CHAPTER FOUR

SAMPLING
Sampling refers to drawing a sample (subset) from a population (the fullest). The theory
of sampling has been developed recently but this is not new. In our every day life we
have been using sampling theory. In all cases we believe that the sample gives a correct
idea about the population.
In research it also refers to choosing a smaller, more tangible number of people to take
part in the research. The usual goal of sampling is to produce a representative sample (i.e.
a sample that is similar to the population on all characteristics, except that it includes
fewer people because it is a sample rather than a complete population).
The foremost purpose of sampling is to gather maximum information about the
population under consideration at minimum cost, time and human power. This is best
achieved when the sample contains all the properties of the population.
Terminology Used in Sampling
Here are some important terms used in sampling:
A sample: A set of element taken from a population (i.e. a finite subset of individuals
defined in a population.)
Sampling unit: The constituents of a population, which are individual’s to be sampled
from the population and can not be further subdivided for the purpose of
sampling at a time. Ex: to know the average income per family the head of
the family is the sampling unit. To know the average yield of wheat each
farmer owner’s yield of wheat is the sampling unit.
Population: the full set of elements or people you are sampling. Technically known as
Universe. The universe can be finite or infinite; in earlier case the number of
items is certain.
Sampling Design: A definite plan for obtaining a sample from the sampling frame. It
refers to technique / Procedure adopted by the researcher.
Census: A measurement of all the units in the population.
Statistics Vs Parameter
Statistics: a numerical characteristics of as ample (i.e. is a number that result from
measuring all the units in the sample.)

1
Parameter: a numerical characteristic of a population (i.e., is a number that result from
measuring all the units in the claimed population)
 Statistics derived from samples are used to estimate population
Sampling error Vs. Non- sampling error
Sampling error: Although the sample is the part of the population it cannot be expected
generally to supply full information about the population. So there may be in most cases
difference between statistics and parameters. The discrepancy between a parameter and
its estimate (statistics) due to sampling process is known as sampling error (i.e., the
difference between the value of a sample statistics (such as the sample mean) and the true
value of the population parameter (such as the population mean).
Non- sampling error: In all research /survey some errors may occur during collection of
actual information. These errors are called non- sampling error.
Reasons for selecting sample: Why sample?
Sampling is inevitable in the following situations:
 Complete enumeration are practically impossible when the population is infinite,
 When the results are required in short time - it would be too time consuming to
study the whole units in the population.
 When the area of survey is wide - it would be too expensive and time consuming
as will to study the whole units in the population
 When resource (Money, time and trained persons) are limited.
 When the items or units are destroyed under investigation for example a
company to demonstrate that their car can survive certain crash tests. Obviously,
the company cannot expected to crash every car, to see if it survives, the company
crash only the sample of cars.
Principles of sampling
Samples have to provide good estimate. The following principle tells us that the sample
methods provide such good estimate.
1. Principle of statistical regularity : A moderately large number of units chosen at
random form large groups are almost sure on the average to possess the
characteristics of the large group.

2
 2. Principle of Inertia of large number : Other things being equal, as the sample size
increase the result tend to be more accurate and reliable.
3. Principle of validity: This states that the sampling methods provide valid estimate
a bout the population units (parameters).
4. Principle of Optimization: The principle takes in to account the desirability of
obtaining a sampling design which gives optimum result. This minimizes the risk/
loss of the sampling design.
TYPES OF SAMPLE DESIGN
There are different types of sample design based on two factors viz. the representation
basis and the Element Selection Technique. On the representation basis the sample may
be probability sampling or it may be non- probability sampling. Probability sampling is
based on the concept of random selection. Whereas non-probability is non-probability is
non-rand sampling.
On the element selection basis the sample may be restricted or unrestricted. When each
sample element is drawn individually from the population at large, then the sample so
drawn is known as unrestricted sample, where as all other forms of sampling are covered
under the term restricted sampling. However, Sample design is basically of two types:
probability and non- probability sampling
Basics sample Design
Element selection Representation Basis
Technique Probability Non-probability
Random Sampling Haphazard/
Unrestricted (simple ) convenience
sampling
Restrict Cluster sampling Purposive sampling
Systematic sampling (Quota, judgment sampling)
Stratified sampling

Probability Sampling
A probability sampling is on where the selection of the units from the population is made
according to known probability. The sample is based on probability theory. Every unit of

3
the population of interest must have a know non-zero chance of being selected in to the
sample.
- Assigns equal probability/ chance to each units of the population (Every element
has equal (non zero) chance of being selected.
- Error of estimation/significance of results obtained can be measured.
- Best technique for representative sample (i.e., produce representative sample).
- Ensure the law of statistical regularity (i.e., on an average, sample chosen will
have the same composition and structure as the universe/ population.)
- Even each combination (possible) of sample will have equal probability of being
picked.
- All choices are independent of one another.
Types of probability sampling
(1) Simple Random sampling: A simple random sample from finite population is a
sample selected such that each possible sample combination has equal probability
of being chosen. It is also called unrestricted random sampling. In this type of
probability sampling each unit in the population is identified, and each unit has an
equal chance of being in the sample. The selection of each unit is independent of
the selection of every other unit. Selection of one unit doesn’t affect the chance of
any other unit
Methods of selection of simple random sampling
(a) Lottery method: This is the most simple and popular method. In this method all the
items of the population are assigned a number on a separate slip of paper of same
size, shape and color. They are folded and mixed in a container. The required
number of slip is selected at random for the desire sample size. If the
universe/population is infinite this method is inapplicable.
(b) Table of Random Numbers: As the lottery method cannot be used, when the
population is infinite, the alternative method is that of using the table of random
numbers. A random number table is so constructed that all digits 0 to 9 appear
independent of each other with equal frequency. If we have to select sample from
population of size N= 100, then the numbers from 001 to 100.

