Restorative Justice, Youth Violence, and Policing: A Review of the Evidence

: Restorative justice seeks to bring those that have created harm together with those that have been harmed, and often stands in contrast to retributive and punitive approaches to justice that centre the state in the responses to crime and harm. Restorative justice approaches are becoming increasingly integrated into parts of the criminal justice system, and this paper examines the evidence for such applications in the context of youth violence and policing. The evidence is built on work conducted for the Metropolitan Police Service, the UKs largest police force with over 30,000 ofﬁcers serving 8 million people in and around London. It does this through a Rapid Evidence Assessment, which utilises the search and sifting principles of systematic reviews on a more limited basis, tailored to the needs of a speciﬁc audience, and conducted within a limited timescale. The results of the assessment are broken down into three areas: beneﬁts, challenges, and deployment considerations. The studies identiﬁed through the assessment suggest that restorative justice and restorative practice can form an important part of an overall strategy to help reduce both incidents of youth violence as well as the longer-term impacts of that violence when it has taken place. We conclude that in the context of violence and young people, effective restorative justice police practice should embrace a whole-system approach that incorporates multi-agency working and consistently engages with young people at risk of becoming violent offenders or victims.


Restorative Justice, Restorative Practice, and Its Applications
As an approach to dealing with crime, restorative justice seeks to bring those that have created harm together with those that have been harmed (Braithwaite 1989;Hand et al. 2012;Rossner 2017). In contrast to retributive and punitive approaches to justice, which centre on state responses to crime and harm, restorative justice places the 'ownership' of conflict with those that have the greatest stake in the event (Dzur and Olson 2004). In practice, that often means bringing victims/those harmed and offenders/the harmer together in highly controlled circumstances, to address the salient issues and consider ways forward (Braithwaite 1989). Restorative justice approaches seek to repair, rebuild, or redress relationship breakdowns, thereby supporting victims to understand and overcome the harm they have experienced; help offenders to appreciate the impact of their actions and in doing so consider their wider offending behaviour; and, where available, engage communities to support both victims and offenders in moving forward. Moreover, the restorative process can facilitate 'bi-directional dialogue' between victims and offenders, enabling victims to have a voice, be heard, and gain answers to questions (Van Camp and Wemmers 2013;Zehr 1990). Further victim-specific benefits of participating in restorative justice include empowerment and agency, being validated as a victim, gaining closure, and reduced levels of stress and fear (Angel 2005;Koss 2014; O'Mahoney and Doak 2017; Van Camp and Wemmers 2013).

•
Direct contact (face-to-face): for example, victim-offender conferences, circles; • Indirect contact (non-face-to-face): for example, letter writing, shuttle work; • Potentially overlapping processes: for example, victim and offender circles that may or may not intersect, surrogate offender interactions; • Discrete processes: for example, healing circles for victims, community or family to repair relationships.
Restorative practices are increasingly being implemented globally, with a growing emphasis on greater use within criminal justice (Marder 2020b;Pali and Maglione 2021). These span a range of different criminal justice systems, for instance Canada, (Roach 2013), Iraq (Ali Al-Hassani 2021), and China, (Zhang and Xia 2021). In Europe, there has been a growing focus on restorative justice as part of criminal justice policy. The 2012 EU Victims' Rights Directive (Directive 2012/29/EU) explicitly mentions restorative justice as a tool for supporting victims of crime. The 2018 Council of Europe Recommendation CM/Rec 2018 (2018) encouraged 'the development of innovative restorative approaches [ . . . ] by judicial authorities, and by criminal justice and restorative justice agencies'. The 2020 EU-wide Strategy on victims' rights (2020-2025) called on member states to promote the use of restorative processes. Additionally, the 2021 'Venice Declaration' on the Role of Restorative Justice in Criminal Matters supported Member States to develop policies and plans for the integration of restorative justice practices.
Both restorative justice and restorative practice are becoming increasingly established and evidence-based elements of the criminal justice system in the UK. Ranging from work in policing (Marder 2020b;Shapland et al. 2017), in probation (Kirkwood and Hamad 2019), in prisons (Dhami et al. 2009;Calkin 2021), in youth offending teams (Criminal Justice Joint Inspection 2012), as well as in other parts of the institutional context such as in forensic mental health settings (Cook et al. 2015), social work, and social care (Parkinson et al. 2018). Outside of the formal criminal justice setting, there are also a wide range of community-based schemes that engage with specific groups, particularly young people, to help address problem/offending behaviours (Eriksson 2009). This includes work in supported housing ) and social work with families, children, and young people (Roche 2006).

