• Editor’s Comment
 • Jan Hindman
 • STABLE-2007 & ACUTE-2007: Improving the Assessment of Dynamic Risk Potential
 • ATSA and Public Policy: Many Steps Forward
 • Clinical and Theoretical Notes on the Change Process for Sexual Offenders
 • Juvenile sexual offenders: Comparison of victim age based subgroups and prediction of treatment outcome and recidivism
 • Board of Directors Election Results
 • Student Committee Report
 • Research Committee Report
 • New ATSA Members
 • Advertisement: Understand Offending
Newsletter Archives

Printer-Friendly Version

Contact the editor or
submit articles to:

Robin J. Wilson
Forum Editor
The GEO Group, Inc.
Florida Civil Commitment Center
13613 SE Hwy 70
Arcadia, FL 34266
Phone: (863) 491-6129
Email: dr.wilsonrj@verizon.net

Association for the Treatment of Sexual Abusers
4900 S.W. Griffith Drive
Suite 274
Beaverton, OR 97005

Phone: 503.643.1023
Email: atsa@atsa.com
Web: www.atsa.com

Vol. XIX, No. 4
Fall 2007
Juvenile sexual offenders: Comparison of victim age based subgroups and prediction of treatment outcome and recidivism

Therese Skubic Kemper, Ph.D.

     Six years ago, as a third year graduate student, I began a clinical practicum at the Florida State University Specialized Treatment Program (STP), which provides services to youth at Arthur G. Dozier School for Boys (AGDS), a high security juvenile justice facility. During this placement, I was to provide individual and group therapy to adjudicated juveniles, many of whom had committed sexual offenses. Although this was a new area for me, I already had limited but successful experience treating individuals with serious mental illness, such as depression, bulimia, and personality disorders. It seemed to me that once I adjusted to the anxiety of a new job and the idea of working with incarcerated kids, treating them (I thought) would be no different. Essentially, point me towards the latest research, I’ll read about the empirically supported treatments, and with some expert guidance from my supervisors and a treatment manual, I’d be ready to forge ahead. 

     The good news, my supervisor shared, was that research was generally showing a positive effect of treatment. The bad news was that whether treatment worked for juveniles was far less conclusive. More bad news was that it was not clear exactly what type of treatment worked, how it worked, for whom it worked, or what specifically should be treated. My readiness to forge ahead was considerably lessened, but I was also curious given that treatment for sexual offenders was already happening in the facility. My thinking was, if there was not a well-established treatment for juvenile sexual offenders, what were they doing here?

     That was the beginning of my interest in the assessment and treatment of juvenile sexual offenders. It was clear at this facility, as it is in treatment programs across the country, that treatment providers cannot sit and wait for a “gold standard” of treatment to be available. Children and families are in dire need of our assistance and victims need protection. Judges are mandating youth to complete sexual offender specific treatment and legislation previously reserved for adult sexual offenders, such as community notification sexual offender registration and civil commitment, is being widely applied to juveniles. As a result, critical decisions about a youth’s release from incarceration, placement with family, or need for further supervision are being made based on judgment of treatment outcomes and assessment of recidivism risk, which are imperfect, at best, for juveniles.

     As a relative newcomer to the treatment and research of juvenile sexual offenders, these realizations about the high stakes of working with these youth are both distressing and exciting. The distress surfaces when I cannot provide families, youth, or judges with more definitive information about the role of treatment or future risk. On the other hand, because research and treatment of juvenile sexual offenders is relatively new compared to many other areas of psychological research, we have the unique opportunity to build knowledge about juvenile sexual offenders from the ground up. Though it is a tedious task, the work currently being done with these youth will provide the foundation for the development of empirically validated and cost effective treatment and risk prediction, which I find to be enormously exciting. In fact, we are well on our way. Treatment, such as multisystemic therapy (Borduin, Henggeler, Blaske, & Stein, 1990; Letourneau, Schoenwald, & Sheidow, 2004), and risk assessment measures, such as the J-SOAP (Righthand, Prentky, et al., 2005) and ERASOR (Worling, 2004), are showing promise and should leave us optimistic about what lies ahead.

