«The Assessment of Female Sexual Offenders Franca Cortoni Université de Montréal & Theresa A. Gannon University of Kent Cortoni, F., & Gannon, T.A. ...»
The Assessment of Female Sexual Offenders
Université de Montréal
Theresa A. Gannon
University of Kent
Cortoni, F., & Gannon, T.A. (in press). The assessment of female sexual offenders. In L. Craig &
M. Rettenberger (Eds). Assessment of Sexual Offenders. Chichester, UK: WileyBlackwell.
The Assessment of Female Sexual Offenders
Just like with male sexual offenders, the assessment of women who have committed
sexual offences is predominantly driven by the need to establish the likelihood of future sexual offending behaviour, identify problematic issues related to their offending, and outline interventions that would reduce their risk of recidivism. Women are also subjected to the same sanctions as males in the criminal justice system, including social control policies (e.g., Sexually Violent Predator laws in the U.S.). As such, it is crucial that their assessment of risk and treatment needs be based on empirically validated approaches. Due to the dearth of information on female sexual offenders’ risk of sexual recidivism and related treatment needs, the assessment of these women has traditionally been conducted using male-based risk assessment procedures.
Basically, the idea was that (1) crime is neutral, and (2) male-based tools are better than nothing.
In more recent years, however, these two premises have been refuted. First, research has now established that gender matters in criminal behaviour. In other words, there exist gender-specific issues that need to be taken into account when assessing women. Second, research shows that female sexual offenders differ greatly from their male counterparts in terms of recidivism rates.
Hence, risk assessment tools validated for males would over-evaluate risk of recidivism in females. Because of these issues, the assessment of risk and treatment needs of female sexual offenders remains a difficult endeavour. The low prevalence of female sexual offending, and their low rates of recidivism, renders difficult the development of women-specific empiricallyvalidated risk assessment tools and practices. Despite these difficulties, there is now a growing empirical foundation from which evaluators can draw to improve the validity of their assessment of female sexual offenders. This chapter reviews this empirical foundation and provides guidelines for the evaluation of female sexual offenders’ risk and treatment needs.
Assessing risk of recidivism Professionals tasked with the evaluation of risk of recidivism of female sexual offenders should be thoroughly familiar with the general principles of assessment for risk of recidivism. A brief review of these principles may be useful here to place into context the difficulties that currently underlie the assessment of risk of sexual recidivism in women. Within the context of criminal justice, the assessment of risk of recidivism is a process that evaluates and attempts to limit the probability that a new crime will reoccur. Integral to the risk assessment are the determination of the type of event being predicted (general offense; criminal violent behaviour;
sexual recidivism); its likelihood of reoccurrence (low; moderate; high); the conditions under which it may occur (dynamic risk factors and related situational factors – e.g., ongoing relationship with co-offender); whether these conditions are present (e.g., have these dynamic risk factors been resolved?); and which interventions (therapeutic or otherwise) might prevent these conditions from occurring. Consequently, a comprehensive assessment of risk of recidivism contains many pieces of information that, when integrated together, provide a full portrait of the individual that informs decision-makers and case managers on the case (see Douglas, Blanchard, & Hendry, 2013 for a review of these steps).
The first step in risk assessment is therefore to identify the likelihood that once an individual has been detected and sanctioned for a criminal act, he or she will do it again.
Fundamental to this task is empirical knowledge about (a) base rates of recidivism and (b) static and dynamic risk factors. Base rates are the proportion of the population that demonstrates the phenomenon of interest. In our case, this means the proportion of female sexual offenders who reoffend with a new sexual crime. Risk factors are individual characteristics of the offender that increase or decrease the probability of recidivism. Static risk factors are aspects in the offender’s history that cannot be changed through an intervention. Dynamic risk factors are those aspects of the offender that are amenable to change. Those are the issues addressed in therapeutic and related interventions (e.g., education) designed to reduce and manage the offender’s risk of recidivism.
