The Relationship Between Family Characteristics and Criminal Behavior Among Sexually Abusive Juveniles and Delinquent Youth

Purpose: Researchers have documented the significant challenges faced by families of adolescent sexual abusers (e.g., Barbaree, Marshall & McCormack, 1998), including parent criminality, intergenerational patterns of sexual victimization, and substance abuse. Additionally, family violence has been theorized as a predictor of juvenile sexual offenses. However, few researchers have investigated the relationship between family characteristics and sexual and nonsexual criminality among sexually abusive youth. Family problems have long been identified as a critical risk factor for understanding delinquency (e.g., Farrington et al, 2000), and exposure to domestic violence was found to be related to sexual offending (poor impulse control and callous-unemotional traits did not mediate the effects of witnessing family violence; Caputo, Frick, & Brodsky, 1999). Moreover, a recent meta-analysis reported that adolescent sex offenders had greater exposure to sexual and nonsexual violence in the family, compared with delinquent youth (Seto & Lalumiere, 2010). While specific family characteristics (e.g., mental health issues) and family violence are prevalent among both sexual and nonsexual offenders, further research is needed to determine what relationships may exist between these characteristics and sexual aggression and delinquency.

Method: Survey data were collected from two racially/ethnically diverse groups of sexually abusive youth (n=332) and delinquent male adolescents (n=170), with an average age of 16.6 (SD=1.53), in court-ordered residential treatment. Data were collected using a set of demographic questionnaires, e.g., regarding criminality committed by family members, a history of family mental health problems, parental abuse of alcohol or drugs, and history of sexual victimization (family problems scale). Additionally, Elliot, Huizinga and Ageton's (1985) Self Reported Delinquency (SRD) measure was used to assess delinquency. Using logistic regression models we tested whether family problems predicted sexual and nonsexual offenses.

Results: Sexually abusive youth committed a greater number of delinquent acts (e.g., set fires and used alcohol) t(448)= -3.78, p<0.001), and had more severe family problems t(495)= -10.84, p<0.001), than the delinquent group. Group differences were also found with regard to family structure (e.g., sexual abusers were more likely to live with a mother or a grandparent). Thirty percent of sexual offenders reported that sexual abuse of children was a family problem, compared with 2.5% of delinquent youth. Logistic regression indicated that the family problems scale score was significant in predicting sexual offender status (χ2=76.37, p < .001), and interestingly, the total delinquency scale score was nonsignificant. The model correctly classified 82.1% of sexual offenders, but only 54% of delinquent youth.

Implications: Implications include that social work practice with juveniles who sexually abuse should consist of multiple modalities (targeting trauma, the home environment, and delinquency), including tested family therapy approaches, e.g., Multisystemic Family Therapy (MST; Borduin & Schaeffer, 2002). Yet, data suggests that treatment programs for adolescent sexual abusers are decreasing their use of family-based approaches (Burton & Smith-Darden, 2001). The finding that specific family problems predicted sexual offender status was in contrast to some previous research. Additional research is needed to confirm whether targeting family characteristics reduces the risk for sexual recidivism, and to further assess nonsexual crimes among sexual abusers.