The use of cognitive behavioural therapy offender behaviour schemes as a means of rehabilitation in the HM Prison service. Does it work? Why was it introduced?

Cognitive behaviour treatment or therapy (CBT) began life as a psychological tool for addressing mental “disorder” and rapidly gained favour with the medical profession as an adjunct to and eventually the preferred substitute for interpersonal therapy and clinical intervention. Its potential for effecting modifications in the attitudes and behaviour of offenders was recognised and began to be practised in earnest in this jurisdiction in a correctional context in the early 1990’s initially by the Probation Service as an initiative in the community. However, it was rapidly adopted by the Prison Service for use in a custodial setting. It has been seized upon by those responsible for criminal justice and the penal system as a potentially cost-effective and socially desirable means of reducing delinquent behaviour (specifically recidivism since in almost every case the subject of the treatment will only become eligible for it as a result of having already offended, usually at a serious level meriting incarceration or participation in a community punishment as an alternative to a custodial sentence).

Custom Criminal Law Essays Order

In order to evaluate the claims that are made for the use of such methods within prison, it is necessary first to examine what CBT is and what theoretical use may be made of it as a means of correcting delinquent behaviour and thus reducing reoffending. To determine the true validity of this use of CBT, it will be necessary to examine its implementation in a correctional setting. Thereafter, in order to assess its validity and the merits of its use, rigorous analysis must be applied to such evidence as is thus far available as to the results of its employment. This process is hampered by the inadequacy of current data, deficiencies in the statistical methods of measuring the outcomes of CBT programmes and a degree of political and institutional bias which may have the effect of distorting an appreciation of its true value. Nonetheless, it is possible to scrutinise a number of studies which have thus far been published in order to determine whether there is any true empirical basis upon which to suppose that this treatment is capable in theory and, more importantly, in fact of having any significant impact upon the acknowledged need to reduce reoffending. However, sight must not be lost of the fact that such studies taken in isolation may result in inappropriate conclusions. For this reason it will be necessary to have regard to other evidence - of a qualitative rather than a purely quantitative nature in order to determine whether CBT is in reality “What Works”.

Definition

It is stressed by Vennard et al that “cognitive behaviourism is not a unified, distinct psychological theory or method but a term given to a range of interventions” derived from three distinct psychological theories:

1) Behaviourism concentrates upon the role of external factors in influencing the behaviour of the individual; thus, for example, peer pressure or a perception that delinquent behaviour will not produce an adverse reaction from those in authority are capable of reinforcing such patterns of behaviour ;

2) Cognitive Theory emphasises the role of the individual’s thought processes such as memory and reasoning in shaping the ultimate actions of the individual ;

3) Social Learning Theory tends to combine the above two approaches. While it is accepted that external factors such as the environment are influential (behaviourism), it is argued that such factors as the observations of the actions of others and, in particular, the observed consequences of those actions may also influence behaviour as a result of the operation of cognition .

It is entirely natural, therefore, that such psychological theories should be applied to an analysis of the behaviour of offenders. In consequence, it is assumed that criminal behaviour is the result of environmental influences and/or the failure of the offender to acquire cognitive skills appropriate to shape his/her behaviour within society and/or the learning of inappropriate lessons from observing the behaviour (and outcomes thereof) of a delinquent peer group. However, it is argued by McGuire that the cognitive behavioural approach does not ascribe criminal behaviour solely to individual psychological attributes but must also take into account the prevailing social conditions (including the effect of a peer group) that will affect individual development. Therefore, because it is considered that delinquent behaviour is learned rather than innate, it is supposed that programmes of behaviour modification can be developed which will address the “cognitive deficits” of the individual: specifically, that the individual can be taught insight into their understanding of and reaction to the world about them, that they can learn the inappropriateness of certain types of behaviour and that they can thereby be equipped with strategies for dealing appropriately with certain types of challenge and responding in a non-criminal manner to a given situation or opportunity. It is thought that methods can be learned for dealing with situations which might provoke anxiety and anger and otherwise give rise to delinquent behaviour (“positive coping responses”). Thus there is an emphasis upon the need for self-evaluation and a resultant self-control in responding to situations which might hitherto have given rise to (further) offending.

A more narrowly defined sub-category of cognitive therapy is therefore to be found in social skills training. This is a form of behavioural therapy which is designed to address character deficits such as low self-esteem and an inability to interact appropriately with other adults or younger people. It will be seen that this may be of particular relevance when addressing the behaviour of, for example, sex-offenders or young people who abuse or react violently to particular social pressures as a result of a pre-existing inadequacy in their character which prevents them from behaving in accordance with the norms which society imposes upon the law-abiding, “responsible” adult. The aim is to equip the individual with a set of skills to allow them, for example, to accept appropriate guidance and correction from adults or others in authority and to resist peer pressure which might otherwise cause a relapse into delinquent behaviour. Practical techniques thus employed might include self-visualisation exercises or role play. Hollin has identified certain key elements which are commonly incorporated into such an approach: problem-solving training; social skills; anger control and moral education. As will be seen, this approach has given rise to a degree of controversy when an attempt is made to evaluate the effectiveness of such programmes since it may be difficult if not impossible empirically to distinguish between the efficacy of such distinct individual elements when they are collectively subsumed into a “multimodal” cognitive treatment programme.

Further, it should be noted that notwithstanding the enthusiasm with which the correctional services have embraced CBT, considerable doubts as to its efficacy remain within the medical profession. Writing in the British Medical Journal on 2nd February 2002 , Jeremy Holmes, a Consultant Psychiatrist in Pyschotherapy opines:

“Cognitive behaviour therapy undoubtedly has much in its favour. It is an attractive, efficient therapy that is relatively easy to learn and deliver and produces good results in many instances. In addition, cognitive behaviour therapy researchers have set standards in detailed descriptions of their methods (“manualisation”), monitoring of adherence, and tailoring treatments to specific disorders that have had a major impact upon psychotherapy practice and research generally.”

However, he warns of a number of deficiencies:

  • The foundations upon which CBT stands “may not be as secure as some of its proponents would have us believe”. In a major National Institute of Mental Health Survey, now some 20 years old, it fared less well as a treatment than interpersonal therapy and clinical management plus medication;
  • The impact of CBT in the long term is uncertain;
  • While CBT appears to function well in clinical trials, there is some doubt whether it is equally effective in real situations than in much-vaunted clinical trials;
  • Even some leading CBT practitioners are beginning to acknowledge limitations in its application.

