Community and custody delivered vocational training and employment programmes
Fox et al. (2020) carried out a Rapid Evidence Assessment of the effect of community and custody delivered vocational training and employment programmes on reoffending. Although REAs on this topic already existed, the review by Fox and colleagues included recently published evidence from the United Kingdom, principally relevant studies published by the Ministry of Justice’s Justice Data Lab.
A systematic search of the literature identified 12,724 records. After screening, we found 46 studies that met our criteria of which 33 high quality studies were synthesised in a meta-analysis. The studies were undertaken across a range of community, custodial, and community and custodial (through-the-gate) settings. The 33 studies were predominantly conducted in the United Kingdom and United States with one based in Israel; all 33 took place between 2000 and 2019. Although the review does not consider the cost of such programmes, the meta-analysis found a combined effect of 9% fewer treated individuals reoffending than those who did not take part in a programme.
The REA was partially funded by the Ministry of Justice.
What is it?
Vocational training and employment programmes are designed to help offenders to learn new skills and to re-integrate into society more easily upon release. Whilst academic education is designed to address subject-specific knowledge and skills, vocational education teaches the knowledge and skills to help individuals to gain specific employment.
Programmes are delivered in either the community or in custody, or in both the community and custody (e.g. through-the-gate). The reviewed programmes often included a range of intervention components, many of which were loosely defined. They often included activities undertaken in conjunction with one another at different points whilst targeting a range of interconnected needs. Examples are:
1) work readiness training including both ‘basic skills’ (e.g. numeracy and literacy, locating information, problem solving, and critical thinking) and ‘life skills’ (‘noncognitive skills’ or ‘soft skills’).
2) job search training and support, which is often a component of work readiness training, but is sometimes offered as a stand-alone service and might include CV writing, interview preparation and job searching.
3) vocational education and training linked to a particular job. It is often linked to apprenticeship schemes and often is delivered partly in a class room and partly in the workplace.
4) job placement or transitional work, involving a temporary or time-limited job which may be paid or unpaid.
5) prison-based work, which involves working in a prison workshop where an outside industry runs a part of its business (e.g. a repair workshop, an assembly line or a call centre) in a custodial setting.
6) support by work coaches, job coaches or case managers who are often part of a work readiness programme or part of the wrap around support for a job placement, but might be offered as a standalone intervention.
7) income support/basic income which accompanies some programmes (generally in the US).
Should it work?
Criminologists have argued that employment is likely to be an important mechanism associated with reductions in reoffending and promoting desistance (e.g. Sampson and Laub 1990, 1995). Evidence from three systematic reviews over the last twenty years (Newton et al. 2018; Visher et al. 2005; Wilson et al. 2000) has indicated that ex-offenders and probationers benefit from vocational training and employment programmes in a number of ways, and that it may be possible that such programmes are associated with a reduction in recidivism.
• Wilson et al. (2000) found that education, vocation and work programmes reduced recidivism, which was lower amongst programme participants (39%) compared to those not involved in programmes (50%). However, almost nine out of 10 studies were rated as methodologically poor by the researchers and they were therefore very cautious about drawing firm conclusions.
• Visher et al. (2005) found limited evidence (with no overall significant impact) to support the impact of employment programmes on recidivism.
• Newton et al. (2018) found some evidence to indicate that programmes are most effective for ex-offenders who pose a high risk of recidivism.
Does it work?
To establish what works, Fox and colleagues conducted a meta-analysis on 33 of the studies found in our search all of which featured level 4 and 5 designs in accordance to the Scientific Methods Scale. These are designs (e.g. RCTs and quasi-experiments using well matched groups) from which the causal effect of the intervention under investigation can be clearly and confidently understood. The meta-analysis included studies conducted in the United Kingdom, United States and Israel, in a range of community and custodial settings, and representing a range of different types of interventions and combinations thereof. The forest plot below is a graphical representation of the meta-analysis and shows the 33 studies.
The summary effect of these studies (a weighted combined effect) shows that, on average, the programmes were associated with a reduction in re-offending. The combined summary effect (risk ratio) for the 33 studies is 0.91, and is statistically significant. This indicates that the vocational training and employment programmes studied were associated with 9% fewer treated individuals reoffending than those who did not take part in an intervention.
Despite this finding, the forest plot below indicates that some studies found individual programmes to be associated with increased reoffending in the treated group, and some confidence intervals suggested the possibility of observing increased reoffending if the study were to be repeated.
How strong is the evidence?
The review found 33 high quality studies which addressed the effects of vocational training and employment programmes on reoffending, making it one of the largest bodies of evidence synthesised on this site. Restricting the included research designs to robust studies (as per the methodology of this website, e.g. randomized, quasi-experimental and matched treatment and comparison groups) highlights effect sizes which are the basis of firm causal inference. A significant number of the studies are from the UK, making the findings particularly relevant to a UK context.
However, the overall summary finding is based on a heterogeneous set of studies, comprising a diverse set of programmes delivered in a range of settings and periods of time. This means that there is not a substantial body of evidence for any one particular programme type or setting.
Is it worth it?
The review by Fox et al. (2020) did not consider the costs of the programmes as this was not within the scope of the review.
Can it be implemented?
The review did not specifically address the considerations to be made when implementing community and custody delivered vocational training and employment programmes. Nonetheless, the fact that these programmes exist, and are prevalent, suggests that implementation does not encounter significant barriers.
What's missing from the evidence?
Given its focus on reoffending, the review does not consider employment and educational outcomes reported by many of the included studies, nor does it include other factors (e.g. increased self-efficacy) which may be important mediators or moderators of other outcomes. Understanding the impacts of programmes on a range of outcomes could be an important step towards developing a more complete understanding of how vocational training and employment programmes are associated with reoffending outcomes. However, this was not within the scope of the review by Fox and colleagues. This combined with the heterogeneity outlined above points to a broader strategic consideration; that evidence synthesis is supported by systematic programmes of research (including replication studies) as the basis for generating a body of evidence to understand what works.