Recidivism rates of female sexual offenders Research shows that the base rates of sexual recidivism of female sexual offenders are very low. Cortoni, Hanson and Coache (2010) conducted a meta-analysis of 10 studies examining the recidivism rates of female sexual offenders. The number of women in the sample was 2,490, and the average follow-up time was 6.5 years. Cumulative sexual, violent and any recidivism were examined separately. The results showed recidivism rates of 20% for any new type of recidivism and 6% for new violent (including sexual) offences. The rate of recidivism for new sexual offenses, however, was 1.5%.
Since the Cortoni et al. (2010) meta-analysis, another large scale study has calculated the recidivism rates of female sexual offenders. Wijkman and Bijleveld (2013) examined the recidivism rates of all adult females (N = 261) over age 18 years convicted for at least one contact sexual offense in the Netherlands between 1993 and 2011. The average follow-up time was 13.2 years. The sexual recidivism rate of 1.1% and the violent recidivism rate of 7.3% are consistent with meta-analytical findings. The rate for any recidivism, however, was slightly higher: 27.6% of the women had committed a new crime.
Besides understanding differences between recidivism and lifetime offending patters, recent findings indicate that attention must also be paid to the varying patterns of recidivism among various subgroups of women that all bear the label of sexual offenders. The accumulating evidence indicates that female sexual offenders represent a diverse group of individuals with differing motivational and offending patterns, as well as differing recidivism rates (e.g., Gannon, Rose, & Ward, 2008; Gannon, Waugh, Taylor, Blanchette, O’Connor, Blake, & Ó Ciardha, 2013; Mathews, Matthews, & Speltz, 1989; Sandler & Freeman, 2007; Vandiver & Kercher, 2004). For example, in Vandiver and Kercher’s (2004) category of homosexual criminals, none of the women had a conviction for a contact sexual offense. Instead, they had all been convicted of such offenses as indecency or compelling the victim into prostitution. Sandler and Freeman (2009) demonstrated that this subgroup of women have very different rates of sexual recidivism compared to traditional contact offenders. Their study of 1,466 female sexual offenders in New York State included a subgroup of 79 women that only had promoting prostitution of a minor as a sexual offense. While the traditional (i.e., contact / child pornography) offenders had a 1.59% rate of rearrest for new sexual offences (22 out of 1,387), the prostitution-only group had much higher rates of rearrest for new sexual offences: 10 out of the 79 women (12.66%) were rearrested for new prostitution-related offences. Further, Cortoni, Sandler, and Freeman (in press) demonstrated that women with only prostitution-related offenses have criminal histories more consistent with general criminality and exhibit more general antisocial features than women convicted of traditional sexual offenses (e.g., rape, sexual assault). These results suggest that evaluators should distinguish between women with traditional sexual offenses from those who only commit prostitution-related offences; the latter group presents very different criminogenic features and recidivism rates than the former.
Static risk factors Broadly speaking, it could be argued that they are two classes of static risk factors that differentially predict recidivism: those related to general recidivism and those related to sexual recidivism. For example, among male sexual offenders, static risk factors for general and violent (non-sexual) recidivism include being at a younger age, being single, and having a history of lifestyle instability, rule violations, and prior criminal history (Andrews & Bonta, 2010; Hanson & Morton-Bourgon, 2005). Static factors specifically related to sexual recidivism include prior sexual offences, and having male, stranger, and/or unrelated victims (Hanson & Thornton, 2000).
Among female sexual offenders, the evidence indicates the same general distinction of factors related to different types of recidivism.
There is sufficient empirical evidence to suggest that the static risk factors for general recidivism in women are similar to those of males. Specifically, the number of prior convictions for any type of offence (misdemeanors, drugs, violence) was related to non-sexual general or violent recidivism (Wijkman & Bijleveld, 2013; Sandler & Freeman, 2009; Vandiver, 2007). As well, a younger age (less than 30) was related to non-sexual recidivism. The finding that prior criminal history is related to future non-sexual recidivism among female sexual offenders is not surprising. It is indicative of an antisocial orientation and is common to all types of offenders, whether males or females (Andrews & Bonta, 2010; Blanchette & Brown, 2006).