It may well be in view of the scepticism that may be applied to results in correctional programmes that these medical doubts will have in the future to be given greater credence.

Implementation of CBT Programmes in the Correctional Services: The Beginnings

The first fully-evaluated programme in England and Wales to apply these principles to offenders commenced in the Probation Service in Mid-Glamorgan in 1990. The programme was known as “Straight Thinking on Probation” (STOP) and included modules on problem solving, social skills, management of emotions, negotiation skills, critical reasoning, creative thinking and values enhancement. These were delivered to probationers over some thirty-five sessions of two hours each. The outcome is summarised by Maguire et al :

“…there was some evidence of fairly short term reductions in offending (35 per cent of programme completers were reconvicted in a year, compared to a predicted rate of 42 per cent; in contrast, a custodially sentenced comparison group with the same predicted reconviction rate showed that 49 per cent were reconvicted within a year of release). There were also more persistent reductions in more serious reoffending among those who completed the programme. These were associated with reported changes in attitude, thinking and behaviour consistent with the rationale of a cognitive-behaviour programme [emphasis supplied]. On the whole, STOP appeared to offer a more effective and constructive sentencing option than other likely sentences for this group of relatively serious and persistent offenders.”
However, initial enthusiasm for even this comparatively modest degree of apparent success should be curbed. There was apparently a bias in selection towards those at high risk of a custodial sentence rather than an attempt at selection made on the basis of tailoring the needs and ability to respond of the subjects to that which the programme had to offer. Nonetheless, this and similar localised initiatives prompted the holding of the first ‘What Works?’ conferences in order to disseminate the concept among managers, practitioners and researchers. Initial implementation and progress within the Probation Service was faltering and would appear to have been hampered by structural changes in the Service. Meanwhile, although commencing such programmes later than in Probation, the Prison Service began to put in place a well-managed and monitored programme of CBT which although initially intended for the rehabilitation of sex offenders were later expanded to the general prison population . The success of these early Prison Service programmes is arguably attributable to a more rigorous Quality Assurance System based upon the use of an Accreditation Panel “with international expert membership to ensure that only programmes with a convincing evidence-base were used, with a comprehensive audit system to inspect the delivery of programmes and ensure integrity”.

Background to Implementation

The practical and theoretical interest in the application of CBT to offending behaviour was drawn together and endorsed by the influential 1997 Home Office Research Study No.171 . This started from the premise that over the preceding decade research had suggested that some forms of rehabilitative work could be successful in reducing reoffending. It was acknowledged that among the range of methods available for working with offenders those falling under the banner of “cognitive-behavioural” were increasingly favoured. It was concluded:

“[Such methods] are widely viewed as offering considerable advantages over more traditional forms of intervention”. Again, the difficulty of defining the exact scope of CBT was acknowledged but it was described as involving “helping offenders to face up to the consequences of their actions, to understand their motives, and to develop new ways of controlling their behaviour”. The first section of the Study sought to arrive at conclusions from the research to date and concluded (pp.vii-viii):

  • “The effectiveness of interventions with offenders varies significantly according to the type of approach adopted. Programmes which seek to modify offenders’ patterns of thinking and behaving are generally more successful than techniques such as group or individual counselling and non-directive therapy.” This is a somewhat trite observation. It is to be expected that a programme which sets out to modify the way in which an individual observes, thinks about and reacts to the world will have a greater effect than les specific regimes which do not possess a targeted outcome and which may involve little more than “befriending” the offender.
  • Even these approaches rarely produce major reductions in reoffending among offenders in general. This supports the view that programme components and styles of delivery need to be closely matched to the needs and learning styles of particular offenders.” This conclusion reflects the criticism of the STOP programme discussed above and supports the view that CBT cannot be regarded as a universal panacea. This is examined in greater detail in the discussion of the types of offender which might predictably benefit from CBT below and foreshadows the next conclusion, namely,
  • The use of cognitive-behavioural methods in programmes which also include training in social skills (for example, teaching sex offenders to form appropriate relationships with other adults) show the most positive results with both juvenile and adult offenders.” This underlines two factors of note: first, the reorganisation of cognition will not, of itself, necessarily result in appropriately modified behaviour unless in addition to the acquisition of a revised insight, the individual is also equipped with appropriate techniques for reacting in accordance with such “corrected” appreciation of a particular scenario in a way which is appropriate to the particular social setting; second, and perhaps more interestingly, there is not found to be a necessarily higher rate of success in terms of the application of this particular method between adults and young people (who might otherwise be supposed to be more susceptible to such adjustment on the grounds of their relative immaturity and the fact that attitudes and behaviour have had less opportunity to become entrenched.
  • Lack of agreement about how to classify the range of techniques encompassed by the cognitive-behavioural approach means that it is not possible to identify which particular forms of intervention are most promising.” This is a recurrence of the criticism of the lack of specific distinction between the different techniques which are placed under the umbrella of the CBT “label”. Again, it will be demonstrated below that this difficulty represents a severe impediment to meaningful evaluation and, in consequence, may gravely impair effective initial implementation.
  • The research shows that the impact of cognitive-behavioural methods of working with offenders can be enhanced by -
    - targetting on factors which have contributed to the offending behaviour;
    - using active, participatory, problem-solving methods of working;
    -matching the intensity and duration of intervention to likely future risk of offending;
    - running programmes as intended, without allowing the goals and methods to shift over time.” It is tempting to be cynical as to the content and expression of this final conclusion: it might be argued that it is obvious that the effectiveness of any form of intervention will be determined by its appropriateness and consistency!

Nonetheless, the authors of the Study proceed to make a number of recommendations to those involved in the management and delivery of offender programmes on the basis of their admittedly “positive, but inconclusive” finding. Principal among these is the need for researchers and practitioners working with offenders to develop and agree the definitions for the various components of CBT methods. Offenders should be allocated to programmes not on the political basis of their likely “tariff” but rather in accordance with an assessment of their particular needs and learning/personal development potential. It is conceded (p.viii) that although cognitive-behavioural methods can yield the most successful outcomes, they are unlikely to work unless other principles which have been shown to succeed in respect of individual elements of the overall programme are followed. Somewhat scathingly, it was concluded that the effective use of CBT required a fuller knowledge of the underlying principles, theories and techniques than most practitioners (with offenders) currently possessed. Particular emphasis was placed upon the need for evaluation of such programmes:

“Programmes…need to make explicit the underlying theory as to how they expect to effect change, the content and method of intervention and the framework for implementation [emphasis supplied]”.