Policing, Young People, Violence, and Restorative Justice
In the context of seeking to grow restorative justice within the criminal justice system, the policing and management of young people has, for some time, been an area in which there has been significant and effective use of restorative approaches. For example, Sherman and Strang (2007, pp. 4, 22) identify several global examples where restorative approaches have been effectively used to reduce recidivism with young people that have committed violent crimes, including in Canberra (Australia), Northumbria (UK), and Indianapolis and in Washington, DC (USA).
In the UK context, McAra and McVie (2010, pp. 182-83) identified both the English-Welsh and Scottish criminal justice systems as ones that have 'embraced restorative justice as part of the perceived solution for youth crimes'. In their 2018 study of Restorative Policing in England and Wales, Clamp and O'Mahony (2019, p. iv) found that where police forces were engaging in restorative practices, they would routinely refer young people (albeit at a slightly lower rate than adults) to restorative services for serious first time and repeat offenses. Marder's (2020a) examination of the institutionalisation of restorative justice into police forces in England, found that restorative approaches were being used effectively in both Durham and Gloucestershire, but that there was a risk of processes being too victim-focused, especially in crimes that involved young people. Payne et al.'s (2021) evaluation of a restorative programme in Gloucestershire, England, demonstrated how approaches that built on interactions between police and young people could significantly improve procedural justice and relationships.
The operationalisation of restorative justice into the policing of youth violence has, however, been patchy. A number of studies in England and Wales have identified that the delivery of restorative justice by police remains inconsistent. Banwell-Moore (2022, p. 7) argues that much of this is due to a 'lack of systematic guidance . . . [that continues to] underpin the work culture of criminal justice organisations . . . and impact[s] on [the delivery] of restorative justice'. Rosenblatt's (2014) study examined the application of restorative justice in YOTs and determined that very few restorative processes take place with young offenders and that many cases continue to focus on low level offences (Hoyle and Rosenblatt 2016). Hoyle and Rosenblatt (2016, p. 25) suggest the need for a more 'paradigm shift' which remains absent due to 'a lack of imagination for envisaging a significant move towards "fully" restorative processes'. Similarly, Shapland et al.'s (2017) restorative policing study aimed to provide a deeper understanding of restorative principles for young offenders. New measures piloted in the Shapland et al.'s (2017) study included: training police response officers in restorative justice; providing direct guidance on how to make referrals (to YOTs and restorative justice service providers); employment of a Restorative Justice Development Officer (to coordinate training, promote restorative justice, and encourage officers to discuss potential cases); and a 'safer schools' initiative. The 'safer schools' initiative encouraged partnership between schools and police to promote restorative justice consistently and comprehensively. The study found that to fully embed restorative justice, as part of a wider youth crime remit, training needed to be continuously developed and updated with refresher sessions (Shapland et al. 2017, p. 67). Moreover, they identified that restorative work requires that police collaborate with partner agencies, which include educational institutions and youth offending services (Shapland et al. 2017, p. 92).
There has been a growing call in the UK for wider deployment of restorative approaches for tackling the harms of youth violence. The Independent Commission on Youth Crime and Antisocial Behaviour (Goldson 2011, p. 10) called for restorative justice to be 'the centerpiece of responses to all but the most serious offences committed by children and young people'. Further to this, the UK Government's 2018 Referral Order Guidance stated that 'youth offender panels [must] operate on restorative justice principles and [ . . . ] should be conducted restoratively and in accord with the principles and ethos of restorative practice' (Ministry of Justice/Youth Justice Board 2018, p. 19). In their report 'Our Generation's Epidemic: Knife Crime', the Youth Select Committee (2019) make a series of recommendations on the greater use of restorative justice for young people involved in serious offenses. Amongst these is a call that 'the next version of the Serious Violence Strategy includes a larger focus on restorative justice ' (2019, p. 48). In 2020, the report from the All Party Parliamentary Group for Restorative Justice (2021, p. 5) (APPG-RJ) highlighted the 'postcode lottery' in the availability of services, the complexity for young people in accessing restorative referrals (2020, p. 10), and the need to 'end to blanket bans for certain offense types ' (2020, p. 19).

Context for the Study
The renewed policy focus has put restorative justice back on the agenda for police forces as a response to youth violence in the UK, with forces seeking to explore how their restorative offer works and for whom that offer is available for. It is within this context that this paper seeks to pull together the extant international research that demonstrates that for the 'recurring shortcomings' in the practical application of restorative justice, that a 'whole system' approach needs to be taken by the youth justice system, including the police, where restorative justice is seen as a legitimate youth violence response and not just as an optional extra tool in the youth justice toolkit. Whilst there is a plethora of research highlighting best and worst 'youth' restorative practice there is scant analysis of restorative justice and youth violence literature in the context of examining what works and why, from an institutional perspective. As a contribution to these discussions, this paper examines the use of restorative justice as an effective youth violence policing response. The research presents a thematic analysis of international best practice of the use of restorative justice in the field of youth justice and specifically youth violence. This paper presents the key findings from a Rapid Evidence Assessment (REA) commissioned by the Metropolitan Police Service (MPS), the UKs largest police force with over 30,000 officers serving eight million people in and around London. The REA provides an examination of current restorative practice related to policing, restorative justice, and youth violence, in order to contribute to the ongoing MPS policy development in the area. That REA selected papers through a rigorous methodological process, examining: the nature of the interventions, including the target populations, the types of interventions used, and the locations in which those interventions took place; about the nature of successful interventions; problems that were identified in the roll-out of interventions; and solutions that might have been identified to overcome those problems. In this paper, we present these findings in two ways: in Section 3 as a summary of the papers selected for the REA and the intervention(s) they detail and in Section 4 as a more detailed discussion on the key issues identified under the headings of 'benefits', 'challenges', and 'delivery considerations'. The paper ends with an overall summary, which argues that restorative justice can be, when implemented effectively and consistently, an effective youth violence response, that requires police forces in England and Wales to adopt a whole system delivery approach. The paper also examines restorative justice from the perspective of young people and young offenders themselves.

Method
To review the evidence related to the effectiveness of restorative justice when dealing with youth violence, this study employed a Rapid Evidence Assessment (REA). A REA utilises the search and sifting principles of systematic reviews but on a more limited basis, tailored to the needs of a focused audience and conducted within a limited timescale (Varker et al. 2015). This study employed a similar approach to that of Nascimento et al. (2022) work to evaluate restorative justice employing a SPIDER research strategy.

