![Restorative Justice in Schools: A Complete Guide for UK Teachers [2026]](https://cdn.prod.website-files.com/plugins/Basic/assets/placeholder.60f9b1840c.svg)
A systematic review of 34 studies found restorative practices improve school climate, reduce suspensions, and develop social-emotional skills. This guide covers implementation, evidence, and practical strategies for UK schools.
Restorative justice in schools saw a 51% reduction in student exclusions in Barnet, whilst non-restorative schools saw exclusions increase by 65% in the same period. This is not a new fad—restorative approaches to managing behaviour have decades of research backing them, yet many UK schools remain unconvinced. This guide explains what restorative justice is, the evidence that supports it, how to implement it in your school, and the common barriers you'll face.
Restorative justice is a philosophy of behaviour management that prioritises repairing harm over punishment. When a conflict or wrongdoing occurs, the school brings together the person who caused harm, the person harmed, and relevant supporters to discuss what happened, why it mattered, and how to make things right.
This approach originates from criminal justice systems in New Zealand and Australia (Braithwaite, 1989), where it was used as an alternative to incarceration. The core question is not "Who broke the rule and what punishment do they deserve?" but rather "Who has been affected, what do they need, and how can the harm be repaired?"
Context shapes restorative practices. Informal chats between learners and staff work. Facilitated dialogues help two parties (Hopkins, 2004). Formal conferences involve stakeholders (Wachtel, 2016). Dialogue, accountability, and relationship repair are central (Zehr, 2015).
In UK schools, restorative justice typically begins with restorative conversations—brief one-to-one check-ins where staff ask open questions: "What happened? Who was affected? What do you need to happen next?" This practice shifts the responsibility for problem-solving from adults to learners, building both accountability and autonomy.
Traditional behaviour management uses consequences to deter misbehaviour. (eg, detention, exclusion). Restorative approaches, however, hold learners accountable relationally (Zehr, 1990; Wachtel, 2016). Learners face the impact of their actions on others and repair harm.
| Aspect | Traditional Behaviour Management | Restorative Justice |
|---|---|---|
| Focus | Punishment for breaking the rule | Repairing harm and rebuilding relationships |
| Who decides the outcome? | Staff apply predetermined consequences | Those affected collectively decide next steps |
| Learner role | Passive recipient of consequence | Active participant in problem-solving |
| Relationship impact | Often damages adult-learner relationship | Aims to strengthen relationships through dialogue |
| Time investment | Quick (decision made by staff in minutes) | Slower (requires dialogue and planning) |
| Outcome | Learner serves consequence and moves on | Relationships repaired; behaviour less likely to repeat |
Restorative chats address minor issues and damaged relationships in UK schools. Formal consequences remain for serious incidents (Hopkins, 2011). This mixed approach balances learner accountability and relationship repair (Wachtel, 2016; Costello, Wachtel & Wachtel, 2009).
Researchers have been testing restorative practices in schools for over 20 years. The evidence base is substantial and mostly positive, though not without limitations.
Systematic review findings. A 2021 systematic review by Fernández-Martín and colleagues examined 34 rigorous studies of restorative practices in schools worldwide. Their key findings: restorative circles were the most commonly used practice (appearing in 26 of the 34 studies), and schools that implemented restorative approaches showed improvements in school climate, reduced suspensions, and gains in learners' social-emotional skills. However, implementation quality varied widely, and schools with stronger training and leadership commitment saw better results (Fernández-Martín et al., 2021).
Large-scale US evidence. A 2023 study by Darling-Hammond and colleagues tracked 485 middle schools over 6 years, with over 3 million learner records. Schools that implemented restorative justice saw improved academic achievement, reduced suspensions, and notably, Black and Latino learners benefited most—suggesting restorative approaches may help narrow racial equity gaps in discipline (Darling-Hammond et al., 2023).
Cluster randomised trial. Acosta and colleagues (2019) conducted a gold-standard cluster randomised controlled trial with 13 middle schools and 2,771 learners. They found restorative practices significantly improved school climate and learner connectedness, even in schools with high baseline conflict. However, effects on individual academic outcomes were modest (Acosta et al., 2019).
Mixed results on absenteeism. A US-focused review by González (2020) found that whilst restorative practices reduce misbehaviour and suspensions, effects on chronic absenteeism are mixed. Restorative justice alone does not solve attendance problems; schools must address underlying causes (transport, safeguarding, poverty, mental health) in parallel.
