A brief review
Abstract
Recidivism rates are very instrumental in the policy making, offender incarceration and treatment. However when these rates are surrounded by uncertainty and are lacking, these rates do more harm than good. This is truer in sexual crimes, particularly in child molestations. While criminological literature emphasizes low recidivism rates for sexual criminals, the public fueled by media, consider it otherwise. Nonetheless recidivism rates are high for child molesters among the sexually offending group. Even among child molesters, there is much difference between the intra-familial and extra-familial groups too. Framework models and tools help specialists in identifying individuals with high risk, which aid in risk assessments. However, the effectiveness of these is again a topic of controversy.
The study of recidivism or repeated offences, with regard to sexual offending is important to the criminal justice system. Although offending by any offender is a matter of public concern, re-offending by sexual offenders is particularly important, given the harm and fear it generates in the community. Sexual offending particularly against children is a very emotive issue, and a matter of huge social concern (Richards, 2011). The term ‘child sexual abuse’ applies to a range of inappropriate behaviors including contact offenses and non-contact offenses. Sexual abuse of children results in serious consequences for victims which includes depression, anti social behaviors, suicidality, alcohol and drug abuse, parenting difficulties, and sexual re-victimization (Zwi et al., 2007). It should be mentioned here that child sex offenders are often driven by compulsion to sexually abuse children, particularly in the face of opportunities. All societies therefore have a responsibility to act on individuals committing crimes and respond appropriately through punishments, detention, and supervision. In determining its response, the society must know the perceived risk of recidivism associated with the crime. Several offenders are particularly managed by their risk of future offending. There are several individual characteristics associated with recidivism in sexual offenders particularly child sex molesters. Although several risk factors have been identified for recidivism risk, the relationship between single risk factor and recidivism is small. Research on recidivism can help in formulating intervention strategies with child sex offenders or child molesters. A better understanding of recidivism and recidivism rates is necessary for better policy making and treatment of offenders.
However there are several issues associated with the measurement of sex offender recidivism. For instance, even the way in which recidivism is measured, can have an influence on the results of the study. Variables like the criteria used, types of offenses studied and the type of the offending population; all have a bearing on the determined recidivism measurements.
In the criminological literature, sex offenders have been found to have lower rates of recidivism, although child sex offenders are very prominent with a comparatively higher rate than other sex offenders. This is against the public and media perception that a sex offender is very certain to reoffend. Low rates of recidivism among sex offenders have been shown by several studies, with the recidivism rates even being lower than other type of offenders (McSherry and Keyzer, 2009). According to the US Justice Department, subsequent to the largest and most comprehensive study ever undertaken, also determined that sex offenders are very less likely to reoffend. The 2003 study of nearly10,000 men for sexual crimes including child molestation found that sex offenders had a re-arrest rate of 25%. Measuring sexual recidivism is a challenging and difficult task, and all studies have their own limitations. Researchers are often required to take key decisions with regard to measuring recidivism among the sex offenders, which impact the study findings. Two important attributes that affect study results are how researchers define recidivism and the time period over which recidivism is measured.
Sex offending has a comparatively higher attrition rate than most other offenses, as most offenses are not reported and only a small number of cases enter the criminal justice system. Most studies rely on reconviction for inclusion into recidivism, while only some studies consider an arrest or a charge as being sufficient for recidivism. Although this approach also has limitations, it is likely to provide a more accurate measure of recidivism (Richards, 2011). The difference between the public perception and the criminological literature is evident from the findings of Jill Levenson, a sex crime researcher from Lynn University of Florida. Jill and her colleagues determined that an average citizen believed that about 75% of sex offenders would reoffend. This situation could be largely attributed to media outcries that project sex offenders as chronic offenders. Despite the reported low rates of recidivism, the media has ironically created hysteria and fear in public minds. This perhaps is largely fueled by the news media. News coverage often emphasizes the dangers associated with convicted sex offenders, pedophiles, that these increasingly get embedded with the public. The ‘Today Show’ ran a series of poorly conceived and designed tests to check on whether strangers would help a child when the child is abducted. ‘Dateline NBC’ began a program intended to lure potential online predators to a house with hidden cameras. The high ratings associated with program saw six more ‘To catch a predator’ specials came up. Good journalism requires stories to be balanced with facts and context. Much of the coverage unfortunately lacks perspective and is too wrong or incomplete, with regard to sexual predators (ATSA, 2008).