Eligibility Criteria
The criteria used for including studies in this systematic review were based on a standardised approach for filtering a large pool of initial documents into a smaller sub-set of papers to be reviewed, according to strict pre-set criteria. Known as the SPIDER strategy, criteria are underpinned by five reviewing components: (S)-sample (the papers selected should be related to the group of interest; (PI)-phenomena of interest (understanding the 'how and why' of behaviours being reported on); (D)-design (the theoretical research framework used); (E)-evaluation (the outcomes or findings of papers/reports); and (R)-research type (whether the study used a qualitative, quantitative or mixed methods approach) (Cooke et al. 2012). To examine the use of restorative justice approaches to tackle violent crimes committed by youths (under 18 years of age), the phenomenon of interest (PI) was centred specifically on the likely benefits, challenges, and delivery considerations for police services implementing or developing restorative justice service provisions.
The sample papers considered for this review were published between 1990 and 2022. They were sourced from international academic peer-reviewed journals and grey literature (both peer-reviewed and non-peer-reviewed) produced by official organisations related to policing or restorative justice and published on their websites (see below for a comprehensive list). Grey literature was specifically included because its value is comparable to that of academic peer-reviewed papers when using qualitative methods (Finfgeld-Connett and Johnson 2013). The grey literature was assessed for inclusion based on the reliability of the site as determined by the professional and expert opinion of the reviewing team (which was composed of academics specialised in restorative justice research) and the certainty of being able to locate the papers if the review needed to be reproduced.

Academic Database Searching Approach
The methods used to identify the academic, peer-reviewed literature included two electronic databases: EBSCOhost and Web of Science. The search was conducted in February 2022 and was limited to English language papers. A systematic list of search terms was chosen to separate relevant papers that represented the phenomena of interest (PI) from the large population of restorative justice papers that were not related to youth violence and the other criteria listed above. The search terms used were as follows: restorative AND (violen* OR "serious violen*" OR "violent crime" OR "violent offend*" OR assault OR "physical assault") AND ("young people" OR "young person" OR "youth offending" OR gang OR child OR youth OR juvenile OR "referral orders" OR "reparation orders" OR "youth offending team*" OR yot OR "youth offending service*" OR yos OR "youth offending panel*" OR "community reparation panel*" OR "family group conferenc*" OR "juvenile justice" OR "juvenile delinquen*" OR "scrutiny panel*").

Non-Academic Database Searching Approach
The following online sources were consulted beyond the academic database searches: the Australian Institute of Criminology-Restorative; the Restorative Justice Council (UK); Restorative Thinking (UK); US Department of Justice, Office of Community Oriented Policing Services (USA); HM Inspectorate of Prisons (UK); Loughborough University (UK); and the London Mayor's Office for Policing and Crime (UK). This search added 10 additional papers to add to the inclusion/exclusion sifting review (there were no duplicates to remove).

Inclusion and Exclusion Sifting Process
The inclusion and exclusion sifting process examined the academic databases and non-academic sources against a set of stringent inclusion/exclusion criteria. The final sample (S) included papers based on phenomena of interest (PI), design (D), and research type (R). The inclusion and exclusion sifting criteria included:

1.
Reporting empirical research, policy analysis or reflection, restorative justice intervention evaluations, applied research, and exploratory research (D). Papers could be qualitative, quantitative or mixed methods (R). Opinion pieces, personal blogs, reviews of literature, and descriptive papers with no empirical component or reflective pieces were excluded.

2.
Centred on an intervention aimed at reducing youth violence or improving victim outcomes (D). Reviewed studies included the varied experiences in the implementation and development process in both criminal and non-criminal justice settings (such as schools).

3.
Describing any restorative justice technique/approach; the search was not confined by a narrow understanding of restorative justice limited to face-to-face conferencing/interventions (PI).
The remaining papers constituted the study sample, and were analysed by a team of reviewers and summarised with a focus on the evaluation (E) using the following headings:

1.
Nature of the intervention a. Target population? b.

2.
What was found to work well? 3.
What problems were identified? 4.
What potential solutions were identified?

Quality Evaluation
A series of measures were implemented to evaluate the quality of qualitative research (Morse et al. 2002). These measures ensured that the transparency of the analytical process was maximised. This included both an internal validity check and a reliability test. Internal validity concerns, specifically systematic (bias) error was addressed by ensuring consistent measurement. This was of particular concern around the sifting process where papers were excluded/included. A benchmarking exercise was used on the first five papers which were analysed collaboratively by the review team to ensure that the inclusion/exclusion criteria were applied consistently. To ensure reliability, a two-stage test-retest process was implemented. First, each review team member re-analysed five percent of the work of another researcher in the team. Second, senior academics on the REA team reviewed a dip-sample of the analysed papers.

Included Studies
An initial 14,182 papers were identified through the academic database search as pertaining to restorative justice within the timeframe allowed by this study. Of these, 263 papers corresponded to the detailed search term criteria on youth violence (as outlined above). Once duplications were removed this was reduced to 176 papers. A further 10 items were then included through the targeted online search, with material coming from the additional targeted online search (see institutions included listed above). Consequently, a list of 186 studies were screened through the inclusion/exclusion exercise, which resulted in 155 exclusions and a final list of 31 studies. These 31 studies were included in the final analysis. The overall figure for each stage of the search are shown in Figure 1.

Details of the Studies Included
Of the final list of 31 sources, 26 were from the academic database search and five were from the additional targeted online search. The five were: Blackley and Bartels (2018) (2010). The studies were spread across the USA (N = 12); the United Kingdom (N = 9); Australia (N = 4); cross-national evaluations (N = 3); and New Zealand, South Africa, and Croatia (each N = 1). A total of 19 of the studies used qualitative methods, 5 used quantitative methods, and 7 used mixed (quantitative and qualitative) methods. Table 1 summarises key details of the included studies.

Details of the Studies Included
Of the final list of 31 sources, 26 were from the academic database search and five were from the additional targeted online search. The five were: Blackley and Bartels (2018), Brathay Trust (2017) (2010). The studies were spread across the USA (N = 12); the United Kingdom (N = 9); Australia (N = 4); cross-national evaluations (N = 3); and New Zealand, South Africa, and Croatia (each N = 1). A total of 19 of the studies used qualitative methods, 5 used quantitative methods, and 7 used mixed (quantitative and qualitative) methods. Table 1 summarises key details of the included studies.