Thus it may be observed that even at the early stage of development of such programmes, doubts were being expressed, if not as to their potential for ultimate efficacy, certainly as to whether CBT represented a sufficiently refined concept to be capable of successful implementation. Perhaps it was this lack of focus that gave rise to the early concerns as to the ability of practitioners successfully to implement such programmes. It may be argued that this ambivalence is reflected in the relative success or otherwise of CBT as later quantified in terms of preventing reoffending as described below. However, it may be the case that it was an initial failure adequately to define the scope and potential of CBT that gave rise to an arguably unimpressive “success” rate. Alternatively, the possibility must be conceded that the failure sufficiently precisely to define the exact operation of CBT has given rise to an ex post facto explanation of the performance of what might in any event have been a concept upon which unfair expectations were heaped from the outset.

Implementation of CBT in Prison

Debidin and Lovbakke argue that in response to the ’What Works?’ movement, the correctional services in England and Wales began in the 1990’s to adopt offending behaviour programmes “as one of the cornerstones of their rehabilitative work”. Accordingly, in 1999, a three-year Crime Reduction Programme (CRP) costing some £250 million was adopted. The main purpose of the CRP was to obtain evidence of the most effective methods for reducing crime and had as its scope a wide range of initiatives pursued by the police and their criminal justice partners including, in particular, the Police and Probation Services. HM Inspectorate of Probation supported the use of offending behaviour programmes in influential reports and these became the foundation for future CBT work within the two services. These identified some by now familiar factors for such programmes to be effective; they should be:

  • “Multi-modal”, i.e. employ a variety of techniques to address the perceived needs of individual offenders;
  • Skills-oriented;
  • Based on the established theories and methods of CBT;
  • Ideally but not invariably community-based;
  • Delivered by staff “adept at pro-social modelling”
  • With effective monitoring and evaluation built-in from the outset.

As observed above, the Prison Service enthusiastically espoused the last of these principles and had previously (in 1996) introduced an accreditation process for its programmes. The General Accreditation Panel was created to formalise the process of implementing programmes in the Prison Service. As a result of the Guide to Effective Practice (Op. Cit.) re-emphasising the need for accreditation, there was established the Joint Prisons/Probation Accreditation Panel which was duly reconstituted as the Correctional Services Accreditation Panel (CSAP) in 2002.

Upon publication of The Impact of Corrections on Reoffending (Op. Cit.), “currently the Prison Service lists nineteen fully or provisionally accredited programmes (including seven drug-treatment programmes) and six programmes not yet accredited or under development. Additionally, two Therapeutic Communities have been accredited in the Prison Service. The Probation Service lists sixteen fully or provisionally accredited programmes (including two drug treatment programmes) and another two under development. As of 2004, the HM Prison Service website lists five categories of programme which can be seen to be largely common to both services:

  • Enhanced Thinking Skills; ETS attempts to develop skills to enable offenders to behave more appropriately. It does not directly address offending behaviour but aims to equip offenders to tackle life’s challenges more effectively.
  • Reasoning and Rehabilitation; This is based on the premise that criminal behaviour is based on the way offenders think, the attitudes they develop and how they interact with others. It assists them to develop skills in effective problem-solving, social perspective, forward thinking and critical and moral reasoning.
  • Think First; Arguably a multi-disciplinary successor to STOP (see above).
  • Controlling Anger and Learning to Manage it (CALM); This course is designed to meet the needs of offenders whose criminal behaviour arises from poor emotional control.
  • Cognitive Self-Change Programme; (see below).

As of 1st April 2004, accredited programmes were being delivered in 112 out of 137 prisons in England and Wales. Fifty-four of these deliver more than one programme, usually a general CBT programme, a sex offender treatment programme (SOTP) and often a third. By contrast, all 42 probation areas of the National Probation Service delivered a general offending behaviour programme (one of four cognitive behaviour programmes) and one or more other programmes addressing aggressive behaviour or substance abuse. These statistics may be argued to betray the continuing bias (apparent in the early stages of development of CBT principles) towards considering such programmes as more appropriate for and more likely to succeed in the community.

The early stages of implementation of CBT programmes by the correctional services might be characterised by a degree of competitiveness between Prisons and Probation possibly exacerbated by the perception discussed above that although Probation was the first to seek to employ such techniques in a co-ordinated fashion, early attempts at accreditation and evaluation might be regarded as lacking in consistency whereas the Prison Service having “entered the fray” at a later stage rapidly came to be regarded as the premier exponent of this method of addressing offending behaviour. Indeed the existence of such tension might explain the insistence of certain commentators upon the fact that while the use of CBT in prisons could not be altogether discounted, its value and efficacy lay primarily in the context of a community sentence. Happily, much of this territoriality has now been displaced by an appreciation that cognitive behavioural modification is properly to be regarded as a process of evolution that can and should span both periods of incarceration and subsequent release into the community on licence. An accredited scheme, the Cognitive Self-Change Programme (CSCP) was implemented in May 2005 . This is described as a programme for “high risk male offenders who have repeatedly committed offences of serious violence. It operates on the principle that the programme is commenced but not completed in prison and that offenders continue the programme post-release by completing what is known as the “Community Component” or “Block 6”. The programme is currently available in 6 male prisons. Thereafter, offenders completing Block 6 are subject to Multi-Agency Public Protection Arrangements (MAPPA). The targeted offenders are those whose repetitive use of violence is part of a general pattern of anti-social behaviour and criminality. In the words of Probation Circular 36/2005:

“The programme aims to reduce violent recidivism by changing offenders distorted thinking processes and individual patterns of antisocial thinking which lead them to violence and criminal conduct, and by reducing the impact of contributory violence risk factors specific to each individual.”

The programme is predicated upon specific “dynamic risk factors” which are to be the subject of attempted adjustment [emphases supplied]:

  • Individual thinking patterns which lead to violence and other anti-social acts;
  • Lack of insight into violent behaviour;
  • Violent fantasy;
  • Poor management of increased arousal or anger;
  • Socio-cognitive skills deficits (such as problem-solving and impulse control).