In line with this, some experts contend that there are also reasons to perceive that the actual recidivism rates, including child molestation rates may not be very different from what the researchers had determined (Wakefield, 2006). This is because frequent offenders are remarkably tracked and monitored and therefore more likely to be caught when reoffending. The high offenders are under extra registration requirements, residency limitations, and community notifications apart from GPS monitoring and potential civil confinement.
However research with the offending population also portrays a similar picture. Sex crime researchers Karl Hansen and Kelly Morton of Public Safety Canada determined a recidivism rate of 14% for a six-year period. They also determined that the rate touched 24% for a period of 15 years. Also the 15-year recidivism rate varied across the offending population, it being 13% for incest perpetrators, 35% for molestation of boys and 24% for rapes. The study based on large scale meta analysis or quantitative review showed results that were not in alignment with public expectations (Arkovitz and Lilienfeld, 2008). Hansen and Morton also found that the total recidivism rate involving sex crimes and non sexual crimes to be about 36% across a period of five to six years.
According to the Association for the Treatment of Sexual Abusers (ATSA) many people expect the re-offending rate for sexual offenses to be around 80 to 90%. However sex offenders are not a homogenous group and sexual offenses are varied. Therefore generalizing sexual re-offending as a single rate is misleading and inaccurate. The re-offense rates needed to be studied for various categories of sex-offenders. Research undertaken by the US Department of Justice and the Canadian Government show a recidivism rate of about 5 to 20% in three to six years of known adult offenders. Some categories like those molesting boys have a higher rate over a longer period (ATSA, 2010).
People who recommend tough legislation against child sexual offenders cite a 2004 Canadian study, ‘Lifetime Sex Offender Recidivism : A 25-year follow up study’. The study led by Canadian researcher Ron Langevin investigated 320 sex offenders between 1966 and 1974, all of whom were referred to a particular clinic for psychiatric evaluations. Langevin reported a sex crime recidivism rate of 61.1%. Taking into account the offender confessions and arrests that did not bring convictions, together with convicted sex crimes, the overall sexual offense recidivism rate was 88.3%. When this methodology was applied to child molesters, the sex offending recidivism rate across 25 years was a staggering 94.1% (Dormin, 2010).
The risk of re-offending and dangerous are related, yet not the same. There are important differences between the probability of a behavior and the costs of that behavior. The assessment of a risk is too difficult due to the complexities of individuals and the number of variables associated with each that needs to be considered. Nonetheless, this requirement is both intellectually demanding while also being professionally satisfying. Although it is difficult to determine reliable and valid predictions of risk, there is an increasing demand to provide clear statements on risk prediction, for varied reasons. This includes for opinion in civil cases, supervision of children, in mental health review tribunals and in child protection cases. Any statement of risk should also give an indication of the potential validity of the assessment, highlighting the strength and shortcoming of the methodologies employed (Briggs, 2007). Webster and Eaves (1997) offer certain common principles aimed at improving predictive accuracy in the assessment of dangerousness and risk.
- Take into account actuarial considerations. The information from historical records should anchor clinical judgment
- Elimination of professional and personal bias
- Predictions should be based on current scientific knowledge
- Conditions associated with assessments needs to be adequate like having sufficient time and all necessary paperwork.
Risk assessment is an area highly dominated by subjective evaluation. A common error is the making of false positive judgments like assuming individuals or clients to be in greater risk. Thus assumptions must be tested and dogma avoided. An important model in risk evaluation is that of David Finkelhor’s “Four Factor’ Model which is a dynamic model of child sexual abuse. The model emphasizes on four factors harming the development of the victimized child namely; traumatic sexualization, betrayal, stigmatization, and powerlessness. The model differentiates motivational factors and maintenance factors (Daugherty, 2011). According to the model, risk assessments should initially include determination of evidence to sexual arousal of children or child related stimuli, evidence of satisfying emotional needs through child abuse. The determination of high risk individuals is normally made based on actuarial assessments. The risk level is estimated through assessments including STATIC-99 and the Rapid Risk Assessment for Sexual Offense Recidivism. However a 100% accuracy of behavior prediction is very rare, though the chances of correctly predicting recidivism increase with the validity of the assessing instruments. In the US, many states have tried to standardize procedures for determining recidivism risk. Researches indicate grounds sufficient to challenge the validity of the tools, with the three best tools used having only moderate levels of prediction (Hanson & Morton, 2007).
