Perpetration and Victimization: The Impact of Question Order on Reports of Intimate Partner Violence and Sexual Violence
Twiddy, S. E., Lynberg, M., & Carley-Baxter, L. (2007, May). Perpetration and Victimization: The Impact of Question Order on Reports of Intimate Partner Violence and Sexual Violence. Presented at American Association for Public Opinion Research Conference, Anaheim, CA.
The context effect of question order on attitudes and opinions has long been studied in the survey research literature. Much of this research has demonstrated that different estimates can be produced depending on the order in which questions are asked, which can be problematic when question ordering leads to different estimates for key questions. However, this concern becomes even more critical in surveys on issues impacting health and public policy such as intimate partner violence (IPV) and sexual violence (SV). Prior research suggests that it is difficult to obtain accurate prevalence estimates of IPV and SV. Few previous studies on these topics have asked about both victimization and perpetration in the same questionnaire. The Pilot Study of the National Intimate Partner and Sexual Violence Survey (NISVS) was a nationally representative random-digit-dial survey of adults 18 years of age and older focused on IPV and SV perpetration and victimization. Prior to fielding the survey, participants were randomly assigned to receive either the set of victimization questions or the set of perpetration questions first. In the analysis, we compared response and refusal rates, as well as negative behavior reports for four topic areas (stalking, sexual violence, physical violence, and emotional behaviors) for both orders of question sets (victimization questions first or perpetration questions first). These results will be used to determine the most effective question order to use for future studies of IPV and SV that include both victimization and perpetration questions.