Download Effect of Community Orders on Vulnerable Offenders: Indigenous & Rural Communities and more Lecture notes Literature in PDF only on Docsity!
COMMUNITY-BASED SENTENCING ORDERS AND PAROLE
A Review of Literature and Evaluations across Jurisdictions
Dr Karen Gelb Dr Nigel Stobbs Professor Russell Hogg
Prepared for the
Queensland Sentencing Advisory Council
by
Queensland University of Technology
April 2019
COMMUNITY-BASED SENTENCING ORDERS AND PAROLE:
A Review of Literature and Evaluations across Jurisdictions
Prepared for the Queensland Sentencing Advisory Council by Queensland University of Technology April 2019
- iii -
Contents
List of abbreviations ................................................................................................................vii Glossary ....................................................................................................................................ix
- Introduction and background........................................................................................... Executive summary x
- 1.1 Scope of the report
- 1.2 Methodology
- 1.2.1 Search strategy
- 1.2.2 Methodological qualification....................................................................................
- 1.3 Definitions: Effectiveness
- 1.4 Definitions: Recidivism
- 1.4.1 Type of reoffending
- 1.4.2 Measures of reoffending: recidivism versus desistance...........................................
- 1.5 Definitions: Costs..........................................................................................................
- 1.6 Structure of the report
- Legislative overview of orders
- 2.1 Imprisonment and parole
- 2.1.1 Legislative structure and composition
- 2.1.2 Practical application – use and breach
- 2.1.3 Costs and resource implications
- 2.1.4 Changing use of imprisonment and parole
- 2.2 Suspended sentences
- 2.2.1 Legislative structure and composition
- 2.2.2 Practical application – use and breach
- 2.2.3 Costs and resource implications
- 2.2.4 Changing use of suspended sentences
- 2.3 Intensive correction orders
- 2.3.1 Legislative structure and composition
- 2.3.2 Practical application – use and breach
- 2.3.3 Costs and resource implications
- 2.3.4 Changing use of intensive correction orders
- 2.4 Home detention orders
- 2.4.1 Legislative structure and composition
- 2.4.2 Practical application – use and breach - iv -
- 2.4.3 Costs and resource implications
- 2.4.4 Changing use of home detention
- 2.5 Community service orders
- 2.5.1 Legislative structure and composition
- 2.5.2 Practical application – use and breach
- 2.5.3 Costs and resource implications
- 2.5.4 Changing use of community service
- 2.6 Probation orders
- 2.6.1 Legislative structure and composition
- 2.6.2 Practical application – use and breach
- 2.6.3 Costs and resource implications
- 2.6.4 Changing use of probation
- 2.7 Community correction orders
- 2.7.1 Legal structure and composition
- 2.7.2 Practical application – use and breach
- 2.7.3 Costs and resource implications
- 2.7.4 Changing use of community correction orders
- Effectiveness of custodial orders
- 3.1 Imprisonment
- 3.1.1 Imprisonment and deterrence
- 3.1.2 Imprisonment and incapacitation
- 3.1.3 Comparing community sentences with custodial sentences
- 3.1.4 Summary of the research on prison
- 3.2 Partially suspended sentences
- 3.2.1 Effectiveness of partially suspended sentences among vulnerable cohorts
- 3.2.2 Factors affecting successful completion of a partially suspended sentence
- 3.2.3 Gaps in research on the effectiveness of partially suspended sentences
- 3.2.4 Summary of the research on partially suspended sentences.................................