     In the mean time, when treatment demands are high and consequences for juvenile sexual offenders are serious we, as treatment providers, have to pull ourselves up by our bootstraps and do the best we can with the resources and knowledge currently available. One bootstrapping strategy used with juvenile sexual offenders has been the application of what is known about and practiced with adult sexual offenders. This has both practical and clinical utility and is certainly a reasonable launching point for work with juveniles. However, while this approach may be used as a temporary guide, it is critical that the assumptions about juveniles and their similarity to adult sexual offenders be examined. One of these assumptions—that there are juvenile equivalents of child molesters and rapists—is one focus of research being conducted at the Florida State University Specialized Treatment Program (STP). This is a particularly important assumption to examine because, with it, may come several other assumptions: that juveniles will exhibit stability in victim choice if they continue to sexually offend, that the groups will show different patterns and levels of sexual deviance, and that recidivism patterns will differ. If application of these assumptions to juveniles is erroneous, treatment effectiveness and risk prediction will be greatly diminished.

     This issue is also important to examine because of the value a valid classification system will have in informing not only treatment and risk prediction, but possibly prevention. Specifically, a valid classification system for juvenile sexual offenders will identify different developmental correlates, specific offending patterns or symptomatic presentation, treatment needs, and outcomes for each specific subgroup. If these factors are identified for particular subgroups of juvenile sexual offenders, children at risk may be identified early and provided with preventative measures. Treatment may also be more cost effective because risk factors specific to the needs of individuals in each subgroup could be targeted; and risk prediction could be improved so that the level of intervention and supervision matches the juvenile’s level of risk and dangerousness. 

     Whether victim age is as valid classification method for juvenile sexual offenders as it appears to be for adults remains an empirical question. The research thus far has been inconclusive as to whether juveniles who offend younger children (child offenders) or those who offend peers or adults (peer offenders) are truly distinct groups. Aside from differences in rates of sexual abuse and victim gender (Worling, 1995), consistent differences between these subgroups have not been found, even on treatment outcomes or recidivism rates (Kahn & Chambers, 1991; Nisbet, Wilson, & Smallbone, 2004; Parks & Bard, 2006; Vandiver, 2006). The research conducted by STP staff at AGDS compared subgroups of child and peer offenders on sexual and nonsexual offense history, abuse history, social skills, impulsivity, treatment outcomes, and recidivism. In addition, a group of mixed offenders—those who had a history of adjudicated sexual offenses against both children and peers—was included. This group was of particular interest because, with few exceptions (see Parks & Bard, 2006), they are often combined with child offenders, are classified as either child or peer offenders based on their most recent offense, or are excluded from research samples. It is possible that this is a distinct group and that combining them with either child or peer offenders clouds true differences between those subgroups.

Our Research  
     Our sample consisted of 198 child offenders (victims were age 12 or younger and 4 or more years younger than the offender), 77 peer offenders, and 21 mixed offenders. Some differences between child and peer offenders were found, as expected, on measures of sexual abuse and victim characteristics. However, the most noteworthy findings were for the mixed group. Admittedly, the mixed group is quite small and this likely constrains the interpretation of the results. But, the mixed group is possibly a severe group and at the very least, worthy of further study.

     Mixed Offenders.  
     Two adjudications (against a child and peer) were necessary for classification as a mixed offender; however, many exceeded this criterion—48% of mixed offenders had three or more adjudicated offenses, compared to 9% of child and 4% of peer offenders. The mixed offenders had a more extensive sexual offense history and a more varied pattern of victim preference, which extended beyond victim age, than other juvenile sexual offenders. In addition, compared to other sexual offenders with two or more sexual offense adjudications, the mixed offenders had more victims than both groups and more sexual offense adjudications than peer offenders. The mixed offenders were also characterized by high rates of abuse, with nearly two-thirds having been abused in some manner and half of them abused sexually.

