In going through a clinical research, it is necessary to determine the sample or representatives that will give validity and generalizability of the study’s output. Among all sampling procedures, random/probability sampling is the best way for preventing any bias (Chin & Lee, 2008). In random sampling, each of the population asked for participation in a controlled trial has equal chance of being selected.
The specific number of population to be recruited in a particular study is the first practical problem investigators face. Determining whether the number of samples is small or large is based upon how the conclusions made from a study will more likely to reflect the true effect of the treatment (Wittes, 2002). In some circumstances, if the subjects of a study is very rare (e.g. when a particular disease happen only once in 100 000 persons), sampling should only be done each time a case in that subject will be present (Fox, Hunn, & Mathers, 2009). But when the study is wide, certain population is needed. But sampling will be “small” or “large” based on the number of events observed. For instance, a sample of 2,000 women on placebo and 2,000 on a new therapy for studying the new drug’s effect in preventing hospitalization of women for hip fracture is “small” since the National Center for Health Statistics suggests that only 20 events are expected to occur in the said group of representatives. The 99 percent of the population will not provide essential information for the effect of the drug therapy (Wittes, 2002). More events are needed to make the study’s output valid for its effectiveness. The sample size for a trial must be large enough to allow reasonable chances of answering the question or providing information for the specific subject being studied. For instance, greater variation in the occurrence of a particular disease will require greater sample size – so that there is greater chance to validate the study.
References
Chin, R., & Lee, B. (2008). Principles and practice of clinical trial medicine. UK: Academic Press.
Fox, N., Hunn, A., & Mathers, N. (2009). Sampling and sample size calculation. Nottingham: The NIHR RDS for the East Midlands/ Yorkshire & the Humber.
Wittes, J. (2002). Sample size calculations for randomized controlled trials. Epidemiologic Reviews, 24 (1), 39-50.