4
 Procedures: Units of the population from which a sample is required are assigned
with equal number of digits. When the size of the population is less than thousand,
three digit numbers 000, 001, 002, --- 999 are assigned. We may start at any place
and may go on in any direction such as a column wise or row-wise in a random
number table. But consecutive numbers are to be used. If any random number is
greater than the population size N the N can be subtracted from the random number
drawn. This can be repeatedly until the number is less than N or equal to N.
Example: In an area there are 500 families. Using the following extract from a table
of random number select a simple of 15 families to find out the standard of living of
the families in that area.

4652 3819 8431 2150 2352 2472 0043 3488

9031 7617 1220 4129 7148 1943 4890 1749
2030 2327 7353 6007 9410 9179 2722 8445
0641 1489 0328 0385 8488 0422 7209 4950

In the above random number table, we can start from any row/column and read three
digit numbers continuously row-wise or column-wise.
Now we start from the third row, the numbers are:
203 023 277 353 600 794 109
179 272 284 450 641 148 908
280.
Since some numbers are greater than 500 we subtract 500 from those numbers and we
rewrite the selected numbers as follow.
203 023 277 353 100 294 109
179 272 284 450 641 141 408
280.
(2) Systematic Random sampling: each unit in the population is identified and
each unit has an equal chance of being in the sample. It is also called Quasi-
random sampling. Selection procedures: systematic sampling involves three
steps:

5
 1st: Determine the sampling interval, which is symbolized by “K” (i.e., it is the
population size divided by the desired sample size).
2nd: Randomly select a number between 1 and k and include that person/unit in
your sample.
3rd: Include all kth elements in your sample. For example, if K is10 and your
random selected number between 1 and 10, for instance 5, then you will select
persons 5,15,25----etc. When you get to the end of your sampling frame you will
have all the people to be included in your sample.
(3) Stratified Random sampling:
All of the methods of sampling the procedure commonly used in surveys are
stratified sampling. This technique is mainly used to reduce the population
Heterogeneity and to increase the efficiency of estimate.

Stratification means division in to groups. In this method the population is divided

in to a number of sub-groups/strata. The strata should be so formed that each
stratum is homogeneous as far as possible. Then from each stratum a simple
random sample may be selected and these are combined together to form the
required sample from the population.

There are two types of stratified sampling. They are Proportional and Non-
proportional. In the proportional sampling equal and proportional representation is
given to subgroups or strata. If the number of items is large the sample will have a
higher size and vice versa. The population size denoted by N and the sample size is
denoted by ‘n’ the sample size is allocated to each stratum. That is given by n/N
=C. so in this method each stratum is represented according to its size.

In Non-proportional sample, equal representation is given to all the sub-strata

regardless of their existence in the population.

6
 Example: A sample of 50 students is to be drawn from a population consisting of
500 students belonging to two institutions A and B. The number of student in the
institution A is 200 and the institution B is 300. How will you draw the sample
using proportional allocation? There are two strata in this case with sizes N 1=200
and N2 =300 and the total population N= N1 + N2 = 500.

The sample size is 50.

If n1 and n2 are the sample size:
n1= n x N1 = 50 x 200 =20
N 500
n2 = n x N2 = 50 x 300 = 30
N 500
The sample size is 20 from A and 30 from B. then the unit from each institution is
to be selected by simple random sampling.

(4) Cluster Sampling: cluster sampling views the units in the population as not
only being members of the total population but as members also of naturally
occurring in a cluster within the population. Cluster sampling is used in a large
geographic sample where No list is available of all units in the population but
the population boundaries can be well-defined.

Example: To obtain information about the drug habit of all high school students in a
state:
You could obtain the list of all school districts in the state and select a
set of school districts.

Then with in each selected school district list all the high schools and
select a SRS of classes.

Within each selected high school list all the high school of classes

Then use the high school students in that class as your sample.

7
 Cluster sampling must use a random sampling method at each stage. Generally it
divides area of interest in to a number of smaller non-overlapping areas /cluster.
Non- Probability/ sampling
It is the one where discretion is used to select representative units from the
population (or) to infer that a sample is representative of the population. This
method is also called Non-Random/judgment / purposive sampling. It mainly used
for opinion survey.
Focus on volunteers, easily available units, or those just happen to be present when
the researcher is done. It is useful for quick and cheap studies, case studies, for
qualitative research, for pilot studies, and for developing hypotheses for further
research.
There are FOUR main types of Non-probability methods/procedures:
(1) Convenience sampling: also called accidental /Haphazard /man in the street
sampling. The researcher selects units that are convenient, close at hand, and
easy to reach.
(2) Purposive sampling: The researchers select the units with some purpose in
mind. The researcher specifies the characteristics of the population of interest
and then locates individuals who match those characteristics. Ex: Experts on
coops, students who lives in dorm on campus.
(3) Quota sampling: The researcher constructs quotas for different types of units. It
also often involves setting quotas and then use convenience sampling to obtain
those Quotas.
(4) Snowball sampling: Each research participant is asked to identify other potential
research participants who have some certain characteristics. You start with one
or few participants, ask them for more, find those, ask them for some and
continue until you have a sufficient sample size. It usually used where no
sampling frame exist.