     The mixed offenders were also noteworthy in terms of the details of their offending history, in that they had a varied and intrusive pattern of offending. Specifically, the mixed offenders were more varied in victim gender and relationship, and appeared to exhibit less discretion in their victim choice than other subgroups. They were the group most likely to victimize both boys and girls and the group most likely to victimize both in and out of family settings. They were also the most likely to abuse multiple victims at the same time, with one-third of the mixed offenders doing so, compared to 10% of child and 5% of peer offenders. Mixed offenders engaged in penile penetration of the victims more often than child offenders and engaged in oral sex more often than peer offenders. Finally, the mixed offenders had the greatest variety of sexual acts in their offense history, suggesting that they engaged in different types of sexual acts (e.g., fondling, penetration, oral sex) with different victims. The mixed offenders did not differ from the other sexual offenders on criminal history variables, including age at first arrest or number of nonsexual charges. It therefore appears that their versatility in offending sets them apart from other sexual offenders only in terms of sexual, and not nonsexual, crime.

     Child Offenders.  
     The child offenders were the largest subgroup, accounting for two-thirds of the sample. Half of the child offenders were victims of abuse and, along with the mixed offenders, experienced higher rates of sexual abuse than did the peer offenders. Over 25% had multiple sexual offense adjudications and they were charged with offenses against an average 1.5 victims. These youth tended to victimize females and relatives; however, nearly one-third had an adjudicated offense against a male. Child offenders were the least likely to engage in penile penetration of the victims, but engaged in oral sex more often than the peer offenders. As a group the child offenders were just over 13 years old at first arrest and were among those with the fewest nonsexual charges in their history.

     Peer Offenders.  
     Peer offenders represented the second largest subgroup, comprising one-quarter of the sample. Over one-third of peer offenders were victims of abuse and, although their rates of sexual abuse were higher than what is typical in the general population, they were lower than the other sexual offenders. Similar to child offenders, they averaged fewer than two victims but nearly 30% had multiple sexual offense adjudications. The overwhelming majority victimized females and they were the group most likely to victimize non-family members and to offend in a public place. Peer offenders were more likely than child offenders to engage in penile penetration of victims, but were least likely to engage in oral sex. Finally, this group was notable in terms of their nonsexual offense history. As a group peer offenders were likely to have been first arrested before age 13 and had an average of over seven nonsexual charges, which was nearly one and a half times the nonsexual charges of the other subgroups and significantly more than the child offenders.

     Involvement in Treatment.  
     The vast majority (88%) of the sexual offenders in this study successfully completed the one year group treatment program at the facility. The mixed offenders had the highest proportion of treatment failures (29%), due to serious conduct problems within the facility or unsatisfactory progress in treatment after multiple attempts in the treatment program. Interestingly, treatment performance was not related to either sexual or nonsexual recidivism after a follow up period of approximately five years. Prior residents of the facility were tracked using a legal research database, Westlaw, which allowed for a nation wide search of adult criminal records. Rates of nonsexual convictions were high (40%) and rates of sexual convictions were relatively low (6%). Both of these rates are consistent with other published data on juveniles (Reitzel & Carbonell, 2006) and with research showing higher nonsexual than sexual recidivism rates in youthful sexual offenders (Freeman, 2007; Reitzel & Carbonell, 2006).

     Groups in this study did not differ in their rates of nonsexual recidivism, which was somewhat surprising given that peer offenders had more extensive nonsexual criminal histories than child offenders and adult rapists generally show higher nonsexual (violent, general) recidivism than child molesters (Freeman, 2007; Hanson & Bussière, 1998). Also somewhat unexpectedly, groups did not differ on rates of sexual recidivism. Despite the mixed offenders’ extensive sexual offense history, they did not have unusually high rates of sexual recidivism. In fact, the child offenders accounted for 16 of the 18 sexual recidivists, with one sexual recidivist coming from each of the other groups. Though this difference was not statistically significant, it is clinically interesting. Child offenders were clearly overrepresented in the group of sexual recidivists, as they accounted for 89% of the sexual recidivists despite accounting for only 67% of the sample.

     Overall, this research offers some support for the validity and clinical utility of classifying juvenile sexual offenders on the basis of victim age. The mixed offenders appeared somewhat more severe in their sexual offense history and did more poorly in treatment than other groups. The child and peer offenders showed differences in sexual abuse history, victim choice, and nonsexual crime, similar to differences found between adult sexual offenders. The results are moderately consistent with the adult and juvenile literature and hint at further avenues for research. For example, mixed offenders may be more sexually deviant than other groups, which could result in extensive future sexual offending and resistance to treatment. Interventions aimed specifically at reducing sexual deviance may be indicated for this group. This research suggests that peer offenders may be similar to adult rapists in their propensity for nonsexual crime and, therefore, treatment targeting general criminal behavior may be most beneficial for them. Finally, child offenders may be at a higher risk for sexual recidivism and future research might focus on whether these offenders are more likely than others to continue victimizing others as they enter adulthood.

