Steps In Sampling Design : A researcher should take into consideration the following aspects while developing a sample design ..
1) Type Of Universe : The first step involved in developing sample design is to clearly define the number of cases, technically known as the universe. A universe may be finite or infinite. In a finite universe the numbe of items is certain, whereas in the case of an infinite universe the number of items is infinite (i.e., there is no idea about the total number of items). For example, while the population of a city or the number of workers in a factory comprise finite universes, the number of stars in the sky, or throwing of a dice represent infinite universe.
2) Sampling Unit : Prior to selecting a sample, decision has to be made about the sampling unit. A sampling unit may be a geographical area like a state, district, village, etc., or a social unit like a family, religious community, school, etc., or it may also be an individual. At times, the researcher would have to choose one or more of such units for his/her study.
3) Source List : Source list is also known as the ‘sampling frame’, from which the sample is to be selected. The source list consists of names of all the items of a universe. The researcher has to prepare a source list when it is not available. The source list must be reliable, comprehensive, correct, and appropriate. It is important that the source list should be as representative of the population as possible.
4) Size Of Sample : Size of the sample refers to the number of items to be chosen from the universe to form a sample. For a researcher, this constitutes a major problem. The size of sample must be optimum. An optimum sample may be defined as the one that satisfies the requirements of representativeness, flexibility, efficiency, and reliability. While deciding the size of sample, a researcher should determine the desired precision and the acceptable confidence level for the estimate. The size of the population variance should be considered, because in the case of a larger variance generally a larger sample is required. The size of the population should be considered, as it also limits the sample size.
The parameters of interest in a research study should also be considered, while deciding the sample size. Besides, costs or budgetary constraint also plays a crucial role in deciding the sample size.
(A) Parameters Of Interest: The specific population parameters of interest should also be considered while determining the sample design. For example, the researcher may want to make an estimate of the proportion of persons with certain characteristic in the population, or may be interested in knowing some average regarding the population. The population may also consist of important sub-groups about whom the researcher would like to make estimates. All such factors have strong impact on the sample design the researcher selects.
(B) Budgetary Constraint : From the practical point of view, cost considerations exercise a major influence on the decisions related to not only the sample size, but also on the type of sample selected. Thus, budgetary constraint could also lead to the adoption of a non-probability sample design.
(C) Sampling Procedure : Finally, the researcher should decide the type of sample or the technique to be adopted for selecting the items for a sample. This technique or procedure itself may represent the sample design. There are different sample designs from which a researcher should select one for his/her study. It is clear that the researcher should select that design which, for a given sample size and budget constraint, involves a smaller error.