephone numbers in a
city quickly.
3. Parameter and Statistic
Any characteristic of a population is a ‘population parameter’. For
instance, students from the science stream of children aged 6-11 years. It is
the true characteristic of the population. Parameters are determined when all
elements in a population are measured. The parameter is never known
with absolute accuracy for large populations, so researchers must estimate
56
it based on samples. Researchers use information from the sample, Research Design
called a ‘statistic’, to estimate population parameters.

4. Sampling Error

Sampling error is not necessarily the result of mistakes made in the

sampling procedure. Instead, variations occur due to the chance selection of
different individuals.

3.10 TYPES OF SAMPLING DESIGNS

Sampling designs are of two types. The first is called probability-sampling
design, and the second is non-probability sampling design. In this unit, we
discuss each type of sampling design in terms of its essential character and
procedure.
i) Probability Sampling Design: Probability sampling provides a statistical
basis for stating that a sample is representative of the ‘target population’. In a
probability sample, every element in the population has a known
chance of being included in the sample. This allows for estimates of the
accuracy of sample findings in approximating what we would find out if
we had conducted a census of the total population.
ii) Non-probability Sampling Design: In non-probability sampling
designs, we do not know whether an element of the population has an
equal chance of being selected. Its probability of selection cannot
be determined, as with probability sampling, where each element has a 50
per cent chance of being selected and a 50 per cent chance of not being
included in the sample. The non-probability samples are drawn based on
judgment regarding the characteristics of the target population and the
needs of a survey. With non-probability sampling, some members of the
eligible target population have a chance of being chosen, and others
do not. Thus, the statistical precision estimates cannot be made with this
sample. The non-probability sampling designs are preferred when there is no
possibility of probability sampling. Whenever feasible, probability-
sampling designs are preferred.

3.11 PROBABILITY SAMPLING PROCEDURES

The following are various probability sampling procedures:

i) Simple Random Sampling

In simple random sampling, each member of the population under study

has an equal chance of being selected. The method involves selecting at
random from a list of the population (a sampling frame). For example,
suppose we need to select 300 people from the Universe (village) study
to assess their access to common property resources. In that case, we
can write each person's name on a separate slip and then put the slip in
a container, and we can conduct a lottery to select 300 people to 57
Basics of Research conduct a study. In the case of an infinite population, the selection of
Methodology each item in a random sample is controlled by the same probability, and
successive selections are independent of each other. In this process,
there is an equal chance for everyone to get selected. One essential
requisite for this kind of sampling is that a complete list of the
population (sampling frame) is needed.
a. There is a chance to get an unbiased sample without much technical
difficulty.
b. It is unbiased
c. Selection will be based on a lottery or through a table of random
numbers, or now through computer-generated random numbers.
The lottery method is adopted for smaller populations or sampling frames.
For more oversized sampling frames, the computer-generated
numbers are selected.
d. It is not a statistically efficient method.
e. It does not provide fair representation to the sub-group population
ii) Systematic Sampling
It is another method of simple random sampling. In this method, the selection
is made from the list. The researcher takes the first unit randomly. After that,
he/she chooses from samples from fixed intervals. For example, If the
researcher can choose every 15th house on one side of the street or every 10th
name from the list, it is called systematic sampling. In this technique, once the
first sample unit is identified randomly, the remaining unit is chosen easily.
i ii ) Stratified Sampling
It is also called proportional random sampling. It is more efficient statistically
because it is an improved version of simple and systematic random sampling. In
this sampling method, the population is divided into specified strata. In this
Stratified Sampling, the population is grouped into homogenous groups (sub-
groups or strata) such as age, gender, class, and location. In stratified
sampling, the researcher controls the relative size of each stratum rather
than letting random processes control it. This guarantees the representation
of different strata within a sample. However, one condition is that the
stratified sampling procedures produce samples more representative of
the population than simple random or systematic samples if the stratum
information is accurate.
iv) Cluster Sampling
The most widely employed probability sample design in survey research is
cluster sampling. It addresses two problems: the lack of a reasonable
sampling frame and the costs of reaching a sampled element or a case.
For example, there is no single list of undergraduates in colleges in a city.
Even if one gets an accurate sampling frame, it would cost too much to
reach many undergraduates as the colleges are spread out geographically
58 in the city. Researchers use a sampling design that involves 'clusters' instead
of a single sampling frame in this case. In this case, the cluster will be Research Design
the college.
A cluster is a naturally occurring unit (e.g., a school with many
classrooms, students, and teachers; a city with zones (East, West, South,
Central, North); states. The clusters are selected randomly, and all
members of the selected clusters are included in the sample, or simply
random, systematic, or stratified samples are taken from each cluster.
Cluster sampling is used in extensive surveys. It differs from stratified
sampling in that with cluster sampling, one starts with a naturally
occurring constituency. The researcher selects from among the clusters
and either surveys all members of the selection or randomly selects from
among them. The resulting sample may not represent areas not covered by
the cluster, nor does one cluster necessarily represent another.
v) Stage Sampling
Stage sampling also known as multi-stage sampling is an extension of cluster
sampling. It involves the selection of a sample in several stages. That is,
taking samples from samples. Suppose we want to survey children's academic
performance in schools in a large city. One type of stage sampling might
be to select several schools randomly, and from each of these schools,
select several classes and select children from within these classes. Another
type of stage sampling could be to select one school in terms of either the
geographic region of the school in the community/city or in terms of the
type of school (public, private aided, private unaided, etc.) and take their
simple random sampling, systematic sampling, or stratified sampling or
even mixture of all these.

