Monitoring and Evaluation of Restorative Justice Research Paper

This sample Monitoring and Evaluation of Restorative Justice Research Paper is published for educational and informational purposes only. If you need help writing your assignment, please use our research paper writing service and buy a paper on any topic at affordable price. Also check our tips on how to write a research paper, see the lists of criminal justice research paper topics, and browse research paper examples.

As restorative justice programs spread throughout the world, it has become more and more important to have adequate evaluations of such programs upon which policymakers, practitioners, and academics can rely. This research paper distinguishes between monitoring, which all programs and practitioners should do themselves, and evaluation, which requires more specialized knowledge and skills. Both monitoring and evaluation need to be appropriate to the aims and context of the program itself and should meet the needs of and be accountable to the stakeholders of the program, including funders, those referring cases to the program, participants, and staff. Evaluation is thus not just “measuring satisfaction” or answering the question “does it make offenders reoffend less?” but must consider all the outcomes desired for the program, whether they have been achieved and, preferably, whether the program is cost-effective, or value for money. The paper provides some guidance on how to set up and conduct evaluations of restorative justice programs, but each program needs to consider its own requirements and aims.

What Is Evaluation?

Evaluation itself is a contested topic and the subject of many academic papers in its own right. Within criminology, evaluation has been the subject of lively exchanges, for example, on whether outcome evaluation is the key or whether the contexts of programs and the theoretical mechanisms through which the outcomes might be expected to occur are equally important (Farrington 1997, 1998; Pawson and Tilley 1998).

Historically, the forms of restorative justice being developed have been highly attuned to their cultural environment. As a result, there are many different types of program or scheme, even when considering only those dealing with crimes and which remain within the sphere of criminal justice (this research paper is restricted to this sector). There is victim-offender mediation, conferencing, indirect mediation, sentencing circles, community panels, and many others. There is even more diversity if one moves away from crime and considers the use of restorative justice in civil disputes, in schools and organizational settings, or in peace-making. It is also important to reflect as to whether the restorative justice program is concerned only with offences which have direct victims (people who are themselves injured or lose property) or whether it is also concerned with indirect victims (family members who are affected, neighbors, the community). The paper will use the term “scheme” to denote a program of work delivered by an organization within a defined geographical area in relation to criminal cases, such that the work is done within an agreed framework or process by one or more mediators or facilitators. It is possible for a scheme to have a national remit, as does the Youth Conference Service of Northern Ireland, which has statutory authority to deal with all cases referred by prosecutors or judges in youth justice cases (Campbell et al. 2006). More typically, however, restorative justice schemes have grown up locally, to serve one town, or one region, often specializing in particular types of referrals.

Within restorative justice schemes, there is considerable local diversity – and this may need to be so if participants (victims, offenders, those supporting each party) are to feel comfortable and safe in the restorative setting and so able to communicate with each other. However, it makes it much more difficult to say which forms of restorative justice are most suitable for whom or for what kind of dispute. We need to develop methods of evaluation which are sympathetic to the local aims and procedures of different restorative justice procedures but which are still rigorous enough to allow us to show whether there are differences in outcomes and to compare schemes.

Evaluation itself is a word which means different things to different people – which is usually the prerequisite for mutual misunderstanding, particularly in a comparative field. First, there is the need to distinguish what will be called “monitoring” from “evaluation.”

Monitoring: A Basic Duty For Each Scheme And Practitioner

Monitoring is an activity which needs to be undertaken by all schemes and by all practitioners. It is the essence of the reflective practitioner: essentially considering at regular intervals what one has done and reflecting on what has worked and what might be changed.

The basic level of monitoring implies record keeping of what cases have been dealt with, from whom referrals have come, who the participants were (including basic demographic details such as age and gender), what happened in the case and the dates of contacts, and when the case was closed. Having such systems is the basis of accountability and a requirement in codes of practice for restorative justice practitioners (e.g., Restorative Justice Council 2011, in the UK). They are important in answering queries from participants (such as “I’ve lost my copy of the outcome agreement: could I have another one?”), from those referring cases and from funders. Without good data systems and monitoring, further evaluation is impossible. Nor is it possible for governments and funders to see whether funds are being used correctly.

