-

In this method, the desired number of sample units is selected deliberately or purposely depending upon the object of the inquiry so that only the important items representing the true characteristics of the population are included in the sample.

Some of the main kinds of non-probability sampling are:

1. Judgment Sampling.
2. Quota Sampling.
3. Convenience Sampling.

We now list some of the advantages and disadvantages of non-probability sampling.

1. If the population size is very small then it is not very cost-effective to do probability sampling. In such cases, it is much easier to simply include sample units based on the discretion of the investigator.
2. In certain situations, it is important that certain units be included in the sample. The investigator can exercise his judgment in such cases and include these units as part of the sample.
3. It takes into account the knowledge, skills, and experience of the investigating researcher. That is not the case in probability sampling which is a completely random process.

1. A serious drawback of this sampling scheme is that it is highly subjective in nature since the selection of the sample depends entirely on the personal convenience, beliefs, biases, and prejudices of the investigator. For example, suppose the investigator wants to conduct a survey to determine the monthly incomes of people living in a particular city. If the investigator wants to prove that the standard of living in the city has gone down he may purposely choose to include only those people who live in poorer neighborhoods and exclude those people who live in richer neighborhoods.
2. This method cannot be used if the sample size is very large since the investigator cannot personally select a large number of units in a reasonable frame of time.
3. If the investigator is not experienced or knowledgeable enough about the subject of inquiry then he may make bad judgments resulting in an unrepresentative sample. This might lead to errors in the study and wrong conclusions.
4. Unlike probability sampling, it is not possible to estimate the standard error since no probabilities are involved in sample selection.