Gomez-Garibello et al. (2022) found restorative justice costs schools about $139 per learner initially. This includes training, resources, and programme setup. Exclusion costs schools £5,000-£50,000, considering alternative provision and support.
The UK Barnet evaluation. Local authority evaluations of schools with established restorative justice programmes in Barnet (2019-2022) showed a 51% reduction in exclusions in RJ-trained schools, whilst comparable non-RJ schools experienced a 65% increase in exclusions during the same period. This suggests significant real-world impact in the UK context, though the absence of a randomised control means we cannot definitively isolate restorative justice as the sole cause.
Overall: Restorative justice is evidence-based. Schools should implement it for relational harm, conflict, and behaviour management—but recognise it is one tool, not a complete solution to all school challenges.
Restorative approaches exist on a spectrum from informal to formal. Here are the main types used in UK schools.
A brief one-to-one dialogue between staff and learner, or between two learners and a mediator. Typically 5-15 minutes. Staff ask open questions: "What happened? Who was affected? What do you think needs to happen now?" This is the most lightweight restorative practice and works well for low-level misbehaviour, conflicts between friends, or a learner who has had a difficult day.
Example: A learner was rude to another during group work. Rather than detention, the teacher asks: "Tell me what happened from your perspective. Who did that affect? What do you think they felt? What could you do to repair things?" The learner offers to apologise and suggest a way to work better together next lesson.
A facilitated group dialogue where 6-20 people sit in a circle and discuss a specific incident or issue. A trained facilitator uses a talking piece (object that passes around the circle; the person holding it speaks) and a structured prompt sequence. Circles take 45-90 minutes and work well for whole-class relationship building, or to address a specific incident that affected multiple learners.
Example: A year group experienced conflict between friendship groups. The school holds a circle where each learner shares their feelings, needs, and what they can commit to. By the end, the group collectively agrees on norms for how they'll treat each other.
More formal than circles, typically involving the person who caused harm, the person harmed, and their supporters (parents, mentors, friends). A trained facilitator guides the conversation. Conferences are used for more serious harm—assault, bullying over time, theft—and take 60-120 minutes. The goal is not to decide punishment, but for the harm-doer to understand impact and collaboratively decide on repair.
Example: A learner bullied another over several weeks. The school convenes a conference with both learners, their parents, and a teaching assistant. The bullied learner explains impact: "I don't want to come to school. I've lost my confidence." The harm-doer's parents are shocked; together, the group agrees on: written apology, supervised group interactions, a buddy system, and weekly check-ins. Three months later, both learners report the relationship has improved and school is safer.
Trained peer mediators help two learners resolve a conflict or misunderstanding. Mediators are often secondary learners trained in active listening and neutral facilitation. This works for disputes between friends, disagreements over fairness, or miscommunications. Peer mediation builds leadership in student mediators and teaches conflict resolution to all learners.
Proactive circles held regularly (weekly or fortnightly) in a class or form group with no specific incident to address. Facilitators use prompts to build connection: "What's something kind someone did for you this week?" or "What's a challenge you're facing?" Community circles prevent harm by building relationships and psychological safety. Learners feel known and cared for, which correlates with better behaviour and mental health.
Implementing restorative justice is a 3-year process, not a 6-week initiative. Schools that have seen success follow a phased approach with strong leadership and adequate training.
Get senior leadership buy-in first. Without headteacher and SLT commitment, restorative justice will fail. Schedule a meeting with your behaviour lead, assistant head, and headteacher. Present the evidence (use this article). Address the common concern: "Won't this be slower and softer than consequences?" Explain that the goal is fewer repeat incidents, not immediate compliance—speed is less important than effectiveness.
Start with staff training. Book restorative justice training for your staff (half-day to full-day). Trainers from organisations like Transforming Conflict or Restorative Approaches Limited offer bespoke school training. Cover: principles of RJ, how to facilitate conversations and circles, when RJ is appropriate vs when formal discipline is needed, managing defensiveness and strong emotions.
Pilot with willing staff. You do not need every teacher on board immediately. Identify 3-4 staff members (form tutors, key workers, behaviour lead) willing to pilot restorative conversations and circles with their tutor groups. They will champion the approach and build confidence in others.
