RESEARCH METHODOLOGY

ASSIGNMENT
SEMESTER: 3

Discuss sampling and the various types of nonprobability sampling?

A sample consists of a small portion of cases a researcher selects

from a large pool of cases and generalizes to the population or a
subset of a population. In quantitative research methodology, a
sample is explained as a set of data collected and selected from a
population by a defined procedure, while sampling refers to the act,
process, or technique of selecting a representative part of a population
for the purpose of finding characteristics of the whole population.

Figure 1: Sample in a population

Depending on the study, the method we use for sampling can differ in
qualitative and quantitative manner. The primary use of sampling in a
quantitative study is to create a representative sample that closely
reproduces features of interest in a larger collection of cases called
population. Typically, the population is very large, making a census
of all the values in the population is impossible. The sample usually
represents a subset of a manageable size. Samples are collected and
are calculated from the samples to help in making inferences from the
sample to the population. The sample is collected from a population,
in which case it is a subset of a population. This helps in generalizing
results of the population through precise sampling procedures in
quantitative research. The procedures here are called probability
sampling.
In most quantitative studies, one wants to see how many categories of
the population falls under categories of interest. Probability samples
can be highly accurate. For large populations, data from well
designed, carefully executed probability sample are often trying to
reach every cause in the population in an accurate manner. To make it
concise, in qualitative sampling, we select cases and treat them as
cases of certain features of the social world. A sample of cases ‘stands
in' for representation of the much larger population's cases.
Sampling proceeds differently in qualitative studies and usually has
a different purpose from quantitative studies. In qualitative studies, in
order to allow us to make statements about the categories in
population, we rarely sample to get a small set of causes that is
accurate to reproduce tendencies of the entire population. Instead, we
identify relevant categories at work in a few cases. In contrast to
quantitative sampling, the central idea of qualitative sampling is to
sample features of the social world. Here, our goal is to deepen our
understanding of the larger process, relationship, or social scene. The
aspects of our sample highlight into key dimensions or processes of
complex social life. We pick a few to provide clarity, insight, and
understanding about the issues or relationships in the social world.
A sample provides us with valuable information or new aspects which
accentuate, enhance, or enrich, key features or situations. We sample
to find theoretical insights, find distinctive aspects of social settings,
events or relationships. We shouldn't overdo the qualitative-
quantitative distinction. Most quantitative studies use probability or
probability like sampling, while most qualitative studies use the
nonprobability method and nonrepresentative strategy.
Sampling strategies:
To avoid sampling in a sloppy manner, systematic sampling is
required while to make a sample appropriate for the study, a sampling
strategy which matches the specific purpose is required. For creating a
representative sample, that mirrors or represents many cases that we
cannot directly examine, two techniques of sampling are present. The
preferred method considered as ‘gold standard' for representative
samples is called the probability sample. A second, simpler way to
produce a representative sample is to use a nonprobability sampling
technique. They have been discussed below:

Probability sampling technique:

In probability sampling, one can determine which sampling units
belong to which sample and the probability that each sample will be
selected in the process. In the probability sampling method, all the
members of the population have a pre-specified and an equal chance
to be a part of the sample. This technique is based on the principle of
random selection, wherein the procedure is designed in a way which
guarantees that each and every individual of the population has an
equal selection opportunity. This helps to reduce the possibility of
bias making the sample fully representative.
Statistical inferences can be made by the researchers using this
technique, i.e. the result obtained can be generalized from the
surveyed sample to the target population. The methods of probability
sampling are defined briefly:

Simple Random Sampling (SRS): A simple random sample is a

subset of a statistical population in which each member of the subset
has an equal probability of being chosen. An example of a simple
random sample can be the names of 25 employees who are chosen out
of a hat from a company of total 250 employees. In this case, the
population is all 250 employees, and the sample is random because
each employee has an equal chance of being chosen in the sample
group.

Stratified Sampling: Stratified sampling refers to another type of

sampling method. With stratified sampling, the researcher divides the
population into separate groups, called strata. Then, a probability
sample (often a simple random sample ) is drawn from each group.
Stratified sampling has several advantages over simple random
sampling. For example, by using stratified sampling, it can be
possible to reduce the sample size required to achieve a precise
sample. Or it may be possible to increase the precision with the same
sample size too.

Cluster Sampling: With cluster sampling, the researcher divides the

population into separate groups, called clusters. Then, a simple
random sample of clusters is selected from the total population. The
researcher conducts his analysis on data from the sampled clusters.
Compared to simple random sampling and stratified sampling, cluster
sampling has advantages and disadvantages. For example, if equal
sample sizes are given, cluster sampling usually provides less
precision than the earlier two methods. On the other hand, if travel
costs between clusters are high, cluster sampling may be more cost-
effective than the other methods at hand.

