Self-administered questionnaires versus face-to-face interviews in assessing sexual behavior in young women.

Effects of mode of assessment, person, and situational variables on the accuracy of self-reports of sexual behavior remain uncertain. To evaluate these influences, 190 young women completed measures of erotophilia and social desirability and then monitored their health-related behaviors with a diary for 8 weeks. They returned on two occasions to complete either a face-to-face interview (FTFI) or a self-administered questionnaire (SAQ) regarding their behavior over the same interval. To check the apparent accuracy of participants' retrospective self-reports, a difference score was calculated by subtracting responses obtained on the FTFI or SAQ from the diary card. Results indicated that both modes of assessment were reliable; reliability did not differ as a function of mode of assessment. However, SAQs elicited less discrepant responses for protected vaginal sex; SAQ and FTFI reports for unprotected sexual behaviors were equivalent. Situational and person variables did not predict accuracy scores, which were impaired at higher frequencies of behavior. Results suggest that both modes of assessment were reliable and SAQs may be more accurate for some sexual behaviors.