Yet new schemes do find it very difficult to remember to set up such systems, which provide not just a case management tool but also allow reports to be derived easily to provide management information (Shapland et al. 2011). There is no “off-the-peg” software currently available for running restorative justice schemes, and it is strongly suggested that new schemes contact others (even others in different countries) to find out their experience in practice. Restorative justice is much more complicated than criminal justice processes, because it is voluntary for both victims and offenders – so there is a need to ensure all the participants, venue, etc., are in place before there is a direct meeting.

Monitoring implies that data will be available from each restorative justice scheme in terms of its activity and some agreed outcome measures. So it will be possible to know how many cases have been referred to the scheme from whom, how many facilitators/mediators dealt with them, when they were referred and when they were closed, and what the process outcome was (one party declined the service, indirect mediation, a meeting between parties, etc.). Some of the latent antipathy towards evaluation by some restorative justice workers seems to have stemmed from the feeling that restorative justice must never be bureaucratized as a product, to be achieved identically each time, and to be able to be counted like tins of beans. This is a valid concern: restorative justice must be flexible and able to meet each group of participants’ needs. The very definition of restorative justice indicates this continuous democratic creation: “a process whereby parties with a stake in a specific offence collectively resolve how to deal with the aftermath of the offence and its implications for the future” (Marshall 1999). If the parties are collectively resolving, then necessarily the content of their discussions and any outcome agreement will be slightly different each time. With monitoring, however, it becomes possible to consider how restorative justice is faring across schemes or across countries (see Shapland et al. 2011, for a review of evaluation results; also Sherman and Strang 2007).

Understanding The Monitoring Figures: Providing A Description Of The Scheme

Given the intrinsic plurality of restorative justice, bare statistics tend to be meaningless. What does it mean to say that one scheme had 30 referrals from prosecutors while another had 150? What does it mean for the participants that 80 % of cases were closed within four months? Monitoring is vital, but collection of figures needs to be amplified by a detailed description of the process of restorative justice being run by that scheme, including:

What kinds of cases does the scheme take? What kinds of cases are referred? What processes are undertaken to see whether cases are suitable?
Who is contacted by whom when a case starts? What is offered? Typically, how does the case proceed?
Do participants meet face-to-face? Just victim and offender with a mediator (or two mediators?) Or with others? How many meetings?
What are typical outcomes? Is there a written outcome agreement (agreed by all parties)?
What happens after any meeting? Is there feedback to those referring cases? Is there feedback to participants about any criminal justice outcomes?

One of the most detailed but most helpful descriptions of what was done by a scheme was Justice Research Consortium Thames Valley (2004) “From soup to nuts” folder. The “from soup to nuts” title describes the typical pattern of an English dinner party – the idea of the folder was that new staff could find out from it how conferences were organized from first receiving the referral or data on the case (soup) all the way through the possible outcomes and processes, to the final feedback contacts with the parties (nuts).

Financial Cost Data

Alongside statistics of cases and this detailed description, it is also important that schemes keep details of their own scheme’s financial operations. It will never be possible to see whether the scheme is value for money unless the financial cost of running the scheme over a certain period of time and undertaking a certain number of cases is known. Value for money, on the British definition of value for money in public services, is achieved if the cost of running the scheme is exceeded by the financial benefits of its work in terms of reoffending prevented or well-being created (if well-being can be measured in financial terms, as it is in health economics and, for example, by the English National Institute for Clinical Excellence). Details of the kinds of financial cost figures that need to be kept are given in Shapland et al. (2008), but essentially they include salaries or fees to personnel, buildings and premises costs, travel costs reimbursed to participants, publicity, insurance, training, and so forth. The costs need to be able to summed over a particular time period, which is the same as that used for case management summary reports, so that it is possible to calculate a cost per case (even though such an idea of an “average” case is difficult for restorative justice, where cases may stop at different stages) or a cost per facilitator.

