By David Riley, Director of Corrections' Psychological Services
The ability to accurately assess an offender's risk has greatly increased over the last 20 years. A variety of tools capable of taking the guesswork out of the risk assessment process are available to staff charged with making decisions about offenders.
Clearly, the judiciary have an investment in this topic due to its impact on processes such as sentencing decisions, release on parole, and issues considered at the time of bail applications.
I've recently delivered presentations to a High Court Judges conference and a series of seminars for District Court Judges on Corrections' risk assessment approaches.
In the early 1990s, the Department was able to explore the relationship between criminal history variables and subsequent re-offending using information held in the government computing facility in Wanganui.
Doing so enabled us to develop the mathematically-based Risk of Conviction (multiplied by) Risk of Imprisonment (RoC*RoI) measures to provide probability estimates of offenders' risk of being reconvicted and imprisoned within a five-year, follow-up period.
While this measure relies only on basic criminal history information, it has been shown to have a high level of accuracy across all age groups, all ethnic groups, both genders and all types of offending except sexual offending.
Additionally, due to the large size of the investigation which drew on more than 133,000 criminal histories, the measure is stable over time and appropriate for a New Zealand offender population.
More recently, Corrections has been able to develop a statistical measure of sex offender risk called the Automated Sexual Recidivism Scale - a reliable and accurate way of classifying sex offenders into one of four categories of risk.
This measure works equally well for people who have sexually offended against children and those who have committed a sexual offence against an adult.
Although both of the above statistical approaches are extremely useful, by relying on historical, unchangeable information, they are insensitive to time-related changes or offenders' efforts to rehabilitate themselves.
For that reason, we have introduced approaches that rely on the assessment of dynamic variables which more directly assess aspects of the individual's functioning, allow for a better assessment of their treatment needs, and provide a way of measuring their progress in the course of treatment.
One example of this approach is the "Violence Risk Scale", employed at the Violence Prevention Unit and used by psychologists in preparing reports to the Parole Board on violent offenders.
A second is the "STABLE-2007", a 13- item checklist for assessing changes in risk of persons convicted of sexual offences.
Work is currently underway to monitor the performance of these measures in a New Zealand context.
It is always desirable to obtain local information as to how well such approaches work, rather than relying on overseas demonstrations of their effectiveness.
The Department's Psychological Services is now investigating a violence risk measure for offenders on community-based sentences.
This will be sensitive enough to determine if changes in a person's circumstances or personal functioning have increased their risk and therefore need for closer management to reduce their risk to the public.
There is absolutely no doubt that structured approaches to risk assessment have moved us far beyond the uncertainty of the outmoded and widely inaccurate "professional judgement". But the area of risk assessment is still best considered a "work in progress".
Psychological Services will continue to refine its approach and incorporate overseas innovations inasmuch as they are applicable to a New Zealand context.
Got a story for Corrections News or want to request the print edition?
Email commdesk@corrections.govt.nz or phone (04) 460 3365.
ISSN 1178-8453