Do Instructions Intended to Reduce False Positives Improve the Measurement of Physical Partner Violence Victimization Among Adolescents and Young Adults?

Abstract

Objective: We examine how instructions to exclude behaviors occurring in playful or joking contextsinfluence the measurement of physical partner violence victimization. Specifically, we demonstrate howsuch instructions influence the prevalence and validity of self-reported victimization. Method: Study 1used a Think Aloud procedure to evaluate thoughts of college students (n =451) reporting victimizationexperiences that occurred during high school. Participants were randomized to report on physical partnerviolence victimization with or without instructions to exclude playful acts. Study 2 experimentallyevaluated whether the instructions affect the criterion validity of victimization scores with measures ofdepressive symptoms and emotion regulation among first-year college students (n = 615). Study 3 soughtto replicate findings from Study 2 in a community sample of 18–25-year-olds (n = 398), usingalternative violence items, response formats, and a different recall period. Study 4 utilized a short-termlongitudinal design to replicate the pattern of findings from Studies 2 and 3, and examine howinstructions influence self-reports of revictimization over a 2-month follow-up among first-year collegestudents (n= 887). Study 5 presents a single-paper meta-analysis that synthesizes prevalence rates acrossthese four studies. Results: Overall, instructions designed to eliminate aggressive acts in joking contextsdid not consistently influence prevalence rates of victimization or improve criterion validity overstandard instructions. Conclusions: Instructions designed to exclude behaviors occurring in playful orjoking contexts do not necessarily produce more valid self-reports of physical partner violence victimization,as compared with standard instructions.