The evidence, however, is quite different when sexual recidivism is examined. Although she had a large sample (N = 471) and a high base rate of sexual offending among women (11%), Vandiver (2007) could not establish any static factor specifically related to the commission of a new sexual offence. Sandler and Freeman (2009) did find a relationship between age and sexual recidivism. In contrast to findings for males, however, being older was linearly related to sexual recidivism among females, but only for those women convicted of prostitution-related offenses.
For females convicted of contact sexual offences, age was not related to sexual recidivism.
Finally, despite the low base rates of sexual recidivism, Sandler and Freeman (2009) found that the presence of a prior child abuse offence of any type was specifically and only related to sexual recidivism. This finding provides the first evidence that static risk factors related to sexual recidivism among women are gender-specific: research has never identified general patterns of child abuse to be related to sexual recidivism among male sexual offenders.
The significance of this factor is as of yet unclear. Perhaps because women tend to be the primary caregivers, they are more likely than men to come to the attention of the criminal justice system for non-sexual abuse of children. Alternatively, it may be that the sexual abuse of children, for these women, is part of a broader generalized pattern of abuse against children.
Further research is needed.
Dynamic risk factors The dynamic risk factors related to sexual recidivism in women are unknown. This is not surprising given the extremely low base rates of sexual recidivism in this group. Here again, evaluators need to be aware that a simple transfer of knowledge from the male sexual offender literature to females is simply not appropriate. The accumulating evidence indicates that while women appear to share some characteristics with men, these characteristics manifest themselves in different ways for women (Cortoni and Gannon, 2011; in press). Further, given their very low base rates of sexual recidivism, it is currently impossible to determine what might be the relationship between these characteristics and a woman’s likelihood that she will commit a new sexual offense. For example, while a woman may present with cognitions that support sex with her victim, there is no evidence to date to suggest that these cognitions augment her risk to commit a sexual offense against a new victim.
On the basis of available evidence, denial or minimization of the offending behaviour, distorted cognitions about the sexual offending and sexual abuse in general, problematic relationship and intimacy deficits, and the use of sex to fulfil intimacy needs have all been found in women (Gannon et al., 2008; Nathan & Ward, 2002; Saradjian & Hanks, 1996; see our later discussion of treatment needs). Sexual gratification, a desire for intimacy (with either a victim or a co-defendant), or instrumental goals such as revenge or humiliation are also associated with female sexual offending (Gannon et al., 2008; Saradjian & Hanks, 1996). An examination of these issues will inform on the woman’s personal circumstances and elements that have likely contributed to the offending behaviour, including whether a co-offender played a role (Cortoni, 2010). It is reiterated though that while these areas may be problematic and in need of intervention, their relationship with sexual recidivism among women is unknown.
Given the similarities in static risk factors for general recidivism among sexual and nonsexual female offenders, Cortoni (2013) hypothesized that both groups of women share the same dynamic risk factors but only for violent (non-sexual) and general recidivism. These dynamic risk factors likely include cognitions supportive of criminal behaviour; relationships with antisocial associates; poor familial relationships and general community functioning; and substance abuse problems (see Blanchette & Brown, 2006 for an in-depth review of this research). It is noted here that some of these factors manifest themselves in gender-specific ways.
Benda (2005) found that family factors such as prosocial family support and the presence of children is a predictor of positive community reintegration for female but not male offenders.
Andrews, Guzzo, Raynor, Rowe, Rettinger, Brews, and Wormith (2012) found a stronger relationship between substance abuse and recidivism for women than for men. Among female sexual offenders, Wijkman and Bijleveld (2013) found that substance abuse and antisocial personality predicted general and violent recidivism.