Understanding ‘What Works’: Accredited Cognitive Skills Programmes for Young Offenders
| Author | Caroline Friendship,Louise Falshaw,Jenny Cann |
| DOI | 10.1177/147322540500500303 |
| Published date | 01 December 2005 |
| Date | 01 December 2005 |
| Subject Matter | Articles |
Programmes for Young Offenders
Jenny Cann, Louise Falshaw and Caroline Friendship
Correspondence: Ms Jenny Cann, Senior Research Officer, Research, Development and
Statistics Directorate, Home Office, 2nd Floor, 2 Marsham Street, London SW1P 4DF.
(Note: this article is submitted in a personal capacity and is independent of any official
view of the Home Office.) Email: Jenny.Cann7Vhomeoffice.gsi.gov.uk
Abstract
This research assessed effectiveness of prison-based cognitive skills programmes for
young offenders in England and Wales, using reconviction as the outcome measure.
Reconviction rates were compared between offenders who started a programme
(N:1,534) and a matched comparison group who did not (N:1,534). Programme
participation did not lead to a statistically significant reduction in reconviction, at one- or
two-years following discharge. However, when offenders who dropped out of a programme
and their matched comparisons were excluded from analysis, the one-year reconviction
rate for programme completers was statistically significantly lower (p:0.05).
Introduction
The ‘What Works’ approach to offender rehabilitation has been prominent since the
mid-1980s. A number of outcome evaluations have indicated that the treatment of
offenders can have a small but statistically significant positive effect in reducing
reoffending (McGuire and Priestley, 1995) and the ‘What Works’ research literature
indicates that interventions incorporating cognitive-behavioural elements produce the
greatest effect (Vennard, Hedderman and Sugg, 1997; Hollin, 1999; Cooke and Philip,
2000; Redondo, Sanchez-Meca and Garrido, 1999). The premise of cognitive models
is that what and how an offender thinks, how they view the world, how well they
understand people, what they value, how they reason and how they solve problems
play an important role in their criminal behaviour (Porporino, Fabiano and Robinson,
1991). Many offenders, it is argued, engage in ‘maladaptive thinking’. Cognitive skills,
therefore, form a primary target for interventions, which aim to replace maladaptive
thinking skills with pro-social ones. Such cognitive-behavioural interventions focus on
changing affect and cognition, teaching offenders the process of consequential thinking
as a precursor to changing offending behaviour (Hollin, 1996).
There are a number of principles outlined by the ‘What Works’ approach to which
adherence is thought to maximise the impact of treatment intervention. These include
the principles of risk, need and responsivity (Andrews and Bonta, 2002; McGuire and
Hatcher, 2001). In other words, the amount of intervention an offender receives must
be matched to their propensity/risk of reoffending, interventions must target specific
needs of offenders known to be associated with criminal behaviour (‘criminogenic
needs’) (Andrews and Bonta, 2002) and the delivery of interventions must be matched
to their learning styles. In support of this, Hollin (1999) suggests that treatment that
engages offenders and is relevant to them will increase responsivity and thus maximise
166
Understanding ‘What Works’: Accredited Cognitive Skills Programmes for Young Offenders
the potential for reducing reoffending. Programme integrity is a key component of
successful interventions (Hollin, 1995), and includes factors such as staff training and
motivation, the existence of and adherence to a programme instruction manual and
ongoing programme monitoring and evaluation.
There is a growing body of evidence regarding the outcome of cognitive-behavioural
interventions – ultimately a reduction in recidivism (Robinson and Porporino, 2000).