Systematic Sampling: With systematic sampling, one tends to create

a list of every member existing in the population. From the list, one
randomly select the first sample element from the first x elements on
the population list. Thereafter, one selects every xth element on the
list present.
This method is different from the simple random sampling method
since every possible sample of n elements is not equally likely to be
chosen.

Nonprobability sampling techniques:

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. In any form of research,
true random sampling is a difficult task to achieve.

Figure 2

Most researchers are bounded by time, money and workforce overdo

and because of these limitations, it is almost impossible to randomly
sample the entire population and it is often necessary to employ
another sampling technique, the non-probability sampling technique.
In contrast with probability sampling, the non-probability sample is
not a product of randomized selection processes. Subjects in such a
sample are usually selected on the basis of their accessibility or by
personal judgment of the researcher.
The downside of the non-probability sampling method is that an
unknown proportion of the entire population was not sampled. This
makes sure that the sample may or may not represent the entire
population correctly. Therefore, the results of the research cannot be
used to obtain generalizations pertaining to the entire population.

Convenience Sampling: This is also known as accidental,

availability or haphazard sampling. The primary criteria for selecting
cases is that they are convenient, easy to reach or readily available.
This method can be used for some explanatory preliminary studies
and some qualitative research. Unfortunately, it often produces very
nonrepresentative samples and isn't recommended for creating
samples representative of the population. By selecting samples based
on convenience, we misinterpret the features of the entire population.
Its because convenience samples are quick, easy and cheap. People
are also ignorant about how to create a good representative sample.
The cons of convenience sampling include possible bias in data
gathering, the possibility of sampling error and lack of generalized
results.
Example: A person on the street interview conducted by T.V.
Channels is usually carried forward by interviewing people who are
convenient to talk to. Such samples may have entertainment value
but are highly misleading as data of a representative population.

Quota Sampling: For many purposes, a well-designed quota

sampling is acceptable as the nonprobability substitute method for
producing a quasi-representative sample. Here, we first identify
relevant categories among the population we are sampling to capture
diversity among the population. Next, we determine how many cases
to get for each category called ‘quota.' Thus we fix a certain number
of cases per category.
Quota sampling is relatively easy. All the selected participants fit an
unfilled quota. Quota sampling has three weaknesses including that
they only capture a few aspects of the diverse population and ignore
others, fixed ratio of samples may not adequately represent the
proportion of cases in the population and some amount of
convenience sampling is used in this process. The following image
shows quota sampling used in terms of gender and age for selecting
participants in a study.

Figure 3 : Quota Sampling

For example, if the basis of the quota is college year level and the
researcher needs equal representation, with a sample size of 100, he
must select 25 1st year students, another 25 2nd year students, 25 3rd
year and 25 4th year students. The bases of the quota are usually age,
gender, education, race, religion and socioeconomic status.

In qualitative research, the purpose may not be to create a

representative sample but to create a sample which the suits the
purpose of study better. Nonprobability sampling often selects cases
with the specific content of the case determining whether it is chosen.
The following techniques are relevant to this case:
Purposive or Judgemental Sampling: This is a valuable sampling
type for special situations. It is used in exploratory or field research. It
uses the judgment of an expert in selecting cases or it selects cases
with a specific purpose in mind. It is informative if the goal is to pick
a representative sample because here, we rarely select the
representative population. It is appropriate to select cases that are
especially informative. Example: We use this to study cases that are
in difficult to reach in the specific population such as prostitutes. It is
difficult to study all the prostitutes, so we locate people who are
prostitutes and study them. We also use purposive sampling to
identify particular types of cases for an in-depth deeper investigation
to gain a deeper understanding of things. The cons of Purposive
Sampling include vulnerability to errors in judgment by the
researcher, low level of reliability and high levels of bias and Inability
to generalize research findings.

Snowball Sampling: We are often interested in the interconnected

network of people and organizations. The crucial feature is that each
person is connected to one and other through a direct or indirect
linkage. Though a person may not directly know everyone in the
network but is influenced by everyone in the linkage. Researchers
represent such networks by drawing a sociogram – A diagram of
circles connected with lines representing linkages. This is also called
network, chain referral, reputational, and respondent-driven sampling.
The method uses the analogy of a snowball which starts small initially
but gets bigger as it rolls on wet snow. This is a multi-stage technique.

