Social research involves the use of quantitative or qualitative study designs that are used to understand some important phenomena in life (Babbie, 2014). The qualitative approach uses descriptive data while the quantitative approach utilizes data that could be used to define some attributes of data points. Sampling is often used in social research to identify the actual number of study participants who should be included in a study. Study subjects are selected from populations, which are very large. Scientists select a certain number of study participants whose measurements or observations could be used to infer about the population (Babbie, 2014; Chen & Zhao, 2010).
Statistical power
The use of sampling helps social scientists to work with a number of study subjects drawn from the population that could give a good statistical power. Small sample sizes lead to less statistical powers. On the other hand, very large sample sizes lead to very good statistical powers. However, the use of large samples has also been associated with very long duration of time needed to complete studies. Therefore, scientists use statistical calculations to determine sample sizes that would be completed within a reasonable period of time (Chen & Zhao, 2010). The importance of a statistical power is that it increases the probability of achieving sample results that are close to the true findings in the population. Therefore, the use of sample sizes with good statistical powers enables the sample results to be used to make critical inferences about the population (Scherbaum & Ferreter, 2009).
Reduction in costs
Another importance of sampling in social science research is the reduction of study costs. For example, a social science researcher would be interested in assessing the factors that make patients not attend public health facilities in a certain location. That location could be characterized by thousands of subjects who could give responses to be analyzed in the study. However, it would require a lot of financial resources to involve all the subjects in the study. Therefore, it would be wise to calculate the right sample size that could give findings with a sufficient statistical power. The right sample size could also involve less financial costs (Chen & Zhao, 2010).
Standard error
Sampling is utilized to achieve results with the right standard error. Standard error denotes the average differences from the calculated mean. The bigger the value of the standard error the bigger is the difference of the calculated parameters from the true population values. Thus, social science scientists use sampling techniques so that they could use sample sizes that would result in small standard errors (Babbie, 2014).
Choice of data collection tools
Sampling is used to determine the methods of data collection that could be used in a research study. For example, studies involving large sample sizes could require the use of telephone interviews or questionnaires that could be sent via mail. However, if a study involves a small sample size, then observation could be used to collect data. The type of data collection method also has an impact on the quality of data that it could collect. For example, the use of questionnaires could result in some questions not answered by study participants while the use of observations could lead to the collection of accurate data (Chen & Zhao, 2010).
Selection of the right subpopulations
Finally, it is critical for social scientists to use sampling so that they could concentrate on subpopulations that could have important variables and ignore irrelevant ones. For example, it would be important for scientists to use only the populations that would have a certain disease rather than use mixed subpopulations, some of which would not have the disease.
References
Babbie, E. (2014). The Basics of Social Research. (6th ed.). Belmont, CA: Wadsworth. Web.
Chen, Z., & Zhao, F. (2010). Determining minimum survey sample size for multi-cell case. International Journal of Reliability, Quality and Safety Engineering, 17(06), 579-586. Web.
Scherbaum, C. A., & Ferreter, J. M. (2009). Estimating statistical power and required sample sizes for organizational research using multilevel modeling. Organizational Research Methods, 12(2), 347-367. Web.