Publication - Research and analysis

What Works to Reduce Reoffending: A Summary of the Evidence

Published: 8 May 2015

This is an updated version of the original review entitled ‘What Works to Reduce Reoffending: A Summary of the Evidence’, published in 2012.

What Works to Reduce Reoffending: A Summary of the Evidence


The following chapter summarizes the key findings from the evidence into what works to reduce reoffending, and then compares the findings of the 2015 edition of the evidence review to the conclusions of the 2011 version.

Desistance is a highly individualised process and one-size-fits-all interventions do not work. There is broad agreement between researchers that generic interventions are ineffective to reduce reoffending, and many studies covered in this review have described varying effects of different interventions. Researchers who focus on RNR state that interventions should be matched to an offender's level of risk, focus on their specific needs and be matched to their individual responsivity characteristics[526]. Researchers who focus on GLM and desistance suggest that users' perspectives should be incorporated into treatment[527]. Thus, whilst there is disagreement about the form in which individualization should take place and the rationale behind it, research suggests that one-size-fits-all interventions are unlikely to be successful, and providing inappropriate interventions may lead to increases rather than decreases in offending.

The evidence is still developing, but a number of studies have found that those serving short prison sentences have higher rates of reoffending than those serving community sentences. A number of quantitative studies have reached this conclusion using data from different countries[528]. Importantly, qualitative research suggests potential reasons why this is the case. Short prison sentences can perceived as meaningless, putting people's lives on hold but not helping them overcome their problems[529]. Imprisonment can also entail losing employment, housing or contact with family. In contrast community punishments may be seen more often as positive and constructive, allowing offenders to get help for their immediate problems such as drug and alcohol use. In Scotland, a presumption against short sentences of three months or less was introduced as part of the Criminal Justice and Licensing Act 2010. This, admittedly tentative, body of evidence is in accordance with this presumption and suggests that offenders should serve community sentences rather than short prison sentences where possible.

More generally, the way in which individuals are processed by the criminal justice system and partner agencies may alter their likelihood of reoffending. Evidence suggests that young people who are diverted from contact with the justice system have lower levels of offending, with impacts potentially lasting well into adulthood. For adults the evidence evaluating strategies of through-care is limited, but that which exists suggests that transitions to life outside prison are smoothest when multiple agencies, including third-sector organizations, work together with prisoners before release to plan transition to the community. There is some limited research to show that receiving visits while in prison may be important in maintaining family bonds, and so it may be beneficial to facilitate family visits for those in prison where possible. Taken together, the evidence seems to suggest that how people are processed by the criminal justice system can impact on rates of reoffending. More research is required, but strategies to reduce contact with the justice system, especially for young people, and efforts to facilitate offenders gaining access to necessary support services may help to reduce reoffending.

There are a number of individual factors which are associated with reduced reoffending. A number of studies have shown that key events in offenders' lives such as marriage, parenthood, finding employment and re-integration in the local community impact upon reoffending[530]. As a result, interventions that help offenders find employment, develop prosocial networks, enhance family bonds and increase levels of self-efficacy and motivation to change may be those more likely to have the strongest positive impact on the risk of reoffending[531].

A number of scholars have argued that desistance from crime is different for women than it is for men, and that women require different interventions to help assist this process. Some researchers have suggested that the process of desistance from crime may be different for women than it is for men. Despite these arguments, there is a lack of evaluations of accredited offending behaviour programmes designed specifically for women. Whilst cognitive-behavioural interventions can be effective with women who offend, some researchers have contended that the often complex and inter-connected needs of women who offend are best met using broader, holistic services[532]. In addition, women-only services may help to reach those who have experienced severe victimization at the hands of men and for whom mixed-sex services may act as a barrier to utilizing available support[533].

Rehabilitative interventions with the strongest evidence base are cognitive-behavioural programmes which address criminogenic needs. There is a substantial amount of literature which shows statistically significant reductions in reoffending for groups of offenders who receive cognitive-behavioural therapy when compared to a control group when interventions are targeted at criminogenic needs[534]. However, the research suggests that the effectiveness of these interventions varies between individuals, and evaluations of projects in the field often show less effectiveness than demonstration projects[535], and can suffer from high rates of participant attrition[536]. As such, more work is required to understand the most effective ways to implement cognitive-behavioural programmes in practice.