Some might comment why bother undertaking the burden of keeping such cost and time data, when the scarce time of personnel could be used in undertaking more restorative justice work? The answer is that cost and time data are not just for the benefit of economists, hard-edged governments wanting value for money, or criminal justice practitioners who hold budgets but also for schemes themselves. They can help to answer questions such as the following: Does direct mediation take longer? Can one split it from the effort needed to be put in to undertake indirect mediation? What are the costs of recruiting, training, and managing volunteers compared to paying a fee per case or employing paid facilitators? All would agree that proper preparation of participants is vital, but over what length of contact should it be encouraged? How much more costly is work in prison (given the need to have security personnel present, get all participants to the prison, etc.) compared to work in the community? These are key questions for the setting up and development of any scheme – and for those funding restorative justice. It is important that new developments or new areas where referrals are taken are coded as such in the data. It can be very difficult to work back through financial data to undertake these costs analyses retrospectively. It is much easier to work out a simple system of codes to allocate to payments as they are made.

Evaluating What The Scheme Is Trying To Achieve: Specifying Aims And Outcomes

Some practitioners and academics tend to be referring to monitoring when they talk about evaluation. Evaluation, however, is a rather more long-term and fundamental exercise – looking at whether the scheme has achieved its desired outcomes. Monitoring is not sufficient. Any scheme also needs to know whether it is achieving its desired outcomes or aims. It also needs to know how its services are being perceived by participants and where it may need to make changes.

Specifying Aims: The First Key Step

In order to start on evaluation, the scheme needs to know what its aims are in sufficiently clear and precise terms so that it is possible to measure whether they have been achieved. Unless the scheme has spent time in working this out and in agreeing what the precise aims are, it is impossible to set out what its outcomes are supposed to be and so evaluation is close to impossible.

This would seem to be obvious. However, the history of crime reduction and criminal justice evaluation is a rather sorry one in this respect – and it is important restorative justice evaluation does not follow the same path. In England and Wales, for example, considerable amounts of money have been invested by government, police, and local authorities on crime reduction initiatives. Yet some of these initiatives have not been implemented as intended, and others have been almost impossible to evaluate (Hamilton-Smith 2004; Hope 2004). One of the reasons has been a lack of precision in the intended aims of the initiatives or a change in those aims during the initiative. The problem is that initiatives designed to reduce burglary may be intended just (a) to reduce recorded offences of burglary; and/or (b) to reduce other linked crime, such as criminal damage; and/or (c) to reduce fear of burglary; and/or (d) to reduce fear of crime more generally; and/or (e) to make residents feel the police or other local authorities are doing their job in the local area better. All these aims are possible; more than one may be desired – but each creates a different outcome measure. If the program does not specify exactly which aims are its priority, the evaluation has real problems.

Restorative justice has potentially as many aims as crime reduction programs. Is a particular restorative justice scheme aiming to:

Achieve particular aims for participants (e.g., increase offenders’ empathy, restore victims’ hurt, provide a sense of closure for victims or offenders, reduce offenders’ reoffending)
Provide a process whereby participants may find their own solutions to situations/conflict (e.g., by allowing each to state the effects of the offence on themselves and ask questions, allowing each to have a fair say in resolution, encouraging a joint problem-solving agreement, providing a safe space for discussion)
Provide a more civilized or cheaper or more participatory alternative to traditional criminal justice.
Or more than one of these?

Each of these aims has been advocated as an aim of restorative justice. In our evaluation of three large restorative justice schemes in England and Wales, the schemes differed as to how the managers perceived their relation to criminal justice, for example, and so whether measuring reoffending was a relevant outcome measure (Shapland et al. 2006a). Facilitators within the same scheme put emphasis on different aims, because typically each scheme had multiple aims.

