A Review of the Recidivism Rates of Adult Female Sexual Offenders

2005 N° R-169

Franca Cortoni

Correctional Service Canada

&

R. Karl Hanson

Public Safety and Emergency Preparedness Canada

Correctional Service of Canada

May, 2005

EXECUTIVE SUMMARY

The purpose of this review was to provide information on the prevalence of female sexual offending, and to establish an understanding of the reoffending rates of these women. To achieve this purpose, a systematic review of available information on female sexual offenders was conducted. There were two goals to this study: 1) to provide an estimate of the proportion of sexual offenders who are women; and 2) to establish the known sexual recidivism rates among female sexual offenders.

The proportion of sexual offenders who were women was estimated based on two general sources of information. The first source of information was the official police or Court reports that detail the gender of the offender. The second source of information was victimization surveys. Information was available for the following countries: Canada, UK, USA, Australia and New Zealand.

The review of rates of sexual recidivism of female sexual offenders was accomplished by conducting a thorough search of published and unpublished literature. Included were conference presentations, government reports, official recidivism data drawn from websites or through direct communication with government agencies, and reports of unpublished studies obtained directly from the researchers. Information was obtained from the U.S., England, Australia, and Canada.

In terms of prevalence, based on official records, the proportion of female sexual offenders ranged from 0.6% (New Zealand inmates) to 8.3% (non-rape sexual offences in the U.S.), with an unweighted average of 3.8% of all sexual offenders. In the victimization studies, the proportion of female sexual offenders ranged from 3.1% (New Zealand, 2001) to 7.0% (Australia, 2002), with an unweighted average of 4.8%. Official reports and results from victimization surveys were consistent with each other. In both cases, results show that women are responsible for between 4 % and 5% of all sexual offences.

The recidivism rate review included a combined sample of 380 female sexual offenders. A weighted average across studies resulted in an observed sexual recidivism rate for female sexual offenders of 1.0% (3/306). The rate of any violent recidivism (including sexual) was 6.3% (15/191) and the rate of any recidivism (including violent and sexual) was 20.2% (68/337). The average follow-up period was 5 years.

In comparison to information on male sexual offenders, a substantial difference in recidivism rates between male and female sexual offenders is noted. Based on meta-analyses of large samples of male sexual offenders, the 5 years recidivism rates would be expected to be 13%-14% for sexual crimes, 25% for any violent crime, and 36%-37% for any new crime (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2004). The differences between the recidivism rates for the male and female sexual offender were statistically significant for all types of recidivism (p < .001).

Using the U.S. data, the only country for which we could conduct direct comparisons between official rates and victimization reports, differential rates of police arrests according to the gender of the offender in relation to victimization survey results were noted. Specifically, 34% of the sexual offences committed by men resulted in police arrest, compared to 57% of the sexual offences committed by female offenders. These findings suggest that further research is needed to understand the personal and criminal justice responses to victimization by female offenders versus male sexual offenders.

Details on the female sexual recidivists were examined to determine if any patterns could be discerned among these women. Only the Canadian research in this review provided details on the 2 female sexual recidivists in their study. The one marker that differentiated these two women was that they had previously exclusively sexually assaulted their victims on their own, without a male accomplice. This particular finding is noteworthy, but requires validation in further research on female sexual recidivism.

Given the low sexual recidivism rates of women sexual offenders, it is suggested that risk assessments of these women be conducted using the risk factors normally found in general female offenders. These factors include antisocial attitudes and associates, substance abuse, problematic relationships, and emotional dyscontrol (Blanchette, 2001).

TABLE OF CONTENTS

• EXECUTIVE SUMMARY
• LIST OF TABLES
• INTRODUCTION
  • Aim of Study
• METHOD
  • Proportion of Sexual Offenders who are Female
    • Official police or Court reports
    • Victimization surveys
  • Recidivism Rates of Female Sexual Offenders
    • Published study
    • Ongoing study
    • Conference presentations
    • Official recidivism data
• RESULTS
  • Proportion of sexual offenders who are female
  • Recidivism Rates of Female Sexual Offenders
• DISCUSSION
  • Prevalence of Sexual Offending by Women
  • Recidivism Rates of Female Sexual Offenders
  • Implications
• REFERENCES

LIST OF TABLES

• Table 1: Proportion of sexual offenders who are female based on officially reported sexual crime.
• Table 2: Proportion of sexual offenders who are female based on victimization surveys.
• Table 3: Sexual and Any Recidivism Rates of Female and Male Sexual Offenders.

