The sexual assault of women is a serious public health problem. Young college women as compared to their non-college peers and the general population are disproportionately assaulted or raped by men indicating a need for effective preventive interventions. Much research has been conducted on understanding the risk factors for engaging in sexual violence perpetration, but little research has focused on protective factors. Moreover, most of this research has focused on individual-level characteristics of the perpetrators and to some degree on peer influences. The ecological systems model posits that human behavior is influenced directly and indirectly by various contextual factors located within multiple environmental levels. Before effective prevention programs can be developed, more research is needed that can enhance our understanding of modifiable factors drawn from multiple levels of the environment that buffer or protect men from perpetration in addition to those that place men at risk for perpetration. The proposed study will define a multilevel protective model of sexual violence perpetration among male college students in Georgia. Data will be collected from a random probability-to-proportionate size sample of 30 colleges on factors characterizing each college's response to sexual violence, compliance to federal mandates regarding reporting of sexual violence incidents, and other aspects of the campus setting across an 18-month study period. Data will also be collected online from 30 college administrators and from a convenience sample of 1800 male college students. Generalized hierarchical linear modeling will be used to model the data. Specifically, the project will 1) describe the developmental trajectory of sexual violence perpetration taking individual, familial, peer and college factors into account;2) identify multilevel protective and risk factors for sexua violence perpetration in the population;3) define multilevel mediation models for protection of sexual violence perpetration considering college-level factors'direct and indirect influence through individual, familial or peer factors;and 4) define multilevel moderation models for protection of sexual violence perpetration considering college-level factors buffering effects on individual, familial or peer risk factors. The results will be used to inform specific policy and program recommendations to respond to the sexual violence of female college students on college campuses.
Public Health Relevance Sexual violence perpetrated against women causes severe physical, interpersonal, and psychological trauma and is costly for individuals and society. College women in particular are at heightened risk for sexual violence. The proposed research will define a multilevel protective model of sexual violence perpetration specific to male college students that will contribute significantly to an empirical knowledge base informing college-level policies and community-based interventions.