Sampling Strategy and Sample Size for a Quantitative Research Plan

Sampling Strategy and Sample Size for a Quantitative Research Plan


Following the needs to determine the benefits of NGOs in South Africa, one important objective of the sample and sampling method will be to ensure generalization of data that proves to be coherent to the study requirements. One important aspect of any research is its ability to generate data from a large population through inferences derived from specific samples. This paper discusses a sampling method and sample size for the research plan on the benefits of NGOs in South Africa, in addition to describing the study population and a complete sample size.

Population and population size

The population group used during the research included adults, both male and female, between the age of 25 and 65, and who have a better understanding of NGOs’ operations, and the specific benefits citizens of South Africa derive from such operations (Adler& Clark, 2012, p. 12). The participants selected for this study might have worked with specific NGOs or have better experiences with NGOs, and this will mean having persons who have participated directly or indirectly in different NGO activities for periods not less than 3 months (Adler& Clark, 2012, p. 12). From a population of 410 respondents, a sample of 35 members will be selected through systematic sampling. The basis of using a sample size of 35 respondents is to improve on the validity of the results and also to ensure that the findings are well distributed (Adler& Clark, 2012, p. 15).



Sampling strategy

This study will explore the concepts of quasi-experiential design, which requires the researcher to use two groups for the purpose of generating data (Maree& Van, 2009, p. 9). The first quasi group will include members who have had face-to-face interactions with NGOs during their operations between the periods not less than three months while the second quasi group will include members who can participate in online surveys (Maree& Van, 2009, p. 10). In all the processes, the main objective of the sampling strategy is to provide supportive information on NGOs performance and its benefit to the whole community of South Africa. With such study specifications, the sampling strategy will focus towards obtaining probability sample (Maree& Van, 2009, p. 12). The probability sampling in this case will increase the chances of having a study sample that is representative and bears all the traits of the study (Maree& Van, 2009, p. 9). Therefore, the need to include probability sample in the study will help the researcher to eliminate instances of having a non-representative sample or data for analysis.

Prior to having a representative sample for investigation, the need to implement systematic sampling will help in determining an interval response (Maree& Van, 2009, p. 13). The interval will be selected for every nth response received during data collection(Maree& Van, 2009, p. 13). The value n represents the integer found as a proportion of the population size and the sample size as applied in systematic methods of sampling. For example, the study aims at using 410 respondents with a sample size of 35 participants. The nth value is obtained by dividing 410 by 35 (410/35), which is approximately 12 (Maree& Van, 2009, p. 14). The researcher will then proceed by picking every 12th person after the randomly selected starting point. The chosen sampling procedure as applied in this study is known as 1-in-12 systemic sampling for the proposed benefits of NGOs in South Africa. The use of systematic sampling procedure applied in this study will help in eliminating possible threat to validity caused by selection bias.

Sample selection for each of the 35 sample group

The use of Cohen’s d and t test for the selection of two independent samples becomes necessary for this study because choosing an exact population to represent the many responses from the entire communities where NGOs operations are felt becomes a problem, and data selection through consensus becomes inadequate (Onwuegbuzie& Leech, 2007, p. 40). Al the necessary processes that could be used to draw appropriate samples of 35 from a population of 410 individuals gave similar results of 35 samples from each group (Onwuegbuzie& Leech, 2007, p. 44).  For example, using a study that closely relates to the concepts of Cohen’s d, that is, subtracting the means of the two sample groups and finding the average of the standard deviations of the sample groups provide two different samples with close to 35 members (Onwuegbuzie& Leech, 2007, p. 44). Following the results of the different methods applied in selecting samples for this study, which advocate for a sample size of 35 individuals, it is therefore rational to use this number of participants as samples with adequate size.

Reasons for choosing a sample size of 35 from a population of 410 individuals

In every research, having a representative sample, which is randomly selected arises with the level of validity and precision of the result required (Charles& Ellis, 2006, p. 28).  A sample size of 35 from a population of 410 individuals is considered valid, convincing and precise and the results obtained in such as case present actual exhibits for other analysis(Charles& Ellis, 2006, p. 28). Otherwise, a sample size of 35 members from a whole population of 410 individuals is preferred by the various methods of sampling that could have been applied. For example, the main methods of sampling, systematic sampling, used in this research advocated having a data whose mean and standard deviations are normally distributed (Charles& Ellis, 2006, p. 31). Also noted is the simplicity of systematic sample, which allows searchers to add a systematic element into the randomly selected subjects without involving much statistical calculations. In this connection, the researcher is well convinced that the data collected on the benefits of NGOs in South Africa will be properly sampled (Charles& Ellis, 2006, p. 32). The systematic sampling avoids cluster sample selection, which unlike systematic sampling ignores the simple rules of equal distancing and randomness as applied in sample selection.

In general, the desires to develop a study that is built on construct ideas can only be possible if the sampling strategies are valid based on the empirical studies conducted on a similar topic (Charles& Ellis, 2006, p. 35). In this connection, it becomes important for the researcher to use appropriate sample size, conduct proper investigation and analyze the necessary evidences based on content and empirical validities (Charles& Ellis, 2006, p. 35). For example, a sample size of 35 individuals ensure content validity, ensures systematic analysis of tests in a way that covers all the samples selected for investigation. Once the samples are selected based on the research specifications (Charles& Ellis, 2006, p. 38), conducting questionnaires or using any other method of data collection becomes easier.




Adler, E. S., & Clark, R. (2012). An invitation to social research: How it’s done. Belmont, CA: Wadsworth.

Charles, C., & Ellis, P. (2006). TITLE: Enhancing Involvement in Treatment Decision Making by Women with Breast Cancer.

Maree, K., & Van, . W. C. N. (2009). Head start in designing research proposals in the social sciences. Lansdowne, Cape Town: Juta.

Onwuegbuzie, A. J., & Leech, N. L. (2007). Sampling Designs in Qualitative Research: Making the Sampling Process More Public. Qualitative Report, 12(2), 238-254.