Antisociality, Sexual Deviance, and Sexual Reoffending in Juvenile Sex Offenders

• Author: Patrick Lussier,Kristie McCann
• Published date: 01 October 2008
• Date: 01 October 2008
• DOI: http://doi.org/10.1177/1541204008320260
• Published By: Sage Publications, Inc.
• Subject Matter: Articles

Youth Violence and
Juvenile Justice
Volume 6 Number 4
Antisociality, Sexual Deviance,
October 2008 363-385
© 2008 Sage Publications
10.1177/1541204008320260
and Sexual Reoffending in
http://yvj.sagepub.com
hosted at
Juvenile Sex Offenders
http://online.sagepub.com
A Meta-Analytical Investigation
Kristie McCann
Patrick Lussier
Simon Fraser University
A meta-analysis was conducted to analyze recidivism rates and assess the role of antisociality
and sexual deviancy in sexual reoffending in juvenile sex offenders. The importance of these
constructs is discussed with regard to current theory and in light of methodological and con-
ceptual limitations of previous studies.
Keywords:
sex offending; meta-analysis; risk factors; juveniles; youth; prediction
Introduction
There is great debate and controversy over the question of persistence in sexual offend-
ing. Some researchers have emphasized the propensity of sexual offenders to reoffend sex-
ually, whereas others have emphasized the fact that these offenders do not restrict
themselves to sexual crimes. What is now known as the generality/specialization debate
(Lussier, 2005; Simon, 1997) has been the subject of several empirical studies looking at
the recidivism pattern of sex offenders (Doren, 1998; Hanson, Morton, & Harris, 2003;
Lussier, 2005). These studies have shown that: (a) sex offenders are not more likely to reof-
fend on their release compared to nonsex offenders; (b) sex offenders who do recidivate are
more likely to commit a nonsexual crime; (c) sex offenders are more likely to recidivate in
a sexual crime than nonsex offenders; (d) most sex offenders will not reoffend sexually,
especially when looking at a short follow-up period, the recidivism rate being only about
10% over a 5-year follow-up period; and (e) sexual recidivism rates tend to increase over
time as a few studies based on serious offenders (i.e., federal inmates, aggressive/repetitive
sex offenders) have shown recidivism rates up to 30% over a 25-year follow-up period.
Based on these empirical observations, it is safe to conclude that there is evidence support-
ing both the generality and the specialization hypotheses regarding criminal activity of
This study was supported by a grant from Social Sciences and Humanities Research Canada and Simon Fraser
University. We wish to thank Dr. Brian Burtch, Dr. Raymond Corrado, and Dr. Jean Proulx for their comments
on an earlier version of this article. Please address correspondence to Ms. Kristie McCann, School of
Criminology, Simon Fraser University, 8888 University Drive, Burnaby (BC), Canada, V5A 1S6, Phone: (778)
782-7259, FAX: (778)782-4520, e-mail: kmccann55@gmail.com
363

---

364
Youth Violence and Juvenile Justice
sexual offenders (Lussier, LeBlanc, & Proulx, 2005). Most of these studies, however, have
been based on samples of adult sex offenders, making it difficult to determine the level of
persistence in juvenile sex offenders.
Theoretical Framework
Integrative models have usually relied on antisociality, sexual deviance or both to
explain sex crimes committed by youths. More than two decades have passed since Ageton
(1983) proposed that juvenile sex offenders are not that different from other juvenile
offenders. The similarities were such that the same causal processes might be in effect in
both types of offenders. Moffitt (1993) proposed a model of antisocial behavior where rape
has been described as being another manifestation of a life-course persistent antisocial syn-
drome. Some argue that this syndrome is the result of successive interactions between a
child with neuropsychological deficits and a criminogenic family environment character-
ized by abuse, neglect, and poor parenting skills. On the contrary, Burton (2003), applying
the principles of social learning theory, hypothesized that sexual offending is specifically
learned through exposure to deviant sexual models in early childhood. The reenactment of
the behavior to which the child has been exposed, followed by physiological, cognitive and
emotional reinforcements, may lead to persistence of the behavior. Knight and Sims-Knight
(2003) instead proposed a model in which early victimization and abuse can lead to sexual
aggression, through the development of psychopathic personality traits (i.e., impulsive, cal-
lous, and unemotional) and high sexualization (i.e., disinhibited sexuality). Psychopathic
traits and high sexualization, in turn may act as disinhibitors favoring the emergence of sex-
ually aggressive fantasies, an important precursor of sexually coercive behaviors.
Therefore, contrary to Moffitt, Burton and Knight and Sims-Knight recognized that juve-
nile sex offenders might be characterized by a specific developmental feature, that is, a ten-
dency to be preoccupied by sexuality and sexual thoughts.
Although empirical studies looking at the developmental antecedents of juvenile sex
offenders have shown a high prevalence of antisocial behavior, the presence of sexual
deviance has been somewhat elusive. For example, studies show that about 50% of samples
of juvenile offenders show a history of conduct problems or nonsex crimes (France &
Hudson, 1993; Seto & Lalumière, 2006). Yet, only a very small subgroup of delinquent
youths will go on to commit sex crimes in adolescence (LeBlanc & Fréchette, 1989;
Zimring, Piquero, & Jennings, 2007). This might be explained by the fact that the risk of
committing a sex crime in adolescence increases with the general frequency of offending.
Indeed, two birth cohorts show that chronic offenders (i.e., those with a minimum of five
arrests), representing about 7% of the sample, were responsible for more than 70% of all
arrests for rape in adolescence (Tracy, Wolfgang, & Figlio, 1990). More recently, a meta-
analytic review by Seto & Lalumière (2006) showed that, generally speaking, juvenile sex
offenders tend to present lower levels of antisocial behavior than juvenile nonsex offend-
ers. A moderator effect might be in effect here, as those youths having offended against
child victims showed lower levels of antisociality compared to those having offended
against peers or adults. On the other hand, their results also showed that sex offenders were

