Chapter Four: Research Design and Sampling

 Research design
 Research design is the conceptual structure within which research would be
conducted.

 It is the general blueprint for the details about the problem, objectives, research
questions, area of study, sampling plan, techniques of data collection and
measurement, and analysis.

 It is the logical and systematic planning and directing the research.

 Research designs indicating the purpose of doing research, intended use and effects of
time.
 There are four parts of research design
1. Sampling design: It describes the various sampling methods to various used for
selecting subject for study.
2. Observational design: It describes the way in which the observations are to be
made.
3. Statistical design: It describes the statistical techniques to be used for analyzing and
interpreting data.
4. Operational design: It describes how the entire operation of research is to be
carried out.
o Research designs vary depending upon the type of research.
o By and large, research designs are framed in three types: Exploratory type of research,
Descriptive type of research, and Hypothesis-testing type of research.
 Important Concepts Relating to Research Resign
 Dependent and independent variables
 A concept which can take on different quantitative values is called a variable.
 As such the concepts like gender, weight, height, income are all examples of
variables.
 Phenomena which can take on quantitatively different values even in decimal
points are called ‘continuous variables’. Eg. Age, Distance, weight, height, etc.
How Far Is There From Your Dorm To Class room? ___ (km)
 If they can only be expressed in integer values, they are non-continuous variables
or in statistical language ‘discrete variables’.
 Discrete variable only takes countable values. Eg. How Many Books Did You
Buy Last Month? ___ (Number). Eg. Number of children, car, houses, etc.
 Categorical variables place the person or thing into a category on the basis of some
qualitative property. Eg. Do You Own A Car? __ Yes __ No; Sex? __ Female __ Male;
etc.
Variable Type Information Type Measurement Type

 Dependent and independent variable

When one variable causes change in another, we call the first variable the independent
variable or explanatory variable.
The affected variable is called the dependent variable or response variable: stimulus,
response.
 The probability of participation off-farm income generating activities is the function of
age, schooling, household size, livestock (in total livestock unit), plot of land owned;
idiosyncratic and covariate shocks; credit and assistance.
 Off-farm income generating activities is dependent variables; whereas the rest of other
variables are independent variables which are affecting participation.
 Extraneous variable: Independent variables that are not related to the purpose of the
study. Eg. In the study of children work effect on educational outcomes; Intelligence may
as well affect educational outcomes, but since it is not related to the purpose of the study
undertaken by the researcher.
 Control: when we design the study minimizing the effects of extraneous independent
variables.
 Experimental and non-experimental hypothesis-testing research
 In experimental hypothesis-testing research you have control and experimental
 treatment groups; and test the hypothesis.
 Whereas, non-experimental hypothesis-testing research there is no
experimental/treatment and control group.
 Treatments: The different conditions under which experimental and control groups are put
are usually referred to as ‘treatments’
 The pre-determined plots or the blocks, where different treatments are used, are known as
experimental units.
 Experimental and Control Group
 An experimental group, also known as a treatment group, receives the treatment
whose effect researchers wish to study, whereas a control group does not.
 Researchers change the independent variable in the treatment group and keep it constant
in the control group. Then they compare the results of these groups.
 Example: You are interested in whether college students perform better in school if they
are paid for their performance. To test this, you divide several students into control and
treatment groups.
o You pay the students in the treatment group for achieving high grades.
o Students in the control group do not receive any money.
 By comparing the average change in their grades over the year, you can find out
whether monetary incentives improve school performance.
 The process of examining the truth of a statistical hypothesis, relating to some research
problem, is known as an experiment.
 Types of Research Design
 Research designs vary depending upon the type of research. The choice of which
design to apply depends on the nature of the problems posed by the research.
 Each type of research design has a range of research methods that are commonly used
to collect and analyze the type of data that is generated by the investigations.
 Different research designs can be conveniently described if we categorize them as:
1. Experimental
2. Quasi experimental
3. Cross-sectional
4. Longitudinal and
5. Case studies.
 Experimental research design: Identifies effects by a random assignment of a treatment
between a control and a treatment group.

