A Gender-Based Investigation of Early Exposure to Violence, Depression, and Dating Violence Perpetration Among U.S. College Students

Background & Purpose

Research shows that children exposed to violence during childhood are at risk for engaging in violence as adults. Unfortunately, few studies have explored the risk mechanisms that connect early violence exposure with dating violence (DV) perpetration. Even less research exists concerning how such risk mechanisms may vary by gender. These knowledge gaps pose significant challenges for the development of violence prevention interventions. Nonetheless, theoretical explanations for these relationships exist. The general strain theory (Agnew, 1992) suggests that people who are exposed to strain (e.g., violence exposure ) are likely to engage in violence to manage their negative emotions (e.g., depression). Also, the theory emphasizes that females and males might experience different emotions and coping methods. Informed by theory, this study uses path analysis to examine how depression might mediate the relationship between exposure to violence and DV perpetration and possible differences between females and males.

Methods
Data were drawn from the International Dating Violence Survey (2001-2006), a rare source of information concerning early violence exposure, depression, and DV perpetration altogether and a unique opportunity to examine DV perpetration. This study included U.S. college students aged 18-23 years old who had dating relationships lasting more than a month (Female = 2,323, male = 1,052). The dependent variable was self-reported physical DV perpetration (eighteen items; revised Conflict Tactics Scales). The independent variable was exposure to violence (fourteen items; Personal and Relationship Profile; PRP; e.g., witnessing violence, experiencing physical abuse). Depression (8 items; PRP) was included as a mediating variable. Path analyses were conducted using Mplus version 7 and restricted maximum likelihood. Missing data were managed with Full Information Maximum Likelihood estimation.

Results
Both models for males and females indicated good model fit [Male: CFI=1.000, TLI=1.027, RMSEA=.000, SRMR=.000, Female: CFI=.996, TLI=.982, RMSEA=.024, SRMR=.008]. Early exposure to violence was positively associated with DV perpetration for both females (β=.19, p<.001), and males (β=.17, p<.001). Exposure to childhood violence had direct effects on depression for both females (β=.36, p<.001) and males (β=.41, p<.001). However, depression symptoms had direct effects on DV perpetration only for females (β=.11, p<.001). Thus, depression partially mediated the relationship between exposure to violence and DV perpetration only for females (β=.04, p<.001).

Conclusions & Implications
Results suggest that though early exposure to violence leads to depression in both young adult males and females, depression links directly to DV perpetration for females only. This result is consistent with previous research showing females exposed to violence during childhood are more likely to show internalizing problems which might lead to DV perpetration. In turn, such findings suggest that depression may be a key prevention target for violence prevention with female survivors of childhood violence. These results also suggest that male perpetrated DV may not be related to depression. While this study begins to show how the risk processes for violence perpetration may vary by gender, future research is needed to consider the other potential risk mechanisms that may lead to DV. Accordingly, this presentation will emphasize theoretically-informed recommendations for future violence prevention research.