- 3.3 Parole
- 3.3.1 Effectiveness of parole among vulnerable cohorts
- 3.3.2 Factors affecting successful completion of parole
- 3.3.3 Electronic monitoring
- 3.3.4 Gaps in research on the effectiveness of parole
- 3.3.5 Summary of the research on parole - v -
- Effectiveness of non-custodial orders
- 4.1 Wholly suspended sentences
- 4.1.1 Effectiveness of wholly suspended sentences among vulnerable cohorts
- 4.1.2 Factors affecting successful completion of a wholly suspended sentence
- 4.1.3 Gaps in research on the effectiveness of wholly suspended sentences
- 4.1.4 Summary of the research on wholly suspended sentences
- 4.2 Conditional suspended sentences
- 4.2.1 Effectiveness of conditional suspended sentences among vulnerable cohorts
- 4.2.2 Factors affecting successful completion of a conditional suspended sentence
- 4.2.3 Gaps in research on the effectiveness of conditional suspended sentences
- 4.2.4 Summary of the research on conditional suspended sentences
- 4.3 Intensive correction orders
- 4.3.1 Effectiveness of intensive correction orders among vulnerable cohorts
- 4.3.2 Factors affecting successful completion of an intensive correction order
- 4.3.3 Gaps in research on the effectiveness of intensive correction orders
- 4.3.4 Summary of the research on intensive correction orders
- 4.4 Home detention
- 4.4.1 Effectiveness of home detention among vulnerable cohorts
- 4.4.2 Factors affecting successful completion of home detention
- 4.4.3 Gaps in research on the effectiveness of home detention
- 4.4.4 Summary of the research on home detention
- 4.5 Community service orders
- 4.5.1 Effectiveness of community service orders among vulnerable cohorts...............
- 4.5.2 Factors affecting successful completion of a community service order
- 4.5.3 Gaps in research on the effectiveness of community service orders
- 4.5.4 Summary of the research on community service orders
- 4.6 Probation
- 4.6.1 Effectiveness of probation among vulnerable cohorts
- 4.6.2 Factors affecting successful completion of probation..........................................
- 4.6.3 The effectiveness of swift, certain and fair sanctions
- 4.6.4 Gaps in research on the effectiveness of probation
- 4.6.5 Summary of the research on probation
- 4.7 Community correction orders
- 4.7.1 Effectiveness of community correction orders among vulnerable cohorts - vi -
- 4.7.2 Factors affecting successful completion of a community correction order
- 4.7.3 Gaps in research on the effectiveness of community correction orders
- 4.7.4 Summary of the research on community correction orders
- 4.8 The effectiveness of community orders generally for vulnerable cohorts...............
- 4.8.1 Aboriginal and Torres Strait Islander offenders
- 4.8.2 Female offenders
- 4.8.3 Offenders with a mental illness
- 4.8.4 Offenders in rural and remote areas
- Conditions attached to non-custodial orders
- 5.1 The effectiveness of treatment
- 5.1.1 Treatment programs that work
- 5.1.2 Programs for specific cohorts
- 5.1.3 Treatment programs that do not work
- 5.1.4 Gaps in research on the effectiveness of treatment
- 5.1.5 Summary of the research on treatment
- 5.2 The effectiveness of supervision
- 5.2.1 Supervision intensity
- 5.2.2 Environmental corrections: a reconceptualisation of supervision
- 5.2.3 Gaps in research on the effectiveness of supervision
- 5.2.4 Summary of the research on supervision
- References
List of abbreviations
ABS Australian Bureau of Statistics ACT Australian Capital Territory ALRC Australian Law Reform Commission BOCSAR Bureau of Crime Statistics and Research CBT Cognitive-behavioural therapy CCO Community correction order CCS Community Correctional Services CJG Community Justice Group CSO Community service order DUI Driving under the influence DWI Driving while intoxicated EM Electronic monitoring GPS Global Positioning System HOPE Hawaii Opportunity Probation with Enforcement ICO Intensive correction order ISP Intensive Supervision Program LSI-R Level of Service Inventory-Revised MST Multi-Systemic Therapy NSW New South Wales NSWLRC New South Wales Law Reform Commission NT Northern Territory NZ New Zealand PAU Probation as usual QCS Queensland Corrective Services QLD Queensland
Glossary
Logistic regression A type of linear model that is used when analysing data that are dichotomous (with two categories of outcome, such as yes/no). Meta-analysis Meta-analysis mathematically aggregates the statistical results of different studies into a single database, allowing the results to be analysed collectively, rather than individually. Overall effects become more evident, and broad patterns within a body of research are revealed with greater clarity and consistency than is possible with traditional reviews. Poisson regression A type of linear model that is used when analysing data that are not normally distributed. Propensity score matching A statistical technique that attempts to estimate the effect of a given intervention while taking account of the factors that predict receiving the intervention in the first instance. Recidivism Reoffending, operationalised in different ways across different studies. Statistical significance The likelihood that a statistical relationship between two variables has not occurred by chance (conventionally measured by whether the probability that the relationship occurred by chance is less than 5%). Survival analysis A set of statistical techniques that is concerned with the time it takes for a particular event to occur. Systematic review An appraisal and synthesis of primary research papers using a rigorous and clearly documented methodology in both the search strategy and the selection of studies.