The programme is long and intensive and comprises some 136 to 257 group sessions spanning 8 to 17 months accompanied by an average of 26 one-to-one sessions in the participating CSCP prison followed by monthly sessions until release of transfer to a non-CSCP establishment. Each of the 6 blocks builds upon the last aiming to provide progressive development up to, through and beyond the date of release. A consideration of the current content of each block demonstrates the application of the CBT principles described above:

1. Foundation sessions that develop skills for the objective observation of thoughts, feelings, attitudes and beliefs. Key tools are introduced;
2. Identifying and understanding the connections; how thoughts, feelings, attitudes and beliefs lead each individual to violence;
3. Identifying new thoughts and intervention strategies, boosting cognitive skills and developing skills for managing individual violence risk factors;
4. Creating a relapse prevention plan and consolidating interventions;
5. Practice in the prison setting;
6. Practice in the community and consolidation of the relapse prevention plan.

The final block consists of a structured intervention in the community after release. Its aim is to facilitate and monitor the post-release practice of the skills learned in prison. An “Offender Manager” is responsible for the delivery of Block 6. The Circular sounds a salutary warning of the potentially adverse implications of lack of resources upon completion of the programme:

“If [Probation] areas do not provide for the delivery of Block 6, difficulties in devising effective risk management plans for offenders on CSCP will impact upon Parole Board decisions concerning their release.”, therefore:

“Inconsistency in the delivery of Block 6 will inevitably have a discriminatory impact upon offenders”.

Thus it may be concluded that in this as in other areas of CBT, the best of intentions and the most effective of methods will be undermined if resourcing is insufficient to keep pace with the dictates of theory.

Assessing Outcomes - Methodology

Before attempting to assess the impact of CBT by reliance upon statistical information, it is essential to consider the methodology which might be employed and the potential deficiencies inherent in such a process. A comprehensive discussion of methodology may be found in The Impact of Corrections on Reoffending (Op. Cit., pp.4 et seq.) While the methodology describes forms the essential precursor to the analytical findings of that comprehensive study, the techniques described may then usefully be applied to a criticism of further Home Office “Findings” published in connection with the efficacy of CBT. The point is well made that:

“Not all evidence is of the same quality and the experience of the last few years has highlighted some deficiencies in the reducing reoffending evidence.”

It is acknowledged that while there is a growing base of evidence arising from ‘What Works’ initiatives in this country, a great deal of evidence in respect of reducing reoffending originates from the United States and Canada. Such sources (while of interest in general terms) should be regarded with distrust given the different socio-economic conditions prevailing in each jurisdiction; in particular in terms of prison regimes, intervention programmes and the labour markets and welfare support systems which will be encountered upon release. It should also be noted that even where there are superficial similarities between prison regime and prevailing social conditions, it must be regarded as almost impossible to compare the detail of intervention programmes in different jurisdictions due to the nature, extent and presentation of the available evidence. Accordingly, outcomes may vary dramatically as a result of factors which are not fully reported or not properly regarded as comparable to similar-seeming methods as described in another penal system. For this reason it is submitted that while there can appropriately be trans-national agreement in principle upon the efficacy or otherwise of a particular therapy, an attempted comparison of the relative success of the outcome of programmes from different criminal justice regimes would largely lack validity and therefore will not be attempted herein.

For any attempt at measuring the outcome of CBT programmes to have any practical validity, it must attempt a practical analysis of whether the intervention has succeeded against the desired measure of whether it has achieved the objectives sought by the correctional services, namely, a real and measurable reduction in reoffending. A helpful “Scientific Methods Scale” against which to grade the validity of research into such outcomes has been developed by Sherman et al in a study for the US Department of Justice. This is set out in The Impact of Corrections on Reoffending (Op. Cit., p.6, Table 1.1) this has been refined by Friendship et al for use in reconviction studies and consists of 5 levels rising according to the perceived efficacy of the measure:

1. Relationship between intervention and reconviction outcome - intervention group only studied; no comparison or “control” group;
2. Expected reconviction rates compared to actual reconviction rates;
3. Performance of comparison group against intervention group without direct connection or comparability between the two (unmatched comparison group);
4. Comparison group and intervention group matched on the basis of perceived similarity (well-matched comparison group);
5. Random assignment of offenders to intervention and comparison conditions.

It will be noted that Level 5 is demonstrably the most valid test since it offers the least likelihood of other factors influencing the outcome and a completely random comparison between the performance of members of each of the subject groups. It is to be noted that reconviction studies in England and Wales have rarely achieved a Level 5 standard. Beyond this, it should be noted that reconviction may have limitations in itself as a measure since even if all other dynamic factors can be controlled and a consistent and regular comparison achieved, there may be other influences at work. The incidence of reconviction will be dependant upon detection and apprehension and cannot therefore represent an absolute standard of reoffending which is the true outcome which such studies are attempting to measure. Further, changes in police practice and prosecution policy may effect further differences in the relationship between reoffending ad reconviction. Finally, reconviction as a measure is an any event something of a “blunt instrument” since on its own it takes no account of any changes in the frequency and severity of subsequent reoffending and, as an absolute test of success or failure can tell us little or nothing of the extent to which the attitude and behaviour of the offender has been influenced by a given programme.

That said, it is important not to be entirely dismissive of measuring reconviction as a methodology. It should be borne in mind that reconviction is systematically and accurately recorded and that, for all its limitations, its is an accessible and empirical measure. Cooke & Michie argue that it can and should be used to be used to inform criminal justice policy; they identify four areas of merit:

  • Reconviction may act as a guide to sentencing policy in at least a crude fashion; for example, by comparing the outcomes after a custodial sentence as against a community-based disposal;
  • It provides at least a simple baseline to guide different regimes of treatment of offenders in custody or more specific interventions;
  • It may act as a broad indicator of the impact of changes in prosecution or sentencing policy;
  • It provides actuarial information which can be employed for the purposes of risk assessment (although it may be argued that this could represent little more than a “crutch” upon which those who are held accountable for changes in penal policy or responsible for decisions as to the release of offenders might lean!)

Outcomes - “What Works?”

