Reviewing Sampling Strategy and Sample Size in a Research Article Research Paper

Sampling Strategy

From September 1999 to June 2004, Homik, Jacobsohn, Orwin, Piesse, and Kalton (2008) embarked on a research study to evaluate the behavioral changes that followed a government anti-drug campaign. The researchers used the stratified sampling strategy to select their respondents. The low possibility of human bias using this sampling strategy is a strong point of the stratified sampling strategy (Lund Research, 2012; Dessel, 2013). The sampling strategy also allowed the researchers to come up with highly representative data, based on the statistical conclusions derived from the sampling strategy (using large samples of respondents included in the study). The use of the stratified sampling strategy also helped Homik et al (2008) to come up with superior data because it not only helped to improve the potential for the researchers to divide the larger population sample into different units, but also improved representations across several strata. One limitation of the stratified sampling strategy was the difficulty associated with re-contacting the respondents, if there was a need to do so (Lund Research, 2012). This is because this sampling strategy requires the use of large samples of people, who may be difficult to re-contact after the first contact (Lund Research, 2012). However, the sampling strategy was appropriate for the “vast nature” of the research phenomenon investigated.

Strengths and Limitations of Sampling Strategy

The strengths and limitations of the sampling strategy adopted by Homik et al. (2008) depend on the fundamental processes that underlie the overall strategy in the first place. Key in the study’s sampling strategy was the use of questionnaires (administered on laptops) and the data collection setting. Using questionnaires as the main data collection technique limited the credibility of the study’s findings because it was difficult to establish if the respondents were telling the truth, or not, using the questionnaire responses alone. Furthermore, phenomenologists believe that most questionnaire designs limit the type of information that most researchers could get in a study (Lund Research, 2012; Dessel, 2013). Conversely, this data collection method limited the study’s validity. Homik et al. (2008) collected data in three phases, separated in 6-month intervals. This duration of data collection was appropriate for the study’s goals (evaluating behavioral changes) because understanding behavioral responses requires more than a couple of months of observation (Dessel, 2013). Therefore, if there were any behavioral changes emerging from the national anti-drug campaigns, the researchers would best capture them in the first few months of their implementation. The researchers did their research during this period. This strategy improved the reliability of their sampling strategy.

Sample Size Critique

The total number of population sampled by Homik et al. (2008) was large enough to provide reliable findings because in all four instances of collecting data, the researchers interviewed 8117, 6516, 5854, and 5126 people. The large sample populations provided an unbiased cross-sectional of young adults in the country. Such large population cohorts also represent a high confidence level. Furthermore, since the researchers collected the views of young people who were targets of the anti-drug campaign, it is plausible to believe that the percentage of the sampled population was similar to the percentage of the desired sample.

Did the Authors Justify the Sample Size?

Homik et al. (2008) say that they sampled the views of their respondents in four successive steps that involved interacting with four cohorts of young people. They justified this sample size because they wanted to have an unbiased representation of the views of American youth towards the government’s anti-drug campaigns. Furthermore, they justified this large sample because they wanted to have an accurate and efficient representation of the views of American youth towards the anti-drug campaigns (Homik et al., 2008). Comprehensively, it was important for the researchers to justify their sample size because it is important for researchers to use reliable and representative samples in research studies (Creative Research Systems, 2012). Furthermore, it is important for authors to show how their sample sizes align with their research goals (Buchner, Faul, & Erdfelder, n.d.).

Role of Sample Size

A credible research should demonstrate the use of an accurate sample size (Creative Research Systems, 2012). A small sample can undermine the credibility of a study because it would produce unreliable findings. Comparatively, while a large sample size could provide accurate results, the cost of collecting information could be high (Dessel, 2013). A study’s sample size is also likely to affect a study’s validity because inadequate responses would mean that a study is unable to measure what it should measure. Similarly, an “adequate” research sample would help researchers to undertake studies that measure what they should measure (Creative Research Systems, 2012). Therefore, having the correct sample size can strengthen, or weaken, a research study by affecting its reliability and validity.

