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


This chapter provides an overview of findings from studies that have followed-up offenders with the aim to investigate what makes some desist from crime (defined as "desisters") and others not (defined as "persisters"). This research tends mostly to be qualitative in nature and draws on offenders' own accounts of the desistance journey to gain a better understanding of the factors that help or impede their efforts to give up crime. The chapter covers the impact of thinking styles, the formation of social bonds, employment, negative external circumstances and contact with the justice system in the process of desistance.

According to some studies but not others, thinking styles are influential in determining whether offending continues or ceases. Desisters do not necessarily face fewer social problems than recidivists but there is evidence to suggest they are more psychologically resilient showing higher levels of self-efficacy and better coping skills. Healy[459] followed-up a sample of 73 adult male probationers in Ireland and investigated differences between those that had stopped offending within a 4-year follow-up period ("desisters") and those that continued to offend ("persisters"). The study found that the two statistically significant predictors of desistance were age at the time of the interview and general attitudes to crime as measured by the CRIME-PICS scale. Desisters tended to be older and less likely to endorse attitudes that were supportive of the criminal lifestyle. On the other hand, those who had offended in the past year were significantly more likely to have currently active thinking styles, for example more commonly endorsing the view that crime is worthwhile. An interesting finding was that both groups reported similar levels of victim empathy, indicating good awareness of the effects of their behaviour on victims. Surprisingly, social circumstances did not emerge as significant predictors of desistance with recidivists and desisters reporting a similar level of criminogenic needs. This finding has been replicated in some studies[460] but not in others[461].

It has been suggested (Healy, 2010) that what differentiates desisters from recidivists is not the number of structural obstacles they encounter but the way they respond to them, with desisters showing higher levels of personal agency, better coping skills and a more positive perception of their lives and future prospects [462]. Maruna compared the life history narratives of 65 English men and women with extensive criminal histories of committing drug and property offences. The desisters in this study were more likely to express the belief that they could control their own futures, whereas the accounts of persisters revealed a fatalistic outlook to life. The study also found that desisters were more likely to take responsibility for their criminal past and see themselves as "good" people. This enabled them to maintain a positive self-image and supported the shift from a criminal to a prosocial identity[463].

The most common triggers of change include the formation of strong social bonds, a developing awareness of the negative consequences associated with crime including the prospect of a lengthy prison sentence, and, in fewer cases, the development of a good relationship with a supervisor and attendance at a rehabilitative programme. The most frequently cited reason for change in Healy's study (cited above) was the formation of strong social bonds with parents, partners and children, a finding similar to that of Bottoms and Shapland in their Sheffield Desistance Study.[464] Similarly, in Scotland, Jamieson et al.[465] found that many women offenders were encouraged in their decision to stop by the support of friends, family, children and loving relationships with law-abiding partners. In Liebrich's follow-up study of probationers in New Zealand, responding to new family commitments was frequently cited as reason for wishing to desist[466]. Strong attachments trigger the motivation to change because they provide emotional support, the prospect of new social roles and models of prosocial behaviour. For example, having children made some participants adopt a new positive perspective and instigated a desire to live up to family responsibilities that was conducive to change. However, it is important to note that having children does not automatically lead to desistance and some studies have found that for some offenders the positive impact of having a child is delayed until children grow older and become more aware of their parents' criminal lifestyles.

The second most commonly reported trigger for change in Healy's study was developing an awareness of the costs of crime including the likelihood of a lengthy prison sentence as a result of repeated contact with the criminal justice system. Many among those who expressed the desire to desist from crime were becoming concerned about spending large portions of their life in prison and were beginning to realise that their current life path was "going nowhere".

Finally, in a smaller number of cases, the apparent trigger for change was some form of external intervention, for example attending a rehabilitative programme or developing a good relationship with a supervisor. Some studies have found that ex-offenders feel empowered when they receive assistance from an outside force who believes in them, for example the significant quantitative finding in relation to offenders on Community Sentences in England and Wales, that offenders who believed that their supervisors understood their needs were significantly less likely to reoffend[467]. Probation supervisors who offered a 'welfare' approach, rather than a strict 'supervision' approach were better received by probationers, which is more likely to support the process of desistance: this suggests that supervision should be linked to wider opportunities and strong welfare supports generally[468]. In contrast, when offenders are categorised as "high-risk" they often lose faith in their ability to change and develop a fatalistic outlook that is not conducive to change[469]. However, in a review of user experiences of supervision in Scotland, McNeill stresses that experiences of supervision vary between supervisors and supervisees, which makes generalizing about experiences of supervision problematic[470].