Method of allocation to Pre-test Intervention Post-test

groups (X)
Random (experimental Measure on outcome ‘Treatment’ Measure on outcome
group) variable (Y) variable (Y)
Random (control group) Measure on outcome No Measure on outcome
variable (Y) ‘treatment’ variable (Y)

 The experimental and control groups need to be similar before the experiment starts.
 Then, participants are assigned at random to experimental and control groups in the
hope that this will result in equal assignment of people to both groups.
 Example: Improved wheat seed given to farmers (experimental group) and the control
group who did not received improved (received local) wheat seed.
 Improvements in the two groups are measured after some time
 If productivity of wheat for experimental is greater than control group, the improved
seed increased the productivity.
 Quasi-experimental research design
 If random assignment is not followed, then the design becomes quasi-experimental.
 Uses a comparison group similar to the treatment group (but there is no random
assignment).
 It is a research design used in situations where two or more groups of participants or
data can be identified, and one is used as the control and the other as the
experimental group.
 When a group is tested for the influence of a variable and compared with a control group
which has not been subjected to the variable.
 Cross-sectional research design
 Cross-sectional field surveys, independent and dependent variables are measured at
the same point in time (e.g., using a single questionnaire).
 includes more than one case;
 collects data at one particular time;
 includes within its research participants groups of people or cases that can be
compared,
 E.g. men and women, people of different ages or ethnicities, or people from different
cities.
 Longitudinal research design
 A longitudinal study follows the same sample over time and makes repeated
observations.
 The same group of people is interviewed at regular intervals, enabling researchers to
track changes over time and to relate them to variables that might explain why the
changes occur.
 Longitudinal research designs describe patterns of change and help establish the
direction and magnitude of causal relationships.
 This allows the researcher to measure change in variables over time.
 Panel studies– these consist of a sample of people, often randomly selected, who are questioned
more than once.
 Cohort studies– these concentrate on a group that shares similar characteristics, such as students
from a particular year of matriculation or people on strike at a certain time.
 Case Study research design
 A case study is a research method involving an up-close, in-depth, and detailed
examination of a particular case.
 A case study is an in-depth study of a particular research problem rather than a
sweeping statistical survey.
 The case study research design is also useful for testing whether a specific theory and
model actually applies to phenomena in the real world.
 In the case study research, data may be collected using a combination of interviews,
personal observations, and internal or external documents.
 The selection of the case may be based on:
  Critical case. Here the case is chosen as one that will enable the researcher to test a
theory or hypothesis.
 Extreme or unique case. Here the focus is on a situation or group that is perceived to
be different to any other and the focus is on its uniqueness.
 Representative or typical case. This is a case that is chosen because it is seen to
represent many other similar cases
 Revelatory case. This is a case that has the potential to shed light on the research
topic.
 It may be that the researcher has access to a situation that has been hidden.
 Sampling Design
 Census and Sample Survey
 All items in any field of inquiry constitute a ‘Universe’ or ‘Population.’
 Population can be defined as all people or items (unit of analysis) with the characteristics
that one wishes to study.

 The unit of analysis may be a person, group, organization, country, object, or any other
entity that you wish to draw scientific inferences about.

 A complete enumeration of all items in the ‘population’ is known as a census inquiry.

 But in practice this may not be true. When the field of inquiry is large, this census inquiry
becomes difficult to adopt because of the resources involved.

