Volunteer Bias in Sexuality Research Using College Student Participants
Wiederman, Michael W., The Journal of Sex Research
Given the private and sensitive nature of sexual information, researchers typically must rely on self-report (Turner, Danella, & Rogers, 1995). Recruiting potential respondents ultimately results in a sample which is less than completely representative of the larger population, thus opening the door for potential volunteer bias (Brecher & Brecher, 1986). To the extent that nonvolunteers systematically differ from volunteers, the data generated are potentially biased (see Catania, Binson, Van der Straten, & Stone, 1995; Catania, Gibson, Chitwood, & Coates, 1990, for reviews of research on volunteer bias in sexuality research).
Others have long noted the apparent reliance on college student participants in psychological research (e.g., Jung, 1969; Smart, 1966), and a recent review indicated that one-half of reports of sexuality research published since 1971 have been based on such student participants (Wiederman, in press), probably because of convenience. What about volunteer bias and college student research participants? One can argue that college students are not representative of adults (Jung, 1969; Smart, 1966). To the extent that college student volunteers for sexuality research are not representative of even the college student population, the validity of such research may be even more suspect than researchers believe.
Nearly 25 years ago, Griffith and Walker (1976) demonstrated that, when given hypothetical experiments for which to volunteer, college students who chose a sexuality study were less inhibited and displayed less sex guilt compared to students who did not volunteer. Subsequently, Catania, McDermott, and Pollack (1986) had college students in introductory social science classes complete self-administered questionnaires, after which they were asked to indicate their potential interest in volunteering for an upcoming study that might be conducted by a member of the research team. The upcoming study was said to involve face-to-face interviews concerning the respondent's sexual attitudes and experiences. Relative to nonvolunteers, those students who volunteered for the interviews were more comfortable with personal disclosure and had more varied sexual experience, although there were no differences between the groups with regard to age, lifetime number of sex partners, current frequency of sexual activity, or number of sexual problems.
Strassberg and Lowe (1995) asked a large sample of college students participating in a self-administered questionnaire study involving sexuality to indicate whether each would be willing to participate in any or all of three additional studies: one involved a face-to-face interview with a same-gender researcher, another involved viewing sexually explicit videos and reporting reactions to them, and the third involved being connected to physiological instruments to measure sexual arousal to sexually explicit videos. Differences between volunteers and nonvolunteers were consistent across each type of study: Compared to nonvolunteers, volunteers displayed less sex guilt and higher levels of sexual experience.
Instead of asking students to indicate hypothetical interest in a future study, two sets of researchers actually manipulated the nature of the study for which students were supposedly volunteering. Saunders, Fisher, Hewitt, and Clayton (1985) posted sign-up sheets describing either a study involving completion of personality questionnaires or a study involving responses to erotica. In both studies, however, students were asked to watch a brief, sexually explicit film and anonymously report their affective reactions. Interestingly, male volunteers for each type of study did not differ with regard to sexual attitudes (i.e., erotophilia) or affective reactions to the erotic film. Female volunteers for the personality study, however, were more erotophobic and indicated more negative affective reactions to the film relative to female volunteers for the erotica study. …