INTRODUCTION

That fact that women engage in sexually abusive behaviours has been established for many years, but it is only in recent years that concerted efforts have been made to study female sexual offenders (Eldridge &amp; Saradjian, 2000; Grayston &amp; De Luca, 1999; Mathews, Matthews &amp; Speltz, 1989; Nathan &amp; Ward, 2002; Saradjian, 1996). To date, there have been few empirical studies of the assessment and treatment needs of this unique population of offenders. In their review of 12 cases of female sexual offenders, Nathan and Ward (2002) comment that any comprehensive theory of female sexual offending should delineate distinct pathways to offending, knowledge of which could aid intervention efforts. Similarly, in their review of the existing literature on women sexual offenders, Grayston and De Luca (1999) note that only sparse data exist on the topic, and that clinicians have developed only tentative theories to guide their work. As women increasingly come to the attention of the criminal justice system for sexual offences, more systematic information is required. One of the most important pieces of information concerns the potential of female sexual offenders to commit new sexual offences. The purpose of the current study is to review the available research on the recidivism rates of female sexual offenders.

Criminal justice systems are primarily concerned with risk of reoffending and the type of interventions required to effectively manage that risk (Andrews & Bonta, 2003). Although there have been tremendous advances in the knowledge of risk assessment for adult male sexual offenders, little is known about risk assessment with female sex offenders. Detected and criminally sanctioned female sexual offenders typically require an assessment of risk of recidivism and of treatment need. Normally, a risk assessment focuses on the static and dynamic risk factors that predict sexual recidivism. Static factors, by their very nature, are unchangeable. Examples of static factors related to sexual recidivism in male offenders include a past history of sexual offending, male victims, stranger victims, and the presence of sexually deviant interest (Hanson & Thornton, 2000). Dynamic risk factors are those factors that are related to sexual recidivism but are changeable. Some examples of dynamic risk factors in male sexual offenders include intimacy deficits, attitudes supportive of sexual abuse, and sexual preoccupation (Hanson & Harris, 2001).

Little is known about the static and dynamic risk factors of women sexual offenders (Hedderman, 2004; Kemshall, 2004). To establish this knowledge, research must demonstrate that such factors are related to recidivism. Consequently, long-term follow-up studies of recidivism are required. In addition, this type of research required information about the recidivism rates of the population of interest. Again, to date, there has been little research that on the recidivism base rates of female sexual offenders. An understanding of these rates would be useful in guiding research development in this area.

To aid in interpreting the recidivism rates of female offenders, the first section of the report reviews the proportion of sexual offenders who are female. Although it is widely known that most sexual offenders are male, the prevalence rates of female sexual offending is uncertain and controversial (Denov, 2003). Some authors have doubted the very existence of females who commit sexually motivated offences. Freund, Heasman, Racansky, and Glancy (1984), for example, declared that ‘pedophilia...does not exist at all in women’ (p.193). The Diagnostic and Statistical Manual, 4th edition state that “except for sexual masochism...paraphilias are almost never diagnosed in females” (American Psychiatric Association, 1994, p. 524). Other authors, however, believe that sexual offending by females is common, but that female sex offenders are “unnoticed, unreported, or diverted from the criminal justice system” (Vandiver & Walker, 2002). Such assertions have typically been reported without reference to empirical evidence.

Aim of Study

The purpose of this study was to provide information on the prevalence of female sexual offending, and to establish an understanding of the reoffending rates of these women. To achieve this purpose, a systematic review of available information on female sexual offenders was conducted. There were two goals to this study:

• 1) To provide an estimate of the proportion of sexual offenders who are women; and
• 2) To establish the known sexual recidivism rates among female sexual offenders.

