Longitudinal Relations Between Dating Violence Victimization and Perpetration: The Moderating Role of School Norms about Dating Violence

Introduction. Dating violence (DV) victimization is experienced by a large percentage of youth and can lead to later perpetration (Gomez, 2011). The current study examined the extent to which psychological and physical DV led to subsequent increases in psychological and physical DV perpetration. Drawing from social learning theory (Bandura, 1977), exposure to norms supporting aggression may increase the likelihood of perpetration. Although research has demonstrated the influence of individual and peer norms on DV perpetration, studies have not examined school norms about DV. Thus, the current study examined the impact of school norms about DV on relations between DV victimization and perpetration.

Method. Participants were drawn from the Multisite Violence Prevention Project (MVPP, 2004) and included ethnically diverse youth who were in a dating relationship within the last 3 months at T1 and T2, in the fall and spring of 6^th grade (N = 2,022; 43% female). The Dating Violence and Norms measure (Foshee et al., 1996) was used to measure T1 physical (α = .89), and psychological (α = .68) victimization, T2 physical (α = .91) and psychological (α = .76) perpetration, and T2 norms about the acceptability of DV (α = .82). Analyses were conducted in Mplus and utilized a multilevel approach where student- and class-level (same school and cohort; n = 74) variables were represented at Level 1 and 2, respectively.

Results. Models employed a hierarchical approach, whereby Level 1 controls (family structure, gender, race/ethnicity, and T1 perpetration) were entered first, Level 1 predictors (T1 victimization) were entered next, followed by Level 2 controls (intervention condition) and predictors (norms about DV). Random slopes were added one-by-one to determine if there was significant variance in relations between victimization and perpetration across classes and if a test of cross-level interactions was warranted. Relations between victimization and perpetration were not significant and did not significantly vary across classes. As such, cross-level interactions were not tested. There was a significant positive association between class norms and psychological, but not physical, perpetration, such that more supportive norms were related to higher class averages of psychological perpetration (β = .59, p < .05).

Conclusions. Previous research has demonstrated the impact of DV norms on physical perpetration (Noonan & Charles, 2009); however, the current finding is novel in that it establishes the relation between class norms about DV and psychological perpetration and underscores the need for prevention efforts to address DV at multiple levels of the school system.