Armstrong (2021) Scotland
An analysis of prior work exploring the reasons why a restorative approach may be of benefit give the perceived failure of conventional criminal justice in addressing the growing problem of child and adolescent harmful sexual behaviour (HSB) in Scotland.
Restorative approaches may be warranted given the perceived failure of conventional criminal justice in addressing child and adolescent HSB in Scotland. Such approaches can empower victims and offer them the opportunity to seek answers to the questions the CJS cannot provide.
Banks (2011) Southern Sudan and East Timor Comparative examination of draft laws in Southern Sudan and East Timor to provide insights into policy choices and the The importance of considering the relationship between international norms of child

Armstrong (2021) Scotland
An analysis of prior work exploring the reasons why a restorative approach may be of benefit give the perceived failure of conventional criminal justice in addressing the growing problem of child and adolescent harmful sexual behaviour (HSB) in Scotland.
Restorative approaches may be warranted given the perceived failure of conventional criminal justice in addressing child and adolescent HSB in Scotland. Such approaches can empower victims and offer them the opportunity to seek answers to the questions the CJS cannot provide.

Banks (2011) Southern Sudan and East Timor
Comparative examination of draft laws in Southern Sudan and East Timor to provide insights into policy choices and the relationship between international norms of child protection and traditional restorative practices.
The importance of considering the relationship between international norms of child protection and traditional and cultural restorative practices.

Barretto et al. (2018) New Zealand
A qualitative analysis of an open-ended survey from a nationally representative sample on public sentiments to address youth justice issues.
Public sentiments showed considerable support for a multi-faceted approach that utilised a combination of rehabilitative, punitive, and restorative forms of justice.

Blackley and Bartels (2018) Australia
Examination of sentencing and treatment practices for juvenile sex offenders in Australia, using examples of judicial reasoning in sentencing.
Multi-systems and ecological approaches to treatment that promote offender rehabilitation and accountability while also providing justice and safety for victims and communities. These included restorative justice conferencing and therapeutic treatment orders, which showed promising results in reducing sexual recidivism.

Bonell et al. (2018) England
Cluster randomised trial, with economic and process evaluations, of the Learning Together intervention compared with standard practice (controls) over 3 years in secondary schools in south-east England.
Learning Together consisted of staff training in restorative practice; convening and facilitating a school action group; and a student social and emotional skills curriculum. Primary outcomes were self-reported experience of bullying victimisation (Gatehouse Bullying Scale) and perpetration of aggression (Edinburgh Study of Youth Transitions and Crime School Misbehaviour Subscale) measured at 36 months. Data was analysed using intention-to-treat longitudinal mixed-effects models.

Brathay Trust (2017) England
Analysis of the Turning the Spotlight Programme led by Cumbria Office for the Police Crime Commissioner and delivering programmes to prevent and reduce incidents of hate crime and domestic abuse.
High numbers of participants from the programmes reported an increased sense of empowerment in relation to keeping themselves safe, including personal development factors such as: increased feelings of self-worth, improved communication skills, increased awareness of self and situational context, and recognition of personal strengths.

Clark (2012) South Africa
Examination of the potential merits of restorative justice as a response to the problem of youth violence, focusing particularly on the 2009 Child Justice Act based on fieldwork in South Africa.
This research draws on both the author's qualitative interview data and a range of surveys with young people conducted by the Centre for Justice and Crime Prevention in Cape Town.
Cossins (2008) Australia, the United Kingdom, and New Zealand Re-analysis of the data reported in Daly (2006) (see below) and comparing restorative justice with other reforms to sexual assault trials.
The research looked to explore whether restorative justice is one of the ways forward in the difficult area of prosecuting child sex offenses.
There is insufficient evidence to support the view that there are inherent benefits in the restorative justice process that provide victims of sexual assault with a superior form of justice. The major concern is that restorative justice will not be able to defuse the power relationship between victim and offender and will re-traumatize victims. Over three-quarters of victims participating directly in youth offender panels were happy with their experience of restorative justice and said that it was effective in achieving reparation for the harm done to them. However, not enough victims are engaging directly with youth offender panels.

Author(s) Location Focus of Study Findings Summary
Daly (2006)  Australia Drawing on the South Australia Juvenile Justice project dataset to analyse youth peer violence ('punch-ups') with a focus on girl-on-girl assaults.
Debate on the appropriateness of RJ for cases of gendered violence is polarized, in part, because there is a lack of empirical evidence and, in part, because of the symbolic politics of justice in responding to violence against women and child victims. However, the conference process was found to be less victimizing than the court process and may produce more effective outcomes.

Daly (2006) Australia
Findings from an archival study of nearly 400 cases of youth sexual assault, comparing conference cases to community service cases over a six-and-a-half-year period.
For conference cases, participation in the Mary Street Programme (RJ) was associated with the lowest prevalence of reoffending (43 per cent), and community service was associated with higher levels (56 per cent).

Froggett (2007) England
Analysis of video data from a creative writing project with young offenders in the context of individuated restorative justice programmes.
It was found that a crucial step in moral learning for these young people is the willingness to self-reflectively acknowledge their own destructiveness in a context which fosters an internal sense of guilt and concern for the hurt caused to others.

Gal (2021) International
Review of existing findings from the fields of RJ, children's rights, psychology, and victimology.
A positive RJ process can provide a constitutive event for children that can affect the way they develop, can strengthen and even repair support systems, enhance resilience, and reduce use of maladaptive coping mechanisms such as self-blame and aggression.

Harden et al. (2015) USA
Analysis of a youth violence prevention and intervention program involving 44 high school-age youth from violence-exposed urban communities in a nine-month, multidisciplinary, after-school program.
Youth in the communities experienced pervasive traumatic stress and in multiple manifestation, and the evaluated the programme showed statistically significant pre-and post-test differences in mean scores for participants on 41 outcome measures (for school, community, family, experience, and self) as compared to just 4 statistically significant pre-and post-test differences in mean scores for the comparison group for measures of community and self.