Integrate restorative approaches into your behaviour policy. The policy update should allow restorative chats for low-level issues. For example, staff may offer learners restorative conversations (Hopkins, 2011). Handle serious incidents through formal discipline first. Restorative approaches may follow safeguarding (Tsang, 2019).
Start with community-building circles. The easiest entry point is weekly 10-minute circles in form time. Use simple prompts: "What's something you're grateful for today?" or "Share one win from this week." This builds relational culture with zero behavioural risk—no harm to repair, just connection. Learners begin to trust the process.
Embed restorative conversations as standard practice. By Year 2, restorative conversations should be your default for low-level misbehaviour (talking in lesson, incomplete homework, minor conflicts). Staff should be confident asking: "What happened? Who was affected? What do you need to happen?" without a trainer present.
Introduce formal restorative circles and conferences. As staff confidence grows, begin facilitating restorative circles for class-wide issues (Year 7 form group conflicts, peer relationships) and conferences for more serious incidents (bullying, assault, theft). You may need to hire an external restorative justice facilitator for conferences; this is worth the cost—a poorly facilitated conference can retraumatise the harmed party.
Create clear referral pathways. Establish who decides when to offer a restorative approach: behaviour lead, form tutors, class teachers? Write a one-page flowchart: "Use restorative conversation if incident is low-level and both parties willing. Use conference if harm is serious and both parties consent." Post it in the staffroom.
Monitor and measure. Track: number of restorative conversations, conferences held, repeat incidents (did the same learner re-offend?), staff confidence (survey), and exclusions (should trend down). Do not expect overnight change—look for 10-15% reduction in exclusions in Year 2.
Integrate RJ into PSHE and tutor time. Use restorative language in all settings. Teach learners the vocabulary: "harm", "affected", "repair", "accountability". In PSHE, run lessons on conflict resolution, empathy, and social-emotional skills—the foundations for restorative thinking.
Expand peer mediation. Train a second cohort of peer mediators. Publicise the service to learners: "Have a conflict? Ask for a peer mediation session." This shifts responsibility for minor disputes from staff to learners, freeing up staff time and building learner agency.
Measure cultural change. Conduct a school climate survey: "Do you feel safe here? Are relationships respectful? Does the school care about repair, not just punishment?" Improved scores indicate culture shift.
Address safeguarding boundaries explicitly. By Year 3, staff should be very clear: restorative approaches are not appropriate for allegations of abuse, serious violence, or criminal behaviour. These must go through formal discipline and child protection procedures immediately. Restorative conferences may follow much later, if both parties consent.
Implementation looks different in primary and secondary schools.
Primary learners have less developed theory of mind and emotional regulation. Restorative practices work best when simple, frequent, and embedded in day-to-day relationships.
Restorative conversations are the primary tool. Teachers ask: "What happened? What were they feeling when you did that?" or "How do you think that made them feel?" Young learners struggle with abstract concepts, so stay concrete: "When you pushed, they fell and got hurt." Always end with action: "What will you do different next time?"
Circles in primary are shorter (5-10 minutes) and use physical objects (talking piece) to keep attention. Prompts should relate to their world: "Who helped you this week?" or "What's something you're proud of?"
Parental involvement is critical. Primary learners' behaviour is still very shaped by home. Restorative conversations are more powerful when parents reinforce at home. A note home: "Your child and another had a conflict today. We've talked about how the other felt. Can you ask them about it and help them think of a way to repair things?"
Secondary learners have stronger theory of mind, peer allegiances, and autonomy drive. Restorative approaches must respect their growing independence and peer relationships.
Restorative conversations work well if framed as dialogue, not interrogation. Learners will shut down if they feel blamed. Frame it: "I want to understand your perspective. What happened from your view?" Acknowledge their autonomy: "I'm not here to punish you; I want you to think through impact and decide what repair looks like."
Circles and conferences are powerful in secondary because peer opinion matters. When a learner hears from peers how their behaviour affected them, it carries more weight than a teacher saying so. Peer mediators are also highly effective in secondary—learners often talk to peers before adults.
Boundaries matter. Secondary learners push back if they perceive unfairness. Be very clear: restorative justice is voluntary, and if they refuse, formal consequences apply. This clarity often makes learners more willing to engage.
Most schools encounter resistance when implementing restorative justice. Here are the big ones.