A substantial proportion of this evidence is derived from meta-analyses,1 and one of
the most recent looked specifically at the effectiveness of behavioural and
cognitive-behavioural intervention in reducing recidivism (Pearson, Lipton, Cleland and
Yee, 2002). Both adult and young offender populations were included and such
interventions were found to be associated with reduced recidivism in both prison and
probation settings. Regarding outcomes specific to young offender populations, a
substantial number of studies have indicated that cognitive-behavioural interventions
can reduce future recidivism. Lipsey (1995) reports findings from a meta-analysis of
nearly 400 comparison group studies of treatments targeting delinquency reduction
(both custodial and community-based) designed to reduce delinquency amongst
12–21-year-old offenders. These included cognitive behavioural programmes, which
were found to be among the most effective. An overall reduction in recidivism of ten
per cent was found (although this was across all interventions considered, not just
cognitive behavioural programmes). Izzo and Ross (1990) analysed 46 juvenile offender
studies and found that programmes incorporating a cognitive component were more
than twice as effective as those which did not. Redondo, Sanchez-Meca and Garrido
(1999) carried out a meta-analysis of 32 European studies which examined the
effectiveness of offender treatment techniques across both adult and young offender
populations. They found that success of cognitive-behavioural programmes was greater
with the latter, although adult populations also benefited more from these programmes
than alternatives.
HM Prison Service introduced cognitive-behavioural treatment intervention for
offenders held in custody in England and Wales in the early 1990s. Two such
cognitive-skills programmes were introduced, Reasoning and Rehabilitation (1992) and
Enhanced Thinking Skills (1993). Reasoning and Rehabilitation (R&R) was developed
during the 1970s (Ross and Fabiano, 1985; Porporino and Fabiano, 2000) and first used
with Canadian offenders. It has been adopted in a number of countries for use with
both custodial and community populations presenting a high risk of reoffending.
Enhanced Thinking Skills (ETS) was developed by HM Prison Service, and is an
adapted version of R&R, used with lower risk offenders.
The benefits of cognitive-behavioural intervention with young offenders in England
and Wales have not yet been expressly evaluated. With regard to imprisoned young
offenders, work outside of England and Wales has indicated that R&R can have
positive effects upon reoffending. For example, Garrido and Sanchez (1991) provided
early evidence that R&R could produce beneficial results with imprisoned Spanish
juveniles (programme participants, n:14, control group, n:17). Similarly, Murphy
1 Meta-analysis is a statistical technique for combining the findings from independent studies. It gives an estimate of overall effect
size, giving due weight to the sample sizes of each study included.
Youth Justice Vol. 5 No. 3
167
and Bauer (1996) found that amongst US imprisoned juveniles, R&R participants
(n:33) had a 16-month reconviction rate of 67 per cent compared with 83 per cent
for a comparison group (n:16). Although the sample sizes in these studies are small,
the findings indicate some early success for R&R with young offenders.
However, the criminological and psychological literature is not entirely supportive of
the efficacy of cognitive behavioural programmes such as R&R in reducing offending
among young offender populations. Whitehead and Lab (1989) for example, concluded
that none of the 50 programmes identified in their meta-analysis, which included
cognitive behavioural treatment, showed particularly encouraging results, and that
correctional treatment as a whole had little effect on recidivism. More recently, a
Colorado Division of Criminal Justice evaluation of R&R delivered to juvenile offenders
found that programme participants showed a 15 per cent higher recidivism rate than a
control group (Pullen, 1996). In general, international experiences of evaluating offender
treatment programmes, including cognitive behavioural examples, have indicated that
reductions in reconviction rates tend to be extremely variable (e.g. Bonta, 2002).
The current study describes the findings from the first evaluation of both R&R and
ETS programmes run with male young offenders (including juveniles) held in custody
in HM Prison Service, in England and Wales. Reconviction is used as the outcome
measure and the impact of programme participation is assessed according to the risk
of reconviction (Beech, Friendship, Erikson and Hanson, 2002).
Method
Participants
The intervention group consisted of 1,534 male young offenders2 who had voluntarily
participated in a prison-based cognitive skills programme, either R&R or ETS, between
1995 and 2000. All had spent at least one year in the community following discharge,
subsequent to a custodial sentence of six months or more, and had been released
between 1996–2000 inclusive. Those receiving a life sentence were removed from the
sample as a reoffence would constitute an infringement of the life licence and thus
would lead to direct recall to prison rather than a conviction through the court system.
Included in the sample were those offenders who began but did not complete a
programme (dropouts). These represented 14 per cent of all programme starters
(n:220). Of those offenders for whom the reason for dropout was known (n:134),
41 per cent had left a programme through their own choice. Other reasons included
transfer to an...
To continue reading
Request your trial