 There are difficulties in measuring whole populations: The large size of many populations;
Inaccessibility of some of the population; Destructiveness of the observation ; and
Accuracy and sampling.
 When field studies are undertaken in practical life, considerations of time and cost almost
invariably lead to a selection of respondents i.e., selection of only a few items.
 The selected respondents constitute what is technically called a ‘sample.
 A sample is a “subgroup of a population”.
 A sample is a group of people, objects, or items that are taken from a larger population for
measurement.
 The sample should be representative of the population to ensure that we can generalize the
findings from the research sample to the population as a whole.
 The selection process is called ‘sampling technique.’ Sampling techniques can be grouped
into two broad categories: probability (random) sampling and non-probability sampling.
 The survey so conducted is known as ‘sample survey’
 Reasons for Sampling
o Reduced cost
o Greater speed
o Greater scope and accuracy
o Representativeness
o Feasibility
 Developing a sampling design, the researcher must pay attention to the following points:
o Sample design is to clearly define the set of objects, technically called the Universe, to
be studied.
o A decision has to be taken concerning a sampling unit before selecting sample.
  Sampling unit may be a geographical one such as state, district, village, etc., or a
construction unit such as house, flat, etc., or it may be a social unit such as family,
club, school, etc., or it may be an individual.
 Source list: It is also known as ‘sampling frame’ from which sample is to be drawn.
 Size of sample: the number of items to be selected from the universe to constitute a
sample.
 Considering the question of the specific population parameters which are of interest.
 Cost considerations
 Sampling procedure: Finally, the researcher must decide the type of sample he/she will
use i.e., he must decide about the technique to be used in selecting the items for the
sample.
 Researcher must keep in view the two causes of incorrect inferences viz., systematic
bias, and sampling error.
 A systematic bias results from errors in the sampling procedures, and it cannot be
reduced or eliminated by increasing the sample size.
 At best the causes responsible for these errors can be detected and corrected.
 Usually a systematic bias is the result of one or more of the following factors:

 Inappropriate sampling frame: a biased representation of the universe, it will result

in a systematic bias.
 Defective measuring device: systematic bias can result if the questionnaire or the
interviewer is biased; systematic bias in the data collected through such a measuring
device.
  Non-respondents: If we are unable to sample all the individuals initially included in
the sample.
 Indeterminacy principle: Sometimes we find that individuals act differently when kept
under observation than what they do when kept in non-observed situations.
 Natural bias in the reporting of data; arises in income data, etc.
 Sampling error can make a sample unrepresentative of its population.

The measurement of sampling error is usually called the ‘precision of the sampling plan’.

If we increase the sample size, the precision can be improved.

Non-sampling Error: are attributed mainly to differential behavior of respondents as well

