A Test of a Model of Sexual Victimization: A Latent Variable Path Analysis
Roodman, Allison A.
MetadataShow full item record
Both a recent narrative review and a meta-analytic review of prevalence rates, indicates that prior sexual victimization increases risk for future victimization (Messman & Long, 1996, Roodman & Clum, in press). The purpose of this study was to examine two competing models of sexual victimization that examined the path between child abuse and later sexual victimization. Hypothesized mediating variables were negative cognitive schemas, dissociation, risky behaviors, and coping strategies. Structural equation modeling was used to examine two competing models of sexual victimization. A sample of 276 college students taking introductory psychology were participants. They anonymously completed a packet of questionnaires that provided the indicator variables for the path models that were tested. Both models tested received minimal support but many of the proposed pathways in the model were not statistically significant suggesting problems with the models. Due to measurement issues with the manifest indicators of the latent factors, any results should be viewed with caution. It appears as though none of the factors in the model mediate the relationship between early and later victimization. However, both models tested demonstrated significant pathways between the factor for child abuse (comprising physical and sexual abuse) and negative cognitive schemas and for child abuse and dissociation. However, the paths from negative cognitive schemas and dissociation to sexual victimization (comprising both adolescent and adult sexual victimization) were not significant suggesting that, although these factors are influenced by child abuse, they do not mediate revictimization. Risky behaviors, as measured by consensual sex and alcohol consumption, do not appear to be influenced by early abuse, but there was a significant pathway between this factor and sexual victimization suggesting that these risky behaviors are independent risk factors for sexual victimization in adolescence and adulthood. In one model there was a significant pathway between child abuse and sexual victimization which is what would be expected given previous findings that suggest past abuse is the best predictor of future victimization experiences (Roodman & Clum, in press). That the other model did not demonstrate this relationship was surprising.
- Doctoral Dissertations