Introduction
Many people often learn with surprise that even without their participation in opinion polls, pollsters still release results on certain characteristics of a population that they are members. People who are ignorant about processes of doing research, often rubbish findings of research studies in which their input was not sought. As unfortunate as it sounds, this group of people are usually oblivious of the sampling process that is part of many research exercises. Selection of suitable units from a population for a particular research activity is normally accomplished by implementing a sampling plan.
Quantitative Sampling
Sampling involves selecting units for study from a population, with particular emphasis on authentic representation. This means that the sample chosen for a research study should have the characteristics that closely and truthfully represent the population from which it’s drawn (Singh, 2007). The representativeness of a sample in quantitative studies counts substantially towards the success of the whole project. The most outstanding characteristics of the population under study should be present in the selected sample. Depending on the sampling procedures used, a researcher might encounter the problem of bias among the selected samples. Unfortunately, there is no definite way of arriving at a representative sample because errors are highly probable. And that is the reason why researchers are supposed to be keen and watchful of errors (VanderStoep & Johnson, 2009).
Researchers are aware of this problem and therefore they always rely on sampling strategies to minimize errors. A sampling strategy consists of the techniques and processes that a researcher employs in order to capture the most salient characteristics of the population and mirror them in the sample (Marshall, 1996). Quantitative research studies are characterized by a sampling design that has the following steps: identification of the target population, choosing the accessible population, formulating eligibility rules, defining the sampling plan, and then recruiting the sample. Some sampling strategies have capabilities to approximate the margin of errors by using statistical methods. Researchers are therefore advised to select sampling strategies that allow minimal errors in their samples. The most common sampling designs include: probability sampling and non-probability sampling (Singh, 2007). These sampling designs are discussed as follows:
- Probability sampling – elements are selected randomly. Probability samples are likely to have a representativeness that is reliable. The elements to be selected are assumed to have equal and independent probability for selection. Probability sampling consists of the following methods: cluster, simple random, systematic, and stratified random (Babbie, 2012).
- Non-probability sampling – non-random ways are used to recruit the units of the sample. This technique has very slim chances of generating representative samples. Despite its weakness, the technique has three applicable methods: purposive, convenience, and quota (VanderStoep & Johnson, 2009).
In this research study, the principal objective is to find out the impact of affirmation action programs in the work places. The quantitative research questions and their hypotheses have already been formulated and therefore we can embark on creating the sampling design. It is important to note that this sampling technique comes with challenges (Marshall, 1996). First, the selection of irrelevant samples will result into a waste of resources because the researcher will be forced to repeat the exercise. Another challenge will come in the form of identifying the employees who are beneficiaries of the affirmative action. Many of them do not like the minority tag. The following is an application of the sampling terminology (Singh, 2007).
- Sample: 250 employees randomly selected from 100 different work places
- Population: All employees working in city X, whose employment was necessitated by affirmative action programs
- Sampling frame: All employees whose names appear in the records of individuals who got employment through affirmative action programs
- Sampling design: Probability sampling
- Unit of analysis: employee who is a beneficiary of any affirmative action program
- Statistic: Average academic level of the employees the in the sample
- Parameter: Gender
Conclusion
A successful sampling plan should generate samples having a high degree of similarity with the general population from which they are drawn. Researchers can achieve the most appropriate sampling if they minimizing the prevalence of errors in their samples. By using sampling strategies such as probability sampling, investigators can successfully accomplish their research studies.
References
Babbie, E. R. (2012). The Practice of Social Research. (13th ed.). Belmont, CA: Cengage Learning
Marshall, M. N. (1996). Sampling for qualitative research. Family Practice. 13(6), 522–525
Singh, K. (2007). Quantitative social research methods. Thousand Oaks, CA: Sage Publications
VanderStoep, S. W. & Johnson, D. (2009). Research methods for everyday life: Blending qualitative and quantitative approaches. Hoboken, NJ: John Wiley & Sons