More research is required to understand the effectiveness of strengths-based intervention programmes and their implications for practice. A number of researchers suggest that interventions to reduce reoffending should focus on individual's strengths rather than just on their criminogenic needs. However, there is debate about how to achieve this in practice. Some researchers suggest that a focus on strengths is an appropriate part of rehabilitation in its own right[537]. In contrast, others propose that factors which are not directly criminogenic should take less emphasis than directly addressing criminogenic needs[538]. This debate in part represents theoretical differences between researchers and there is little evidence available by which to directly compare the two approaches. Some authors have suggested that risk-based and strengths-based approaches do not differ much in practice[539]. This is a developing area and further research is required to outline the impact of strengths-based interventions in practice.

Supervision can be an important factor in helping offenders desist from crime. A number of qualitative studies showed that offenders value getting support to solve practical problems, being listened to and believed in. In addition, the importance of the quality of the relationship between supervisors and probationers was noted in a number of studies[540]. Practitioners must have strong interpersonal skills, be able to exercise discretion and have the ability to be flexible and innovative in response to complex and varied needs[541]. Consistency of supervision and face-to-face meetings are also important factors of effective supervision. However, supervision may not be helpful when it amounts to simply reporting at social work offices and intensive supervision that is not accompanied by some form of support in addressing criminogenic needs is unlikely to lead to reductions in reoffending[542]. As such, more work is required into establishing how best supervision can support offenders desist from crime.

Offenders' relationships - with supervisors, family, friends and the community - are considered important to the process of desistance. Researchers have suggested that offenders' relationships are important in the process of desistance[543]. This research suggests that wherever possible, support from family, friends and supervisors should be incorporated into interventions for offenders. However the research in this area is still developing, and more research is required into how social support can be incorporated into interventions[544].

There is limited, but mostly positive, evidence for the effectiveness of reparative and restorative programmes in reducing reoffending. Some studies have shown that restorative justice conferencing can have a significant impact in reducing reoffending, particularly for adult offenders[545]. However, the total number of robust studies evaluating the effectiveness of such programmes is small, and there is less clarity about the effects, for example, of indirect mediation. Further work is required to fully understand the potential of reparative and restorative programmes in reducing reoffending.

Factors outside of the control of the criminal justice system affect reoffending. Researchers have paid to factors outside of the control of both the criminal justice system and the individual offender in driving reoffending. Structural factors, such as lack of stable employment in sectors likely to employ ex-offenders, available housing and community factors, such as low social cohesion[546], can affect the chances of an individual reoffending. When transitioning from prison to the community, gaps in service provision can hamper attempts to desist from offending[547]. It is therefore imperative that agencies from different government (and third) sectors work together effectively to assist those transitioning back into the community. Not all structural factors are amendable to change by the criminal justice system, but it is important to note that government agencies must work effectively together to support offenders who may face challenges in multiple areas.

Comparison with the findings of the 2011 version of the review

The 2011 review concluded that:

  • Parenthood, marriage, re-integration, employment are important in reducing reoffending
  • Desistance is an individualized process
  • Practical support is important in reducing reoffending
  • The relationship between offender and supervisor can be important to desistance
  • Skilled supervision of offenders is required
  • There is strong evidence for cognitive-behavioural programmes
  • Criminogenic needs must be addressed by interventions
  • Women who offend may require gender-specific programmes and holistic approaches to treatment

The 2015 review has not found any research which has questioned the substantive conclusions of the 2011 review. In some areas, where previous evidence was tentative, they have been strengthened. In particular, whilst the evidence is still not considered conclusive, a number of studies published since the 2011 review have suggested that community sentences can be more effective than short-term prison sentences in contributing to reduced reoffending.

Finally, this review has expanded a small number of areas not covered extensively in the previous review, including the importance of local and structural factors in reoffending and issues regarding the implementation of intervention programmes in practice. Research in these areas is still developing, and more work is required before their impact on what works to reduce reoffending is fully understood.


Email: Justice Analytical Unit