Prof.

Ajay Kumar
Professor & Head
Department of Sociology
A. N. College, Patna

What is Sampling? What are it‟s types?

For studying a social problem it is difficult to study the whole universe of
the problem under study. It is because it is both costly, time consuming and
complex as well as financially not viable. It is therefore convenient to pick
up a sample out of the universe proposed to be covered by the study. But
sampling needs much care. It should be representative and picked up in a
manner that it represents the universe as a whole.

When a social researcher undertakes a research work, he is to decide

basically two important things namely:-

1. What will be the scope of the study? and

2. What will be it‟s population or universe?

He may decide to occur the whole population concerned with the subject, if
he has time, energy, resources and capacity. If that is done, that is known
as census method of study. On the other hand he can pick up a small unit
out of the whole study. Such a unit is expected to be representative of the
whole population. Then it is called Sampling method.

P. V. Young in her book „Scientific Social Survey and Research‟ has

written, “A statistical sample is miniature picture or cross section of the
entire group or aggregate from which the sample is taken. The entire
group from which a sample is chosen is known as „the population‟,
„universe‟ or „supply‟.” According to her :-

1. A sample is a smaller part of the universe.

2. The selection of this smaller part is conducted through a definite
process, that is called sampling.
3. Sampling is a representative or miniature picture of the universe.

1
 4. Facts collected through sampling method is almost the same for the
universe.
According to F. N. Kerlinger, “Sampling is taking any portion of a
population or universe, as representative of that population or universe”.
This definition reveals that the sample is a small representative part of a
universe. Mildred Parton in his book, „Fairchild Dictionary of Sociology‟
has written, “Sampling method is the process or method of drawing a
definite number of individuals, cases or observations from a particular
universe, selecting part of a total group for investigation”. According to
him sampling method is used by a social scientist for investigation of a
group by drawing definite number of units.

From the above mentioned definitions it is obvious that sampling method

for the study of a social problem is really good. It helps in saving time,
money and energy. The sample should be picked up in such a manner that
it represents the entire population to be studied. Sampling technique is
adopted with certain basic assumptions and basis.

The main types of sampling are the following:-

1. Random Sampling: Random sampling is the basic probability

sampling design. A random sample is selected by a process that gives
to each element in the population an equal chance of being included
in the sample. It also makes the selection of every possible
combination of cases in the desired size, equally likely. For example,
that one wants a simple random sample of 100 students from 2000
students of a college. Firstly, write the name of each student on cards
of equal shape and size. Put the cards in a bucket and mix them
thoroughly and let an illiterate person to pick one. Each of the card
has now the same chance of being selected.

Following are the few methods of Random sampling:-

a) Lottery method – This is the simplest way of making the

selection. The number of items in a data are written on cards
of equal dimensions and they are thrown into a bucket. Now a
casual observer selects the number of items required in the
sample. Here, there is no probability in favour of any item.

2
 b) Card method – In this method pieces of wood, tin or
cardboard of equal dimension, with number 0, 1 or 2 printed
on them, are used. Now a list of items of the universe prepared
are divided in 0, 1 and 2 categories. The pieces are rotated in
the drum and then requisite numbers are drawn by an
impartial person. Now, if the pieces drawn are 20 from list 0,
10 from list 1 and 15 from list 1 and 15 from list 2.

c) Regular Interval Method – In this procedure units are broken

up in Numerical, Alphabetical or Geographied sequence. Now,
we may decide to choose every fifth name.
d) Tippet‟s Numbers – On the basis of population statistics,
Tippet has constructed a random list of four digits each of
10,400 institutions. These numbers are result of combining
41,600 population statistics reports. Upon the basis of
Trippet‟s numbers it is very easy to select samples.

e) Grid Method – This method of random sampling is used when

we select a particular region from a large universe. For this
method we use the map of that area and a transparent glass.
We mark the desired number of spot on the transparent glass.
Then we put the glass on the map and select the marked area.

2. Stratified Sampling: In stratified sampling the population of the

universe is first divided into different strata. Such strata may be
based on sex, caste and education. Thereafter required number of
sample are selected from each strata through Random method.
Stratified sampling has two types:-

a) Proportionate – In proportionate sampling the cases drawn

from each strata in the same proportion as select the same
proportion.

b) Disproportionate – In disproportionate sampling the cases are

drawn from the various strata on a disproportionate way. This
type of sample is economical.

3
3. Purposive Sampling; The basic assumption behind purposive
sampling is that with the exercise of good judgement and appropriate
strategy one can handpick the cases to be included in the sample and
thus develop samples that are satisfactory in relation to one‟s
research needs. A common strategy of purposive sampling is to pick
cases that are judged to be typical of the population in which one is
interested.

4. Quota Sampling: In Quota sampling the entire data is split up into

as many blocks as there are investigators and each investigator is
asked to select certain items from his block and study. Its basic object
is the selection of a sample that is a replica of the „population‟ in
respect of which are wished to generalize. Quota sampling by and
large affords the insurance that diverse elements in the population
are included in the sample and that these elements are taken account
of in proportions in which they obtain in the population.
5. Multi Phase Sampling: In this Type of sampling items are selected
in different stages at random. The basic information recorded from
the whole sample makes it possible to compare certain characteristics
of the sub sample with that of the whole sample. One additional merit
is that it also facilitates stratification of the subsample since the
information collection from the first phase sample can sometimes be
gathered before the sub-sampling process takes place.

6. Extensive Sampling: This method is virtually same as census

except that irrelevant items are left out.

7. Convenience Sampling: This is hit or miss procedure of study.

The investigator selects certain items from the domain as per his
convenience. No planned effort is made to collect information. But as
is characterized by hit or miss method, at times clear hit are scored.

8. Systematic Sampling: This type of sampling is a variation of

simple random sampling. It requires that the populations, i.e., a list
of its member be ordered in such a way that each element of the
population can be uniquely identified by its order.

4
9. Cluster Sampling: In cluster sampling the sampler first samples
from the population certain large grouping, i.e., clusters. The
sampling of clusters from the population is done by simple or
stratified random sampling methods. From these selected clusters,
the constituent elements are sampled by random procedures.

10. Accidental Sampling: In accidental sampling, the researcher

simply reaches out and picks up the cases that fall to hand continuing
the process till such times as the sample reaches a designated size.

The selection of a sampling procedure from the above mentioned

technique depends upon the nature, scope, number of units, etc., in a
domain. However, mostly random and stratified sampling techniques
are used, rest are the rarely preferred.