A cautious review of the scientific and criminological literature will reveal the fact that it is almost impossible to determine the actual recidivism rates, although it is important too. Recidivism rates are mostly based on counting the sex offenders released into the community, who are caught and convicted. A study by Prentky et al. (1997) covering 115 child molesters determined that child molesters continued to be at risk of reoffending, even several years after their discharge. They also determined that the recidivism rate was under estimated (Leadership Council, 2006). A review by the American Psychological Association in 2003 also emphasized that sexual offenses against children, have an inherent risk of recidivism by offender, even if the offenses had occurred long ago. Very few sex offenders are caught for the second time, which doesn’t rule out offending by the others. Once imprisoned, a child sex offender learns to be more careful in his further molestation encounters, from that point onwards. These findings all point out to the unreliability of the low recidivism rates associated with sex offenders particularly child molesters. ATSA itself estimates that a released sex offender adds about 5.2 victims in the 25 years subsequent to his release. Thus, proclaiming that only a small percentage of child sex offenders reoffend, the criminal justice fraternity wrongly ignores the plight of the children who suffer sexual abuse. This stand has been of substantial gain for the child molesters, who thus continue to convince people that they are not a big danger to the society (Steele, 2010).
The relationship between age and sexual recidivism too varies across sexual offending groups. Child molesters are generally older than rapists who target woman. While the recidivism rate of rapists decreases with age, extra-familial child molesters hardly showed any reduction in recidivism rate until they crossed the age of fifty. The relativity of age of the child molesters and recidivism rate is largely unknown (Karl, 2010). However based on the fact that anti-social behavior reduces with age, it is very likely that child molestation recidivism rates also decline, although it would be minimal. Extra-familial child molesters are very likely to have deviant sexual practices, which in turn contribute to a higher recidivism rate, even extending to late adulthood. Another aspect of child molester recidivism rates which raises eyebrows is the consideration of intra-familial child molesters as being at a lower risk for reoffending in comparison to extra-familial child molesters. Molestation of unrelated victims has a high re-offense risk. A study by Firestone and colleague reported a recidivism rate of 15% for extra-familial child molesters, over a period of 7.8 years. This is in comparison to another study of 251 intra-familial molesters where a lower sexual crime recidivism rate of 6.4% was determined across a period of 6.7 years. This low rate of recidivism among incest offenders is however due to the different assessment instruments and management of incest offenders in comparison to extra-familial offenders. However this perception of intra-familial child molesters being at a lower risk of re-offense is questionable and challenged under following grounds (Eher and Ross, 2006):
- Clinical issues: Despite there being evidence to show that intra-familial child sexual offenders are less deviant, a comparative study of high and low deviant molesters show that 70% of low deviant and 40% of highly deviant groups are constituted by intra-familial child molesters. A prominent phallometric study has also made a notable observation that incest and extra-familial offenders are equally aroused by pedophilic stimuli.
- Criminological issues: Studies undertaken by Bartosh et al. (2003) did not find any substantial recidivism rate difference between intra and extra-familial offenders. With regard to violent crimes, re-offense rates were significantly high among intra-familial offenders.
- Time difference: The time between the commission of a re-offense and conviction is considerably longer in intra-familial child molester cases. There is therefore a bias between the timelines, in the duration until the instant of failure estimation. There is thus considerable doubt whether the reconviction data interprets the same story for both intra- and extra-familial groups.
One may therefore conclude that a risk of conviction might be lower in intra-familial child molesters, but not the risk of re-offense. Many offenders who initially seemed incestuous were seen to be more pedophilic as they had other non-incest victims too. There are also justified doubts whether pedophilic offenders, both incestuous and non-incestuous really need to be treated as two different groups.
When deliberating on the lesser than perceived rates for recidivism, including that for child molesters, we should remember that recidivism research is complex and intricate, involving several players. A recent follow up on a 1998 meta-analysis determined several factors associated with sexual recidivism that are not relevant to offending in the general public. Measures of anti social personality were seen as indicators of sexual recidivism. Employment instability was also seen to predict sexual recidivism (Krueger, 2007). Among the factors that failed to predict recidivism with acceptable accuracy are social skill deficits, loneliness, and general psychological problems like depression and low self-esteem.
