Types of Sampling


The definition of sample given here is a very simplified, easily understandable yet standard definition.

Sample is referred to a small number of people, selected from the larger number of people which more or less resembles the characteristics of the larger population from which the sample is selected.

The technique of selection of the smaller number of people from the larger population is known as sampling.

Sample Size

Selecting a sample size is guided by the following:

  • unit of analysis.
  • size of universe.

Universe here refers to the larger population from where the sample is selected.

Here, two things to keep in mind:

  • more the unit of analysis, bigger the sample.
  • smaller the unit of analysis, smaller the sample.

Types of Sampling

There are broadly two types/techniques/methods of sampling:

  • Probability Sampling.
  • Non Probability Sampling.

Probability Sampling ensures that the chances of being selected as the sample of each and every unit in the universe is more or less equal.

All units will not be selected as the sample but they would have an equal chance of being selected.

Probability Sampling is possible only if listing of elements of the universe is available.

The Types of Probability Sampling are:

Simple Random Sampling: It ensures selection of required number of sample units from the universe randomly without bias, preference or judgement.

Disadvantage: if not selected properly, 1. chances of unrepresentative is high; 2. changes of missing some groups, communities etc.

Systematic Random Sampling: Samples are selected at regular intervals. The first element of the sample is selected randomly, and subsequent samples are selected at regular intervals.

This sampling method ensures better spread of the sample across regions, different categories, communities etc.

This sampling technique is good for small universe, and representative is normally high.

Disadvantage: 1. Prerequisite of listing of all the elements in the universe. No sample can be drawn without listing being available; 2. Not feasible for very large universe.

Multi Stage Systematic Random Sampling: This method is a refined version of systematic random sampling method.

In this method, the sample is selected using different stages/steps.

This sampling technique is applicable in geographically and numerically large universe.

Stratified Random Sampling: In this sampling technique, members of the population/universe are first divided into smaller groups called strata. Then, random samples are selected from each stratum.

This sampling method captures the key population characteristics in the sample.

Disadvantage: it is ineffective in the study of cross-section of people, locality etc.

In Non Probability Sampling, the chances of being selected in the sample of each and every unit of the universe are unequal.

Sometimes, this method systematically excludes one or more sections, groups, locations etc from being selected as the sample.

The types of Non Probability Sampling are:

Convenience Sampling Method: Sample for the survey is selected as per the convenience of both the interviewer/researcher as well as the respondents.

This sampling technique is very easy and less time consuming.

Disadvantage: The chances of the sample being unrepresentative are very high.

Snowball Sampling Method: The entire sample is not selected at one go. Subsequent samples are selected based on the reference of the previously selected sample.

The sample keeps adding, making it bigger and bigger.

This sampling method is useful for research for which the universe is small and there is hardly any listing available for universe.

Disadvantage: There is always an uncertainty regarding the final number of samples.

However, when the researcher does not have any idea of the universe or in cases of undesirable issues such as issues related to drug addicts, habitual drinkers etc., it is helpful. The question of representativeness does not arise here.

Cluster Sampling or Quota Sampling: The sample is drawn first by fixing quota for different sections which the researcher aims to study, then the sample is drawn randomly or purposively.

This sampling technique is useful if the sample to be studied is small.

Also read survey research and its role in Political Science research.

Survey Research and Its Role in Political Science Research


Importance of Survey as a Research Tool

  • It helps us with evidence, which is very important in any research.
  • We may be aware of some facts, but there is a  need to verify those facts. The findings of the survey may confirm or refute it.

However, survey research is not ideal for controversial or very sensitive issues.

Besides, if the questionnaire is not properly designed, reliability is less.

How Can We Judge Reliability of Survey Findings?

We can judge the reliability of survey findings on the following basis:

  • Representativeness- proportion;
  • Questionnaire design- neutral, not bias;
  • Data collection method- neutrality.

Size of sample, its selection instruments, field work and data collection- these things also determine survey reliability.

Role of Surveys in Political Science Research

Accurate pictures of the political world are not easy to come by. Describing political activities of millions of citizens, thousands of groups and hundreds of institutions are never easy.

To comprehend these phenomena, political scientists need observational tools as powerful as those in the natural sciences. Scientific surveys are one of these tools.

The following are the points in support of surveys as a tool in Political Science research:

  • widely in use since 1940s;
  • surveys are powerful data collectors and accurate magnifiers of information;
  • suitable for almost any topic, wide range of applicability;
  • highly reliable, representative, unbiased;
  • survey methodology is an extraordinary powerful approach to studying the social and political world;
  • capacity for confirming theories about politics; and
  • policy relevance.

Also read types of sampling