Sampling Strategy and Sample Size in Female Condom Research
This paper critically evaluates the sampling strategy and sample size used in Choi et al.'s (2008) randomized controlled trial examining female condom skills training among women attending family planning clinics in the San Francisco Bay Area. The paper discusses the nonprobability nature of the study's sampling design, the geographic and socioeconomic limitations of the 409-participant sample, and how these constraints may have influenced the study's results. It also contrasts probability and nonprobability sampling methods, drawing on Doherty (1994), and considers whether tools such as G*Power might improve sample size determination in similar exploratory research designs.
- Overview of the Study and Its Sample: Describes Choi et al.'s RCT design and participant pool
- Sampling Size Limitations and Justification: Examines constraints and researchers' justification of sample
- Probability vs. Nonprobability Sampling: Contrasts sampling types using Doherty's framework
- Geographic and Cultural Constraints on Generalizability: Discusses how location limits study's broader applicability
- Implications for Research Design: Reflects on nonprobability sampling in exploratory research
✍️ How to write this paper — guide, tools & examples ▾
What makes this paper effective
- Grounds its methodological critique directly in the source study, citing specific page numbers and study details to support each evaluative claim.
- Clearly distinguishes between probability and nonprobability sampling using an external theoretical source (Doherty 1994), demonstrating awareness of sampling theory beyond the study itself.
- Acknowledges real-world constraints — cost, logistics, geographic access — that make nonprobability sampling a practical necessity in exploratory research, showing nuanced judgment rather than simple criticism.
Key academic technique demonstrated
The paper demonstrates methodological evaluation: rather than simply summarizing a study, it critically assesses the study's sampling decisions, traces their consequences for generalizability, and situates those decisions within established sampling theory. This involves moving between descriptive summary and analytical critique, a key skill in research methods writing.
Structure breakdown
The paper opens with a summary of Choi et al.'s study design and sample characteristics, then identifies specific limitations of the sample size and selection process. It introduces theoretical definitions of probability and nonprobability sampling for contrast, applies that framework to the study under review, and closes by acknowledging geographic and cultural factors that further constrain the study's generalizability while defending the practical rationale for nonprobability approaches in early-stage research.
Overview of the Study and Its Sample
The study by Choi et al. (2008), entitled "The Efficacy of Female Condom Skills Training in HIV Risk 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 in four San Francisco Bay Area cities: Concord, Mountain View, Santa Cruz, and San Francisco. A total of 409 women elected to participate.
Women in the experimental group experienced a four-session female condom skills training intervention, while the control group underwent a four-session general women's health promotion intervention. Participants answered questionnaires at baseline, at three months, and at six months (Choi et al., 2008, p. 1841). This design is significant because the sampling size was constrained by several factors: the women's attendance at clinics in specific geographic areas, their willingness to answer detailed and intimate questions about their lives, and their availability and inclination to participate in relatively time-intensive educational sessions.
Sampling Size Limitations and Justification
The sampling size appeared to be based on the number of women the researchers could recruit, given the practical limits of the study, rather than on the total number of women attending the clinic — a figure not reported in the research article. The sample is relatively small, particularly given that it was drawn from a narrow geographic area and a specific socioeconomic group. The women were primarily low-income, as evidenced by their reliance on a clinic as their primary source of healthcare.
To justify the sample, the researchers noted that the group was ethnically diverse and fell within a relatively narrow age range, and that participants exhibited similarities in male condom use, number of partners, and lifetime history of sexually transmitted infections (Choi et al., 2008, p. 1844). The randomized control research design was intended to ensure a balanced distribution of participants across both the experimental and control groups.
Probability vs. Nonprobability Sampling
Probability sampling is conducted according to a highly specific methodology. Researchers first select the population of interest, enumerate the units of that population (such as gender, socioeconomic class, and lifestyle), and then determine the sample size using a probabilistic algorithm (Doherty, 1994). As Doherty (1994) explains, "Interviewers will have no choice about who they are to interview. The algorithm specifies who is to be in the sample."
In contrast, the nonprobability sampling used by Choi et al. (2008) follows a far less formal methodology. A classic nonprobability example, as Doherty (1994) describes, is "the case where we want to get information about the urban population, but can afford to sample in only one city" — and thus a representative sample is selected informally rather than algorithmically. In this study, a representative sample drawn from four California clinics was used to represent low-income women who could benefit from instruction in female condom use. Cost and logistical constraints influenced this choice more than a desire for statistical rigor.
References
Choi, K., et al. (2008). The efficacy of female condom skills training in HIV risk reduction among women: A randomized controlled trial. American Journal of Public Health, 98(10), 1841–1848.
Doherty, M. (1994). Probability vs. nonprobability sampling in sample surveys. The New Zealand Statistics Review, 21–28.
Create your account
Always verify citation format against your institution’s current style guide requirements.