It is important to realize that each aim implies that specified people should judge whether it has been achieved. The first aim above, for example, is primarily about participants’ perceptions, rather than schemes’ or referrers’ perceptions. It needs to be evaluated using data from participants or relating to participants. The second relies on participants as well but could also be measured by observers at restorative justice meetings and through questions to facilitators. The third could be judged by all of these, but it can be argued that if it is a publicly funded service, then criminal justice system professionals and policymakers should also be judging its worth against other alternative processes in criminal justice.

Exactly which aims are seen by the scheme as paramount will depend upon its history but also upon its ideology and the theoretical mechanisms through which it believes restorative justice is impinging on the participants and others. Aims are inseparable from theory. Many theoretical strands have combined to make up the rich tapestry of modern restorative justice, and they differ according to the key purposes they see for restorative justice (Shapland et al. 2006b). Zehr (1990), for example, stresses its healing and conflict reduction potential and this is a dominant strand in victim-offender mediation, for example, in Nordic countries. Australasian restorative justice has stressed the democratic, participatory, emotional, and procedural potential of restorative justice conferencing (Braithwaite 1999; Harris et al. 2004). Recent schemes in England and Wales, following the dominant crime reductive imperative of national and local criminal justice agencies and government, have sought its potential in reducing offending, though our own evaluation has always stressed that both reducing offending and responding to victim needs must be equal aims (Shapland et al. 2011).

For restorative justice, specifying theoretical mechanisms is even more important than for crime reduction. This is because restorative justice is an intrinsically process-based initiative. It cannot be seen in evaluation terms as a black box model, where participants are put in one end, a mediator added, one shakes them all together, and outcomes come out at the other end. For restorative justice workers, it is precisely what happens in the middle, during the process, which is important. How exactly is it expected that victims will be reinstated or restored? What will be a sign of this? Are there mediations or conferences where this occurs more often or where conditions seem more fruitful? How can we make those occur more often? Is it something the mediators say? Or the surroundings? Or what? Similar questions can be asked in relation to offenders.

Setting Out Realistic Outcome Measures: The Difference Between Outcomes And Outputs

Evaluation is primarily about measuring outcomes – representing the final aims of the scheme. However, commonly, measuring outcomes takes some considerable time. If one aim of a scheme is to reduce reoffending, this is typically done by looking at reconviction rates or rearrest rates (in the USA). It is necessary to leave a period of time for reoffending to occur (1 year minimum currently in the UK and in the USA) and for official systems to record it (i.e., cases to reach conviction and sentence). It is important that reoffending be checked to see if it is being reduced in relevant ways (is it most relevant to measure likelihood of reoffending, or frequency, or seriousness, or all three?), but this tends to be a complex and costly business, because of the ways in which official conviction or arrest data are held. Would it then be sensible to measure intermediate goals called “outputs” more often, restricting evaluations looking at the final reconviction outcome to periodic sweeps every 5 years or so? Are there relevant outputs that could be measured? For reoffending or official conviction, it may be that successfully completing a release program from prison or a community program may be a proper output measure (though there is always the danger of contamination of the results if the criminal justice program workers know which offenders have been through restorative justice and are either favorable or very unhappy about it).

Restorative justice has been accused more recently of just relying on “satisfaction” measures (i.e., sending small questionnaires to participants immediately after closing the case asking if they were “satisfied” with the process and the result of the restorative justice). There is nothing wrong with sending out satisfaction questionnaires – they provide an opportunity for participants to feedback their experiences and some idea for supervisors as to whether any particular type of case or individual facilitator is providing problematic. They are a typical, simple output measure of participant perceptions. What is important, though, is that periodically it is checked that when an independent researcher does a more in-depth interview with participants at different times after the restorative justice is finished, it is considered whether there are different views about preparation, processes, outcomes, and feedback by the scheme about the result of the case and whether “satisfaction” immediately after the case ends continues for some months afterwards. Few restorative justice evaluations have been able to look in detail at all these elements, but Shapland et al.’s (2011) evaluation of victims’ and offenders’ views found that views on different aspects of the process and outcomes tended to be highly correlated and that there was not a big dip in positive views with time. They also found, though, that schemes might be regarded highly in terms of facilitator performance and outcome agreements but still fall down, for example, on feedback to participants about whether offenders had completed agreement items or what had happened to cases later in criminal justice.