It must be appreciated that just as the implementation of CBT programmes is still in a state of comparative infancy, the use of reconviction data to evaluate the efficacy of the same is therefore inevitably relatively immature. Many reconviction studies have yet to have sufficient time to run to be regarded as valid and such studies are further hampered by the fact that reconviction rates two years after release from prison are only available three years after the offender has been on the programme. Nonetheless some results are beginning to emerge. Friendship et al conducted an evaluation of the Reasoning and Rehabilitation and Enhanced Thinking Skills programmes in 30 prisons. The treatment group consisted of 670 adult male offenders serving a custodial sentence of two years or more who had voluntarily participated in either of the two programmes. Their reconviction rates after treatment were compared with a group of 1801 adult male offenders serving sentences of comparable duration. A reduction in the reconviction rate of 14% was found in respect of low to medium risk offenders and some 11% in respect of high risk offenders. There was a minor difference in the direction of the result in respect of low and high risk offenders when they were treated as distinct groups but this was not regarded as of overall significance. In summary, the results appeared to demonstrate that reconviction rates in both categories fell after participation in R & R and ETS programmes. This finding was hailed by the Home Office on the basis that:

“Based on the number of prisoners expected to complete a cognitive skills programme in 2002-2003, this reduction represents almost 21,000 crimes prevented.“

Again, however, it might be argued that there is a problem in the initial selection: while such results appear encouraging, they may be devalued by the failure to take into account differences in the groups between which comparison was made. Admittedly, an honourable attempt at matching was made: a profile of the treatment group was drawn up using 6 empirically matching variables:

  • Sentence length;
  • Year of discharge;
  • Number of previous convictions; and
  • Possibility of reconviction

This profile was used to identify a large group of untreated offenders using the Inmate Information System and each offender was matched to the treatment group as closely as possible (although it is accepted that there still remained significant variations such as an age within a range of 5 years). Then, for each treated offender, three untreated offenders were randomly selected. It is submitted that in addition to the inevitable variations which would nonetheless occur, a striking disparity between the two groups exists in the fact that participation in the programmes was voluntary; further, one of the stated criteria upon which members of the treatment group were selected was “motivation to participate”. Given that the entire ethos of CBT is the adjustment of attitudes and behaviour, such an obvious willingness to participate must, it is argued, be capable of having a significant impact upon outcome. The initial motivation of the vast majority of the treatment group must - as a matter of common sense - be greater than that of almost ever member of the control group. With respect to the distinguished authors, the claims made for the results contained in Home Office Findings No.161 namely:

 Vindication of the work of the Joint Prison Probation Accreditation Panel in ensuring that standards of programme and design are maintained across the prison estate; and
 That the study “forms the basis for cost-effectiveness analysis of offender programmes” which can be calculated by putting their cost to the Prison Service against the savings to the community and criminal justice system through lower reconviction rates and a reduction in offences committed;
may be regarded as at best somewhat ambitious and at worst outrageously self-serving.

A second study by Falshaw et al is understandably less bullish. This study involved a treatment group of 649 adult male offenders who had voluntarily participated in an accredited prison-based cognitive skills programme (again, either R & R or ETS) between 1996 and 1998. The comparison group consisted of 1947 adult males who had not participated in a cognitive skills programme during their custodial sentence. All offenders had received a custodial sentence of 6 months or more (but less than life) and at least two years had passed since their discharge from prison. The basis of selection of the comparison group was the same as in the previous study with an attempt to create as close as possible a match generating some 3 comparison group participants for each member of the treatment group. In this case the match variables were subtly different including ethnicity and offence type. The expected reconviction rates were similar for both the comparison and treatment groups and to this extent the samples were well matched. To the probable dismay of the Home Office, the two-year reconviction rates between the comparison and the treatment groups showed no significant differences in any of the risk bands. It was concluded:

“These results were unexpected because an earlier evaluation of the programme for the period 1992-1996 (before accreditation) had shown that treated offenders had lower reconviction rates than matched comparisons in all risk groups.”

A somewhat desperate exercise was then embarked upon in an attempt to identify further factors which might explain the differences between the findings of the two studies. Such further analysis examined the following:

  • The impact of “dropouts” from the programme;
  • Time between treatment completion and release;
  • The effect of commencing the programme nearer to the beginning of the accreditation period;
  • The differences between R & R and ETS;
  • Time to first reconviction;
  • Frequency of reconviction.

all to no avail! This represents a particularly stinging rebuke to the accreditation process which had been highly praised in the earlier study and which it was expected would therefore be responsible - by dint of further reform and implementation - for an even greater “success rate”. Accordingly, the search began for excuses. It was suggested that recent international experience of evaluating offender programmes showed that reductions in reconviction rates are extremely variable. It was argued that while the early cognitive skills programmes adhered to ‘What Works’ research evidence, the introduction of accreditation of prison cognitive skills programmes was expected to improve the quality of programme delivery and therefore effectiveness and that it might therefore be the case that the high level of motivation shown by staff who volunteered to deliver the early programmes may have had an impact on the initial evaluation results. Further, prior to accreditation, attendance upon such a programme was not part of an offender’s sentence plan and was less likely to influence early release from custody. While attendance remained voluntary, the potential impact upon the prospects of release may have meant that while individuals were still motivated to volunteer to attend, their motivation to change may not have been so great. Resources were singled out for potential blame: this period of research coincided with a dramatic expansion in the implementation of cognitive skills programmes (from 30 prisons with 746 programme completions during the earlier study to 72 prisons and 2837 completions in the latter. It was opined:

“There is little published evidence available on the effect of large scale expansion but speculation among experts (!) suggests that treatment quality might be compromised.”

Worst of all, it was argued that the closer matching of the treatment and the comparison groups between the first and second study - previously the cause for actuarial self-congratulation - might have been responsible:

“…whilst the current groups of offenders were well matched on a small number of static risk factors they could differ on dynamic risk factors. These include level of thinking skills and motivation to change as well as other dynamic risk factors relevant on release such as housing, employment and drug use.”

While on one level the results may appear disappointing and the attempts to vindicate the validity of the earlier study by explaining the latter as an aberration may even be regarded as risible, this apparent “set back” may in fact give rise to a much-needed refinement of the evaluation process. At the conclusion of Home Office Findings No.206, the authors are forced to rephrase the now familiar question as, “What works with whom?” As has been observed above, the efficacy of evaluation based upon too narrow and limited a range of criteria is highly questionable. Hitherto, the individual characteristics of the offenders of the type which might make them more highly motivated to change or more susceptible to the treatment to which they were subject had not been given significant weight if indeed they were taken into account at all. It was declared:

“Research is planned to investigate “What works with whom” by monitoring the changes an offender makes during programme participation. The characteristics of the offenders who make the most progress will be identified.”

