• The article is here in PDF format: seto_cs_2010-2011.pdf

[Introduction]

There has been increasing attention to the problem of online sexual offending, particularly the use of Internet and related digital technologies to obtain, distribute, or
produce child pornography, or to contact potential child victims to create opportunities for sexual offending (e.g., attempting to arrange a meeting with a minor met online, for sexual purposes).

There has been a particular focus on child pornography offending, which involves child exploitation and can increase the incidence of contact sexual offending by increasing demand for new content and thereby increasing the production of such images.

Though they continue to represent only a small proportion of total child exploitation crimes, the number of arrests for online sexual offenses has increased greatly in the past 10 years

• (Bates & Metcalf, 2007; Motivans & Kyckelhan, 2007; Wolak, Finkelhor, & Mitchell, 2005, 2009).

There is specific public and professional concern about the likelihood that online
offenders also commit contact sexual offenses offline (e.g., Lam, Mitchell, & Seto,
2010).

There are two forms of this question:

• (a) What is the likelihood that an online offender has a history of offline sexual offending?
• (b) What is the likelihood that an online offender will go on to commit an offline sexual offense in the future?

In this article, we report meta-analyses addressing each of these questions. The results of these two quantitative reviews are relevant for risk assessment and management
because, for the first question, a high degree of overlap suggests risk assessment measures developed with offline sexual offenders are also likely to be valid.

This is clearly so for online offenders who already have a known contact sexual offense history, and is likely to be true for online sexual offenders if they are similar to offline sexual offenders in crime-related characteristics (see Babchishin, Hanson, & Hermann, in press).

For the second question, information about the recidivism rates of online offenders can help guide policy and practice decisions; for example, high recidivism rates might suggest that more intensive (and expensive) responses are warranted, whereas low recidivism rates suggest that other, higher risk populations of sexual offenders are a higher priority for law enforcement and other social responses.

[... ... ... ...]

Discussion

In our first meta-analysis, we found that approximately 1 in 8 online offenders have a known contact sexual offense history at the time of their index offense, based on official records of arrests, charges, or convictions. The prevalence was higher when self-report information was used, with approximately half of the online offenders admitting to a contact sexual offense, consistent with the observation that official records are a conservative estimate of actual offending (even if some of the selfreported offenses are false confessions and did not actually occur).

Although there is considerable overlap between online and offline offending, our results suggest there is a distinct group of online offenders whose only sexual crimes involve illegal (most often child) pornography or, less frequently, illegal solicitations of minors using the Internet. Knowing about criminal history, however, does not directly address the question of future risk to commit contact sexual offenses. After all, almost all of the sexual offenders followed in the studies reviewed by Hanson and his colleagues (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005) had committed contact sexual offenses, and the composite rate of detected sexual recidivism was 13% after an average of 5 to 6 years’ follow-up.

Longer-term follow-ups suggest approximately a third of sexual offenders with child victims will be detected for new sexual offenses after 20 to 30 years of opportunity (Hanson, Steffy, & Gauthier, 1993). Though other offenders in these follow-up studies will have committed offenses that were not officially detected, these data are not consistent with the idea that committing a contact sexual offense means that the offender will do it again.

Our second meta-analysis found that online offenders rarely go on to commit detected contact sexual offenses. During the follow-up period (up to 6 years), less than 5% of the online offenders were caught for a new sexual or violent offense. Two studies found no sexual recidivists.

The observed rates will increase with longer follow-up periods and not all new
offenses are detected. Nevertheless, these rates are substantially lower than the recidivism rates of typical groups of offline sexual offenders. It is quite possible, however, that some online sexual offenders have relatively high recidivism rates. Eke and Seto (2008) found that those online offenders who already had a history of offline offenders showed sexual recidivism rates higher than the expected base rates for typical sexual offenders (A. J. R. Harris & Hanson, 2004). In contrast, the online offenders who had no history of contact offenses almost never committed contact sexual offenses, despite a comparably high likelihood that they were sexually interested in children (Babchishin et al., in press).

[... ... ... ...]

Implications

Given that many online offenders are strongly aroused by child pornography (Seto et al., 2006), our results suggest that pedophilic interests do not necessarily result in contact sexual offenses against children. Many of the online offenders in our study are likely to be sexually interested in children, but only half are known to have acted on these sexual interests.

Those individuals who act on their pedophilic interests are likely to have personality traits and life circumstances that facilitate antisocial behavior and criminality (see Seto, 2008). Further research is needed to articulate the risk factors for sexual offenders who are neither pedophilic nor particular antisocial (e.g., a significant
portion of incest offenders).

The low recidivism rates of online offenders may be used by some readers to minimize the seriousness of the online crimes committed. We believe this would be a
mistake. Child pornography is a serious crime because it contributes to the sexual
exploitation of children by creating demand for content, it offends community standards and values, and it is viewed by many members of the public as a serious crime (Lam et al., 2010).

It would also be a mistake to fail to differentiate online offenders by the risk they pose. Although the research on risk factors is limited, we believe that the risk factors for online offenders are likely to be the same risk factors found for offline offenders (i.e., sexual deviancy, antisocial orientation, and intimacy deficits).

Until research suggests otherwise, we recommend that valid measures of these risk
factors should be used by the police, courts, correctional systems, and clinicians to
prioritize interventions for individuals involved in online sexual offenses.

Download: 

seto_cs_2010-2011.pdf