The objection: "This is soft. Learners won't take it seriously. We'll lose authority."
The reality: Restorative approaches hold learners more accountable, not less. A learner in detention has 1 hour to sit and think. A learner in a conference must face the person affected, hear their pain, and collaboratively solve the problem. This is harder than detention.
How to address it: Invite sceptics to observe a restorative conversation or circle. Let them see the learner engaged, thinking critically, and taking responsibility. One observation converts more staff than ten conversations about theory.
The objection: "We don't have time for this. Teachers are already stretched."
The reality: Restorative conversations take 5-15 minutes. They replace punishment conversations that take 5-15 minutes anyway. Over time, fewer repeat incidents mean less overall time spent on behaviour.
How to address it: Calculate the time saved. If exclusions drop from 40 to 20 per year, your school saves approx. 40 hours of reintegration meetings, SEN review meetings, and alternative provision coordination. Use that evidence to justify training time upfront.
The objection: "Learners won't engage. They'll refuse to apologise or they'll be insincere."
The reality: Some learners will be reluctant initially. This is normal. The key is offering choice: "You can have a conversation with me, or we can go straight to a consequence. What would you prefer?" Many learners choose dialogue once they realise it's fair and focuses on understanding, not blame.
How to address it: Never force a learner into a restorative process. If they refuse, apply the standard consequence. However, be clear: "If you'd like a chance to repair things, I'm open to that." Often, once the emotion cools, learners come back requesting restorative dialogue.
The objection: "Some staff facilitate conversations well. Others make it worse."
The reality: Quality varies. A poorly facilitated conversation where the adult does all the talking or doesn't listen to the learner's perspective defeats the purpose.
How to address it: Invest in ongoing training, not one-off induction. Annual refresher sessions, peer observations, and reflective practice (staff discussing "how did that conversation go?") build consistency. Have one lead facilitator who coaches others.
The objection: "Is restorative justice safe for allegations of abuse or assault?"
The reality: No. Restorative justice is never appropriate as a first response to abuse or serious violence. Child protection and formal discipline must come first. However, restorative approaches may be offered much later (months or years) if both parties consent, as part of healing and reintegration.
How to address it: Build this distinction into your policy explicitly. Train staff on red lines: "If there is evidence of abuse, violence, or criminal behaviour, you go to the designated safeguarding lead and behaviour lead immediately. Restorative processes come later, if at all." This clarity protects both learners and staff.
Learners with social, emotional, and mental health needs (SEMH) are disproportionately excluded from school. Restorative approaches can support these learners, though implementation requires adjustments.
Reduces anxiety. Learners with anxiety often avoid conflict because they fear adult anger. Restorative conversations, framed as understanding rather than interrogation, reduce anxiety. A learner thinks: "They want to understand, not blame me" rather than "I'm in trouble."
Restorative practice builds belonging. SEMH learners often feel disconnected. Community circles foster relational safety (Wachtel, 2016). Regular sharing creates psychological safety, preventing behavioural problems (Cameron & Thorsborne, 2001).
Teaches emotion regulation. Restorative conversations model how to name emotions, reflect on impact, and problem-solve. Learners with poor emotion regulation begin to internalise these skills through repeated, low-pressure dialogue.
Increases agency. Many excluded learners have internalised a narrative: "I'm bad. I always mess up." Restorative approaches shift this: "This behaviour caused harm, and you have the power to repair it." This builds agency and hope.
Adapt facilitation for emotional intensity. An emotionally dysregulated learner may not be ready for a dialogue immediately after an incident. Wait 30-60 minutes for emotions to cool. Start with a brief conversation with a trusted adult (key worker, learning mentor) before introducing peers.
Use shorter, simpler language. Learners with lower processing speed or communication difficulties need shorter prompts: "What happened?" (pause for answer) "How did that affect them?" (pause) rather than multi-part questions.
Involve parents or carers proactively. SEMH learners often benefit from parental support. Ring home immediately: "Your child had a conflict today. We're working on repair. Can you support at home?" This consistency improves outcomes.
Link to trauma-informed practice. Many SEMH learners have trauma histories. Restorative approaches align with trauma-informed teaching principles: safety, choice, collaboration, trustworthiness. A conversation framed as "I'm here to help you think this through" feels safer than "You broke the rule."