Harland (2011) Northern Ireland
Qualitative study carried out by the Centre for Young Men's Studies with 130 marginalized young men aged 13 to 16 from 20 different communities across Northern Ireland addressing themes of violence, conflict, and safety.
Multi agency approaches with local community-led programs and peacebuilding processes can help to engage marginalized young men in peacebuilding, better prepare them for living in a multicultural society, and help alleviate the fear, apprehension, suspicion, and distrust of others.

Author(s) Location Focus of Study Findings Summary
Kelly (2017)  USA Data from National Centre for Educational Statistics study of US schools (n = 2648), used to identify use of interventions intended to reduce school violence.
Identified interventions intended to reduce school violence. Schools that used Conflict Transformation Education interventions and Restorative Justice/Discipline interventions were predictive of lower rates of violence reported in schools. Schools which reported using both mental health and restorative justice/discipline together reported lower rates of disciplinary actions.

Kim and Gerber (2012) Australia
Australian data from Reintegrative Shaming Experiments between 1995 and 1999 examining juvenile offenders' perceptions on preventing reoffending, repaying the victim and society, and the degree of repentance.
There was no significant relationship between RJ conference and the offenders' own perceptions on the prevention of future offending. However, those who experienced RJ conferences are significantly more likely to perceive that they were able to repay the victim and repay society than those who had experienced traditional court processing.

Lane et al. (2007) USA
Randomized experiment, comparing youths in the experimental group, who had interventions that were restorative, with those on routine probation, using interviews with youths in both the experimental and control groups 1 year after random assignment.
Youths who had the restorative treatment were less likely to have taken drugs recently and less likely to have been involved in violence or homicide.

London Assembly Police and Crime
Committee (2016) England London Assembly Police and Crime Committee report on causes of youth serious crime and how best to prevent offending Approaches to tackling serious youth violence should include RJ, which would build London's focus on young victims.

Mann (2016) USA
Examination of the perceptions of school discipline administrators on implementing Peer Jury as an alternative school discipline strategy. Analysed the impact on attendance, instruction, recidivism of negative behaviour, and the disproportionality in the issuance of sanctions.
Peer Jury approaches in schools: (1) promotes leadership, accountability, ownership, and civic engagement; (2) increases student attendance and instructional time; (3) decreases discipline problems and negative behaviour, including recidivism rates; (4) are recommended as an alternative to suspension and expulsion; (5) can support parents and community engagement; (7) addresses disproportionality of sanctions issued; (8) are an effective discipline option.

Mateer (2010) USA
Using a case study format with interviews of involved administrators, teachers, and juvenile justice practitioners to document how a junior high school community recovered from a planned copycat to the Columbine shooting.
Restorative justice practices used in the school were uniquely suited to the event and responsive to the healing needs of the community at the time. In this situation not only was the harm repaired, but the community used the pain created by the harm to create a transformation improving the school.

Movsisyan (2014) USA
A qualitative case study which included interviews of students from grade 9-12 who had experience with school violence and with the assistant principal.
Youths at the school felt that the use of restorative justice and other similar communication approaches helped them to feel safe and not fear violence at school. The youths stated that because they felt safer, they were less prone to engage in violence to defend themselves.

Moyer et al. (2020) USA
Qualitative interviews and observations used to construct a case study of the successful campaign by Voices of Youth in Chicago Education to pass SB100, a progressive Illinois law aimed at ending the school-to-prison pipeline.
Storytelling empowered young people, particularly historically marginalised groups, and strengthened their relationships within the campaign, enabling them to see that their experiences were shared by others and therefore part of a larger systemic problem.

Ohmer et al. (2010) USA
Description of an exploratory study of a pilot training program the authors developed to facilitate residents' ability to intervene in neighbourhood problems in a low-income community in Atlanta, Georgia.
After the programme trial, willingness to use direct intervention in problem situations increased from 3 to 9 out of 10. Restorative practices equipping people with skills to approach a situation with peaceful and non-threatening strategies and gave confidence to tackle conflicts more directly.

Payne and Welch (2018) USA
Analysis of data from a nationally representative sample of schools to examine school conditions that influence the use of restorative responses to violence and misbehaviour.
Restorative justice consistently produced positive effects, regardless of school characteristics. If implemented more broadly within schools, restorative justice may substantially reduce student offending, increase perceptions of safety, enhance learning, promote positive school climate, and dismantle the school-to-prison pipeline that is exacerbating inequality and disadvantage for certain students.

Peurača and Vejmelka (2015) Croatia
Analysis of a peer mediation programme based on interviews with three experienced experts in the field of peer mediation in Croatian schools.
The study found that there is the need to work on improving the implementation of peer mediation in schools, through: Innovative and comprehensive promotion of peer mediation; quality education of children and adults in peer mediation; evidence-based practice; multidisciplinary and cross-sectional cooperation; well planned and systematic collection of data and availability of results; systematic evaluation of programmes.

Smokowski et al. (2018) USA
Analysis of the North Caroline Youth Violence Prevention Center programme including 'Positive Action', administered in 13 middle schools for 3 years; 'Parenting Wisely', an online program provided to 300 parents; 'Teen Court', a community-based restorative justice alternative provided to 400 adolescents; and additional county-level data on levels of youth violence.
The efforts of this university-community partnership was associated with a 47% reduction in non-school-based offenses, a 31% reduction in undisciplined/delinquent complaints, and an 81% reduction in the use of corporal punishment, along with smaller reductions in school-based offenses, short-term suspensions, and assaults. In addition, some county-level indicators of violence decreased.