Factors associated with sustained abstinence from offending include strengthening social relationships, developing new social networks, finding suitable employment and improved emotional well-being. As the quality of offenders' relationships with the important people in their immediate social circles improves, they are more likely to want to live up to others' expectations and sustain a crime-free lifestyle. Strong family bonds can encourage desistance by giving structure to offenders' lives and by acting as sources of informal monitoring and support. Also, when offenders develop strong emotional ties with members of their wider network they are more likely to take into consideration the feelings of others when considering a reversion to crime. Being trusted by significant others and the wider social network has proven to be a strong motivating factor for sustained desistance from crime[471]. In this regard, it is important that the Sheffield Desistance Study in England found that immediate social circumstances were significantly related to desistance, independently of past offending history (although a substantial history of offending did act as a slowing effect on desistance).[472]

In a qualitative study in Scotland, Weaver found that social capital and relationships were central to the process of desistance[473]. Offenders can desist in order to improve relationships which are incompatible with continued offending. As a result, the author suggests that services for offenders should incorporate offenders' social relationships, for example by including peer support and allowing users to have input into how services are designed. Volunteering is also proposed as an avenue to build social capital. Some researchers have suggested that existing interventions do not pay sufficient attention to offenders' existing sources of social support[474], although empirical tests of programmes designed to utilize existing social support are lacking.

Taking up new employment and recreational opportunities can also encourage desistance by providing access to more prosocial social networks. As McNeill and Whyte note, without access to social capital, it may be difficult to begin and maintain desistance[475]. By securing a job or a stable relationship, offenders start to realise that they have a future and are accepted and trusted by others, which leads to increases in self-esteem and positive identity change[476].

Farrall (2002)[477] investigated the effect of probation supervision on subsequent offending among a sample of 199 male and female probationers aged 17-35 that were spread across six English probation services. In this study, probationers attributed their desistance primarily to finding suitable employment and/or a stable partner rather than any help they got from their probation officer, which suggests that offender supervisors should proactively try to assist offenders with finding employment and improving family relationships if they are to increase their chances of desisting from crime. However, looking back later on their journey to desistance, in 2014 these offenders ascribed a greater effect to their supervisors' suggestions and nudging[478].

In Burnett's follow-up study of 130 property offenders released from custody in England and Wales, desisters were more likely to have secured stable employment and accommodation and rate their personal relationships as good compared to recidivists[479]. Changes in social circumstances are also often accompanied by improvements in emotional well-being that have been positively linked to desistance[480].

Desistance attempts fail when external circumstances such as financial problems make offenders feel trapped in a criminal lifestyle, when there is a change in social circumstances, for example a failed relationship, and when offenders are insufficiently committed to change or feel ill-equipped to solve the problems they encounter. It is important to recognise that the journey to desistance can follow a 'zigzag' rather than a linear pathway, and many will continue to drift between conformity and offending for some time. The majority of participants in Healy's study attributed their ongoing offending to external circumstances such as financial problems and addiction, which they felt unable to overcome[481]. The number and extent of obstacles to desistance predicted reoffending in the Sheffield Desistance Study.[482] Financial problems have also been cited by other studies as a major criminogenic need for women, with many women prisoners being financially dependent on their families after release[483].

Contact with the criminal justice system can induce positive changes for some but engender reoffending for others, which illustrates the subjectivity of the desistance process and the variability in the quality and usefulness of such contact. In Healy's study some participants claimed that contact with the criminal justice system induced change whereas others thought it engendered reoffending[484]. This illustrates that it is the offender's interpretation of the event that matters in bringing about change more than the event itself. For case management, desistance research stresses the importance of consistency and commitment in the case management team, and the value of face-to-face meetings between case management teams and offenders[485].

Short-term prison sentences can be perceived as pointless, serving neither to rehabilitate nor punish offenders. Qualitative research has investigated the experiences of some who have served multiple short term prison sentences in Scotland[486]. Short term prison sentences were routine for many of the people interviewed and they perceived multiple sentences as part of an on-going single experience of punishment. People serving short term sentences typically did not report any rehabilitation or punishment effect of their sentence. The authors warn that short-term sentences may weaken social bonds on the outside, disrupting natural processes of desistance. The authors also feel that short-term sentences do not allow for people to build their capacities: for example, those serving sentences of less than six months were unable to participate in rehabilitative programmes which worked around a twelve-week model. The interviewees perceived these sentences as pointless, and they were a source of anger and hopelessness. It should be noted that these findings are from interviews with just 22 prisoners serving short-term sentences in Scotland, and so may not be representative of the experiences of all of those who serve short prison sentences.[487]


The above review of qualitative and quantitative studies suggests that the onset and maintenance of desistance depends, to a large extent and for a significant proportion of offenders, upon them developing prosocial thinking styles, higher levels of self-efficacy, and prosocial bonds. Interventions that target these areas are, therefore, more likely to be successful in reducing reoffending. Many of these studies have also stressed that the process of desistance varies between individuals, and researchers have recommended that service users' input should be incorporated into rehabilitation programmes in order to tailor services to users' needs[488]. A final important theme coming from this body of research is that the quality of the relationship between probationers and supervisors can be important in the process of desistance, as well as probationers' relationships with family and peers. Attempts should be made to encourage the formation and maintenance of strong relationships between probationers and supervisors, as well as probationers and family, peers and their communities, but not to the exclusion of practical support.


Email: Justice Analytical Unit