as interviewers. Source from difference in response, difficulties in defining, difference in
interpretations, and inability in recalling information and so on.
 Sample size determination: act of choosing the number of observations to include in a
sample.
 Sample size used in a study is usually determined based on the cost, time, or
convenience of collecting the data, and the need for it to offer sufficient statistical
power.
 Sample size depends on degree of homogeneity, degree of confidence required, cost,
number of sub groups to be studied, practicality, Prior information, Rule of Thumb:
based on past experience with samples that have met the requirements of the
statistical methods, using formula such as Cochran’s and Yamane Formula, etc.
 Sampling Technique
 Depending upon the nature and scope of the investigation and the situations under which
the study is being carried out, appropriate sampling technique is being chosen.
 Available sampling techniques can broadly be categorized into two categories:
o Probability sampling and
o Non-probability sampling
 Probability sampling is a technique in which every unit in the population has a chance
(non-zero probability) of being selected in the sample, and this chance can be accurately
determined.
 Every item of the universe has an equal chance of inclusion in the sample.
 Called random sampling’ or ‘chance sampling’ or Lottery method.
 Probability method of sampling which gives the probability that our sample is
representative of population.
 Populations are not always quite as uniform or one-dimensional as, say, a particular type
of component in a production run, so simple random selection methods are not always
appropriate.
 Therefore, the first question to be asked is about the nature of the population: is it
homogeneous or are there distinctly different classes of cases within it, and if so, how are
they distributed within the population (e.g. are they grouped in different locations, found
at different levels in a hierarchy, or are they all mixed up together)?
 Types or Techniques Probability Sampling: There are a number of techniques of taking
probability sample.
 But here only six important techniques have been discussed as follows:
o Simple random sampling
o Systematic sampling
o Stratified sampling
o Multiple or Double sampling
o Multi-stage sampling
o Cluster sampling
 Simple random sample is one in which each element of the population has an equal and
independent chance of being included in the sample i.e. a sample selected by
randomization method.
 Simple random sampling is used when the population is uniform or has common
characteristics in all cases (e.g. medical students, international airports, dairy cows).
 Simple random sampling is the most widely used, simplest method of drawing sample
from a population.
 Advantages
 It requires a minimum knowledge of population.
 It is free from subjectivity and free from personal error.
 It provides appropriate data for our purpose.
 The observations of the sample can be used for inferential purpose.
 Disadvantages
 The representativeness of a sample cannot be ensured by this method.
  This method does not use the knowledge about the population.
 The inferential accuracy of the finding depends upon the size of the sample.
 Systematic Sampling
 Systematic sampling involves a random start and then proceeds with the selection of
every 𝒌𝒕𝒉 element from that point onwards;
where k = N/n, where k is the ratio of sampling frame size N and the desired sample size
n, and is formally called the sampling ratio.
 In this approach, as before, a specific sample size is determined.
 Then, the size of the sample is divided by the total eligible population to determine the
sampling fraction.
 Advantages
 This is a simple method of selecting a sample.
 It reduces the field cost.
 Inferential statistics may be used
 Sample may be comprehensive and representative of population.
 Observations of the sample may be used for drawing conclusions and generalizations.
 Disadvantages
 Knowledge of population is essential.
 Information of each individual is essential.
 This method can’t ensure the representativeness.
 There is a risk in drawing conclusions from the observations of the sample.
 Stratified Sampling: applied when a population from which a sample is to be drawn does
not constitute a homogeneous group.
 The sampling frame is divided into homogeneous and non-overlapping subgroups
(called “strata”), and a simple random sample is drawn within each subgroup.
 Strata are formed on the basis of common characteristic(s) of the items to be put in
each stratum.
 This means that various strata be formed in such a way as to ensure elements being
most homogeneous within each stratum and most heterogeneous between the
different strata.
 Strata are purposively formed and are usually based on past experience and personal
judgement of the researcher.
 Advantages
 It is (more precisely third way) a good representative of the population.
 It is an improvement over the earlier.
 It is an objective method of sampling.
 Observations can be used for inferential purpose.
 Disadvantages
 It is difficult for the researcher to decide the relevant criterion for stratification.
 Only one criterion can be used for stratification, but it generally seems more than
one criterion relevant for stratification.
 It is costly and time consuming method.
 Selected sample may be representative with reference to the used criterion but not for the
other.
 There is a risk in generalization,
 Multiple or Double or Repetitive Sampling
 When employing a mailed questionnaire, double sampling is sometimes used to
obtain a ‘more representative sample.
 This is done because some randomly selected subjects who are sent questionnaires
may not return them.
 To eliminate this bias, a second sample may be drawn at random from the non-
respondents and the people interviewed to obtain the desired information. Thus this
technique is also known as repeated or multiple sampling.
 Advantages
o This sampling procedure leads to the inferences of free determine precision based on
a number of observations.
o This technique of sampling reduces the error.