3.12 NON-PROBABILITY SAMPLING

PROCEDURES
The following are the non-probability sampling procedures:
i) Accidental or Convenience Sampling
This is also called the 'man-on-the-street' survey. It involves choosing
the individuals/cases readily available on the street, marketplace, school,
or cinema theatre until the required sample size is obtained. This
survey can produce ineffective, highly unrepresentative samples and is
not recommended. When a researcher haphazardly selects a convenient
sample, he can quickly get a sample that misrepresents the population. Such
samples are cheap and quick, although biases and errors are plenty. An
example of this kind of survey is the kind of interviews, television
programmes conduct on the street.
ii) Quota Sampling
Quota sampling has often been described as the non-probability equivalent of
stratified sampling. In quota sampling, the researcher divides the
population into sub-groups or categories such as men and women, reserved castes 59
Basics of Research and non-reserved castes, arts and commerce, younger and older, and then decides
Methodology the proportion of individuals in each category or sub-group. Thus, the number of
respondents is fixed in various sample categories.
iii) Purposive or Judgmental Sampling
In purposive sampling, researchers handpick the cases to be included in the
sample based on their judgment of their typicality. That means the
researcher's judgment is used in selected cases with a specific
purpose. Purposive sampling is appropriate in the following three
situations. First, a researcher uses it to select unique cases that are exceptionally
informative. Second, a researcher may use purposive sampling to select
members of a difficult-to-reach, specialized population. Third, purposive
sampling is preferred when a researcher wants to identify particular
types of cases for in-depth investigation.
iv) Dimensional Sampling
This is a further refinement of quota sampling. It involves identifying
various factors of interest in a population and obtaining at least one
respondent of every combination of those factors.
v) Snowball Sampling
It is also called a network or chain referral or reputational sampling, and it
is a method for identifying and sampling (or selecting) the cases in a
network. Snowball sampling is based on an analogy to a snowball which
begins small but becomes more extensive as it rolls on wet snow and
picks up additional snow. Friendship networks are the most important
sources of this kind of sampling. Further, professional associations and
chat groups may also provide bases for such sampling.
Sample Size
A question often asked is about the size of the sample. How large should
it be? The best answer could be that it depends. That means there is no
clearcut answer to this question. However, the survey sampling literature
does give prospective survey researchers general advice. For some people,
a sample of thirty is the bare minimum if the researcher plans to
undertake some form of statistical analysis of the data. However, the
sample size depends on the researcher's plans, how accurate the sample
has to be for his purposes, and population characteristics. It may be
stated here that a large sample alone does not guarantee a
representative sample. Further, when we increase the sample size, we
increase costs. Larger samples mean increased costs. A researcher may keep in
mind all these considerations while judging the size of the sample.
CHECK YOUR PROGRESS 1

Note: i) Use the space given below for your answers.

ii) Check your answers with those given at the end of the unit.
60
1. Write a short note on the principles of research design. Research Design

……………………………………………………………………………….

……………………………………………………………………………….

……….………………………………………………………………………

………………….……………………………………………………………

2. Write a short note on sampling techniques.

……………………………………………………………………………….

……………………………………………………………………………….

……….………………………………………………………………………

………………….……………………………………………………………

3. Write a short note on types of research design.

……………………………………………………………………………….

……………………………………………………………………………….

……….………………………………………………………………………

………………….……………………………………………………………

3.13 LET US SUM UP

The Unit introduced the learners to the significance of developing research
design. In this unit, we have discussed the need for research design and
principles and various types of research design. We have also discussed the
need for sampling, sampling techniques and different types of sampling
procedures.

3.14 KEY WORDS

Completely : CRD involves only two principles namely the
Randomized Design principle of replication and the principle of
(CRD) randomization of experimental designs. The
essential characteristic of the design is that subjects
are randomly assigned to experimental treatments or
vice-versa.

Research Design : The research design is the arrangement of

conditions for data collection and analysis in a
manner that aims to combine relevance to the
research purpose with the economy in procedure.

61
Basics of Research
Methodology
3.15 SUGGESTED FURTHER READING/
REFERENCES
Aldridge, A. and Levine, K. (2001). Surveying the Social World -
Principles and Practice in Survey Research. Open University Press:
Buckingham.
Bryman, A. (2008). Social Research Methods. 3rd Edition, Oxford University
Press., New York.
Cohen, L., Manion, L., and Morrison, K. (2007). Research Methods
in Education. Routledge: London and New York
Kothari, C.R. and Garg. G. (2019). Research Methodology: Methods and
Techniques., New Delhi: New age Publishers.
Malik, SM. (2014). Sampling Methods and Estimation of Sample Size (Unit
19) In: MGS 005 Research Methodologies in Gender and Development
Studies, IGNOU, New Delhi. Available at
https://egyankosh.ac.in/handle/123456789/8711
Moser, C. A. and Kalton, G. (1971). Survey Methods in Social Investigation.
The English Language Book Society: London
Sarnamohan, V. (2014). Non-Probability Sampling (Unit 2) In: Block 4
Quantitative Methods. MGS 005 Research Methodologies in Gender and
Development Studies, IGNOU, New Delhi.
Sarnamohan, V. (2014). Probability Sampling (Unit 1) In: Block 4
Quantitative Methods. MGS 005 Research Methodologies in Gender and
Development Studies, IGNOU, New Delhi.
Selltiz, C., Jahoda, M., Deutsch, M. and Cook, S.W. (1964). Research
Methods in Social Relations.
Singleton Jr., R.A. and Straits, B.C. (2017). Approaches to Social Research,
6th Edition. Oxford University Press, Oxford.

3.16 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 1

1. Please refer to section 3.4

2. Please refer to section 3.7

3. Please refer to section 3.5

62