Use restorative circles carefully. Full circles may be too intense for a learner in crisis. Instead, use smaller circles (4-6 people) or one-to-one conversations. Build toward whole-group circles as the learner's regulation improves.
Restorative justice is powerful, but it is not a panacea. Be realistic about what it can and cannot achieve.
If a learner is being abused at home, a restorative conversation will not stop the abuse. Restorative justice addresses school-based harm and relationships. Serious safeguarding concerns require child protection procedures, not restorative dialogue.
Restorative justice is slower than traditional consequences. A learner can receive detention instantly; a restorative conference requires scheduling, facilitator training, and willingness from both parties. In the UK, where schools are under pressure to show quick behaviour improvements, this can feel frustrating. However, the research suggests that slower repair leads to fewer repeat incidents—so it is faster overall.
A learner can be forced into detention. They cannot be forced into a genuine restorative dialogue—they will sit silent and the process fails. If a learner refuses, you apply the standard consequence and move on. This means restorative justice works best for about 70-80% of incidents; the remainder require formal discipline.
If a learner is missing school because of transport insecurity, hunger, or untreated ADHD, a restorative conversation will not solve those problems. Restorative justice must be paired with whole-family support: referrals to counselling, breakfast clubs, SEND assessment, family liaison work.
The evidence base for restorative justice is strong, but most studies are in secondary schools and majority-white communities. Research on primary, special schools, and diverse communities is limited. Implementation quality varies widely—some schools have seen dramatic reductions in exclusions, others have seen minimal change. The variable findings likely reflect differences in staff training, whole-school commitment, and baseline culture.
Restorative justice is a powerful, evidence-based approach to behaviour management that repairs harm and rebuilds relationships. It is not a quick fix; schools that commit to it over 3 years see marked improvements in school climate, reduced exclusions, and better outcomes for vulnerable learners. Here are the essentials:
Schools like those in Barnet have demonstrated that restorative justice works in the real world, not just research. The investment is substantial—time, training, facilitator support. But the payoff—fewer excluded learners, better relationships, healthier school culture—is worth it.
Restorative justice shows promise, but limitations exist. Most research comes from secondary schools (mostly white) in North America and Australia. Little research exists for primary schools or diverse, deprived schools. Implementation varies; some schools cut exclusions by 70%, while others see little change. Training, commitment, and school culture likely affect results, (Hopkins, 2011; Morrison, 2006). We need more research to find what ensures success. Studies often measure exclusion rates and school climate, (Cameron & Thorsborne, 2001); adult outcomes need study. Schools must have realistic expectations; restorative justice is a helpful approach but not a cure-all, (Reimer, 2011).
Restorative justice in schools saw a 51% reduction in student exclusions in Barnet, whilst non-restorative schools saw exclusions increase by 65% in the same period. This is not a new fad—restorative approaches to managing behaviour have decades of research backing them, yet many UK schools remain unconvinced. This guide explains what restorative justice is, the evidence that supports it, how to implement it in your school, and the common barriers you'll face.
Restorative justice is a philosophy of behaviour management that prioritises repairing harm over punishment. When a conflict or wrongdoing occurs, the school brings together the person who caused harm, the person harmed, and relevant supporters to discuss what happened, why it mattered, and how to make things right.
This approach originates from criminal justice systems in New Zealand and Australia (Braithwaite, 1989), where it was used as an alternative to incarceration. The core question is not "Who broke the rule and what punishment do they deserve?" but rather "Who has been affected, what do they need, and how can the harm be repaired?"
Context shapes restorative practices. Informal chats between learners and staff work. Facilitated dialogues help two parties (Hopkins, 2004). Formal conferences involve stakeholders (Wachtel, 2016). Dialogue, accountability, and relationship repair are central (Zehr, 2015).
In UK schools, restorative justice typically begins with restorative conversations—brief one-to-one check-ins where staff ask open questions: "What happened? Who was affected? What do you need to happen next?" This practice shifts the responsibility for problem-solving from adults to learners, building both accountability and autonomy.
