Non-probability sampling is a form of sampling technique in which the subset of population is not achieved by chance or probability but by some logic. Here, we should know that there is difference between sampling and sample; sampling refers to the selection of subset from population whereas, sample is the subset of population which represents the whole population. There are two types of sampling which are: Probability Sampling and Non-Probability sampling. In non-probability sampling technique, each respondent of the population does not have equal chance or probability to be taken into sample, even some members would have zero probability of being taken into the sample. Selection criteria of the sample members are totally biased and it depends upon the judgments and convenience of the interviewers. This sampling method is also divided into different types which are described below:
In this method, researchers decide the sampling respondents on the basis of convenience. For example, a researcher wants to select 10 students from a class for interview while using convenience sampling. Now selecting 10 students from the first row would be convenience sampling because researcher is selecting respondents on the basis of convenience. In most circumstances following may be true:
Majority of respondents do not respond to fill the questionnaire.
Sample units may be distributed sparsely.
Few respondents do not cooperate in filing the questionnaires.
Few interviewers may not be serious in selecting the sampling units as per assumed sampling plan.
Probability sampling technique has more accuracy in the terms of confidence level but it has some limitations to be fully executed. So, convenience sampling is used to avoid such limitations. Researchers select the sample on the basis of time and cost limitations such as: departmental stores, super stores, restaurants, telephone directory, newspaper subscribers list etc.
This is also one of the non probability sampling technique in which population is classified in to various groups on the basis of different criteria such as: age (young age, middle age, old age). In the quota sampling technique, proportion of sampling units is selected from each criterion, which makes the population. Fore example, if a researcher wants to select a sample of 50 employees from an organization while using the quota sampling, he/she can do so by assigning quotas to various departments such as: 10 employees from marketing, 10 from finance, 10 from supply chain, 10 from production, 10 from human resource. In this sampling method, researcher can also adopt any other non probability sampling technique such as: convenience or judgment sampling method, for selecting the required respondents.
In the judgment sampling, sample respondents would be taken by the opinion of the researcher or with the advice of some expert. At the first stage experts guide the researcher in selecting sample members then the researcher on the basis of past experience applies intuitive judgment to select sample. This sample method leads to the personal biasness. If this method is applied sincerely then better results would be achieved. Selective sampling members of the sample are taken from the population which avoids the inclusion of other units, for this reason this technique is also known as purposive sampling. Judgment sampling method is used for sampling the population in exceptional events, where the units of population are not equally qualified for sampling frame and also for the units of sample.
Snowball sampling is a non probability sampling method in which few initial respondents are selected randomly and further members are taken through referral process. In snowball sampling, one sample is selected from other sample through references. Therefore it is also known as reference sampling. In such cases sample size will improve. This is a convenient and cost effective sampling technique and can be applied in situations where the development of sampling frame is time consuming and difficult.