METHOD

Proportion of Sexual Offenders who are Female

The proportion of sexual offenders who were women was estimated based on two general sources of information. The first source of information was the official police or Court reports that detail the gender of the offender. Information was available for the following countries: Canada, UK, USA, Australia and New Zealand. The second source of information was victimization surveys. Details about these sources are provided below.

Official police or Court reports

Kong, Johnson, Beattie, and Cardillo (2003). This report provides information on the extent and nature of sexual offences in Canada based on data held by the Canadian Centre for Justice Statistics (Statistics Canada). The sexual offences included in this report are those reported to the police as documented in the Uniform Crime Reporting Survey, and Adult and Youth Court Surveys.

U.S. Department of Justice (2003a). The Uniform Crime Reports are produced by the Federal Bureau of Investigation of the U.S. Department of Justice. For the purpose of the current study, the relevant information concerned the number of males and females arrested for “Forcible Rape” and “Sex Offences (excluding Forcible Rape and Prostitution)” in 2002.

Home Office (2004). The report provides statistics of the number of offenders found guilty or cautioned in England and Wales between 1993 and 2003. The numbers used in the current study are for male and female offenders found guilty of sexual offences in 2003.

Australian Bureau of Statistics (2005). The Criminal Courts publication reports on all cases adjudicated by the Supreme, Intermediate, and Magistrates' Courts of Australia for the period July 1, 2003 to June 30, 2004. Statistics concerning gender were only available for offenders adjudicated at the Higher Courts for “sexual assault and related offences”.

New Zealand Department of Corrections (2003). This report describes the characteristics of offenders serving prison sentences November 15, 2001. The sexual offences are identified as “sexual violence” and “other sexual”.

Victimization surveys

Badgley (1984). This random population survey was conducted in February, 1983, with the aim of estimating the proportion of the Canadian population who had been victims of sexual abuse. Participants completed anonymous questionnaires in their own homes (1002 males, 1006 females). Sexual abuse was defined as unwanted sexual acts.

U.S. Department of Justice (2003). The National Crime Victimization Survey is an ongoing survey designed to assess criminal victimization. Questions concerning Forcible Rape and Sexual Assault were included among questions about other types of crimes (e.g., common assault, car theft). The data reported are the 2002 estimates based on a sample of 76,050 individuals from 42,000 households. Information concerning the gender of the offenders was drawn from Table 38 (single-offender victimizations) and Table 44 (multiple-offender victimizations). Because the number of offenders in multiple-offender victimizations was not specified, it was assumed that these offences involved two offenders.

Finkelhor, Hotaling, Lewis and Smith (1990). This national survey was designed to estimate the proportion of U.S. residents who were victim of sexual abuse. Phone interviews were conducted in July, 1985. Sexual abuse was assessed with four questions concerning unwanted sexual activity prior to age 18.

Australian Bureau of Statistics (2003). The Australian Crime and Safety Survey collects information from individuals and households about their experience of selected crimes and crime-related risk factors. The data presented in this report was from the April, 2002, survey of 41,200 individuals. Gender information for sexual offenders was only available for the female victims of sexual offences (1% of women).

Fergunnon, Lynskey, & Horwood (1996). This study reported on the sexual abuse experiences of a cohort of 1,019 individuals born in Christchurch, New Zealand in mid-1977. Children had been interviewed at annual intervals up to age 16, and again at age 18. Participants were asked whether, before the age of 16, anyone had attempted to involve them in a series of 15 sexual activities “when they did not want this to happen”.

Morrie, Reilly, Berry & Ransom (2003). The New Zealand National Survey of Crime Victims is a comprehensive survey of criminal victimization, first conducted in 1996. The data used in the current report was from the 2001 survey of 5,147 individuals. Sexual crimes were defined as sexual interference, sexual assault, or any unwanted sexual activity. Information concerning the gender of the sexual offenders was available for the 57 victimization that had occurred after January 1, 2000.

