The current study was part of a larger research project, The Children with Sexual Behavior Problems Longitudinal Study, designed to study wards of the Department of Children and Family Services (DCFS) who have exhibited sexual behavior problems. The study's sample included 131 males aged 9-17.

Two different models were hypothesized to differentiate two types of offenders: those who target younger children and those who victimize peer or older females. It was hypothesized that a history of sexual abuse would predict internalizing symptoms and social competence deficits, which would then increase the likelihood of abusing younger children. It was further hypothesized that exposure to domestic violence would relate to negative attitudes toward women and aggression to increase the likelihood of victimizing same-age or older females. Additionally, a history of physical abuse and community violence exposure were predicted to relate to trauma and aggression, which would increase the likelihood of offending against peer or older females. Correlations and MANOVAs examined associations and differences among variables. Using the MPlus software program, a pathway analysis tested each pathway and indirect effects of the mediator variables.

Although no significant differences emerged in the MANOVAs, the patterns of mean comparisons indicated that contrary to the overall significant associations between community violence and trauma, depression, and anxiety in the full sample, males who perpetrated against younger males reported the highest levels of exposure to community violence but the lowest levels of internalizing symptoms. Finally, although the overall model fit of each pathway was good, only one significant pathway emerged: witnessing community violence predicted trauma symptoms.

It is possible that increased power with a higher number of participants in each of the four groups would have yielded more substantive findings. The patterns of mean differences, however, may have useful clinical utility. Finally, it must be considered that classification of adolescents exhibiting sexually abusive behavior may need to diverge from that used with adults; the fluidity of adolescent development challenges the idea that demonstrating sexually abusive behavior represents a fixed pattern of behavior as documented in adult sex offenders.