Sexual abusers of children have a comparatively higher risk of recidivism, than any other type of sexual offenders. A 1995 recidivism study by the Department of Corrections in Alaska showed that offenders receiving treatment had longer survival rates before their re-offence compared to those who did not receive any treatment. Those who undertake advanced levels of treatments were seen to have higher survival rates before re-offense, compared to those with lesser treatment levels. Despite all measures employed, re-offense persists. Reducing the risk of recidivism in child molestation requires addressing the characteristics associated with the risk. Recidivism data has its own limitations, due to which no definitive conclusions can be made with regard to the recidivism rates. The definition, methodology, and interpretation in recidivism studies are obviously not consistent across all studies, which have corresponding effects on the study results (Bonnar-Kidd, 2010). These inconsistencies can be attributed to several factors which cannot be ignored, equalized, or neutralized across the studies. It is not clear if the sexual recidivism rates, particularly that of child molesters point to a lack of opportunity to offend, non-detection of subsequent crimes or to habilitation. Several studies in sexual recidivism have been undertaken in international literature. While recidivism rate research could be hampered by several factors, there is however little controversy on managing it. Many insightful meta-analyses have shed light through consistent findings, using different methodologies, over several decades (Gelb, 2007). There is now a general consensus among researchers that sexual recidivism is associated with deviant sexual interests and antisocial behavior or lifestyle instability.
As of now, there is almost no definite empirical evidence for identifying treatment effectiveness with regard to sexual offense recidivism prevention treatments. Experienced clinicians are often unable to differentiate between sexual offenders including child molesters who benefitted from the treatment and those who did not. Future offenses can only be prevented by addressing problems that are currently evident. The dynamic risk factors are characteristics that are subject to change and when changed can result in a corresponding increase or decrease in recidivism rates. These dynamic factors are important although there is a lack of research evidence connecting these factors to sexual recidivism. However long term follow up studies could identify stable risk factors, than rapidly changing risk factors, all contributing to development of reliable recidivism rates.
References
Arkowitz H & Lilienfeld S.O. (2008) Once a sex offender, Always a sex offender?. Maybe not. Scientific American http://www.scientificamerican.com/article.cfm?id=misunderstood-crimes
ATSA (2008) Research and statistics debunk: Common misconceptions in Zoth L (Ed.)Sex offenders and public policy. Gale Cengage Learning. pp 43-48
ATSA (2010) The registration and community notification of adult sexual offenders http://www.atsa.com/registration-and-community-notification-adult-sexual-offenders
Bonnar-Kidd K.K., (2010) Sexual offender laws and prevention of sexual violence or recidivism. American Journal of Public Health. 100(3): 412-419
Briggs D., Doyle P., Gooch T., & Kennington R. (2005) Assessing men who sexually abuse: a practice guide. Jessica Kingsley Publishers, London. pp145-152
Daugherty L. (2011) How does child sexual abuse happen?. http://www.cleananpress.com/abuse/happens.htm
Dormin C (2010) Facts and fiction about sex offenders. University of Cincinnati. http://www.corrections.com/news/article/24500-facts-and-fiction-about-sex-offenders
Eher R & Ross T (2006) Reconsidering risk for reoffense in intrafamililial child molesters: New aspects on clinical and criminological issues. Sexual Offender Treatment, Vol 1:2
Gelb K (2007) Recidivism of sex offenders research paper. Sentencing Advisory Council, Melbourne pp-26-30
Hanson K.R & Harris A.J (2008) Where should we intervene? Dynamic predictors of sexual offense recidivism in Holmes R.M and Holmes S.T (Eds.) Current perspectives in sex crime pp 300. Sage Publications
Hanson R.K & Morton K.E (2007). The accuracy of recidivism risk assessments for sexual offenders: A meta-analysis. Department of Justice, Canada
Karl H (2010) Age and sexual recidivism: A comparison of rapists and child molesters. Department of the Solicitor General Canada
Krueger J (2007) Research Bulletin: Sex offender populations, recidivism and actuarial assessment. New York State Division of Probation and Correctional Alternatives. P4
Leadership Council (2006) Recidivism: How often do Child Molesters go on to reoffend?. http://www.leadershipcouncil.org/1/res/rcd.html
McSherry B & Keyzer P 2009. Sex offenders and preventative detention: Politics, policy and practice. Leichhardt: Federation Press
Richards K (2011) Trends & issues in crime and criminal justice. Australian Institute of Criminology. http://www.aic.gov.au/publications/current%20series/tandi/421-440/tandi429.html
Steele H (2010) 25% Recidivism Rate – really? http://innocentjustice.org/2010/25-recidivism-rate-%E2%80%93-really/
Wakefield H. (2006) The vilification of sex offenders: do laws targeting sex offenders increase recidivism and sexual violence? J Sex Offender Civil Commitment Sci Law 2006;1:141–149.
Zwi et al. 2007. School-based education programmes for the prevention of child sexual abuse. Oslo: Campbell Collaboration