Recidivism Rates of Female Sexual Offenders

The review of rates of sexual recidivism of female sexual offenders was accomplished by conducting a thorough search of published and unpublished literature. Included were conference presentations, government reports, official recidivism data drawn from websites or through direct communication with government agencies, and reports of unpublished studies obtained directly from the researchers. This search yielded one published study (Broadhurst & Loh, 2003); one ongoing study (Hanson, Harris & Thomas, 2005); 2 conference presentations (Peterson, Colebank & Motta, 2001; Williams & Nicholaichuk, 2001); and 2 official sources of recidivism data (UK Home Office and Florida State, US).

Published study

Western Australia. Broadhurst and Loh (2003) examined the probability of re-arrest for sexual offenders in the state of Western Australia. All sexual offenders who were arrested between 1984 and 1994 were included in the study. The study included 43 female sexual offenders. Offence information was based on original police charges and coded from the Western Australia Police Service. Recidivism for the female sexual offenders was reported in Footnote 1 (p. 134) concerning re-arrest for a new sexual offense and re-arrest for a non-sexual violent offense. The average follow-up time for the total sample was 5.7 years. It is unknown whether offenders in the study had received offence-specific treatment.

Ongoing study

Dynamic Supervision Project (Hanson et al., 2005; Harris & Hanson, 2003). Offenders were selected from all provinces and territories of Canada as well as the states of Alaska and Iowa. The aim of this prospective study was to test the validity of a system of risk assessment for sexual offenders on community supervision (probation or parole). Assessments were conducted between 2001 and 2004, with recidivism information provided on an ongoing basis by the officers supervising the cases (up to March, 2005). Of the 987 offenders included in this project, 6 were female. The average follow-up period for the female sexual offenders was 2.5 years, and ranged from 2 months to 40 months.

Conference presentations

Canadian Federal Offenders. Williams and Nicholaichuk (2001) conducted a follow-up of 72 female sexual offenders who received Federal Sentences (2 years or more) in Canada between 1972 and 1998. Due to deportation or continued incarceration, recidivism data could only be obtained for 61 of the cases. Recidivism was coded from Royal Canadian Mounted Police records, a national database that contains all charges and convictions on every offender in Canada. The average follow-up period was 7.6 years, although it ranged from 10 months to 22 years.

State of Kentucky, U.S. Peterson et al. (2001) examined the recidivism rates of 115 American female sexual offenders. In this study, all the women had been or continued to be in treatment for their sexually offending behavior. The average follow-up was 5.5 years, and ranged from 2 months to 10 years. Recidivism was coded from official Kentucky Court records.

Official recidivism data

Home Office Statistics (1998-2003). In a series of reports to the UK Parliament, the Home Office provides information on the reconviction rates of offenders from England and Wales. Recidivism was defined as new convictions within 2 years of discharge from prison. The data used in this review covers the period from 1994 to 1999, during which a total of 81 female sexual offenders were released from prison. No information was provided, however, as to whether offenders received offence-specific intervention.

Florida State, U.S. In 2003, Holley and Ensley produced a recidivism report on inmates released from Florida prisons between 1995 and 2001. Seventy four female sexual offenders were included in the study. Recidivism was defined as conviction for a new felony offence during a 60 months follow-up period. Recidivism probabilities were calculated through survival analysis. There was no information about whether offenders received offence-specific interventions prior to release.

RESULTS

Proportion of sexual offenders who are female

Table 1 shows the proportion of sexual offenders based on police, court and prison information. The proportion of female sexual offenders ranged from 0.6% (New Zealand inmates) to 8.3% (non-rape sexual offences in the US), with an unweighted average of 3.8% of all sexual offenders.

Table 2 shows the proportion of sexual offenders who are women based on victimization surveys. In the victimisation studies, the proportion of female sexual offenders ranged from 3.1% (New Zealand, 2001) to 7.0% (Australia, 2002), with an unweighted average of 4.8%.

It is noted that both official reported information and information obtained via victimization surveys were consistent with each other. In both cases, results show that women are responsible for between 4 % and 5% of all sexual offences.

