Department of Corrective Services

The Clinical Governance Unit (CGU) Research and Evaluation Team (R&E) was formed following a recommendation from the Mahoney Inquiry into the Management of Offenders in 2005.  This changed the face of research in the Department as, prior to this, the only formalised research mechanism within Offender Programs was the Offender Programs Edith Cowan Project (OPEC).  This was established in 2001 under a MOU with ECU.  It resulted in Professor Alfred Allan of ECU and Deborah Dawson of Offender Programs constructing a research specific unit resourced by 40 volunteer students form Perth’s universities.  This arm of the CGU R&E Team is responsible for a retrospective longitudinal study of some 2000 sex offenders and validation studies of risk instruments used in Western Australia.

The Mahoney recommendation has resulted in the creation of an evaluation arm of R&E, involving 8.4 research officers who have responsibility for evaluations of programs in prisons and the community.  In 2009 the Queensland and New Zealand Evaluation Frameworks were adapted to guide Western Australian evaluation.

The work of the Research and Evaluation Team is based on principles developed in the ‘What Works’ literature.  “’What works’ analyses are about accumulating evidence and developing evidence based policy and programs.” (Pawson 2006).

The following principles are from the ‘What Works’ literature:

• Risk Principle: that the intensity of intervention should match the risk level of offenders.
• Need Principle: that intervention should target the underlying causes of reoffending.
• Responsivity Principle: that efforts should be made to increase offenders’ responsiveness to intervention.
• Program Integrity Principle: that program delivery should be clearly manualised and facilitators should adhere to the manual.

Pawson, R (2006) Evidence-based: a realist perspective. London: Sage

The general goal of program evaluation is to measure effectiveness.  A three-stage offender program evaluation strategy has been adopted.

This is monitored by the Clinical Governance Program Development and Standards team. There are three main elements to ensuring that program integrity is maintained. These are:

• program accreditation
• facilitator accreditation
• site auditing

Short term measures of program impact make an important contribution to the

evaluation process by:

• ensuring that the program is impacting the specified target criminogenic needs
• informing continued program improvements by identifying unmet needs in offender populations and in this way contributing to future intervention design, development and knowledge about the integration and sequencing of interventions
• measuring performance at local and state levels
• examining area and regional differences
• examining the effectiveness of the program for different types of offender
• ensuring that offenders selected for a program are the most appropriate for that program.

Measuring Short Term Impact

• Pre and post testing
• Focus groups with each program group
• Module by module and end of program reviews conducted by facilitators

Determination of the impact of programs on recidivism rates is, by its nature, a long term strategy. Rehabilitation programs that target high risk offenders can take up to nine months to complete. It is also important to allow time for a sufficient number of offenders to have completed the program for analyses to be undertaken. After program completion, prisoners must finish serving out the remainder of their sentences. Once released, the minimum period at liberty – used by most research studies, correctional organisations and government reporting agencies (for example, the ABS) at state, national or international levels – is two years. Sex offender program evaluations generally take longer as the research indicates a much longer time at risk prior to sexual re-offending compared with other populations. Research also recommends that data continue to be collected to allow recidivism rates to be calculated over variable periods.

Measuring Long Term Impact

The primary measure of recidivism is conviction and return to corrections, although there may be scope for future consideration of other indices such as arrests and order breaches. Information on the type of re-offence is recorded to enable determinations of whether offenders return for committing similar crimes to which they previously were convicted or whether it is another type of offending. Offender Services’ information systems provide the means by which offender characteristics, program completion, and recidivism data are collected. As the amount of data available increases, more elaborate analyses are undertaken. For example:

• examining the effects of aftercare programs and services on reconviction rates.
• examining the seriousness of the re-offence. Simply examining re-offence rates may miss the fact that offenders who have done a program are reoffending with much less serious offences.
• examining multiple programs effects – that is, examining the impact of treatment on offenders who have completed more than one program.
• analysing the way offenders with different characteristics respond to programs – when programs are found to be less effective with certain types of offenders then modifications to or replacements of the program can be made for these offenders.
• statistical tests of significance can be used to determine if a difference exists between the treatment and comparison groups. In addition, effect sizes will also be examined to ensure that not only is there a difference between the treatment and comparison group, but that this difference is of a magnitude to justify the investments being made in treatment.
• confirming the observed short-term measures predict recidivism – if these measures are found to be good predictors then more confidence can be placed in these short-term measures when evaluating other newly introduced programs. Both the predictive validity for psychometric tests for specific treatment programs can be tested.
• comparisons can also be made with recidivism trend data from other Australian jurisdictions and from international sources.

Clinical Governance Unit
Level 4, 141 St Georges Terrace
PERTH  6000  WA.
Telephone: (08) 9264 1015
Facsimile:   (08) 92641077