Sampling Techniques

It would normally be impractical to study a whole

population, for example when doing a questionnaire survey.
Sampling is a method that allows researchers to infer information
about a population based on results from a subset of the
population, without having to investigate every individual.
Reducing the number of individuals in a study reduces the cost
and workload and may make it easier to obtain high quality
information, but this has to be balanced against having a large
enough sample size with enough power to detect a true
association.
Sampling

Measuring a small portion of something and then making

a general statement about the whole thing.

Process of selecting a number of units for a study in such a

way that the units represent the larger group from which
they are selected.
Why We Need Sampling?
Sampling makes possible the study of a large, (different
characteristics) population.

Sampling is for economy

Sampling is for speed.

Sampling is for accuracy.

Sampling saves the sources of data from being all consumed.

General Types of Sampling

1. Probability sampling

2. Non-probability sampling
Probability Sampling Techniques
The purpose of probability sampling is to get a sample that represents the
population from which it was drawn.

In this sampling technique, everyone has a chance of being selected or

sampled before the sampling begins.

The five most common probability sampling techniques are:

❑ simple random sampling

❑ stratified random sampling
❑ cluster sampling
❑ systematic sampling
❑ multi-stage sampling
Simple Random Sampling

A sampling technique that involves drawing a

sample from a population such that everyone in the
population has an equal chance of being selected.

a. Lottery Method
For example, placing the names of all your
students in a hat and blindly drawing ten names from
the hat.
b. Table of Random Numbers
A random number table is a series of digits
(0 to 9) arranged randomly in rows and columns, as
demonstrated in the small sample shown below. The
table usually contains 5-digit numbers, arranged in rows
and columns, for ease of reading. Typically, a full table
may extend over as many as four or more pages. You will
find random number tables in most statistical textbooks.
Random number tables have been in existence since
1927 and are generated by a variety of methods.
Stratified random sampling
A sampling technique that involves first dividing
the population into subgroups based on known
demographics and then taking a simple random sample
from each subgroup.
For example, if you know that JILCF-Math has a
population of 60% women and 40% men (a known
demographic) and you wanted a sample of 100
students; you could randomly sample 60 women and 40
men. Thus, your sample has the same stratification as
the population, something that is not guaranteed by a
simple random sample.
Cluster sampling
A sampling technique that involves dividing the population into
groups called clusters, randomly selecting a certain number of clusters,
and then sampling every person (or thing) within each cluster.
This technique is useful when the population of interest is spread out
over a large area or gaining access to individuals proves too difficult.
For example, suppose you wanted a sample of 1st graders
from around the Philippines. How could you get a list who these 1st
graders are? With thousands of schools located across the country,
you probably could not.
However, it would be easy to obtain a list of all the schools.
Once you have this list of schools (clusters), you randomly sample the
desired number of schools before collecting data on all the first graders
in each of the chosen schools
Systematic sampling
A sampling technique that involves choosing
every nth person in a list or that you encounter
after randomly choosing a starting point within
certain parameters.
For example, suppose you wanted a sample
of 50 students arriving at the Cultural Center for
graduation.
If you know 500 students are expected to
attend, you would need to sample every tenth
person to arrive (500 divided by 50). However, to
make your sample random in nature, you would
have to choose a random starting point between 1
and 10
Multi-stage sampling
A sampling technique that involves combining two or more of the
above techniques in order to avoid the limitations associated with any
particular sampling technique.
For example, suppose you wanted a sample of 1st graders from
around the Philippines. How could you get a list who these 1st graders
are? With thousands of schools located across the country, you probably
could not.
However, it would be easy to obtain a list of all schools. Once you
have this list of schools (clusters), you randomly sample the desired
number of schools before collecting data on all the first graders in each of
the chosen schools. However, choosing every 1st grader from your
randomly chosen clusters might still prove to be overwhelming. If you
randomly chosen 25 schools, you would have to interview thousands of
first graders. To further limit the size of your sample, you could randomly
choose 10 students from each of your randomly chosen clusters (schools).
Thus, you would have combined cluster sampling with simple random
sampling.
 Non-probability samples have the distinguishing
characteristic that subjective judgments played a role in
the selection of the sample wherein, every participant has
an unknown chance of being selected.
This is unlike probability sampling which uses
randomization to ensure selection without subjectivity and
has the characteristic that every element in the population
has a known nonzero probability of being included in the
sample.
Unequal chance of being included in the sample (non-random)

Non random or non - probability sampling refers to the sampling process

in which, the samples are selected for a specific purpose with a pre-
determined basis of selection.

