Individual, familial, and social characteristics of adolescent child molester

This study examined variables that are associated with adolescent child molesters in comparison to both non-sex offending delinquents and to adolescents who committed sexual assaults against peers or adults. The variables were selected in part from past research that investigated the relevance of a contemporary theory of delinquency to sexual assaults by adolescents, and in part from past research that has shown that other variables might have theoretical significance for adolescent child molestation. Specifically, (1) age of the offender; (2) prior experience of childhood sexual abuse; (3) quality of bonds to parents, to schools and to peers; (4) friendship patterns; and (5) attitudes toward violence and violent sex were selected for study. Seventy-five adolescent offenders were recruited from agencies within the Greater Toronto Area that provide services for teenagers who are involved in offending behaviours. Offenders were assigned to three groups: (1) adolescents who committed sexual assaults against young children, (2) adolescents who committed sexual assaults against peers or adults, and (3) adolescents who committed non-sexual offences. Socio-demographic information was collected through structured interviews with the offenders. Subsequently, offenders completed a battery of questionnaires that assessed their quality of bonds to parents, to schools and to peers. They also completed a questionnaire on attitudes toward violence and violent sexual behaviour. Results showed that prior experience of childhood sexual abuse, differentiated adolescent child molesters from non-sex offending delinquents and from adolescent who committed sexual assaults against peers or adults. Adolescent child molesters also differed from adolescent who committed sexual assaults against peers or adults on age of the offender and number of opposite sex friends. Analyses showed all other variables (i.e., bonds to mothers; bonds to schools; bonds to peers; and attitudes toward violence and violent sex), separately, did not distinguish adolescent child molesters from non-sex offending delinquents and from adolescents who committed sexual assaults against peers or adults. Logistic regression analyses revealed that a constellation of variables (i.e., age of the offender, prior experience of childhood sexual abuse, bonds to schools, and number of opposite sex friends) reliably distinguished adolescent child molesters from non-sex offending delinquents. In addition, a subset of this constellation of variables (i.e., age of the offender, bonds to school, and number of opposite sex friends) reliably distinguished adolescent child molesters from adolescent who committed sexual assaults against peers or adults. Overall, the results support a multiple risk perspective in understanding adolescent child molestation. Implications of the findings, and recommendations for future research were discussed.