The Independent Commission on Youth Crime and Antisocial
Behaviour (2010) England, Wales, Northern Ireland Inquiry prompted by concern about deep-rooted failings in the response to antisocial behaviour and crime involving children and young people The Commission proposed a major expansion of restorative justice in England and Wale to include youth conferencing as the centerpiece of responses to all, but the most serious offences committed by children and young people.

US Department of Justice (2016) USA
Descriptions of approaches taken to combat youth violence submitted by mayors and other officials in 30 US cities of varying sizes and representing every region of the country.
A successful response calls for strong partnerships between mayors and police chiefs of the kind that community restorative policing concepts have been shown over the past two decades to strengthen.

Utheim (2011) USA
Examination of the use of restorative practices for navigating conflicts among court-involved youth at an urban high school.
A restorative ethos allows social actors the opportunity to reclaim their human agency as participants of social conflicts. The sense of communal belonging and common humanity that restorative processes aim to inspire was often captured in conversations with senior or former students.

Wallis et al. (2013) England
Analysis of a Youth Offending Service innovative groupwork programmes, one for young people who have committed violent offences and the other for car crimes.
The young people that undertook the course saw value, and very few came back to the attention of the YOS for similar offences.

Discussion
This rapid evidence assessment of research examining whether restorative justice is an effective mechanism for responding to youth violence identified that although there are clear benefits to using restorative justice mechanisms for youth violence cases, there are also some challenges to implementation. However, there is also clear evidence that these challenges can be overcome, and existing research identifies some of the key delivery considerations that support successful restorative justice implementations into police services.

Benefits
The focus in this review is on evidence of the applicability of restorative justice mechanisms to youth violence (including for serious and sexual offences) and the benefits for both victim restoration and offender reformation. This review identifies clearly that restorative practice facilitates the promotion of peer leadership, accountability, ownership, agency and civic engagement. For instance, building positive intrapersonal relationships between young people, police, staff, teachers, administrators and community members through open communication and training all stakeholders (including young people) in conflict resolution and restorative practice has been demonstrated to be key to addressing violence, reducing the number of violent incidents, the use of disciplinary and exclusion measures, and improving perceptions of safety (Mateer 2010;Movsisyan 2014). Indeed, implementing restorative justice holistically has the potential to 'substantially reduce student offending, increase perceptions of safety, enhance learning, promote positive school climate, and destroy the school-to-prison pipeline that is exacerbating inequality and disadvantage for certain students' (Payne and Welch 2018, p. 237).
Since the introduction of Youth Offending Teams (YOTs) in the United Kingdom, through the Crime and Disorder Act 1998 and the Youth Justice and Evidence Act 1999 restorative justice has become one of the guiding principles of youth justice. That means that all YOTs should be based on restorative principles and the resulting community sentences given should also be restorative. However, restorative justice is often seen as being applicable to less serious crimes or anti-social behaviours involving young people. The research consulted makes it clear that restorative justice should not be confined to being a tool by which to deal with minor youth anti-social behaviour and offending (Daly 2006). In serious youth offences, including sexual violence, restorative justice has been found to be a highly effective response for victims and offenders alike (Daly 2006).
Restorative justice conferencing is associated with lower recidivism compared to court sentences (including community sentences); conferencing is associated with the lowest levels of recidivism (43 per cent) compared to service reoffending rates (56 per cent) (Daly 2006). There were no significant differences for extra-or intra-familial cases compared to courts, however, individuals going through a restorative justice process were more likely to admit responsibility without suffering long-term ostracism and victims were more likely to recount their experience of how the offence affected them (Daly 2006). It was also found that implementing a 'restorative justice victim empathy assault awareness course' (within YOTs) added to a substantial reduction in re-offending and enabled participants to reflect meaningfully on their use of violence and therefore reconsider the use of violence altogether (Wallis et al. 2013).
More specifically, restorative justice was identified as being particularly important for use with harmful sexual behaviours (HSB) perpetrated by children or youth. Where restorative justice practitioners worked collaboratively with child psychologists, social workers, and youth justice officers, the impacts were particularly beneficial in improving the life chances and outcomes for victims, children, or young persons who engage in HSB, and their families (Armstrong 2021). The key finding from this review identified that there is a need to recognise that youth sex offenders who have committed serious offences, including against other children, remain vulnerable due to their age and all responses need to take this into consideration (Blackley and Bartels 2018). There is considerable evidence to suggest that the criminal justice system is not the right response for youth who engage in HSB. Instead approaching harmful sexual behaviours as a child protection issue requires a therapeutic response and this is something that restorative justice practice incorporates.
The studies examined showed that restorative justice processes provide a holistic approach to offender reformation. This included opening a dialogue between a victim of crime and the responsible party that created the recognition that the parties belong to the same moral community. This is achieved by offering young people the opportunity to reflect and acknowledge their own destructiveness in a manner that fosters guilt and concern for the hurt caused to others (Froggett 2007). This process of 're-sensitisation' (Clark 2012, p. 87) is essential and counters the personal experiences of offending youth who have had repeated exposure to violence in their schools, communities, and homes. These experiences can result in youth becoming desensitised to violence and being unable to (fully) appreciate and understand the impact of their violent behaviour on others (Clark 2012). This recognition of belonging to a shared moral community and re-sensitisation meant that young people involved in restorative justice processes felt more inclined to repay the victim/society for the harm caused (Froggett 2007;Clark 2012;Kim and Gerber 2012). Perhaps even more importantly, participating in restorative justice conferences shows evidence that young offenders are more likely to repent and have the confidence in their ability to make amends (Kim and Gerber 2012). The studies consulted showed an increased likelihood of repairing the relationship between the harmed and the harmer (Froggett 2007).
There was also robust evidence showing the benefit of restorative justice for victims of violent crimes. Victim restoration begins with empowering victims of violent crimes, including victims of sexual violence, and this was enabled though the implementation of restorative processes which give individuals confidence that they can keep themselves safe (Armstrong 2021;Brathay Trust 2017). Victim empowerment was accompanied by RJ participants reporting improved feelings of self-worth; better communication skills; greater awareness of self and of situational contexts; and better recognition of personal strengths (Brathay Trust 2017). These feelings promoted through restorative conferencing and dialogue, offered survivors of violence the opportunity to seek answers to questions that the traditional criminal justice system does not provide (Armstrong 2021).