Minimum Evaluation Is Not Optional

Some evaluation periodically is essential for all schemes. Without minimal evaluation which goes beyond monitoring, even a very major and potentially very successful program will leave nothing behind. It may have benefited the participants and criminal justice practitioners who have been involved, but their memories will fade in time and no one will be able to build on those useful foundations to develop or start new programs in the future. The minimum evaluation which is needed for all schemes and for all work funded through the public purse therefore seems to me to be:

Monitoring.
Process evaluation which provides a clear description of the scheme and its processes: a how to do it manual.
Identification of clear outcome measures: what the scheme aimed to do and how it expected these outcomes to occur (its theoretical view).
Evaluation using outcome measures periodically, supplemented as necessary by output measures (such as participants’ feedback and satisfaction) regularly. Ideally, this evaluation will be conducted by independent researchers or data gathered by the scheme analyzed by researchers independent of the scheme and its funders.

Further Important Questions For Evaluation

Funders and developers of restorative justice, as well as schemes themselves, however, will have many further questions. Answering these will often, however, require more specialized help from different academic disciplines. They may be matters which each country or scheme will want to ask at intervals, in a planned program of developing restorative justice through evaluation. They are certainly elements on which countries trying to develop restorative justice will wish to try to acquire comparative data, so that it is possible to think about different forms of restorative justice and different procedures which provide the most suitable platforms for participants’ particular needs.

Consideration Of Human Rights And Legal Issues

The first specialist area relates to the fact that restorative justice done in connection with criminal justice will raise many legal and human rights issues. This is so whenever the action at the heart of the justice process involves behavior which is against the criminal law, though the precise issues will differ when the restorative justice is never dealt with formally by criminal justice (community restorative justice), when it is diversionary (i.e., the case no longer will be dealt with by criminal justice personnel), or when it feeds into criminal justice processes (e.g., presentence or prerelease restorative justice).

Some key issues are:

Examining the scheme’s processes to ensure participants have adequate information and explanation about the process and its outcomes (including whether outcome agreements can be enforced and what happens if they are not completed) to be able to give informed consent on participation. This includes any likely effect on criminal justice proceedings (see Campbell et al. 2006; Daly 2003 for analyses of such effects in relation to young participants).
Considering power imbalances between participants not only in terms of the consequences of the offence but in criminal justice terms. This can be very complicated in terms of victims and offenders (Shapland et al. 2006b), even when the offence does not itself present unusual power issues (participants know each other well, live near each other, the offence involves interpersonal power).
Thinking how participants may be able to maximize communication within the restorative justice process (translators, dealing with disabilities, etc.). Communication is one of the key benefits both victims and offenders experience from restorative justice, compared to criminal justice (Shapland et al. 2011). Similarly, one of the reasons given for dissatisfaction in conferences, in the few cases in which this occurred in the same research, was when communication was difficult to achieve. If the aim of restorative justice is to facilitate communication and participation, then personal or situational difficulties need to be minimized.
If the case involves a serious offence or is still within criminal justice, then it may need to be ensured that offenders (sometimes others as well) can access legal representation in order to advise them on whether to participate in restorative justice.
The question of how open the restorative justice event itself is a key one. Normally, events (direct mediation meetings, conferences, circles) are private to the participants and facilitators. This has correlated with the dominant use of restorative justice allied to criminal justice for young offenders, for diverted cases, and for minor criminal cases where it is seen as similar to civil dispute resolution. The typical paradigm in these instances would have been a closed process if the case had been dealt with through normal criminal justice processes. However, often, these events are also open to observation by those training as facilitators and sometimes to other practitioners – though permission should be sought from the participants for this. Normally, they are not videoed or recorded, though a report is often prepared on the case to be sent to those referring it to the scheme. There are major questions about accountability in relation to restorative justice, which has not often been considered (Dignan et al. 2007). If, for example, there were to be a disturbance during the event, or a complaint afterwards about the mediator/facilitator, what mechanisms exist to answer those questions? Is there a need to have some form of recording to ensure accountability to participants? Moreover, if restorative justice does increasingly deal with more serious offending, is there a need for recording to provide accountability to judges, referrers, those hearing appeals, etc.? Recording itself does not necessarily breach privacy, as long as that recording remains within the group of those properly concerned with that case.