A third study was undertaken by Cann et al . Again the subjects were those who had undertaken R & R and ETS programmes but on this occasion the treatment groups consisted of 2195 adult males and 1534 young offenders. These were closely matched to comparison groups - this time of the same numbers but again anticipated two-year reconviction rates were very close. Again the overall result may be regarded as disappointing: there were no differences in the one- and two-year reconviction rates between the adult males who started a prison-based cognitive skills programme and their matched comparison group just as there was no difference between the one- and two-year reconviction rates of young offenders who started such a programme and their comparison group. However, when programme dropouts were excluded from the analysis, the one-year reconviction rate for both adult men and young offender programme completers were “significantly” lower than for the matched comparison groups. There was a 2.5% difference in reconviction for the adult male programme completers and a 4.1% difference for the young offender programme completers. This is to be contrasted with the lack of effect of the exclusion of programme dropouts upon the results of the second survey. Disappointingly, however, it was found that these differences in the reconviction rate were not sustained at two years post-release. The study was in conclusion forced once again to be candid as to the defects inherent in the matching process and the fact that possibly not enough was known about certain characteristics of the offenders to make the analysis of comparison between the two groups meaningful:

“This study was limited by the act that the comparison groups were matched to the programme participants on a small number of static demographic and criminal history risk factors only. There may be differences between the programme participants and the comparison groups in dynamic risk factors, such as cognitive deficits and motivation to change, but this data was not available for the comparison groups. Other dynamic risk factors relevant on release from prison were not recorded for either programme participants or comparison groups, e.g. participation in probation programmes, employment and drug use.”

Further, notwithstanding the defence of the use of reconviction date by Cooke and Michie (supra), it concluded:

“This study also relied on official recorded conviction data which is only a proxy measure of reoffending (Lloyd, Mair & Hough, 1994 )”.

By contrast with the self-congratulatory style of Home Office Findings No.161, this study concedes that “the Prison Service is still in the process of understanding ‘What Works’ with offenders and identifying those who would benefit the most from programme participation”. The current programme of research includes a qualitative evaluation of cognitive skills programmes. This is expected to provide some information regarding the levels of dynamic need in programme participants such as motivation to change. It is anticipated that this will help to identify relevant dynamic variables for comparison group matching.

It is instructive to make a comparison between the above studies and the analysis conducted in respect of the pre-accreditation R & R programme carried out by the Probation Service in Mid-Glamorgan described above which it will be recalled was hailed as the forerunner of subsequent corrective cognitive therapy programmes. In the instance of that STOP programme, subsequent analysis found favourable differences between predicted and actual reconviction rates for each group . It has been conceded that while the results of this study might be read as generally supportive of an association between treatment and less serious reoffending in the short term, “the numbers included in this study are too small for tests of statistical significance to be meaningful and the results might be explained by other differences between the groups apart from participation in treatment”.

Some limited encouragement for the champions of CBT in prison may perhaps be derived from an early study of pre-accreditation sex offender treatment in prison which ran in 23 prisons in England and Wales between 1992 and 1994. This study found a significant reduction in reconviction when sex offence and violent reconviction rates within two years of release were combined. Overall this resulted in a 3.5% reduction in reconviction for sex offence or violent reconvictions. However, once again while such a finding is potentially encouraging, it is generally understood that (other than very high risk cases) the reconviction rate for sex offences is in any event comparatively low.

What conclusions is it therefore possible safely to raw from such studies? Debidin and Lovbakke (Op. Cit., pp.47 et seq.) conducted a comprehensive review not only of the programmes described above but also of a significant number of comparable Probation Service programmes. This notwithstanding, they too were forced to the overall conclusion that “the evidence on the effectiveness of these programmes is inconclusive overall”. It should be conceded, however, that (with the notable exception of Home Office Findings No.206) analysis of outcomes for completers and non-completers in programmes produces a more favourable pattern of findings. Typically, completers perform better than non-starters and that those who start but fail to complete fare significantly worse than other groups. It has been observed however that analysis by reference to completion status does not meet the right criteria for drawing conclusions because it does not exclude the possible effects of selection. Those who were sufficiently motivated to complete the programme might possibly have been expected to perform better in any event as a result of unconsidered characteristics such as motivation and general cognition. In fairness, the Home Office is beginning to concede that factors such as those described above including, in particular:

  • The impact of the rapid expansion of programmes;
  • The problem of targetting programmes upon those offenders who light be expected to benefit; and
  • The difficulties caused to both actual outcome and the subsequent statistical analysis thereof

must be effectively addressed. A recent submission by the Prison reform Trust to the Home Affairs Select Committee expressed concerns as to targetting based specifically upon Home Office Findings No.206:

“The study suggests that there is evidence of a shift in programme targetting towards lower risk offenders who are not suitable for cognitive behavioural treatment. This trend has been brought to the attention of PRT during visits to some prisons. The Prison service should examine how cognitive skills programmes are being targeted and delivered.”

The announcement of the Key Performance Indicators for the Prison Service for 2005-2006 included a restructuring of the cognitive behaviour development programme. A study is currently being undertaken within the Prison Service which will examine the relationship between offender characteristics and prison factors and the outcome of treatment programmes.

A Qualitative Approach?

The quantitative studies described above produce on the whole truly disappointing results in respect of the measurable efficacy of CBT in prisons (and indeed beyond). Should this therefore compel the conclusion that intervention of this type is a failure in adjusting the attitudes and behaviour of offenders and reducing reoffending? This too is too stark and simplistic a conclusion to draw. Clarke et al set out to obtain a subjective understanding of “What Works” by conducting 113 interviews with prisoners and members of prison staff from six prisons. This study was based upon the required refinement (discussed above) of the original question to “What works for whom?”. It was resolved at the outset that “in order to develop a full understanding of what constitutes effective practice, it is essential to explore what social and institutional factors enable prisoners to use the treatment programmes successfully and how prison and resettlement contexts can encourage or inhibit that process”. As in the empirical studies, the subjects had undergone either R & R or ETS programmes. The sample group was diverse in that the offenders comprised both programme graduates (adult male offenders over 21) who had recently completed a cognitive skills programme and, interestingly, “reconvicted graduates”, i.e. those who had completed such a programme but had since been reconvicted and returned to prison. In addition to the above and members of prison staff (both those who had been involved in a programme and those who had not), further interviews were conducted with those who had been released from prison after completing a cognitive skills programme and, under supervision in the community, had not yet reoffended. It appears that there was a general recognition that the programme had made some contribution to behaviour adjustment with interviewees commenting upon the fact that they regarded their newly-acquired thinking skills as having had a positive impact upon the way in which they approached other people and social situations. However, few identified a desistance from reoffending as a direct consequence of undergoing the programme. Prison staff however reported a perceptible improvement in the attitudes and behaviour of programme participants. Interestingly, the “reconvicted” graduates appeared anxious to blame extraneous factors such as resettlement difficulties and addiction problems as a cause of their reoffending rather than the failure of the programme (although some did concede that they had failed to appreciate the need to apply their new-found skills to situations in which they might be led into reoffending). This might tempt the conclusion that cognitive readjustment will fail if the programme graduate is confronted by former influences or present difficulties and that it is therefore ineffective as a means of preventing reoffending. It is submitted that this might not be so if one considers the responses of those who had not reoffended (“desisting graduates”). This group acknowledged that they had benefited from extraneous supportive influences such as resettlement and acquiring employment but also recognised the need (absent from the previous group) to apply their cognitive skills in daily life. However, it must also be conceded that there was a widespread acceptance within this category of offender of the view that the programme had “worked for them because they had reached a stage in their lives at which they realised that they genuinely wanted to change”. An important lesson to be learned from this study is the institutional factors which are perceived as impacting upon the efficacy of the programme. The possibility of underreporting as a result of the rapid expansion of programmes has already been discussed. In addition, participants in this study identified factors such as inability to attend classes due to staff shortages (or even lateness due to the time of unlocking) as having a detrimental effect upon progress within the programme. It was universally recognised by interviewees that a supportive institutional culture was essential extending not just to such practical matters as those referred to above but also in more general terms to a positive endorsement of the aims and value of the programmes by staff and senior managers.