Figure 4 - Sociogram

Example: The figure above represents the example aong with the
sociogram. To study friendship network among teenagers in a
community, We ask 3 teenagers their names, and names of their four
best friends, this goes on so forth and we identify a large network of
friendships. Each is indirectly related to the original teenager. The
process stops if the network is closed or outside the limit of study.
The disadvantages include that The researcher has little control over
the sampling method, representativeness of the sample is not
guaranteed and the sampling bias is also a fear of researchers when
using this sampling technique.
Deviant Case Sampling: This is also known as extreme case
sampling. We use this when we are interested in cases that are
different from the dominant or mainstream characters of other uses.
Similar to purposive sampling, a variety of techniques to locate cases
with specific characteristics are used. The goal is to collect cases that
are unusual, deviant, and aren't representative of the whole. We can
sometimes learn more about social life by considering cases that fall
outside the general pattern of behaviour. Example: If we want to
study high school dropouts, the previous research suggests that a
majority of dropouts come from low income, single-parent families
and tend to be racial minorities. The family environment where
parents or siblings are themselves dropouts matter. They can also
engage in illegal behaviour. We might study dropouts who do not
come under the specific criteria and learn more reasons for dropping
out. The cons include the difficulty to find deviant cases and that it
provides inaccurate generalizations.

Sequential Sampling: In this, we continue to gather cases until the

new amount of information ends or a certain diversity of cases is
reached. Thus one gathers cases until the saturation point. It requires
to continuously gathering all cases. Sequential sampling is a non-
probabilistic sampling technique, initially developed as a tool for
product quality control. The sample size isn't fixed in advance, nor is
the timeframe. The process begins, first, with the sampling of a single
observation or a group of observations. These are then tested to see
whether or not the null hypothesis can be rejected. This sampling
method is hardly representative of the entire population. Its only
chance of approaching representativeness is when the researcher
chooses to use a large sample size significant enough to represent a
big fraction of the entire population. Example: We locate and plan an
in-depth interview of 60 widows over 60 years of age who have been
living without a spouse for 10 or more years. Depending on our
purpose, getting more 20 widows whose experience and background
differ from the first 60 may be unnecessary.
Theoretical Sampling: Here, what we sample comes from a
grounded theory. A growing theoretical interest guides the selection
of sample cases. The researcher selects cases based on the new
insights the sample can provide. Some disadvantages are associated
with this sampling method. It is a highly systematic method and thus
the application of theoretical sampling requires more resources like
time and money as compared to other sampling methods. It is a very
complicated method and isn't easy to understand. To achieve
proficiency in developing these categories researcher proceeds to
another location to increase his knowledge in the category which
sounds very complex and may be problematic.
Example: A field researcher who observed people at a particular site
during weekdays can observe them during other time periods to have
a fuller picture and to observe whether important conditions are the
same.

Adaptive Sampling: This is a design that adjusts based on early

observations. Adaptive sampling is a sampling technique that is
implemented while a survey is being fielded—that is, the sampling
design is modified in real time as data collection continues—based on
what has been learned from the previous sampling that has been
completed. Its purpose is to improve the selection of elements during
the remainder of the sampling, thereby improving the
representativeness of the data that the entire sample yields.

To conclude, nonprobability sampling can be used when

demonstrating that a particular trait exists in the population. It can
also be used when the researcher aims to do a qualitative, pilot or
exploratory study. It can be used when randomization is impossible to
obtain like when the population is almost limitless. It can also be used
when the research does not aim to generate results that will be used to
create generalizations pertaining to the entire population.

REFERENCES:
Neuman, W.L. (2006). Social Research Methods: Qualitative and
Quantitative Approaches (6 the Ed.) Boston: Pearson Education.
https://www.merriam-webster.com/dictionary/sampling
https://en.wikipedia.org/wiki/Sample_(statistics)
https://zhanglab.ccmb.med.umich.edu/literature/Hastings1970.pdf
https://onlinecourses.science.psu.edu/stat100/node/18
https://www.socialresearchmethods.net/kb/sampprob.php
http://www.investopedia.com/terms/s/sample.asp
http://stattrek.com/statistics/dictionary.aspx?definition=Stratified%20
sampling
https://greengarageblog.org/7-pros-and-cons-of-convenience-
sampling
https://explorable.com/non-probability-sampling
http://www.betterevaluation.org/en/evaluation-options/sequential
https://greengarageblog.org/7-pros-and-cons-of-convenience-
sampling
http://methods.sagepub.com/reference/encyclopedia-of-survey-
research-methods/n4.xml
https://en.wikipedia.org/wiki/Theoretical_sampling
https://explorable.com/sequential-sampling