SAMPLING TECHNIQUES:

The sampling process is defined as the selecting of sufficient numbers of suitable elements to carry out a research process.  The sampling process includes two main types of sampling, probability sampling and non-probability sampling.

PROBABILITY SAMPLING:

This type of sampling is applicable when elements in the population have a known chance of being chosen as subjects. Probability sampling can be restricted or unrestricted probability sampling.

  • Unrestricted or Simple Random Sampling:

In this sampling, every element in the population has a known and equal chance of being selected as subject. Consider a jar full of 100 marbles, out of which we need a sample of 10 marbles. If we were to draw out 1 marble from this population, we know that this marble will have 1/100 chance of being drawn, the second marble will have 1/99 chance of being drawn and so on.

  • Restricted Probability Sampling:

Unlike simple random sampling, this type of sampling technique offers a more efficient way to select a sample. Below are the types of restricted probability sampling:

  1. Systematic sampling:

This design involves drawing every nth element in the population beginning with a randomly chosen element between 1 and n. for example, if we want a sample of 40 houses from a total population of 300 houses in a particular city, then we could sample every fifth house starting from a random number from 1 to 5. Let us assume that the random number was 5, then houses labelled 5, 10, 15, and 20 would be chosen for the study until the 40 houses were selected.

  1. Stratified Random Sampling:

As the name suggests, stratified random sampling involves stratification or segregation of population, followed by random selection of subjects from each stratum. The population is firstly divided into mutually exclusive groups that are relevant to the study and then sample is drawn from it. Examples of this type are stratifying customers on the basis of income levels, age groups and residence in certain localities.

Cluster Sampling:

These are the samples gathered in chunks or groups (hence clusters) from a population. In this sampling, the target population is first divided into clusters and then a random sample of cluster is drawn from it. From this chosen cluster either all the elements are chosen or a sample is chosen from amongst it. A type of cluster sampling is area sampling in which clusters comprise of geographical areas such as countries, cities or city blocks. For example if you want to study the lifestyle trend of residents living in Minnesota, you would chose a cluster sample from Minnesotan city population and then choose a sample from this cluster for your study.

  1. Double Sampling:

This sampling is used when further information is required from the subset of a group from which some information has already been extracted. An example of this is questioning a group of customers about the choice of a particular cold drink they prefer when they have already responded to the question of whether they prefer cold drinks over water is double sampling.

NON-PROBABILITY SAMPLING:

This type of sampling requires no probabilities to be attached to samples being chosen for the study. Many statisticians use this technique as an inexpensive alternative to probability sampling. Below are the types of non-probability sampling techniques:

  1. Convenience Sampling:

As the name suggests, convenience sampling involves collection of information from members of the population who are conveniently available for observation. An example of this is choosing a group of random customers in a shopping mall to conduct a study of their views on taste of a noodles brand.

  1. Purposive Sampling:

In contrast to obtaining information from respondents who are readily available, sometimes respondents are chosen who can provide the needed information, either because they are the only ones who have that information or they fulfill some criteria set by researcher. This purposive sampling takes two forms, judgement sampling and quota sampling:

  • Judgement sampling:

This involves choosing the subjects who are in the best position to respond to researcher due to the information that they have. For example, if the researcher wants to find out what it takes for women managers to make it to the top of the hierarchy, he would approach those women who had been previously at secretarial positions but later progressed to top positions because they have gone through the experience and can respond effectively.

  • Quota sampling:

In this sampling, certain quota is assigned to the chosen group of subjects that may have been chosen randomly. For example to understand the work attitude of workers in an organization, a quota can be set. If there are 40% blue-collar workers and 60% white-collar workers and if a total of 20 workers are needed to respond to questions, a quota of 8 blue-collar and 12 white-collar workers can be set. After that the first conveniently available 8 blue-collar workers and 12 white-collar workers can be chosen for the study.

Research Week