Survey Sampling Methods

Print Date: 30.04.24

Population: An aggregate of all individuals or items (actual or possible)

of interest in any particular study defined on some common
characteristics is called a population.
In statistical enquiry we usually need information about a particular
group of individuals or items. This group is known as a population.

Sample: A representative part of the population is called a sample. The

number of individuals in the sample is called sample size.

Survey: Survey is an important means of collecting information. It is of

two types:
(a) Census: If every member of the population is surveyed, this is called
a census. When the population is small, the census could be
a straight forward exercise. However, when the population is large,
conducting a census can be very time-consuming and expensive
exercise.
(b) Sample survey: Sample survey is a statistical investigation of a
population based on a sample of predetermined size.

If the sample is representative of the population, a sample survey can

give an accurate indication of the population characteristic being
studied.

Advantage of sample survey over census

Following are the points in favor of sample survey over census:

1. Less cost: Since sample survey is based on a part of the population, it
is very much less expensive in terms of money and man-hour than a
complete count.
2. Less time-consuming: Being small in scale, sample survey takes less
time to complete.

1
3. Greater scope: A sample survey is likely to permit the collection of
data in a wide range of phenomenon. A more thorough investigation
is possible in a sample survey rather than a census.
4. Greater accuracy: The results obtained in a sample survey are more
reliable and accurate than that of a complete count because:
• highly trained and efficient personnel are used
• more careful supervision is made in field work
• quality equipments are used
5. Flexibility: If the investigation requires highly trained personnel or
even specialized equipments, census is not practicable and hence
sample survey is more flexible and appropriate.
6. Feasibility: Quite often sample surveys are more feasible than census.

Study population refers to the group of subjects for which we have

observations, while target population refers to the larger group to which
we would like to make inferences.

Element: An element is an object or individual on which the

measurement is taken.

Sampling Unit: A sampling unit or simply unit may be taken as a well

defined, distinct and identifiable element or group of elements on which
observation can be made.

Sampling frame: Sampling frame covers all elements of the target

population. The list or the map or other acceptable material which serves
as a guide to the population to be covered is known as sampling frame.

A sampling frame is organized and arranged in such a way that every

unit occurs once in the list and no unit is excluded from the list.

Sampling: Sampling is a process or technique of taking sample from the

given population.

2
Types of Sampling: Sampling may be classified under the following
three heads:
(i) Subjective or purposive or judgment sampling
(ii) Probability sampling
(iii) Mixed sampling

(i) Subjective or purposive or judgment sampling: In this sampling

• the sample is selected with some definite purpose,
• the choice of the sampling units depends entirely on the discretion
and judgment of the investigator.
This sampling is of little use and can not be recommended for general
use.

(ii) Probability Sampling: Probability sampling is the scientific method

of selecting samples according to some law of chance in which each unit
in the population has some definite pre-assigned probability of being
selected in the sample. This is the most widely used sampling technique.

Mixed Sampling: If the samples are selected partly according to some

laws of chance and partly according to a fixed sampling rule (no
assignment of probabilities) they are termed as mixed samples and the
technique of selecting such samples is known as mixed sampling.

Major Steps in a Sample survey

Sample survey is a very efficient technique to prove relevant information

for drawing inference about a large population. Principal steps involved
in a sample survey are discussed below:

1. Objective of the survey: Objective of the survey should be in clear and

concrete term. What it wants and how it is going to use the results,
should be clearly defined in objective

3
2. Defining the population to be sampled: The target population from
which the sample will be chosen, should be defined and identified in
clear and unambiguous terms.

3. The sampling frame and the sampling unit: The sampling frame
should be well defined and the sampling units must cover the entire
population and they must be distinct, unambiguous and non-
overlapping.

4. Data to be collected: The data must be relevant and pertinent to the

purpose of the survey.

5. The questionnaire or schedule: A set of questions relevant to the

objectives of the study is called questionnaire or schedule.

6. Methods of collecting data: At this stage we are to choose what

method(s) will be used to collect information e.g., interview method,
mail questionnaire method etc.

7. Selection of proper sampling design: The next step is to select a

sampling technique from a number of sampling designs available.
Appropriate sampling design leads to obtain reliable estimates of the
parameters.

8. Sample size: Determination of sample size is a very important but

difficult task for any sample survey.
The major determinants for the sample size are:
• the size of the sub population for which relatively precious
estimates are required
• human and financial resources
• time available
• the required level of reliability

9. Pre-testing of survey materials: Pre-test is a trial which allows to test

4
 the questionnaire in the field and to check on the management of field
operations. The results of the pretest usually suggest whether any
modification is needed before a full-scale data collection is
undertaken.