Traditional behaviour management uses consequences to deter misbehaviour. (eg, detention, exclusion). Restorative approaches, however, hold learners accountable relationally (Zehr, 1990; Wachtel, 2016). Learners face the impact of their actions on others and repair harm.
| Aspect | Traditional Behaviour Management | Restorative Justice |
|---|---|---|
| Focus | Punishment for breaking the rule | Repairing harm and rebuilding relationships |
| Who decides the outcome? | Staff apply predetermined consequences | Those affected collectively decide next steps |
| Learner role | Passive recipient of consequence | Active participant in problem-solving |
| Relationship impact | Often damages adult-learner relationship | Aims to strengthen relationships through dialogue |
| Time investment | Quick (decision made by staff in minutes) | Slower (requires dialogue and planning) |
| Outcome | Learner serves consequence and moves on | Relationships repaired; behaviour less likely to repeat |
Restorative chats address minor issues and damaged relationships in UK schools. Formal consequences remain for serious incidents (Hopkins, 2011). This mixed approach balances learner accountability and relationship repair (Wachtel, 2016; Costello, Wachtel & Wachtel, 2009).
Researchers have been testing restorative practices in schools for over 20 years. The evidence base is substantial and mostly positive, though not without limitations.
Systematic review findings. A 2021 systematic review by Fernández-Martín and colleagues examined 34 rigorous studies of restorative practices in schools worldwide. Their key findings: restorative circles were the most commonly used practice (appearing in 26 of the 34 studies), and schools that implemented restorative approaches showed improvements in school climate, reduced suspensions, and gains in learners' social-emotional skills. However, implementation quality varied widely, and schools with stronger training and leadership commitment saw better results (Fernández-Martín et al., 2021).
Large-scale US evidence. A 2023 study by Darling-Hammond and colleagues tracked 485 middle schools over 6 years, with over 3 million learner records. Schools that implemented restorative justice saw improved academic achievement, reduced suspensions, and notably, Black and Latino learners benefited most—suggesting restorative approaches may help narrow racial equity gaps in discipline (Darling-Hammond et al., 2023).
Cluster randomised trial. Acosta and colleagues (2019) conducted a gold-standard cluster randomised controlled trial with 13 middle schools and 2,771 learners. They found restorative practices significantly improved school climate and learner connectedness, even in schools with high baseline conflict. However, effects on individual academic outcomes were modest (Acosta et al., 2019).
Mixed results on absenteeism. A US-focused review by González (2020) found that whilst restorative practices reduce misbehaviour and suspensions, effects on chronic absenteeism are mixed. Restorative justice alone does not solve attendance problems; schools must address underlying causes (transport, safeguarding, poverty, mental health) in parallel.
Gomez-Garibello et al. (2022) found restorative justice costs schools about $139 per learner initially. This includes training, resources, and programme setup. Exclusion costs schools £5,000-£50,000, considering alternative provision and support.
The UK Barnet evaluation. Local authority evaluations of schools with established restorative justice programmes in Barnet (2019-2022) showed a 51% reduction in exclusions in RJ-trained schools, whilst comparable non-RJ schools experienced a 65% increase in exclusions during the same period. This suggests significant real-world impact in the UK context, though the absence of a randomised control means we cannot definitively isolate restorative justice as the sole cause.
Overall: Restorative justice is evidence-based. Schools should implement it for relational harm, conflict, and behaviour management—but recognise it is one tool, not a complete solution to all school challenges.
Restorative approaches exist on a spectrum from informal to formal. Here are the main types used in UK schools.
A brief one-to-one dialogue between staff and learner, or between two learners and a mediator. Typically 5-15 minutes. Staff ask open questions: "What happened? Who was affected? What do you think needs to happen now?" This is the most lightweight restorative practice and works well for low-level misbehaviour, conflicts between friends, or a learner who has had a difficult day.
Example: A learner was rude to another during group work. Rather than detention, the teacher asks: "Tell me what happened from your perspective. Who did that affect? What do you think they felt? What could you do to repair things?" The learner offers to apologise and suggest a way to work better together next lesson.
A facilitated group dialogue where 6-20 people sit in a circle and discuss a specific incident or issue. A trained facilitator uses a talking piece (object that passes around the circle; the person holding it speaks) and a structured prompt sequence. Circles take 45-90 minutes and work well for whole-class relationship building, or to address a specific incident that affected multiple learners.
Example: A year group experienced conflict between friendship groups. The school holds a circle where each learner shares their feelings, needs, and what they can commit to. By the end, the group collectively agrees on norms for how they'll treat each other.
