Questionnaire response bias and face‐to‐face interview sample bias in sexuality research

Abstract

Response bias in self‐administered questionnaires (SAQs) and sample bias in face‐to‐face interviews (FTF) are examined. SAQ study results suggest (a) nonresponse bias is of chief concern in studies of masturbation, (b) partial responders fail to answer sexuality questions because they are uneasy making personal disclosures of sexual information, and (c) this latter threat varies inversely with sexual knowledge and experience indices. No significant order effects were found within SAQs, a result likely related to subjects' high educational levels. The FTF interview study showed (a) volunteers, relative to nonvolunteers, are more willing to disclose sexual information independent of overall self‐disclosure tendencies, (b) completing SAQs with sexuality items did not affect volunteering, (c) results from FTF interviews overestimate the breadth of sexual behaviors and knowledge, and (d) the FTF interview method may be less advantageous to use than SAQs. Data are also presented which support the reliability and construct and discriminative validity of a self‐disclosure measure specific to the sexual context. Within the limits of the present sample the findings suggest relatively less bias than might be expected in research on sensitive topics like sexuality.