"It Still Happened": A Mixed-Methods Analysis of College Students' Rationales for Endorsing Acts of Violence Victimization

Objective: Measures of relationship violence have been criticized for failing to distinguish intentional acts of aggression from behaviors that occur in a playful context. However, efforts to reduce this concern by modifying the questionnaires' instructions have not reliably reduced reporting rates or improved the criterion validity of the measures. This experimental study investigated how respondents who were randomly assigned to one of three instruction conditions perceived and responded to a measure of relationship violence. Method: Undergraduate students (N = 150) reported on partner violence victimization using the Conflict in Adolescent Dating Relationships Inventory (CADRI; Wolfe et al., 2001). They received either the standard instructions, instructions to exclude acts that occur while joking around or roughhousing, or instructions to include only intentional acts. Participants then took part in a think aloud procedure and a structured cognitive interview to examine their reasons for endorsing or not endorsing items. Results: Instructions designed to eliminate reports of behaviors that occur in playful contexts or are not intended to harm did not reduce overall rates of victimization. Respondents' perceptions of their partners' intent and the impact of the behavior on them were particularly salient when deciding whether to endorse items, but many participants reported behaviors simply because they occurred. Conclusion: These findings underscore the importance of victims' subjective perceptions and experiences in determining whether to endorse a particular behavior and offer insight into why instructions not to report certain kinds of behaviors do not consistently decrease rates of violence or improve the measure's validity.