Measuring Sexual Aggression

Abstract

This study examined the reliability of a German version of the Sexual Experiences Survey (SES) using a retest design. A total of 114 respondents were asked to complete the SES twice within a period of 3 to 4 weeks. Overall percentages of consistent responses at the two data points were high, with a mean score of 95% across all items. Separate analyses for positive and negative responses revealed the high overall consistency to be due in large part to the high consistency of negative responses that dominated the response distribution. Positive responses, indicating experiences of sexual victimization or aggression, showed less stability over the two data points. Moreover, a distinct asymmetry in inconsistent responses was found, with retractions at Time 2 being far more frequent than revelations of previously undisclosed experiences. The implications of these findings are discussed in terms of the ability of the SES to detect sexual aggression and victimization.