Alcohol and sexual risk behaviors as mediators of the sexual victimization-revictimization relationship

Abstract

Objective: Women who experience sexual victimization, whether in childhood, adolescence, or adulthood, are at elevated risk of sexual revictimization. The mechanism responsible for this robust association is unclear, however. The present study proposed and tested a prospective, mediated model that posited that the association between adolescent and college victimization is mediated via 2 types of risk exposure in the first semester of college: alcohol-related and sexual risk behaviors.

Method: Female adolescents (N = 469) were recruited from the community at the time of high school graduation. They completed baseline assessments as well as follow-ups at the end of the first and second semesters of college.

Results: Consistent with hypotheses, adolescent sexual victimization was associated indirectly, via high school risk behaviors, with increased first-semester college risk behaviors (i.e., sexual partners, hookups, heavy episodic drinking, and heavy drinking contexts), which were, in turn, strongly predictive of sexual victimization experiences in the first year of college. College risk behaviors partially mediated the significant association between adolescent and first-year college victimization; however, even women without prior victimization faced elevated risk of college victimization with higher levels of college risk behaviors.

Conclusions: Women who have experienced adolescent sexual victimization engage in higher levels of risk taking in college, thereby increasing vulnerability to college victimization. Intervention to reduce these primarily alcohol-related risk-taking behaviors may reduce vulnerability to college sexual victimization.

Figure 1

Conceptual prospective model of first semester college risk behaviors as mediators of sexual revictimization during first year of college. T0 is the baseline at the end of senior year in high school; T1 is the end of the first fall semester in college; and T2 is the end of the first spring semester in college.

Figure 2

Standardized parameter estimates for hypothesized mediational path model. Chi-square = 64.65, df = 20, N = 469, p < .001, CFI = .97, TLI = .92, RMSEA = .07. Error terms and correlations are omitted for clarity. HED, heavy episodic drinking. Loadings marked “nt” were not tested for significance since these parameters were fixed to 1.0 during estimation to identify the latent variables. ns, not significant at p < .05. ***p < .001.