Validation of an Idea through Personal Experiences

Knowing the correct sample size to use in research is a difficult process. The difficulty stems from knowing how much is enough? Similarly, even if researchers try to have a large sample size, the response rate could be low, thereby undermining the validity and reliability of the research findings. In this regard, I believe that most researchers should be preoccupied with making sure that their sample sizes reflect the acceptable confidence levels and have a low margin of error. Similarly, they should make sure that the standard of deviation is within the acceptable limits. I believe that by evaluating these factors, they would ensure that their findings represent the majority view of the population they intend to sample. I also believe there should be more emphasis on these three factors (confidence level, standard of deviation and standard of error) as opposed to the actual number of people who participate in a research study. Bearing these factors in mind, is there a formula that could merge these factors to help researchers know what population sample to use in their studies?

Economics behind Public Health Interventions

Feasibility of the Intervention and Long-term Savings

The proposed study seeks to investigate the efficacy of the interventions introduced to manage the 2012 Sierra Leone cholera outbreak. The study is economically feasible because it only seeks to evaluate what these interventions have managed to achieve since their introduction. Furthermore, the study would only use a representative sample of the interventions used, thereby reducing the cost of undertaking a full-scale survey of the existing interventions. The long-term savings that are likely to accrue from the study stem from its ability to help us know which interventions have worked and which ones have not. This way, it would be easier for health workers to establish which interventions they need to stop and which ones they should continue supporting (Wong et al., 2011). Such an approach would help in resource allocation processes because they could allocate more resources towards interventions that work and fewer resources towards those that do not work (Haines et al., 2013). This way, there would be long-term savings on health expenditure that would further improve the efficacy of existing health interventions.

How Potential Health Outcomes outweigh Cost of Intervention

Understanding the efficacy of existing interventions for managing cholera would create better health outcomes through increased preparedness in managing cholera. Furthermore, it would improve the country’s response to the outbreak by identifying interventions that work and which ones do not work. This effort does not only strengthen emergency responses to similar crisis, but also minimizes the possibility of another outbreak occurring. This means that Sierra Leone would be better prepared to manage another crisis. In fact, the death toll and resources used to manage future outbreaks would reduce, thereby saving the country the huge human and economic cost associated with cholera. This way, the cost of the study is low, compared to its economic benefits (Micklethwaite, Brownson, O’Toole, & Kilpatrick, 2012).

Justification of Intervention

Understanding the importance of public health interventions to manage cholera is important in improving Sierra Leone’s preparedness for another cholera outbreak because it satisfies three principles of ethical deliberation in public health as explained by of Upshur (2012). For example, the proposed study would satisfy the harm principle, which advocates for the introduction of public health interventions aimed at doing the least harm on human communities. Understanding which public health interventions work, and which ones do not, upholds this principle because the process would highlight which interventions have the most positive impact on the country’s cholera management efforts, and which ones do not (Upshur, 2012). Since the proposed study would also explain the reasons for the success and failure of public health interventions, it would also be easy to identify the least restrictive means for managing cholera (another ethical principle of public health) as the above cost-effective analysis demonstrates. The intervention would also uphold the transparency principle of public health because through the open access to public health information, it would be easy to establish the successes, or failures, of existing public health interventions. These insights justify the proposed intervention.

References

Buchner, A., Faul, F., & Erdfelder, E. (n.d.). G Power. Web.

Creative Research Systems. (2012). Sample Size Calculator: The Survey System. Web.

Dessel, G. (2013). How to determine population and survey sample size?Web.

Haines, T.P., Hill, A.M., Hill, K.D., Brauer, S.G., Hoffman, T., Etherton-Beer, C., & McPhail, S.M. (2013). Cost effectiveness of patient education for the prevention of falls in hospital: Economic evaluation from a randomized controlled trial.BMC Medicine, 11(1), 1–12. Web.

Homik, R., Jacobsohn, L., Orwin, R., Piesse, A., & Kalton, G. (2008). Effects of the national youth anti-drug media campaign on youths. American Journal of Public Health, 98(12), 2229–2237. Web.

Lund Research. (2012). Stratified random sampling. Web.

Micklethwaite, A., Brownson, C. A., O’Toole, M. L., & Kilpatrick, K. E. (2012). The business case for a diabetes self-management intervention in a community general hospital.Population Health Management, 15(4), 230–235. Web.

Upshur, R.E. (2012). Principles for the justification of public health intervention. Can J Public Health, 93(2), 101-3.

Wong, J.B., Coates, P.M., Russell, R.M., Dwyer, J.T., Schuttinga, J.A., Bowman, B. A., & Peterson, S.A. (2011). Economic analysis of nutrition interventions for chronic disease prevention: Methods, research, and policy.Nutrition Reviews, 69(9), 533–549. Web.