中華心理衛生協會

 Purpose: Proponents of theoretical models of sexual assault emphasize that these models are comprehensive, and they claim that these models can explain various types of sexual assault regardless of the perpetrator's or the victim's race, gender, or age. However, these models have not been empirically validated with heterogeneous populations. We examined 2 sexual assault models-a self-developed 4-factor model, and Malamuth et al.'s (1995) confluence model--to assess the comprehensiveness of these 2 models in explaining forced intercourse and multiple forced intercourse behaviors in 2 highly heterogeneous populations: male college students (community sample) and male inmates (institutional sample). The self-developed model incorporates 4 risk factors: adverse childhood experiences, deviant sexual arousal, sexual promiscuity, and psychopathic personality traits. The confluence model includes variables related to 2 pathways of sexual assault formation: hostile masculinity and sexual promiscuity. Methods: We collected data using a cross-sectional, anonymous self-report questionnaire. A total of 1,015 male college students and 1,029 male inmates participated in the study. The questionnaire included variables proposed by the 2 aforementioned sexual assault models. Data analysis primarily involved regression analysis and structural equation modeling to verify the fit between each of the sexual assault models and the data. Results: Our analysis resulted in 3 major findings: (1) Both sexual assault models can explain sexual assault and multiple sexual assault behaviors among both male college students and inmates. (2) The self-developed 4-factor model demonstrated greater explanatory power for sexual assault and multiple sexual assault behaviors than the confluence model for both populations, with early adversity, deviant sexual arousal, and sexual promiscuity exhibiting higher explanatory power for sexual assault. (3) Both models demonstrated greater explanatory power for sexual assault and multiple sexual assault behaviors for the male college students than for the inmates. Conclusions: These findings confirm that both sexual assault models can be applied to explain sexual assault behaviors across these 2 highly heterogeneous populations, providing empirical support for assertions regarding the encompassing nature or inclusiveness of sexual assault theory. However, our research sample did not include women, and the types of sexual assault did not encompass child sexual abuse. Future research is needed to determine whether female sexual assault behaviors and child sexual abuse can be explained by these sexual assault theories in order to solidify the argument that these theories are truly inclusive. For practical application, we recommend that because of the impact of early adversity on the likelihood of sexual assault behavior, government and social agencies should proactively allocate resources to high- risk families and intervene early to mitigate the negative consequences of early adversity. Furthermore, as deviant sexual arousal and sexual promiscuity significantly contribute to the likelihood of sexual assault behavior, we recommend that cognitive-behavioral therapy be employed to treat individuals with deviant sexual arousal, and that public awareness campaigns be used to highlight the negative consequences and potential costs associated with early sexual promiscuity.