Critique of sampling strategy and sample size in research

¶ … Sampling Strategy and Sample Size in a Research Article

The study by Choi (et al. 2008) entitled "The efficacy of female condom skills training in HIV risk: A reduction among women: A randomized controlled trial" examined the extent to which training women in how to use the female condom at family planning clinics could potentially promote the use of this form of birth control. The study population was selected from women attending clinics at four San Francisco Bay Area cities (Concord, Mountain View, Santa Cruz, and San Francisco). 409 women elected to participate in the study. The women in the experimental group experienced a 4-session female condom skills training intervention vs. The control group which underwent a 4-session women's general health promotion intervention. Women answered questionnaires at baseline, at 3 months, and at 6 months (Choi et al. 2008: 1841). This is significant because it stresses how the sampling size was relatively limited by a number of factors, namely the women's presence at clinics in very specific geographic areas; their willingness to answer detailed and intimate questions about their life; and the fact they had time and the inclination to participate in relatively time-intensive educational sessions.

The sampling size seemed to be based upon the number of women the researchers could solicit to participate, given the limits of the study rather than upon the number of women who attended the clinic as a whole, which was not noted in the research article. The sampling size seems relatively small, particularly given the fact that it was selected from a narrow geographic area and socio-economic group. The women were mainly low-income, given the fact they were relying upon a clinic as their primary source of healthcare. To justify the sample, the researchers note that the group as a whole was ethnically diverse and of a relatively narrow age range and exhibited similarities in both male condom use, number of partners and lifetime history of STDs (Choi et al. 2008: 1844). The randomized control research design was intended to ensure a balanced sampling in both experimental and control groups.