Recidivism Rates of Female Sexual Offenders

Table 3 shows the recidivism rates of the female sexual offenders included in this review. A weighted average across studies resulted in an observed sexual recidivism rate for female sexual offenders of 1.0% (3/306). The rate of any violent recidivism (including sexual) was 6.3% (15/191) and the rate of any recidivism (including violent and sexual) was 20.2% (68/337). The average follow-up period was 5 years.

In comparison to information on male sexual offenders, a substantial difference in recidivism rates between male and female sexual offenders is noted. Based on meta-analyses of large samples of male sexual offenders, the 5 years recidivism rates would be expected to be 13%-14% for sexual crimes, 25% for any violent crime, and 36%-37% for any new crime (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2004). The differences between the recidivism rates for the male and female sexual offender were statistically significant for all types of recidivism (p < .001).

Table 1: Proportion of sexual offenders who are female based on officially reported sexual crime.

Table 2: Proportion of sexual offenders who are female based on victimization surveys

Table 3: Sexual and Any Recidivism Rates of Female and Male Sexual Offenders

DISCUSSION

Prevalence of Sexual Offending by Women

It is clear that sexual offending is much more common among males than females. Compared to male offenders, results from this review indicate that women commit sexual offences at a ratio of approximately 1 to 20 based on both official reports and victimization surveys. In our review, the comparison between official reports of female sexual offending and reports from victimization surveys showed remarkably similar proportions of female sexual offenders.

Previous literature reviews have suggested that sexual offending by females is not necessarily rare, but it is under reported (Denov & Cortoni, in press). The current review suggests victims of male sexual offenders and victims of female sexual offenders report their experience to the police at similar rates. If there is a differential underreporting of female versus male sexual offenders, then this effect must operate at the level of individuals failing to identify the behaviour of female sexual offenders as a crime (the victimization survey level).

We were able in one instance to obtain the number of arrests for sexual offences, and the results of the victimization survey for an identical year, thereby permitting some more detailed comparison. Specifically, in the United States in 2002, the police arrested almost 88,000 persons for sexual offences (82,000 males; 6,000 females see Table 1). In contrast, in the same year, the results from the U.S. victimization survey show estimated that approximately 250,000 sexual offences were committed that year (U.S. Department of Justice, 2003b) of which 4.2 % of the offenders would be expected to be females (approximately 10,500 females and 239,500 males). These results indicate that less than half of all sexual offences in that year in the U.S. resulted in police arrest. More importantly, differential rates of police arrests according to the gender of the offender in relation to victimization survey results were noted. Specifically, 34% of the sexual offences committed by men resulted in police arrest (82,000/239,500), compared to 57% of the sexual offences committed by female offenders resulted in police arrest (6,000/10,500). The meaning of this finding is unclear and may not be statistically reliable considering that only 4 females were used to create the estimate of the rate of female sexual victimization. Nevertheless, these findings do suggests that further research is needed to understand the personal and criminal justice response to victimization by female offenders versus male sexual offenders.

Recidivism Rates of Female Sexual Offenders

Our review showed that the sexual recidivism rate of female sexual offenders is extremely low. One percent of the female sexual offenders included in this review had come to the attention of the criminal justice system as a result of a new sexual offence. This rate is much lower than the overall 13.4% sexual recidivism rate observed for male sexual offenders (Hanson & Bussière, 1998).

In contrast to the extremely low rates of sexual recidivism, the general recidivism is much higher. In this review, 20% of the female sexual offenders where known to reoffend (6% violently). It appears that female sexual offenders, like male sexual offenders, engage in a variety of offending behaviours.

Rates from individual jurisdictions were examined to determine whether further information could be gleaned on the few women who did commit a new sexual offence. It was found that only two studies reported any sexual recidivism among female sexual offenders: Home Office found 1 out of 81 (1.2%) and two of the Canadian federally sentences women (3.3%; Williams & Nicholaichuk, 2001). No additional information was available on the one female sexual recidivist from the U.K. Williams and Nicholaichuk (2001) reported that the only factor that clearly differentiated the two sexual recidivists from all the other women who did not commit a new sexual offence was that they were the only two offenders who had exclusively engaged in solo offending. In other words, the majority of the women in that study had committed sexual offences in company of a male accomplice. The two women who did commit further sexual offences had previously sexually offended on their own against unrelated victims. Of course, we cannot validate a risk marker based on 2 cases. This particular finding is noteworthy, and bears further scrutiny in future research about the dynamics of female sexual recidivism.