Executive summary
In October 2017, the Queensland Sentencing Advisory Council (QSAC) received a Terms of Reference from the Attorney-General and Leader of the House to examine community-based sentencing orders, imprisonment and parole options. This reference was made in response to recommendations made in the recent review of Queensland’s parole system (Sofronoff, 2016). In particular, QSAC was tasked with considering recommendations 2-5 of the report (Sofronoff, 2016, p. 23):
- Recommendation 2: Court ordered parole should be retained.
- Recommendation 3: A court should have the discretion to set a parole release date or a parole eligibility date for sentences of greater than three years where the offender has served a period of time on remand and the court considers that the appropriate further period in custody before parole should be no more than 12 months from the date of sentence.
- Recommendation 4: A suitable entity, such as the Sentencing Advisory Council, should undertake a review into sentencing options and in particular, community based orders to advise the government of any necessary changes to sentencing options.
- Recommendation 5: Court ordered parole should apply to a sentence imposed for a sexual offence. In response to Recommendation 4, QSAC commissioned a literature review to assist in informing its broader advice to the government. This report presents the findings of the review. This report contains two components:
- a review of each order to present its legislative and practical framework; and
- a review of existing literature to identify the effectiveness of various sentencing options. Key findings The use of imprisonment and parole
- Prison is the most expensive criminal justice system response to crime, costing more than $3.4 billion in Australia in 2017-18.
- Despite these costs, the use of imprisonment has increased throughout Australia over the past decade, in contrast to a decrease in the use of custodial terms in other countries.
- xii - The effectiveness of imprisonment
- Although imprisonment is undoubtedly effective at punishing offenders and denouncing criminal behaviour, research shows that it is not effective as a deterrent to further offending and it appears to reduce reoffending via incapacitation only to a limited extent.
- There appear to be minimal differences in reoffending between custodial and non- custodial sentences. However, the evidence on this issue is mixed, with some research showing higher rates of recidivism following incarceration.
- At best, imprisonment has a marginal impact on recidivism. At worst, imprisonment increases the likelihood of reoffending. The effectiveness of partially suspended sentences
- There is no robust research on the effectiveness of partially suspended sentences. What little research exists finds that recidivism rates are higher following a partially suspended sentence than after a wholly suspended sentence.
- There is no research on the impact of partially suspended sentences among vulnerable offenders.
- Recidivism rates following a partially suspended sentence appear to be lower among older offenders and those with no criminal history, but the evidence for this is weak. The effectiveness of parole
- The effectiveness of supervised parole relative to unsupervised release remains the source of considerable debate. Nonetheless, there is reasonable evidence that parole is more effective at reducing recidivism than unsupervised release.
- There is some evidence that more active supervision can reduce recidivism, but only if the focus is rehabilitation, rather than compliance.
- There is very limited research on the effectiveness of parole for vulnerable cohorts. It appears that parole might be less effective for Aboriginal and Torres Strait Islander offenders. While parole generally appears to be more effective for female offenders than for men, it appears that Caucasian women and older women are more likely to complete parole and probation successfully, while those with substance abuse problems are less likely to be successful. Offenders with a mental illness have been shown to be more likely than healthy offenders to return to custody for either a new crime or a technical violation.
- Although evidence is mixed for some factors, there is consistent evidence that reoffending on parole is more likely among parolees who are young, male, Indigenous, with a criminal history. There is no clear consensus on the effectiveness of court-ordered versus board-ordered parole.
- Most of the research shows that electronic monitoring while on parole reduces recidivism cost-effectively, especially when used as a genuine alternative to
- xiii - imprisonment and with high-risk sex offenders. Evidence on net-widening with electronic monitoring remains inconclusive. The effectiveness of wholly suspended sentences
- Wholly suspended sentences have a small but significant effect on reducing recidivism when compared with imprisonment, especially for repeat offenders. While there is no research on the effectiveness of wholly suspended sentences among vulnerable cohorts, they are considered useful for offenders who are unable to access other community orders, such as those in rural and remote areas.