Challenges
Despite the evidenced benefits of the application of restorative justice in youth justice outlined above, the studies, within this review, identified several key challenges for implementation and practice of restorative justice as a response to youth violence. These challenges include professional scepticism and resistance, implementation and capacity, inclusion of participants, training, and addressing structural socio-economic underlying factors.
The reviewed studies highlighted that collaboration, support and 'buy-in' of criminal justice professionals and victim services is essential to the implementation and success of restorative justice programmes, the growth in the number programmes available, and adequate generation of referrals (see Armstrong 2021, p. 6). Scepticism and resistance from other professionals (including victim advocacy groups); a lack of case workers; staff retention levels; and slow referrals can hamper the success of restorative justice programmes (Armstrong 2021;Brathay Trust 2017). Implementation and delivery of restorative justice requires multidisciplinary and cross-sectional cooperation of institutions, civil society organisations and local communities and networking with international organisations (Peurača and Vejmelka 2015). Insufficient organisational and individual professional support, 'buy-in' and resistance and scepticism can result in disengagement with the restorative justice process (Brathay Trust 2017; Armstrong 2021) and prevent restorative justice being systematically and institutionally embedded into youth justice practice (Utheim 2011, p. 292).
One of the major challenges identified is that the 'promise' of restorative justice 'to give victims a voice is often unfulfilled' (Gal 2021, p. 144). Too few victims (especially when the victim is themselves a child) engage directly with youth offender restorative justice interventions (see Criminal Justice Joint Inspection 2012; Gal 2021). Whilst the youth justice system is (meant to be) underpinned by the principles of restoration, responsibility and reparation, by its very nature youth justice is 'offender focused' (Armstrong 2021) which results in victims not being systemically included in restorative justice interventions (Gal 2021;Armstrong 2021). There is a tendency for contact with victims to be seen more as an 'administrative process' rather 'than a fundamental part' of the youth justice system response to violent crime (Criminal Justice Joint Inspection 2012, p. 10). Participation in restorative justice is constrained by professionals 'exclusionary' decision-making processesvictims (and offenders) are routinely excluded and not offered the opportunity to participate in a restorative justice intervention (Banwell-Moore 2019; Criminal Justice Joint Inspection 2012). Moreover, the Criminal justice Joint Inspection review of restorative justice (2012, p. 40) identified that the youth justice system's 'offender focus' meant that 'victims were much less likely to become involved' due to perceiving the process to be 'primarily for the benefit of the offender'. The review argued that 'more needs to be done to improve the quality of the initial contact that youth justice professionals have with victims' (Criminal Justice Joint Inspection 2012, p. 40).
There is a lack of understanding of restorative justice amongst youth justice and victim services professionals. Many professionals do not have the necessary practical training to conduct restorative conferences or to deal with victims within youth justice referral order panels (Criminal Justice Joint Inspection 2012). Facilitating restorative justice requires trained and highly skilled individuals (Mann 2016;Ohmer et al. 2010). Training and awareness of restorative justice needs to take place within criminal justice and victim service organisations and across disciplines, particularly in school settings (Mateer 2010). Mateer (2010) argues that even when professionals' positions are not primarily focused on restorative justice, exposure to those values enables them to use restorative principles when the need arises. It is important for people in leadership to be trained even if their primary responsibility is not restorative justice (Mateer 2010). Widespread specialist training of all youth justice professionals 'in the principles and practice of restorative justice' would ensure that restorative justice 'becomes the centerpiece of responses' to youth violence (The Independent Commission on Youth Crime and Antisocial Behaviour 2010, p. 67).
Another challenge to restorative justice as a response to youth offending is that there is a tendency to see the 'problem' of violence as an individual issue and therefore provide only individual solutions. The concept of violence stemming from structural, societal constructs is often not considered as part of the restorative process (Brathay Trust 2017). "It could be argued that young people are thus disciplined via restorative practices to adhere to a paradigm in which structural problems, such as racism or poverty, are rescaled to the individual" (Schulz et al. 2021(Schulz et al. , p. 1240. Therefore, serious attention must be paid to the critical capacities of staff to use restorative practices in ways that do not rationalise 'violence' and 'hot emotions' as products of 'pathological' individuals (Schulz et al. 2021(Schulz et al. , p. 1241. Transforming and challenging myriad forms of violence requires providing youth with opportunities to enhance individual agency and examine and collectively rethink the power relations in which they are enmeshed (Schulz et al. 2021(Schulz et al. , p. 1240. Indeed, many young perpetrators are themselves victims (Brathay Trust 2017).