In addition to the above, there is also anecdotal evidence available as to the perceived efficacy of the programme. Writing in The Observer , David Rose reported his findings after being given access to programme sessions and interviewing participating prisoners at four prisons over a period of two months. It is inappropriate here to reproduce the detail of the personal accounts given by prisoners but it is fair to summarise the responses as positive overall and ranging from a perception that the programme had effected a minor improvement in general attitude to an identification of the process as being a major contributory factor to a determination to overcome issues and influences which had led to offending in the past and a resolve not to reoffend. A note of caution should be sounded: as in the case of the “instrumentalist participant” described below, it is possible to conclude that the positive feedback upon the programme to the journalist was self-serving on the part of those who while still incarcerated have a vested interest in displaying a positive attitude and declaring reform.

“What Works - for whom?”

A recurrent concern herein has been the targetting of offenders as candidates for CBT and it has been seen that the Home Office, confronted by the apparent failure of programmes to have a demonstrable empirical effect upon reoffending rates has initiated further work on the identification of offenders who might benefit from such therapy. In the early days of CBT, the matching of offenders to programmes was crude. Hedderman and Sugg asked study participants to provide information on the types of offenders who were targeted by programmes. Such information as was returned revealed attempts to match participants to programme-types by reference to the type of offence committed: for example, young offenders would frequently be directed toward substance-abuse programmes and violent offenders toward those with a considerable anger-management element. There is a superficial merit in such an approach but it overlooks the general benefit (if indeed there is one) of cognitive therapy per se. More recently, there has developed an appreciation that it is possibly the pre-existing, personal characteristics of the offender which should dictate whether or not they are likely to respond positively to CBT. The study by Clarke et al (supra) concluded:

“The motivation of individual programme participants was seen as a key factor in explaining the treatment impact and outcomes of cognitive skills programmes. Motivation for programme participation and individual change are influenced by a complex combination of the individual’s characteristics, the institutional context in which programmes are delivered and how the individual engages with the programme.”

In possibly the first attempt to categorise programme attendees by such criteria, the authors identified four distinct types:

  • The Hostile Participant: this was an individual who manifested hostility toward the process exemplified by disruptive behaviour in class, general demeanour and “deviant” behaviour toward the programme;
  • The Instrumentalist: this was a participant who was perceived as being on the programme for cynical and self-serving reasons such as a desire to obtain a favourable parole report and who might therefore be demonstrating outward co-operation without the
  • The Sceptic: This was a candidate who may have begun as an instrumentalist but who, as a result of adjustments in thinking was beginning to engage with the programme following an appreciation that it might facilitate further personal development;
  • The Self-Developer: This was the most motivated class of individual who was probably highly appreciative of the need to change even before embarking upon the programme.

Thus there is strong evidence to suggest that the selection of offenders for CBT should be dictated by the characteristics of the individual rather than by the type of offence committed. The difficulty with the above findings is that they are borne of observations of responses to the programme after selection and commencement. It is submitted that work should be undertaken in order to determine whether it is possible to pre-identify the cognitive and motivational characteristics of potential candidates. A set of criteria should be developed by which such traits may be measured (acknowledging that this will never be an exact science but accepting that it must represent a improvement upon present practice and offer a greater prospect of ultimate success. It may also be necessary to address other issues. While it may be equitable to roll out such programmes to all institutions, there is already evidence that too-rapid expansion has prejudiced the quality of delivery. It is unfair to the individual if the resourcing and institutional culture of his particular establishment is such that the quality of a CBT programme will be impaired to such an extent as to lessen his prospects of success. Similarly, the heavy preponderance of males within the programmes studied may be reflective of the prison population at large but there must at least be a suspicion that the current under representation of females betrays a degree of discrimination. Finally, there should be an emphasis on the possibility of effecting permanent change. At present, admission onto programmes is regulated in part by the proximity of an individual to a parole date. There is a danger that this will in itself promote an instrumental approach of the type described above. A promising candidate for change should be allowed the earliest possible opportunity to commence the process since this in itself offers a greater prospect of changes in behaviour and attitude becoming established, permanent and better able to withstand the pressures of the external influences which will bear upon the offender upon release.

Conclusion

Cognitive behavioural therapy schemes were introduced in prisons in the early 1990’s following the acclaim of this type of therapy by the medical profession and (apparently successful at the time) introduction by the Probation Service with the specific intent of adjusting the attitudes and behaviour of individuals so as to lessen the likelihood of reoffending upon release. While on the one hand this may be regarded as a laudable attempt to pursue the aim of rehabilitation as opposed to punishment, the political glee which greeted supposed early evidence of success in achieving this objective indicates that other motivations possibly exist. The claim made in the first evaluation described above that an apparent drop in rates of reconviction could be directly translated - by reference to the number of prisoners currently engaged in cognitive behaviour schemes - into a calculable reduction in the number of offences which would be committed in the future indicates a different agenda. In the current climate there is great political capital to be made from a claim that offending rates have dropped and also by claiming vindication of the much vaunted “prison works!” declaration. There is also a strong sense in which it is being implied that there are considerable resource implications flowing from the success of such programmes: a decreased crime rate would ease the burden upon the police and other parts of the criminal justice system and a reduced return to prison of recidivists would clearly relieve the highly publicised strain upon prison accommodation.