---

McCann, Lussier / Predictors of Recidivism in Juvenile Sex Offenders
365
more likely to be characterized by a less conventional sexual behavior (e.g., early age of
onset, fewer sexual partners) and more atypical sexual interests (e.g., presence of deviant
sexual interest) compared to juvenile nonsex offenders. Cortoni and Marshall (2001) sug-
gested that juvenile sex offenders may be more inclined to be sexually preoccupied because
of a tendency to use sexuality as an inappropriate way of coping with stressful situations
and negative emotional states. Although these results are interesting in understanding par-
ticipation in sex crime, they do not necessarily apply to persistence in sexual offending.
This is especially important considering that the majority of delinquent youths do not
become persistent offenders.
Persistence in Sexual Offending
Few investigators have looked at the base rate of reoffending in juvenile sex offenders
(e.g., Alexander, 1999; Beckett, 1999; Caldwell, 2002; Weinrott, 1996; Worling &
Långström, 2006; Zimring, 2004). Based on these few investigations, it is possible to draw
some tentative conclusions about base rates. First, there are some disagreements as to the
similarities/differences between the rates of sexual reoffending between juveniles and
adults. Indeed, Caldwell suggested that recidivism rates of juvenile offenders were lower
than those reported for the adult offenders. This is in keeping with Weinrott’s earlier obser-
vation that most juvenile sex offenders do not reoffend, at least not sexually. More
recently—and contrary to Caldwell’s conclusion—Worling & Långström found that the
sexual reoffending rate was about 15% using a new charge as a criterion to define recidi-
vism, and about 14% when using more conservative criteria such as convictions and court
records. Second, Caldwell pointed out that, on average, sexual crimes accounted for about
25% of the detected recidivism, suggesting that persistent juvenile sex offenders are also
generalist in their offending. Similarly, Worling & Långström found that the average gen-
eral recidivism rate (i.e., any crime type) for juvenile offenders was about 42%, or about 3
times higher than reported for sexual recidivism (see also Weinrott, 1996; Zimring, 2004).
Third, based on the previous reviews, it appears that there is much heterogeneity across
studies as to the general and the sexual recidivism rates of juvenile offenders. For example,
Caldwell found that if the general recidivism rates vary between 2.8% and 90% across stud-
ies, it varies between 1.8% and 19.6% for sexual recidivism (see also Weinrott, 1996;
Worling & Långström, 2006). This heterogeneity appears to be related to methodological
considerations such as (a) the definition of the reoffending; (b) the length of the follow-up
period; and (c) the type of sample used (treatment studies versus nontreatment studies).
Nontreatment studies with longer follow-up periods based on charges (rather than convic-
tion) appear to show higher recidivism rates.
There are few reviews of the risk factors of persistence of sexual offending in juvenile
offenders, and most pertinent studies have been qualitative in nature (e.g., Caldwell, 2002;
Carpentier, Proulx & Lussier, 2005; Zimring, 2004). Worling and Långström (2006) recently
reviewed the empirical and...