Evaluation To Develop A Better Appreciation Of What Restorative Justice Is Achieving

In order to discern whether restorative justice schemes are delivering their aims, there will need to be periodic investment in more in-depth and specialized evaluation. Just one research design may not suit all aims. For example, the current most stringent research design to measure decreases in reoffending is a randomized controlled trial (RCT) (see Sherman et al. 1997). In the evaluation of three large schemes in England and Wales, one of the schemes evaluated, Justice Research Consortium, used this design, and this enabled the evaluators to provide results on reoffending devoid of selection effects in participants volunteering to participate in restorative justice (Shapland et al. 2011). However, the use of RCT is rare in criminal justice research.

The use of an RCT is valuable in considering reoffending, but it does not by itself establish what it is about restorative justice which was found particularly beneficial by those victims (and offenders) who experienced it. An RCT can only look at whether victims who experienced restorative justice were more satisfied, or more helped, than victims who only experienced criminal justice, not the mechanisms by which this was occurring. Considering what elements of restorative justice were most helpful needed to be assessed by detailed interviewing of victims (and offenders) at different points after the process ended and comparison of their answers with observation of the conferences they attended (where this was possible) (Shapland et al. 2011).

Moreover, an RCT is unhelpful in determining cost value parameters, unless it is accompanied by accurate time measurements by schemes of the time taken on experimental group cases, compared to that on control group cases – and this was not always the case (Shapland et al. 2008). This is because the time taken on control group cases randomized out of the experiment after consent by victim and offender is obviously far less than that on experimental group cases – and one needs to know both in order to work out whether the financial cost of any decreases in reoffending attained by the experimental group over those achieved by the control group was greater than the cost of the scheme.

The choice of evaluation methods, therefore, depends very much on exactly which aim is being evaluated. Evaluations so far have concentrated upon participants’ views, whether the restorative justice was completed and agreements made, whether outcome agreements were fulfilled, and reoffending (see Shapland et al. 2011 for a review of the findings in relation to youth and adult restorative justice). A concise summary of the results would be that, overall, participants tend to say they are highly satisfied with processes and usually with outcomes. Where outcome agreements were made, they were fulfilled at least to the same extent that similar criminal justice outcomes (e.g., financial penalties, community work) were occurring in that country at that time. There is some limited, but positive evidence on reoffending, particularly for direct meetings.

However, there are key remaining areas where evaluation in different countries is still lacking. They include:

Why do some potential participants decline to take part? Most evaluations so far have been confined to those who do take part.
What are participants’ expectations of restorative justice – and were they fulfilled? Answering this question suggests interviewing participants before and after participation, which has rarely been done.
What happens to those who do not agree to participation in restorative justice? Do they still obtain victim/offender support services?
Which kind of procedure is most suitable for which kinds of cases and participants? There is some limited evidence that direct meetings (mediation or conferencing) produce greater satisfaction and greater drops in reoffending than indirect (shuttle) mediation (Shapland et al. 2011), and that conferencing is more effective in reducing power imbalances between participants, because the presence of victim and offender supporters is a mitigating one (Strang 2002; Shapland et al. 2011), but there have been very few direct comparisons between procedures.
What would happen if there were no scheme – so to what extent is this scheme and its way of working better than none, or alternative methods available in that country?

In many ways, we are only at the beginning of thinking through the most effective ways of evaluating restorative justice and of exploring its potential for all its participants and the communities in which they live. As we accumulate more results of monitoring and more specialist evaluation results, hopefully that future may unfold. Continuing comparative ways of working, evaluating and discussing evaluation results will continue to play a crucial role.

