|
...from files of child sexual offenders from a community-based sexual offender treatment program. Offenders with single or multiple victims were compared, as were offenders who did or did not offend against victims of both genders, and offenders with only intra-familial or extra-familial victims versus offenders with victims in both relationship categories. Variables that discriminated single-victim and multiple-victim offenders were similar to those identified in actuarial risk assessment scales, with the exception of history of childhood sexual abuse. With the exception of physical abuse history, the same variables discriminated specific offender groups according to victim gender and victim relationship, although in different combinations. There was limited support for the notion of specific risk variables.
**********
A considerable amount of empirical work in sexual offender risk assessment over the last decade has sought to identify those factors that are associated with sexual recidivism (e.g., Hanson & Harris, 2001; Hanson & Thornton, 1999), which has resulted in a proliferation of actuarial risk assessment measures designed specifically for use with perpetrators of sexual offences. (1) These measures vary both in terms of their content and in the extent to which they have been validated for use with different populations. All of the measures provide a composite score or rating of the overall level of risk that an individual offender presents, often reported in terms of the probability of someone with a similar offence history offending again within a set period of time. Of available actuarial instruments, the Rapid Risk Assessment for Sexual Offence Recidivism (RRASOR; Hanson, 1997) and Static-99 (Hanson & Thornton, 1999) are best supported in terms of studies of predictive validity and validation with different populations.
The RRASOR was initially developed to enable the prediction of sexual offence recidivism using a small number of variables that were easily scored. Based on meta-analysis and multiple regression, the variables chosen were prior sex offences, current age, victim gender, and relationship to victim. The RRASOR has high inter-rater reliability and considerable evidence of predictive validity from at least 17 studies (Doren, 2002). The Static-99 was developed by adding additional static risk factors to the RRASOR in an attempt to improve coverage of risk variables. The Static-99 has high inter-rater reliability and evidence of predictive validity from at least 15 studies (see Doren, 2002).
Specific Risk Assessment
Specific risk assessment can be defined as judgments concerning risk of offending for a specific individual in a specific set of circumstances. A specific risk question often posed is about the risk presented by an offender to a particular potential victim or type of victim, rather than the risk of the offender committing any further offence. Specific risk assessment is required when a known offender requests access to children. For example, an incest offender may seek to return to the family home following a period of imprisonment. Second, an incest offender who sexually abused his female child may be denied access to his victim, but may request access to his male child. Greenberg, Bradford, Firestone, and Curry (2000) have argued that:
child protective agencies, parole officers and judges are often faced with perplexing decisions on whether to allow various degrees of access of incest perpetrators with a previous and subsequently now empowered older victim or their other non-abused biological children from the same family as the victim (p. 1493).
Third, specific risk assessment may be required when offenders enter new relationships and seek to move into new families. Fourth, specific assessments of risk may be required by child protection agencies when new allegations of child sexual abuse are made against ex-offenders, but there is insufficient evidence to prove a case. Actuarial risk assessment measures may have limited utility in answering specific questions of this kind. While they provide estimates of how likely it is that offenders with similar profiles will re-offend, they do not offer information about the risk presented to a particular type of child.
For example, although the evidence for the usefulness of the RRASOR and Static-99 for predicting sexual recidivism is based on multiple studies and large data sets, there is much less evidence for their usefulness in the prediction of specific types of recidivism. Sjostedt and Grann (2002) examined the prediction accuracy of the RRASOR and Static-99 for recidivism in general, and for recidivism of a variety of offence characteristics, such as familial victims, stranger victims, victims under 15 years of age, or re-offending within 1 month. Although their initial database held more than 1,200 cases, the low recidivism rate of 6% over an average follow-up time of 5.68 years meant that their sample contained only 75 recidivists. When they examined recidivism related to specific offences, their sample contained only three re-offenders within 1 month (0.2%) and approximately 38 re-offenders (3%) against victims under 15 years of age. This study is significant in that it suggests that although actuarial risk assessment methods provide general estimates of risk of re-offending, they have not been developed in ways that allow the assessor to discriminate between different types of offenders or offences.
In order to assess specific sexual offender risk questions, it may be necessary to construct specialised instruments to assess the risk presented to particular types of victim, given the known characteristics of an offender. For example, an instrument could assess the degree of risk of sexual abuse an offender presents to male victims, given a history of abusing female victims?
The stability of victim preference is a central issue here. Empirical studies suggest that the stability of victim preference is unclear. Guay, Proulx, Cusson, and Ouimet (2001), for example, found that victim preferences remained stable across repeated sexual offence occasions. These findings were not supported, however, by Heil, Ahlmeyer, and Simons (2003). They found that relatively few sex offenders abuse only one type of victim, concluding that although sex offenders may have 'a preferred victim, this preference can change over time and may be expanded when the preferred victim is unavailable' (p. 233). If victim preference tends to be stable, then the rate of offenders crossing victim types should be low, and certain characteristics may differentiate those offenders who cross victim types from the majority. If victim preference is unstable, then the rate of offenders crossing victim types may be high. In this case, specific risk assessment instruments may not be feasible, and specific risk assessment questions would reduce to the issue of general risk of sexual recidivism.
The aim of the present study was to examine the feasibility of the process of developing specific risk assessment instruments by identifying those variables (if any) that discriminate between different sub-groups of offenders based on victim type. In addition, variables associated with general sexual 'recidivism' were also identified in order to compare them with variables associated with crossing victim types. As has been described above, even large samples of offenders in recidivism studies may mean that few cases are available to investigate specific risk questions (Sjtstedt & Grann, 2002). Instead of recidivism data, the present study used historical data contained in clinical case files to identify differences between offenders who crossed victim types and those who did not. In place of comparing recidivists and non-recidivists, offenders with single victims were compared with offenders who had multiple victims (many of whom were recidivists). By this method, specific risk...
NOTE: All illustrations and photos
have been removed from this article.
More articles from Psychiatry, Psychology and Law
The relationship between the adjustment of Australian police officers ..., April 01, 2006
Hunter Area Health Services v. Presland: liability of mental health se..., April 01, 2006
How processing resources shape the influence of stealing thunder on mo..., April 01, 2006
'Locked up without guilt or sin': the ethics of mental health service ..., April 01, 2006
The detection of malingered psychopathology and cognitive deficits: em..., April 01, 2006
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
%20%20&position=1)