Implications

The results from this review have implications for those professionals working with female sexual offenders. First, it appears that evaluators should be more concerned about the risk of non-sexual recidivism than sexual recidivism in female sexual offenders. Second, the substantial difference in recidivism rates suggests that risk tools developed on male sexual offenders are unlikely to apply to females. Simply extrapolating from the male sexual offender literature to assess risk in female sexual offenders is likely to lead to invalid risk appraisal and unintended consequences. For example, in Washington State in the U.S., female sexual offenders are now being civilly committed as sexual predators, despite the lack of evidence regarding risk posed by these women (Aylward, Christopher, Newell, & Gordon, 2003).

The low sexual recidivism rate means that extremely large samples are required to establish empirically validated risk markers for sexual recidivism among female offenders. Only time and the accumulation of knowledge will permit such empirical validation. Until then, in light of the higher rate of general recidivism for these women, it is suggested that an evaluation of risk takes into consideration dynamic risk factors related to general recidivism in women. These factors include antisocial attitudes and associates, substance abuse as precursor to offending, problematic relationships, and emotional dyscontrol (Blanchette, 2001). As these factors are also commonly found in general female offenders, it appears reasonable at this time to also consider these factors when assessing and treating female sexual offenders.

REFERENCES

American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders, (4th ed.). Washington, DC: Author.

Andrews, D. A., & Bonta, J. (2003). The psychology of criminal conduct (3rd Ed.). Cincinnati, OH: Anderson.

Australian Bureau of Statistics. (2005). Criminal courts, Australia, 2003-2004. (No. 4513.0). Canberra, Australian Capital Territory: Author.

Australian Bureau of Statistics. (2003). Crime and safety, Australia. (No. 4509.0). Canberra, Australian Capital Territory: Author.

Aylward, A., Christopher, M., Newell, R.M., & Gordon, A. (2002). What about women who commit sex offenses? Paper presented at the 22^nd Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, St. Louis, Missouri, U.S.A.

Badgley, R. (Chairman). (1984). Sexual offences against children Vol. 1. Report of the Committee on Sexual Offences Against Children and Youths. Ottawa: Supply and Services Canada.

Blanchette, K. (2001). Classifying female offenders for effective intervention: Application of the case-based principles of risk and need. Unpublished manuscript, Carleton University.

Broadhurst, R. & Loh, N. (2003). The probabilities of sex offender re-arrest. Criminal Behaviour and Mental Health, 13, 121-139.

Denov, M. (2003). The myth of innocence: Sexual scripts and the recognition of child sexual abuse by female perpetrators. Journal of Sex Research, 40, 3, 1-12.

Denov, M., & Cortoni, F. (in press). Adult female sex offenders. In C. Hilarski & J. Wodarski (Eds.). Comprehensive mental health practice with sex offenders and their families. New York: Haworth Press.

Eldridge, H., & Saradjian, J. (2000). Replacing the function of abusive behaviors for the offender: Remaking relapse prevention in working with women who sexually abuse children. In D. R. Laws, S. M. Hudson & T. Ward (Eds), Remaking relapse prevention with sex offenders: A sourcebook (pp. 402-426). Thousand Oaks, CA: Sage Publications.

Fergusson, D.M., Lynsky, M.T., & Horwood, L.J. (1996). Childhood sexual abuse and psychiatric disorder in young adulthood: 1. Prevalence of sexual abuse and factors associated with sexual abuse. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1355-1364.

Finkelhor, D., Hotaling, G., Lewis, I.A., & Smith, C. (1990). Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse and Neglect, 14, 19-28.

Freund, K., Heasman, G. Racansky, I.G., & Glancy, G. (1984). Pedophilia and heterosexuality VS. homosexuality. Journal of Sex and Marital Therapy, 10, 193-200.