The sampling method in research needs to be understood by researchers because it is almost impossible for us to conduct research by involving everyone in the population. The sampling method was used to obtain a population representation.
In social research, the accuracy of the sampling process determines the degree of bias that can be generated. This article discusses how to perform a sampling technique. In applying sampling, researchers need to understand the types. This article describes the types or types along with examples.
Research Methods: Types and Examples
In general, there are two sampling methods or techniques in social research, namely probability sampling and non-probability sampling. Both have their respective types. The wrong sampling method can result in a high degree of bias. Of course it also creates validity problems.
Non-probability sampling
The main characteristic of this sampling technique is that the sample is taken without giving equal opportunity to individuals in the population to be selected as the sample. This type is very relevant to be used for early stage research. For example, to read the general trend of social phenomena before conducting a more in-depth and specific research. Several sampling techniques of this type are:
Based on subject availability
For example, research on mall visitors, for example. Researchers collect information from informants who are snacking. The research was carried out at that time. Since it was impossible to survey or interview all mall visitors, the researchers randomly selected informants who happened to be passing by and were willing to participate.
Purposive sampling or purposive sampling
This technique is very helpful if the researcher has sufficient knowledge about the characteristics of the population under study. For example, researchers want to know the first experiences of traveling abroad for women who do solo travel. Researchers joined a group of female travelers. So the sampling is done by only selecting those in the group.
Snowball sampling or snowball sampling
This technique is very relevant if the researcher does not know the informant network to be studied. However, the first informant who was contacted was an important actor in the issues discussed, so the researcher asked the next informant’s comments from the previous informant. For example, research on street beggars in urban areas. The first informant selected has an acquaintance who has the potential to become the next informant. From here, researchers can collect more and more data, such as a snowball rolling from the top of the mountain to the bottom, where it gets bigger and bigger.
Quota sampling
This technique is based on the proportional count of informants selected from the selected unit. For example, researchers conduct research in the context of a population in a country. From the state unit, researchers compare opinions in a gender perspective. Then the number of men and women taken as a sample is determined proportionally. Thus, the sample taken represents the proportion of gender in a country.
Research Design: Types and Examples
Probability sampling
This sampling type is taken by giving each individual in the population the same opportunity to be selected as a sample. It is widely believed that this type of sampling method is capable of significantly reducing bias. However, the degree of bias depends on the relevance of the sampling technique chosen in this type.
Simple random sampling
This technique is applied by identifying the population to be studied by giving a label. Usually the label is a number. For example, if the researcher wants to study a population of 2000. However, only 100 samples will be taken. Then the entire population is numbered from 1 to 2000, then a random number is drawn of 100. Those who are randomly selected become informants. This technique is very relevant to use for research in demographically homogeneous populations.
Cluster sampling or cluster sampling
This technique is done by creating a cluster to determine the sample. For example, a study of the views of interfaith community leaders on the meaning of tolerance in Yogyakarta. Researchers create clusters based on their religion. Next, the researchers listed the figures in each religion who had published popular books on religion. List of figures identified as selected informants.
Syatematic sampling or systematic sampling
This technique is similar to random sampling. The difference is in the informant selection process. If informants random sampling is determined randomly, systematic sampling is determined in a systematic way. For example, the researcher has a list of 2,000 people. If only 100 people will be selected as informants, then the researcher determines for himself that for example, those who are in the order of number 10, 20, 30, and so on are selected.
Stratified sampling or stratified sampling
From the terms used it is clear, namely strata. Here the researcher creates groups or clusters of the population based on strata. For example, the subject of research being carried out is students at a private university. Researchers make strata based on length of study. So the researcher has a list of how many informants were in the first, second, third year, and so on. Informants at each stratum are taken proportionally so that they represent students as the population being studied.