Delivery Considerations
The results from the included studies provided useful guidance on issues around design and delivery of restorative services intended to reduce youth violence both in volume and impacts. Schulz et al. (2021) demonstrate that restorative programmes dealing with youth violence should not be seen as simplistic solutions to complex structural problems, but when performed properly they provide a capacity to challenge and transform violence by supporting young people (and those supporting them) to come together to address these issues. Similarly, Utheim (2011, p. 292) argues that when restorative justice practices are implemented as 'systematized and institutionalized restorative praxis', they can substantially reduce youth violence and offending. Common across these studies, and within much of the other work identified as part of this study, is the importance of such schemes as whole systems approach that involve partnership working and collaboration between services. Such integrated approaches incorporate various groups including: school students, school and other agency staff, community members, professional support agencies, and police officers. Where such services exist, they often have a positive effect on: discipline issues and actions including suspension; exclusion rates; incidents of violence; negative behaviour; bullying; mental health issues and wellbeing; smoking, alcohol and drug use; police contact; and a reduction in recidivism (Mann 2016;Kelly 2017).
Both Harden et al. (2015) and Lane et al. (2007) argue that restorative justice or restorative practice programmes that target youth violence need to be comprehensive, involve multiple agencies, and be developed as a phased or staged approach. Evidence for the importance of this systematised and multi-stakeholder approach was found across several papers in different country contexts. For example, a US study (US Department of Justice 2016, p. 69), demonstrated that restorative justice programmes reduced the number of youth offenders on probation attached to gang terms and conditions when those programmes were combined with strong 'individualised client-specific services, evidence-based practices, restorative justice practices, cross-agency data sharing, re-entry services, and parent support'. In this scheme, officers would work with families and engage community and public assets to support crisis situations. Interventions include 'mediations and home visits to address the most serious violence offense suspensions and chronic truancy offenders' (US Department of Justice 2016, p. 71). 'School' police officers, as part of a collaborative multidisciplinary team, were able to sign-post and refer the 'at need' young people to the correct services (US Department of Justice 2016).
Elsewhere in the US, Smokowski et al. (2018) demonstrated the cumulative benefits of packaged programme delivery in the North Carolina Youth Violence Prevention Center identify. In this example, the programme comprised three evidence-based elements aims at decreasing youth violence; a universal 'Positive Action' school-based program; a parenting skill course, 'Parenting Wisely'; and a community-based restorative justice programme, 'Teen Court'. This integrated programme reduced non-school based offences by 47%; school-based offenses, short-term suspensions, and assaults all decreased; participants in the online parenting skills education had more confidence and less conflict with their children; participants of the restorative justice alternative for youth offenders reported improved mental and behavioural health; and there was a decrease across some indicators of violence at the county level (Smokowski et al. 2018).
In Croatia, peer mediation in schools similarly demonstrated that the most effective approaches in tackling youth violence were restorative approaches that incorporate multidisciplinary and cross-sectional cooperation of institutions, civil society organisations, local communities, and networks of related international organisations (Peurača and Vejmelka 2015). Similarly, in Northern Ireland, Harland (2011) identified the benefits of multi-agency approaches in which youth workers in partnership with local community-led programs and peacebuilding processes play an important role in engaging with the wider community to address conflict and violence.
The engagement of law enforcement officials in these multi-dimensional restorative youth-violence projects was important. However, restorative services for young people at risk of, or having engaged in violent acts, often involved law enforcement officials in a variety of contexts. For example, several studies talked about the importance of either officer (sometimes specialist school's officers) focusing on building trust in communities, particularly communities that were marginalised (US Department of Justice 2016). Officers would often target locations with high levels of youth violence, truancy, and suspensions, with restorative justice intervention programmes. This also contributed to victims making changes to staying safe, particularly in circumstances that include domestic abuse education and access to other services and relevant support (Brathay Trust 2017).
For such schemes to work, the evidence suggests that an efficient referral system, robust exit routes for those that might not complete the programmes, and awareness and signposting to other forms of support, services, and resources are required (Brathay Trust 2017;Mateer 2010). The issue of awareness was common across many of the studies in this assessment, not just in the context of signposting to associated support services but on the benefits, purpose, and practices involved in delivering restorative justice and practices to young people in the context of violent behaviours (Mateer 2010;Movsisyan 2014;Ohmer et al. 2010).

Conclusions
The studies selected for this review suggest that restorative justice and restorative practice can form an important part of an overall policing strategy to help reduce both the incidents of youth violence as well as the longer-term impacts of that violence when it has taken place. It is notable that this includes sexual violence and other serious physical violence, that these approached have the potential to also reduce reoffending, and to reduce reoffending offence severity over more commonly used retributive youth justice sanctions.
It is also clear that, in order to maximise the impact of these approaches in policing and the criminal and youth justice system more widely, it is important to adopt a whole-system approach in which restorative justice and practice is embedded into the organisational culture of schools, youth justice services, police forces, and other agencies that deal with victims, young perpetrators, and the communities in which they live. Such approaches, when conducted properly, can aid in breaking the 'school to prison pipeline', can improve victim satisfaction and wellbeing, and can support young perpetrators of violent crime in gaining a greater awareness of the impact of their behaviour.
In the context on the MPS ongoing review around the use of restorative justice, including in cases of youth violence, it is clear that there is a growing demand in UK policing /policy for greater use of restorative justice. The UK Government's (Ministry of Justice/Youth Justice Board 2018) Referral Order Guidance, Youth Select Committee (2019) report on restorative justice use with knife crime, the Restorative Justice APPG (2021) report on expanding the use of restorative justice all point towards a renewed emphasis on the integration of restorative approaches in policing. At the international level, an increasing number of countries are adopting various restorative principles into their criminal justice systems and into the mechanisms working with young people (Pali and Maglione 2021;Roach 2013;Ali Al-Hassani 2021;Zhang and Xia 2021). In Europe, this has taken the form of increasingly direct language over the integration of restorative justice into policy (Marder 2020b Taken together it is clear that, whilst not without delivery considerations, restorative justice and restorative practices can offer valuable tools to those seeking to support young people at risk of, or dealing with the consequences of, violence. We recognise that, in the context of violence and young people effective restorative justice police practice should embrace a whole-system approach that incorporates multi-agency working and consistently engages with young people at risk of, having undertaken, or having been victims of, violent crimes.

Data Availability Statement:
The origical Rapid Evidence Assessment on which this work was based can be found at: https://eprints.glos.ac.uk/11071/.