     In contrast to, and despite these important differences, some similarities between the groups make these subgroups appear somewhat arbitrary and suggest that victim age-based subgroups of juveniles may not be as distinct as are adult rapists and child molesters. No differences were found on measures of physical abuse, social skills, or impulsivity, and subgroups exhibited some commonalities in their patterns of sexual offending and recidivism. Furthermore, there are inherent difficulties in determining who constitutes a juvenile version of a child molester when, in reality, the offenders themselves are children. As a result, we must be cautious when making downward extensions of adult treatment methods and policies to juveniles. Our research findings also make it clear that detailed examination of a wide range of developmental factors, sexual deviance, criminal history, and long-term offending patterns will be necessary before we can validate juvenile sexual offender subgroups and identify their specific treatment needs.

Closing Thoughts  
     Six years ago, I wanted answers about who to treat, what to treat, and how to treat juvenile sexual offenders. Although I was surprised to find that starting up in this area would not be as straightforward as familiarizing myself with the latest treatment manual, I have been rewarded with the opportunity to observe the rapid increase in knowledge about juvenile sexual offenders. Recently, I have been fortunate to have the support of supervisors and the AGDS administration to integrate research and clinical work in my dealings with youth. This type of combination of research and practice maximizes our potential to improve the services we provide, while protecting victims and making a meaningful contribution to the knowledge of juvenile sexual offenders. It is my hope and expectation that our efforts in this line of work will result in fair and effective practices for juveniles who sexually offend.

*This research was funded partially through a Florida State University Dissertation Award and the Association for the Treatment of Sexual Abusers’ Falconer Grant for graduate student research.


Borduin, C. M., Henggeler, S. W., Blaske, D. M., & Stein, R. J. (1990). Multisystemic treatment of adolescent sexual offenders. International Journal of Offender Therapy and Comparative Criminology, 34, 105-113.


Freeman, N. (2007). Predictors of rearrest for rapists and child molesters on probation. Criminal
Justice and Behavior
, 34, 752-768.


Hanson, R. & Bussière, M. (1998). Predicting relapse: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348-362.


Kahn, T. & Chambers, H. (1991). Assessing reoffense risk with juvenile sexual offenders. Child Welfare, 70, 333-345.


Letourneau, E. J., Schoenwald, S. K., & Sheidow, A. J. (2004). Children and adolescents with sexual behavior problems. Child Maltreatment, 9, 49-61.


Nisbet, I., Wilson, P. H., & Smallbone, S. W. (2004). A prospective longitudinal study of sexual recidivism among adolescent sex offenders. Sexual Abuse: Journal of Research and Treatment, 16, 223-234.


Parks, G. A. & Bard, D. E. (2006). Risk factors for adolescent sex offender recidivism: Evaluation of predictive factors and comparison of three groups based upon victim type. Sexual Abuse: A Journal of Research and Treatment, 18, 319-342.


Reitzel, L. R. & Carbonell, J. L. (2006). The effectiveness of sex offender treatment for juveniles as measured by recidivism: A meta-analysis. Sexual Abuse: A Journal of Research and Treatment, 18, 401-421.


Righthand, S., Prentky, R., Knight, R., Carpenter, E., Hecker, J.E., & Nangle, D. (2005). Factor structure and validation of the Juvenile Sex Offender Assessment Protocol (J-SOAP). Sexual Abuse: A Journal of Research and Treatment, 17, 13-30.


Vandiver, D. M. (2006). A prospective analysis of juvenile male sex offenders. Characteristics and recidivism rates as adults. Journal of Interpersonal Violence, 21, 673-688.


Worling, J. (1995). Sexual abuse histories of adolescent male sex offenders: Differences on the basis of the age and gender of their victims. Journal of Abnormal Psychology, 104, 610-613.


Worling, J. R. (2004). The Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR): Preliminary psychometric data. Sexual Abuse: A Journal of Research and Treatment, 16, 235-254.

[Back to Top]