10. Organizing the field work: Efficient organization of field work is a

pre-requisite for the success of any survey.
• Field workers must be given proper training
• Appropriate measures must be taken so that the field workers are
regularly supervised and their works be monitored
• Quality work should be ensured

11. Data management: A well prepared data management plan is of

prime importance.

12. Data editing and checking: Once the data is at hand, it should be
edited and checked for any erroneous and inconsistent entries. Error
may occur by the enumerator or due to misreporting or ignorance of
respondent or at the time of editing which may leads to inconsistent
results.

13. Data Analysis:, Edited and corrected data are then tabulated and
statistical analysis is made to obtain the desire estimates of the
parameters.

14. Report writing: Report writing includes:

• to explain the findings of the study
• lessons learned
• to highlight the policy implications and
• to suggest necessary actions.
• to make guideline for all future surveys.

5
 Errors in Sample Survey

The errors involved in the collection, processing and analysis of the data
may broadly be classified as:
(a) Sampling error
(b) Non Sampling error

(a) Sampling error: The error which arises entirely due to sampling is
called sampling error. Sampling errors occurs due to
• faulty selection of sample
• faulty demarcation of sampling units
• Improper choice of sampling techniques

Note: Increase of sample size usually results in the decrease in

sampling error.

(b) Non sampling Error: The error which occurs due to other than
sampling is called non sampling error.
Errors in implementation of the survey and errors of measurement
are collectively called non sampling errors. Non sampling errors
arise from the facts:
• Faulty planning, definition, or inappropriate objection of the
survey
• Data specification being inadequate and inconsistent
• Improper editing, coding or recording information
• Response and non-response bias
• Failure in reporting or recording information
• Improper execution of the survey design

6
Sampling with replacement and without replacement
A sample may be drawn with replacement (SWR) or without
replacement (SWOR). If one unit is drawn and returned it to the
population before the second unit is drawn is known as Sampling With
Replacement (SWR) and if the 1st unit not returned to the population
before the 2nd one is drawn the sampling is called Sampling Without
Replacement (SWOR)
• For SWOR, the number of units remaining after each draw will be
reduced by one and the probability of drawing any remaining unit
will be increased
• For SWR, the total number of units for each draw remains
unchanged and the probabilities remain unaltered.

Different types of Random Sampling:

1. Simple Random Sampling (SRS)
2. Stratified Random Sampling
3. Cluster Sampling
4. Systematic sampling
5. Multistage sampling
6. Quota sampling

Simple Random Sampling (SRS): In this sampling each unit of the

population has an equal and independent chance of being selected in the
sample.
Thus from a population of N units, the probability of drawing any unit at
the first draw is 1 , the probability of drawing any unit in the second
N
1
draw from the remaining (N-1) units is and so on.
N −1

Method of selection of a simple random sample

1. Lottery method
2. Mechanical Randomization or Random Numbers method

7
Advantages of SRS
• every unit has an equal chance of being selected,
• Personal bias is completely eliminated,
• Precise estimates are obtained.

Disadvantages of SRS
• It requires a very up-to-date frame
• Administrative in-convenience (sampling units may be widely
spread in different areas), very costly to collect data etc
• For a given precision, SRS requires a large sample size.

Stratified Random Sampling: This sampling is used when the

population is split into distinguishable layers or strata which are quite
different from each other such as Age groups, occupational groups,
income groups etc.
Let a population of N units is divided into subpopulations of
N1, N 2 , . . . , N L units respectively. These subpopulations are non-
overlapping and together they comprise the whole of the population i.e.
N1 + N 2 + . . . + N L = N

Systematic Sampling:
Systematic sampling is a commonly employed technique if the complete
and up-to-date list of the sampling units is available.
This consists in selecting only the first unit at random, the rest being
automatically selected according to some predetermined pattern
involving regular spacing of units.
Let us suppose that N sampling units are serially numbered from 1 to N
in some order and a sample of size n is to be drawn such that

N
N = nk  k =
n
where, k usually called the sampling interval which is an integer.

8
Systematic sampling consists in drawing a random number, say, i  k and
selecting the unit corresponding to this number and every kth unit
subsequently.
Thus the systematic sample of size n will consist of the units i, i+k, i+2k,
.... , i+(n-1)k

The random number `i' is called the random start.

Cluster Sampling
In this sampling the total population is divided, depending on problem
under study, into some recognizable sub-divisions which are termed as
clusters and a simple random sample of these clusters is drawn. We then
investigate every unit in the selected clusters.

For example: If we are interested in obtaining the income or opinion

data in a city, the whole city may be divided into N different blocks or
localities and a simple random sample of n blocks is drawn. The
individuals in the selected blocks determine the cluster sample.

Note: In cluster sampling elements are homogeneous between clusters

and heterogeneous within clusters but in stratified random sampling
elements are homogeneous within strata and heterogeneous between
strata.