CHAPTER 2: COMMUNITY RESPONSE TO SEX OFFENDERS - FEAR VERSUS EDUCATION

If we look at a comparative analysis from the Bureau of Justice Statistics we see sex offense as a subdivision of violent crimes. In these statistics the variables do not break down type of sexual offender but the method of the offense indicating type of violence.

(BJS, March 1996 p.2)

Bureau of Justice Statistics concluded 7 in 10 offenders with child victims reported they were imprisoned for a rape or sexual assault and an estimated 18.6% of inmates serving time in State prisons in 1991 for violent crimes. About 61,000 offenders nationwide had been convicted of a crime against a victim under the age of 18. Approximately 58% were under the age of 13. It was also reported that 4 in 10 child victims of violence suffered either a forcible rape or another injury (Greenfield, 1996, p. 3). The Bureau of Justice statistics indicated that in 1994 the rate for sexual assault by age per 1,000 persons were 4.4 assaults under the age of 25, 2.1 for ages 25 - 49 and 0.1 for ages 50 or older (Perkins, Klaus, 1996). According to these statistics as people get older, they are less likely to be victims of sexual assault.

Studies concerned with the rate of recidivism vary widely depending on the source of the study. Problematic for these types of studies is the wide range of variables when addressing sex offender populations. Variables range from the type of offense; age of victims; number of victims; cross gender victimization; whether treatment was received in an incarcerated setting or community based setting; varying types of treatment models within these settings'; whether treatment was refused; motivation of offender during the treatment process; and age of the offender. There is currently a strong movement to refine assessment tools to predict the sex offender's risk for future recidivism. In this chapter we will discuss some of the findings and complexities of this predictive task.

R. Karl Hanson (Hanson, 1997) developed the current recidivism rate scale that is used within the state of Vermont. It is a brief actuarial risk scale for sexual offense recidivism. In developing this tool, Hanson took into consideration the cost of applying the tool within the criminal justice system and ease of use (Hanson, 1997). The purpose of this instrument development was to identify those factors that have been consistently related to sex offender recidivism. It was based on several studies including a 10-year follow-up study of Canadian federal offenders released in 1983. The study pool consisted of 316 sexual offenders and 2,709 non-sexual offenders; a follow-up study of Canadian provincial offenders released between 1958 and 1975 in a 15 to 30 year follow-up with a study pool of 191 child molesters and 137 non-sexual offenders. Hanson and Bussiere's research (1996) suggested that sexual recidivism can be predicted by a different set of factors than those that predict general or non-sexual violent recidivism. The initial indicators for recidivism among sexual offenders including the definitions for the terms were as follows:

1. Prior sex offenses:

This includes prior convictions. It does not include self-disclosure of other non-conviction offenses. If an arrest did not result in a conviction than it is not included.

2. Any stranger victims:

Those victims that did not have a previous relationship with the offender prior to the offense.

3. Any prior offenses:

Prior convictions of a non sexual nature whether they be violent or non violent offenses.

4. Age (young-less than 25):

This indicates the age at the time of release for incarcerated sex offenders.

5. Never married:

Married includes both legal and common-law relationship or stable relationship over a two-year duration.

6. Any non-related victims:

Related victims include biological, step-relationships, cousins, siblings, parents, foster children and situations where the victim is living in the same home as the offender.

7. Any male victims:

Those who have ever offended against a male whether that male be a child or adult.

The current scale has been down sized to include the four factors. The definitions remain the same as above and each item is scored as "0" if the criteria did not apply and "1" if the criteria apply to the offender. The resulting score (0 - 5) is then placed within the context of estimated sex offender recidivism rates based on 5-Year Follow-up and 10 Year Follow-up studies (Hanson, 1997).

1. Prior sex offenses.

2. Age at release.

3. Victim gender.

4. Relationship to victim.

For statistical purposes the estimated sex offense recidivism rates are as follows:

In another cross-validation study of the Violence Risk Appraisal Guide (a tool used to predict violence among high-risk offenders) sampled 288 sex offenders that included 142 child molesters, 88 rapists and 58 individuals who fit both criteria. The individuals who fit both categories were considered to have had histories of serious offenses, although not all were sexual crimes but those that evoked violence, had poor childhood adjustment and lack of adult social adjustment. In this study, N = 104 had received lengthy treatment (M > 5) in maximum-security psychiatric facility (Rice, Harris, 1994, p. 233). One hundred and ninety two of the subjects received phallometric sexual preference measurements. Results showed that rate of violent recidivism was high, with 166 of the 288 or 58% of the subjects committing another violent offense. The rate of sexual recidivism was 35%, with 101 men recorded with a new sex offense. The sample characteristics and background variables were as follows:

Note: For one-way among-group comparisons, *p <.05, **p<.01, ***p<.001. (a) Items in the VRAG (b) Additional items selected by the Cox proportional hazards event history analyses (Rice and Harris, (1997) p.234)

Analysis of the data showed that rapists had faster recidivism rates than child molesters. Those included in the mixed victim group had a recidivism rate faster than either rapists or child molesters. The results showed that rapists were more likely to re-offend during the early years of follow-up while non-rapist sex offender recidivism increased during the later years of follow-up (Rice, Harris, 1997, p. 239). Rice and Harris were mainly concerned with the prediction of future violence among violent offenders regardless of sexual content for future recidivism due to such legal maneuvers as plea-bargaining and the difficulties in determining motivation.

Furby, Weinrott and Blackshaw examined 42 studies, which were published in 1989 in the Psychological Bulletin. They concluded that the variation in samples and methodology made it impossible to come to an accurate and stable conclusion regarding recidivism rates among varying types of sex offenders. This alone makes it near impossible to compare and contrast findings of the numerous studies that are currently available (Furby, Weinrott, Blackshaw, 1989 p. 27). They were, however, able to conclude that the longer the follow-up the greater the number of later criminal activity (sexual or non-sexual). The researchers believed there is no evidence that recidivism will not occur as a result of treatment. There is an indication that recidivism rates may be higher for varying types of sexual offenders.

Ogloff, Wong and Greenwood (1990) undertook a study within a therapeutic community program for personality disordered individuals, many who were classified as psychopaths. The findings indicated that the psychopath was more likely to not complete group treatment due to lack of commitment and motivation to change their behavior. Their initial motivation to attend community treatment may be less altruistic than the non psychopathic sex offender since courts, parole boards and probation consider favorably those criminals who state their consent to participate in such programs. Once in the program, the psychopathic sex offender does not actively commit to the self-change required. The result is that they are discharged early from the treatment program having not gained maximum benefit, or they are remanded back to the prison system for punitive measures (Ogloff, Wong, Greenwood, 1990).

Evaluation of the therapeutic community programs for violent psychopathic offenders by Marnie E. Rice of the Mental Health Centre in Ontario, Canada added further insight into the problem of the psychopathic personality in the community programs.

While Rice does not discount that some type of therapeutic intervention might benefit the psychopathic offender a regime of stringent monitoring and supervision with zero tolerance of aggression is suggested.