- Although the evidence is sparse, it appears that wholly suspended sentences might be more likely to be completed by older offenders and those convicted of property offences. The effectiveness of conditional suspended sentences
- Although the evidence is very limited, it suggests that people on suspended sentences with conditions might be more likely to reoffend than those without conditions. Without further research, though, this conclusion remains tentative. No further conclusions may be drawn about conditional suspended sentences due to the lack of evidence. The effectiveness of intensive correction orders
- Research shows that there is no difference in the effectiveness of intensive correction orders when compared with supervised suspended sentences. There is good evidence, though, that intensive correction orders are more effective at reducing recidivism than either periodic detention or short terms of imprisonment, especially among offenders classified as high risk. There is no evidence on the effectiveness of intensive correction orders among vulnerable cohorts.
- Reoffending following an intensive correction order appears to be more likely among men, Indigenous offenders, those with criminal histories and those classified as high risk. The effectiveness of home detention
- Home detention can place unintended burdens on other members of the household. Nonetheless, home detention can ease reintegration following prison, facilitate reconnection with pro-social family and activities, and deter future offending. While there is little research on the effectiveness of home detention among vulnerable cohorts, it may be useful for offenders who are unable to access other orders.
- Intensive case management, using a mix of surveillance and rehabilitative strategies, appears to be important for successful completion of home detention. Findings on other factors affecting completion of home detention have been inconsistent.
- xv - The effectiveness of community orders generally for vulnerable cohorts
- Community orders are seen as more appropriate than terms of imprisonment for Aboriginal and Torres Strait Islander offenders, for whom prison can be particularly harmful. But community sentences need to be more accessible and more flexible to provide greater support and to mitigate against higher breach rates. Conditions of community sentences, as well as support and services, need to be culturally appropriate.
- Community orders are also seen as more appropriate for female offenders, who have multiple and complex needs. Community sentences should offer multi-agency wrap-around support and services designed specifically for women, including practical help with issues such as health, housing, childcare and employment.
- Offenders with a mental illness appear to have worse outcomes on community orders than offenders without a mental illness, primarily because they have more of the key risk factors for recidivism. Community sentences for this cohort should therefore focus on the same core correctional principles and interventions as are used for offenders without a mental illness.
- Offenders in rural and remote areas have less access to community sentences, so are more likely to be imprisoned. The availability of community sentences should be expanded to reach this cohort, including a coordinated multi-agency approach to support disadvantaged people in these areas. The effectiveness of treatment
- There is now a large body of evidence that offender treatment is an effective way to reduce recidivism and improve a range of outcomes for offenders. The most successful programs are those which adopt the risk-need-responsivity approach to rehabilitation, involve cognitive behavioural therapies or include drug treatment, such as therapeutic communities.
- Treatment effectiveness may depend on the nature of participation, with programs having only limited effect when offenders are mandated or coerced into treatment.
- Treatment programs for specific types of offender appear to be mostly effective. The evidence is less positive, however, about the effectiveness of treatment programs for reducing reoffending among family violence offenders and offenders with a mental illness.
- Treatment programs that are gender-informed are more effective than gender- neutral programs for female offenders. Similarly, interventions that are culturally appropriate are more effective for Indigenous offenders.
- Programs that are solely based on discipline, surveillance or punitive practice – without any rehabilitative support – do nothing to reduce recidivism and may instead increase the likelihood of reoffending.
- xvi - The effectiveness of supervision
- The evidence shows mixed support for the effectiveness of supervised release. Supervision without adequate rehabilitation services and support – that is focused on enforcement – does not reduce recidivism. When combined with critical rehabilitation services such as mental health treatment, drug treatment and housing assistance, supervision focused on service delivery is effective at reducing reoffending.
- The evidence on high-intensity supervision is mixed, with much of the evidence indicating that its heightened surveillance acts to increase both recidivism and technical violations. However, when coupled with therapeutic interventions, high- intensity supervision can be effective, especially for high-risk offenders.
- Although the evidence is sparse, low-intensity supervision, used for low-risk offenders, does not appear to increase recidivism, so may be a cost-effective tool for managing large, low-risk offender cohorts.
- While evidence is limited, it suggests that an environmental corrections approach – whereby supervision is used to reduce offenders’ opportunities to reoffend – can effectively reduce recidivism.
Orders relating specifically to youth, pre-sentence programs, specialist courts and low-end orders are excluded from this review. An overview of the legislative framework of each of these orders is presented in Chapter 2.
