Volunteer bias in research employing vaginal measures of sexual arousal

Abstract

Demographic, sexual, and personality characteristics of volunteers and nonvolunteers for a laboratory study of sexual arousal were compared. Subjects were 296 female students from an introductory psychology course who had volunteered for a study on sexuality and personality. After completing several questionnaires, subjects were presented with a written description of an experiment that involved viewing sexually explicit videotapes and measurement of sexual arousal with a vaginal photoplethysmograph and indicated in writing whether or not they would be interested in participating in the experiment. Chi-square analyses revealed that a greater percentage of volunteers had experienced sexual trauma and that fewer women in this group reported objections to viewing sexually explicit films than nonvolunteers. A discriminant function analysis revealed that the volunteers masturbated more frequently, had more exposure to commercialized erotica materials, were exposed to these materials at an earlier age, and reported less sexual fear than nonvolunteers. Other personality, demographic, and sexual characteristics did not differ across the groups. This study demonstrates that the external validity of studies employing vaginal measures of sexual arousal is limited. Researchers must use caution when discussing the generality of findings based on genital measurement of sexual arousal.