Factors Associated with the Sexual and Nonsexual Offending of Adolescent Sex Offenders

Abstract

There can be no doubt that sexual assault, especially involving children, is one of the foremost concerns of our society today. However, not only are children often the victims of sexual assault, in many cases those considered by the law to be children (10 to 18 years of age) are also the perpetrators of sexual assault. This thesis reports on three studies that investigated factors associated with the sexual and nonsexual offending of adolescent sex offenders. In Study 1, a prospective longitudinal approach was taken to identify predictor variables for sexual recidivism among young men who had committed sexual offences as adolescents and were followed up into adulthood. The sample consisted of 303 clients whose information was stored on an archival database within the New South Wales Department of Juvenile Justice. The study found although sexual recidivism rates before age 18 were 25%, adult rates of sexual recidivism were much lower (4.5%). Adult rates of nonsexual recidivism, however, were high (61%). An older age at initial assessment, having a history of nonsexual offences, and a higher number of charges for the index sex offence, were all associated with an increased risk of adult sexual recidivism. Study 2 compared developmental characteristics (abuse histories, conduct problems), individual characteristics (callous unemotional traits; externalising and internalising behaviours), and family functioning (parental involvement; parental supervision and monitoring; positive parenting practices) of 108 male adolescent sexual offenders. These measures were subjected to Principal Components Analysis and yielded three main components, designated Negative Environment, Positive Environment and Transgression. Three subgroups of participants were then identified on the basis of their criminal histories: specialist offenders (n = 47); versatile offenders (n = 33), and short-duration offenders (n = 28). MANCOVA revealed a significant multivariate main effect for offender subtype. Versatile offenders were significantly different from the other groups on the Transgression component, which contained elements of abuse victimisation (physical and sexual) as well as abuse perpetration (conduct problems). This group also had significantly more adult victims than either of the other groups. For Study 3 a group (n = 12) of male youth were recruited from specialised adolescent sex offender treatment programs in New South Wales and Queensland. Participants took part in a recorded semi-structured interview and a verbatim transcript of the interview was produced for analysis. The construct of “Relational Templates” emerged from the data as a core category around which their experiences of sexual and nonsexual offending were organised. When viewed as an attempt to engage another in sexual relations, the sexual offending of this group was seen to share similarities to their approach to nonsexual relationships, including their nonsexual offending or delinquency. The thesis considers the findings of these studies with respect to etiological theories for adolescent sexual offending and discusses the implications of these findings for treatment programs and public policy.