1.2 Methodology
This report contains two components, requiring two different approaches. The first – the legislative review – involved legal analysis of each order to present its legislative and practical framework. The second – the literature review – was undertaken as a desktop review of existing literature on community-based sentencing options. It was not intended as a systematic review, but as a broad search of sound evidence on the effectiveness of various sentencing options. 1.2.1 Search strategy Both scholarly research and grey literature 1 were searched for relevant material. The search related to the effectiveness of the following orders: imprisonment, imprisonment with probation, imprisonment with court-ordered parole, imprisonment with board-ordered parole, partially suspended sentences, wholly suspended sentences, conditional suspended sentences, intensive correction orders, home detention, community service, probation and community correction orders. Research on the effectiveness of treatment and supervision conditions was also included. Key search terms included the name of each type of order plus key words such as recidivism, effectiveness/effect, evaluation/evaluate and cost. Searches were repeated with the addition of various terms such as women , mental illness and the like when looking for research on vulnerable cohorts. Questions of framework and conditions for each order type were addressed via legislative instruments and public information sites such as the Victorian Sentencing Advisory Council’s. Such sites, plus government reports and court annual reports, were also examined for information on the use of each order over time. While the focus of the literature search was on research produced in Australia, research from the United Kingdom (UK), Canada, the United States (US),^2 New Zealand and Europe was also considered, as appropriate. (^1) Grey literature includes reports produced by government bodies and other organisations that are not part of the academic system. (^2) Despite differences in the legal system, the United States is a valuable source of research. There is a huge research literature from the US that is methodologically very strong and, unusually, includes cost-benefit assessments. Some of the world’s leading authors on the effectiveness of sentencing orders have written extensively on US experiences and research, ensuring that this is a valuable source of evidence.
1.2.2 Methodological qualification This review focuses on studies that are sufficiently sound that they provide reliable and meaningful information. In particular, priority has been given to studies that use robust statistical methods such as randomised controlled trials 3 or quasi-experimental designs using propensity score matching^4 with a control group. For those issues in this review for which such strong evidence is unavailable, this is noted and a more cautious tone is adopted.
1.3 Definitions: Effectiveness
Sentences in Queensland may be imposed for one or more of the following purposes: (i) punishment, (ii) rehabilitation, (iii) deterrence, (iv) denunciation and (v) community protection. The effectiveness of sentences may be assessed against any of these. Punishment and denunciation are purposes relating to the concept of just deserts, such that the sentence becomes an end in itself. These sentences are thus effective in their own right. For the remaining sentencing purposes – rehabilitation, deterrence and community protection – the sentence is imposed in order to achieve a utilitarian purpose, such that the sentence is a means to an end (reducing further offending). The most common definition of effectiveness is based on the future behaviour of the individual offender, and whether the sentence has reduced the likelihood of further criminal behaviour. One of the major aims of any criminal justice system is to enhance the safety of the community by reducing reoffending. But the measurement of the effectiveness of any system in achieving this aim is fraught, not only in terms of accurately measuring reoffending, but even in the task of defining the meaning of effectiveness. Effectiveness is often defined in absolute terms: has the criminal justice system (or any specific component of the system) successfully prevented further offending? That is, the system has been proven effective only if offenders do not go on to commit any further crimes. But a more nuanced view of effectiveness suggests that the criminal justice system has been effective in its aim of enhancing public safety if the nature or frequency of subsequent offending changes: if subsequent offending is less serious or less frequent than previously. (^3) A randomised controlled trial design is considered the ‘gold standard’ of research. It involves purely random assignment of subjects to different interventions in order to examine the impact of the intervention on the outcome of interest. The random nature of the assignment implies that factors that might influence the outcome are evenly distributed among the different groups; any resulting differences in outcomes are thus wholly attributable to the intervention itself. As one cannot simply randomly assign sentences to examine their effectiveness, pure random designs are rare in research on the effectiveness of sentences. Instead, quasi- experimental designs with a matched control group are able to approximate the effect of random assignment. (^4) Propensity score matching is a statistical technique that attempts to estimate the effect of a given intervention (in this review, a specific sentencing order) while taking account of the factors that predict receiving the intervention in the first instance. It is considered a strong statistical technique that, in the absence of true randomisation, provides reliable results, as it aims to minimise any confounding impact on the outcome through a process of matching subjects with like characteristics.