The sample is not a proportion of the population and there is no system in

selecting the sample. The selection depends upon the situation.

No assurance is given that each item has a chance of being included as a sample

There is an assumption that there is an even distribution of characteristics within

the population, believing that any sample would be representative.
Types of Non-Probability Sampling
Judgment or purposive or deliberate sampling

Convenience sampling

Quota sampling

Snow Ball Sampling

1. Judgment or purposive or deliberate sampling
In this method, the sample selection is purely based on the judgment of
the investigator or the researcher. This is because, the researcher may lack
information regarding the population from which he has to collect the
sample. Population characteristics or qualities may not be known, but
sample has to be selected.

In this method of sampling the choice of sample items depends primarily

on the judgment of the researcher. In other words, the researcher
determines and includes those items in the sample which he thinks are
most typical of the universe with regard to the characteristics of research
project.
For example, suppose 100 boys are to be selected from
a college with 1000 boys. If nothing is known about
the students in this college, then the investigator may
visit the college and choose the first 100 boys he meets.
Or he may select 100 boys all belonging to III Year, or
he might select 25 boys from BSBA course, 25 from
BS Psychology courses, 25 boys from Education courses
and 25 from Engineering courses. Hence, when
only the sample size is known, the investigator uses
his discretion and select the sample.
The use of judgment sampling is justified by following premises:

If there are a small number of sampling units is in the universe,

judgment sampling enables inclusion of important units.

Judgment stratification of population helps in obtaining a more

representative sample in case research study wants to look into
unknown traits of the population.

Judgment sampling is a practical method to arrive at some solution to

everyday business problems.

Limitations:

The judgment sampling involves the risk that the researcher may
establish conclusions by including those items in the sample which
conform to his preconceived ideas.

There is no objective way of evaluating the reliability of sample results.

2. Convenience sampling
Convenience sampling is commonly known as unsystematic, accidental
or opportunistic sampling. According to this procedure a sample is
selected according to the convenience of the investigator.

In this method of sampling the choice of sample items depends

primarily on the judgment of the researcher. In other words, the
researcher determines and includes those items in the sample which he
thinks are most typical of the universe with regard to the
characteristics of research project.

A type of non probability sampling which involves the sample being

drawn from that part of the population which is close to hand. That is,
readily available and convenient.

For example, suppose 100 car owners are to be selected. Then we may
collect from the RTO's office the list of car owners and then make a
selection of 100 from that to form the sample.
A convenience sampling may be used in the following
cases:

i) When universal set is not well defined,

ii) When sampling unit is not clear, and

iii) When complete list of the source is not available.

3. QUOTA SAMPLING
In this method, the sample size is determined first and then quota is
fixed for various categories of population, which is followed while
selecting the sample.

In this method the quota has to be determined in advance and

intimated to the investigator. The quota for each segment of the
population may be fixed at random or with a specific basis. Normally
such a sampling method does not ensure representativeness of the
population.

Example: - Suppose we want to select 100 students, then we might say

that the sample should be according to the quota given below : Boys
50%, Girls 50% Then among the boys, 20% college students, 40%
SHS students, 30% JHS students and 10% elementary school
students. A different or the same quota may be fixed for the girls.
MERITS OF QUOTA SAMPLING

Reduces cost of preparing sample and field work, since ultimate

units can be selected so that they are close together.

Introduces some stratification effect.

DEMERITS OF QUOTA SAMPLING

Introduces bias of investigator is not involved at any stage, the

errors of the method cannot be estimated by statistical
procedures.

Since random sampling is not involved at any stage, the errors of

the method cannot be estimated by statistical procedures. Quota
sampling is most commonly used in marketing survey and
election polls.
4. SNOWBALL SAMPLING
It refers to Identifying someone who meets the criteria
for inclusion in the study.

Selection of additional respondents is based on

referrals from the initial respondents.