However, it has been seen that attempts thus far to demonstrate the success of CBT by reference to empirical studies of rates of reconviction cannot be said to show a particularly significant improvement either as a result of the original programmes or their subsequent accreditation. Indeed, it might be argued on the basis of such studies that there is no perceptible impact. This may be the result of faulty methodology: it must be conceded that an analysis of reconviction rates alone is at best a blunt statistical instrument and at worst an inappropriate indicator. At the very least other factors should be taken into consideration and there is a convincing argument that a qualitative approach may be a preferable method. (This will not, of course, allow the populist argument that CBT is merited solely - if at all - on the basis of crime reduction.) Nonetheless it is submitted that it is too soon to abandon such schemes. There is considerable evidence of success in changing the attitude and behaviour of offenders even if this does not ultimately produce a favourable adjustment in the crime figures. Rehabilitation should be regarded as a wider concept than simply the prevention of reoffending and it may be thought that where offenders are reporting an improvement in their life skills and their ability to cope with previously challenging situations and where prison staff are reporting improvements in the actions and demeanour of those in their charge this is a commendable end in itself. Further, it is not correct to equate reoffending with recidivism without more. As has been observed the use of simple reconviction statistics is an unsatisfactory measure of actual recidivism since conviction is also dependant upon factors such as the efficiency of the police and the current policy of the Crown Prosecution Service. Even if these factors did not justify the continuation of CBT in prisons, it is too early to conclude that it cannot ever have a positive impact upon reoffending. Considerable further work is required to develop more efficient targetting of offenders to participate in cognitive behaviour programmes and other factors which might currently be adversely influencing the success of such programmes such as resourcing and institutional culture should be addressed. While it may be true to say that the “jury is still out” on the question of whether CBT can have a significant impact upon reoffending, the past and present dearth of other strategies to reduce crime by previous offenders is itself a compelling reason to continue the attempt at least until it may be definitely concluded that such programmes are not having a tangible effect and should therefore be regarded as squandering resources which might more appropriately be used elsewhere.

Bibliography

Ausubel, D., Theories and Problems of Child Development, (1958)

Blackburn, R., The Psychology of Criminal Conduct: Theory, Research and Practice, (1995)

Cann, J., Falshaw, L., Nugent, F. & Friendship, C., Understanding What Works: accredited cognitive skills programmes for adult men and young offenders, Home Office Findings 226 (2003)

Clarke, A., Simmonds, R. & Wydall, S., Delivering cognitive skills programmes in prison: a qualitative study, Home Office Findings 242 (2004)

Cooke, D. & Michie, C., Predicting recidivism in a Scottish prison sample, Psychology, Crime and Law, 4, 169-211

Davies, K., Lewis, J., Byatt, J., Purvis, E. & Cole, B., An evaluation of the literacy demands of general offending behaviour programmes, Home Office Findings 233 (2004)

Falshaw, L., Friendship, C. & Bates, A., Searching for ‘What Works: an evaluation of cognitive skills programmes, Home Office Findings 206 (2003)

Friendship, C., Beech, A. & Browne, K., Reconviction as an outcome in research: A methodological note, British Journal of Criminology, 42, 442-444 (2002)

Friendship, C., Blud, L., Erikson, M. & Travers, R., An evaluation of cognitive behavioural treatment for prisoners, Home Office Findings 161 (2002)

Friendship, C, Mann, R. & Beech, A., The prison-based Sex Offender Treatment Programme - an evaluation, Home Office Findings 205 (2003)

Hansard, Prison and Probation Services: Cognitive Skills Programmes, (3 Mar 2004: Column WA111), www.parliament.the-stationery-office.co.uk

Harper, G., & Chitty, C. (Eds.), The impact of corrections on reoffending: a review of ‘what works’, Home Office Research Study 291, (2nd Ed., 2005)

HM Inspectorate of Probation, Strategies for Effective Offender Supervision, (1998a)

HM Inspectorate of Probation, Evidence-Based Practice: A Guide to Effective Practice (1998b)

Hollin, C., Cognitive-behavioural interventions with Young Offenders, (1990)

Hollin, C., McGuire, J., Palmer, E., Bilby, C., Hatcher, R. & Holmes, A., Introducing Pathfinder Programmes into the Probation Service, Home Office Findings 177 (2002)

Holmes, J., All you need is cognitive behavioural therapy?, BMJ 2002;324:288-294

Lloyd, C., Mair, G. & Hough, M., Explaining reconviction rates: A critical analysis, Home Office Research Study 136 (1994)

McGuire, J., “Cognitive Behavioural Approaches”: An introductory course on theory and research, Liverpool University Course Manual (1996)

Maguire, M., Morgan R., Reiner, R. (Eds.), The Oxford Handbook of Criminology, (3rd Ed., 2002)

Probation Circular 36/2005, Public Protection: Cognitive Self-Change Programme, (20 May 2005)

Raynor, P. & Vanstone, M., Straight Thinking on Probation (STOP): The Mid-Glamorgan Experiment (1997)

Rose, D., It’s official - prison does work after all, The Observer, May 5 2002, www.observerguardian.co.uk

Soothill, K., Francis, B. & Fligelstone, R., Patterns of offending behaviour: a new approach, Home Office Findings 171 (2002)

Submission from the Prison reform Trust to the Home Affairs Select Committee Inquiry into the Rehabilitation of Prisoners (2004), www.prisonreformtrust.org.uk

Shadish, W., Cook, T. & Campbell, D., Experimental and quasi-experimental designs for generalized causal interference, (2002)

Sherman, L., Gottfredson, D., MacKenzie, D., Eck, J. Reuter, P. & Bushway, S., Preventing Crime: What Works, What Doesn’t, and What’s Promising, (1997)

Vennard, J., Sugg, D. & Hedderman, C., Changing Offenders’ Attitudes and Behaviour: What Works?, (1997) Home Office Research Study 171

www.bmj.bmjjournals.com

www.hmprisonservice.gov.uk

www.probation2000.com/pit/circulars

 

order a law essay

Related criminal law essays

Please note: The above essays and dissertations were written by students and then submitted to us to display and help others. Thanks to all the students who have submitted their work to us.

Back to Free Law Essays


order