o This method maintains the procedure of the finding evaluate the reliability of the
sample.
 Disadvantages
o This technique of sampling cannot be used for a large sample. It is applicable only
for small sample.
o This technique is time consuming, costly, and requires more competition.
o It’s planning and administration is more complicated.
 Cluster Sampling
 When the population is very wide or big, say for countrywide survey, it may not be
feasible to take sample units directly from the population itself.
 Then cluster sampling technique is applied.
 Then some clusters are selected at random out of all the clusters.
 In such cases, auxiliary/secondary information like block list, village list, and
subdivision lists is used in probability sampling.
 Advantages
o It may be a good representative of the population.
o It is an easy method.
o It is an economical method
 o It is practicable and highly applicable in education.
o Observations can be used for inferential purpose.
 Disadvantages
o Cluster sampling is not free from error.
o It is not comprehensive.
 Multi-Stage Sampling
 Multistage sampling is the extension of clustering in more than one stage.
 Instead of taking blocks/ villages directly as clusters, one may take districts at the
first stage, then blocks/villages from the selected districts and farmers from the
villages, etc.
 A multistage sampling is a two-stage/three-stage, etc., sampling depending upon the
number of stages or clusters.
 Advantages
o It is easier to administer than most single stage designs mainly because of the fact
that sampling frame under multi-stage sampling is developed in partial units.
o A large number of units can be sampled for a given cost under multistage sampling
because of sequential clustering, whereas this is not possible in most of the simple
designs.
 Disadvantages
o It is a difficult and complex method of samplings.
o It involves errors when we consider the primary and secondary stages.
o It is again a subjective phenomenon.
 Non-Probability Sampling
 Non-probability sampling is a sampling technique where the samples are gathered
in a process that does not give all the individuals in the population equal chances of
being selected.
 Non-probability sample is not a product of a randomized selection processes.
 Why use non-probability sampling method?
 It is also useful when the researcher has limited budget, time, and workforce.
 It can be used when randomization is impossible like when the population is
almost limitless.
 It can also be used when the researcher aims to do a qualitative, pilot, or
exploratory study, etc.
 Advantages of Non-Probability Sampling
o Possibility to reflect the descriptive comments about the sample.
o Cost-effectiveness and time-effectiveness compared to probability sampling.
o Effective when it is unfeasible or impractical to conduct probability sampling.
 Disadvantages of Non-Probability Sampling
o Unknown proportion of the entire population is not included in the sample group i.e.
lack of representation of the entire population
o Lower level of generalization of research findings compared to probability sampling
o Difficulties in estimating sampling variability and identifying possible bias.
 Types of Non-Probability Sampling
 Convenience Sampling
 Quota Sampling
 Judgmental Sampling
 Snowball Sampling
 Convenience Sampling: samples are selected because they are accessible to the
researcher.
 It is called accidental or opportunity sampling, this is a technique in which a sample
is drawn from that part of the population that is close to hand, readily available, or
convenient.
 Advantages
o It is very easy method of sampling.
o It is frequently used in behavioral sciences.
o It reduces the time, money, and energy i.e. it is an economical method.
 Disadvantages
o It is not a representative of the population.
o It is not free from error.
o Parametric statistics cannot be used.
 Judgmental Sampling: selecting a sample on the basis of researcher’s own
knowledge of the population, its elements, and the nature of aims of the research.
 It is called Purposive non-probability sample.
 Advantages
o Use of the best available knowledge concerning the sample subjects.
o Better control of significant variables.
o Sample groups data can be easily matched.
o Homogeneity of subjects used in the sample.
 Disadvantages
o Reliability of the criterion is questionable.
o Knowledge of population is essential.
o Errors in classifying sampling subjects.
o Inability to utilize the inferential parametric statistics.
o Inability to make generalization concerning total population.
 Quota Sampling
 In this technique, the population is segmented into mutually exclusive subgroups
(just as in stratified sampling), and then a non-random set of observations is
chosen from each subgroup to meet a predefined quota.
 Under quota sampling the interviewers are simply given quotas to be filled from
the different strata, with some restrictions on how they are to be filled.
 Advantages
o It is an improvement over the judgement sampling.
o It is an easy sampling technique.
o It is most frequently used in social surveys.
 Disadvantages
 It is not a representative sample.
 It is not free from error.
 It has the influence of regional geographical and social factors.]
 Snowball Sampling: Network Sampling or chain sampling
 Snowball sampling is usually done when there is a very small population size.
 In this type of study, the researcher finds out initially one or two or a few
respondents for the study purpose and then takes the help of these respondents to
find more and more respondents on specific aspects.
 Problems in Sampling
 Survey errors
 Sampling errors occur because of variation in the number or representativeness of the
sample that responds.
 Sampling errors occur as a result of calculating the estimate (estimated mean, total,
proportion, etc) based on a sample rather than the entire population.
 Non-sampling error
 Wrong population is being sampled.
 Non response error ; common in self-administered surveys.
 Interviewer’s error.
 Non-coverage error: Omission of part of the target population.
 Instrument error
“End of Chapter Four”