Bibliography:

Braithwaite J (1999) Restorative justice: assessing optimistic and pessimistic accounts. Crime Justice: Rev Res 25:1–127
Campbell C, Devlin R, O’Mahony D, Doak J, Jackson J, Corrigan T, McEvoy K (2006) Evaluation of the Northern Ireland youth conferencing service. NIO research and statistical series report no. 12. NIO, Belfast, at http://www.nio.gov.uk/evaluation_of_the_ northern_ireland_youth_conference_service.pdf
Daly K (2003) Mind the gap: restorative justice in theory and practice. In: von Hirsch A, Roberts J, Bottoms AE, Roach K, Schiff M (eds) Restorative justice and criminal justice: competing or reconcilable paradigms? Hart Publishing, Oxford
Dignan J, Atkinson A, Atkinson H, Howes M, Johnstone J, Robinson G, Shapland J, Sorsby A (2007) Staging restorative justice encounters against a criminal justice backdrop: a dramaturgical analysis. Criminol Criminal Justice 7:5–32
Farrington D (1997) Evaluating a community crime prevention program. Evaluation 3(2):157–173
Farrington D (1998) Evaluating “communities that care”: realistic scientific considerations. Evaluation 4(2):204–210
Hamilton-Smith N (2004) The Reducing Burglary Initiative: design, development and delivery. Home Office Research Sudy 287. Home Office, London
Harris N, Walgrave L, Braithwaite J (2004) Emotional dynamics in restorative conferences. TheorCriminol 8:191–210
Hope T (2004) Pretend it works: evidence and governance in the evaluation of the reducing burglary initiative. Criminal Justice 4(3):287–308
Justice Research Consortium Thames Valley (2004) From soup to nuts: a manual for restorative conferencing. Thames Valley, Bicester
Marshall T (1999) Restorative justice: an overview. Home Office Occasional Paper, London, Home Office
Pawson R, Tilley N (1998) Caring communities, paradigm polemics, design debates. Evaluation 4(1):73–90
Restorative Justice Council (2011) Best practice guidance for restorative practice. Restorative Justice Council and Ministry of Justice, London
Shapland J, Atkinson A, Atkinson H, Chapman B, Colledge E, Dignan J, Howes M, Johnstone J, Robinson G, Sorsby A (2006a) Restorative justice in practice: the second report from the evaluation of three schemes. The University of Sheffield Centre for criminological research occasional paper 2. Faculty of Law, Sheffield, at http://www.sheffield.ac.uk/law/ research/clusters/ccr/occasionalpapers
Shapland J, Atkinson A, Atkinson H, Chapman B, Colledge E, Dignan J, Howes M, Johnstone J, Robinson G, Sorsby A (2006b) Situating restorative justice within criminal justice. Theor Criminol 10:505–532
Shapland J, Atkinson A, Atkinson H. Chapman B, Dignan J, Howes M, Johnstone J, Robinson G, Sorsby A (2007) Restorative justice: the views of victims and offenders. Ministry of Justice Research Series 3/07. Ministry of Justice, London
Shapland J, Atkinson A, Atkinson H, Dignan J, Edwards L, Hibbert J, Howes M, Johnstone J, Robinson G, Sorsby A (2008) Does restorative justice affect reconviction? The fourth report from the evaluation of three schemes. Ministry of Justice Research Series 10/08. Ministry of Justice, London
Shapland J, Robinson G, Sorsby A (2011) Restorative justice in practice. Routledge, London
Sherman L, Strang H (2007) Restorative justice: the evidence. Smith Institute, London
Sherman LW, Gottfredson D, MacKenzie D, Eck J, Reuter P, Bushway S (1997) Preventing crime: what works, what doesn’t, what’s promising. US Department of Justice, Washington, DC
Strang H (2002) Repair or revenge: victims and restorative justice. Clarendon, Oxford
Zehr H (1990) Changing lenses: a new focus for crime and justice. Herald Press, Scottdale