Sampling strategy is a process or procedure of selecting a portion of a population which can be used to make inference on that population (Frankfort-Nachmias & Nachmias, 2008). Sampling strategies are categorized into two designs: probability sample designs and non-probability sample designs (Frankfort-Nachmias & Nachmias, 2008, pp 168-169). Probability sample designs are characterized as having a higher degree of representativeness of the population being study. This characteristic can strengthen quantitative research studies by reducing the risks to study validity. Variations of probability sample designs include simple random sample, systematic sample, stratified sample, and cluster sample. Non-probability sample designs, on the other hand, are considered as not having the representativeness of the population being studied, and can weaken the study validity (Frankfort-Nachmias & Nachmias, 2008, pp 168-169). Variations of non-probability sample designs include convenience sample, purposeful sample, and quota sample. In my research plan, I want to explore the factors that are responsible for the increasing incidence of HPV infections among young people in the United States given that the Food and Drug Administration (FDA) has since 2006 approved vaccines that can prevent the disease. I hypothesize that political orientation is a factor in HPV vaccine initiation. To test this hypothesis, the HPV vaccination rates of each of the 50 states and the District of Columbia will be compared with the political party affiliation of the population in each state. The sampling strategy being employed is non-probability sampling (Purposeful sample). HPV vaccination rates for each state will be taken from the CDC’s 2011 National Immunization Survey – Teen (NIS-Teen) (Centers for Disease Control and Prevention [CDC], 2012, pp 4) and political affiliation data will be taken from the 2011 political affiliation tracking surveys by Gallup (Gallup, Inc., 2012, pp 3). The two sources of my study data used probability sampling with 95% confidence interval in their surveys (CDC, 2011, pp 4; Gallup, Inc., 2012, pp 1). This can perhaps minimize the threat to validity posed by non-probability sampling since the underlying sampling strategy used when the data was initially collected used probability sampling. For each strategy that you did not choose, state why that one is not appropriate for your research questions, hypotheses, and variables. Selecting a sampling strategy is predicated on the type of study being conducted. Conducting a correlation analysis using non-probability sample is sufficient to test the hypothesis that political affiliation is a factor in HPV vaccine initiation. Any of the four variations of probability sampling strategy (simple random sample, systematic sample, stratified sample, and cluster sample) could be used in my analysis as well. Probability sampling has a higher degree of representativeness of the population being study because of its reliance on the use of randomization in selecting sampling units from a population (Laerd Dissertation, 2012). It also reduces the risks to study validity. However, probability sampling was not used as a sampling strategy in my study for the following reasons: a) it is time-consuming, and b) it is expensive to conduct (Lesser, n.d.). This scholar-practitioner project is limited by time that may not be aligned with the time required to conduct a probability sampling. Since the data required to conduct the study analysis could be obtained using a sampling strategy that is less time-consuming and inexpensive, I opted for a non-probability sampling strategy. Sample size computation To derive the appropriate sample size for this project, G*power 3, a free statistical software developed by (Faul, Erdfelder, Lang, & Buchner, 2007) was used to compute the sample size. With a moderate effect size of .4, an alpha of .05, and a default statistical power of 80%, the appropriate sample size calculated for this project is 46 when conducting a correlation statistical test.
References
Centers for Disease Control and Prevention. (2012). National and State Vaccination Coverage Among Adolescents Aged 13–17 Years — United States, 2011. Retrieved fromhttp://www.cdc.gov/mmwr/preview/mmwrhtml/mm6134a3.htm?s_cid=mm6134a3_e%0D%0AFaul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191. Frankfort-Nachmias, C., and Nachmias, D. (2008). Research methods in the social sciences (7th ed.). New York: Worth. Gallup, Inc. (2012). More states move to GOP in 2011. Retrieved from http://www.gallup.com/poll/152438/States-Move-GOP-2011.aspx#3 Laerd Dissertation. (2012). Probability sampling. Retrieved fromhttp://dissertation.laerd.com/probability-sampling.php Lesser, V. (n.d.) Advantanges and disadvantages of probability and non-probability based surveys of the elderly and disabled. Retrieved from http://ncat.oregonstate.edu/pubs/TRANSED/1081_Surveys.pdf