More formal than circles, typically involving the person who caused harm, the person harmed, and their supporters (parents, mentors, friends). A trained facilitator guides the conversation. Conferences are used for more serious harm—assault, bullying over time, theft—and take 60-120 minutes. The goal is not to decide punishment, but for the harm-doer to understand impact and collaboratively decide on repair.
Example: A learner bullied another over several weeks. The school convenes a conference with both learners, their parents, and a teaching assistant. The bullied learner explains impact: "I don't want to come to school. I've lost my confidence." The harm-doer's parents are shocked; together, the group agrees on: written apology, supervised group interactions, a buddy system, and weekly check-ins. Three months later, both learners report the relationship has improved and school is safer.
Trained peer mediators help two learners resolve a conflict or misunderstanding. Mediators are often secondary learners trained in active listening and neutral facilitation. This works for disputes between friends, disagreements over fairness, or miscommunications. Peer mediation builds leadership in student mediators and teaches conflict resolution to all learners.
Proactive circles held regularly (weekly or fortnightly) in a class or form group with no specific incident to address. Facilitators use prompts to build connection: "What's something kind someone did for you this week?" or "What's a challenge you're facing?" Community circles prevent harm by building relationships and psychological safety. Learners feel known and cared for, which correlates with better behaviour and mental health.
Implementing restorative justice is a 3-year process, not a 6-week initiative. Schools that have seen success follow a phased approach with strong leadership and adequate training.
Get senior leadership buy-in first. Without headteacher and SLT commitment, restorative justice will fail. Schedule a meeting with your behaviour lead, assistant head, and headteacher. Present the evidence (use this article). Address the common concern: "Won't this be slower and softer than consequences?" Explain that the goal is fewer repeat incidents, not immediate compliance—speed is less important than effectiveness.
Start with staff training. Book restorative justice training for your staff (half-day to full-day). Trainers from organisations like Transforming Conflict or Restorative Approaches Limited offer bespoke school training. Cover: principles of RJ, how to facilitate conversations and circles, when RJ is appropriate vs when formal discipline is needed, managing defensiveness and strong emotions.
Pilot with willing staff. You do not need every teacher on board immediately. Identify 3-4 staff members (form tutors, key workers, behaviour lead) willing to pilot restorative conversations and circles with their tutor groups. They will champion the approach and build confidence in others.
Integrate restorative approaches into your behaviour policy. The policy update should allow restorative chats for low-level issues. For example, staff may offer learners restorative conversations (Hopkins, 2011). Handle serious incidents through formal discipline first. Restorative approaches may follow safeguarding (Tsang, 2019).
Start with community-building circles. The easiest entry point is weekly 10-minute circles in form time. Use simple prompts: "What's something you're grateful for today?" or "Share one win from this week." This builds relational culture with zero behavioural risk—no harm to repair, just connection. Learners begin to trust the process.
Embed restorative conversations as standard practice. By Year 2, restorative conversations should be your default for low-level misbehaviour (talking in lesson, incomplete homework, minor conflicts). Staff should be confident asking: "What happened? Who was affected? What do you need to happen?" without a trainer present.
Introduce formal restorative circles and conferences. As staff confidence grows, begin facilitating restorative circles for class-wide issues (Year 7 form group conflicts, peer relationships) and conferences for more serious incidents (bullying, assault, theft). You may need to hire an external restorative justice facilitator for conferences; this is worth the cost—a poorly facilitated conference can retraumatise the harmed party.
Create clear referral pathways. Establish who decides when to offer a restorative approach: behaviour lead, form tutors, class teachers? Write a one-page flowchart: "Use restorative conversation if incident is low-level and both parties willing. Use conference if harm is serious and both parties consent." Post it in the staffroom.
Monitor and measure. Track: number of restorative conversations, conferences held, repeat incidents (did the same learner re-offend?), staff confidence (survey), and exclusions (should trend down). Do not expect overnight change—look for 10-15% reduction in exclusions in Year 2.
Integrate RJ into PSHE and tutor time. Use restorative language in all settings. Teach learners the vocabulary: "harm", "affected", "repair", "accountability". In PSHE, run lessons on conflict resolution, empathy, and social-emotional skills—the foundations for restorative thinking.
Expand peer mediation. Train a second cohort of peer mediators. Publicise the service to learners: "Have a conflict? Ask for a peer mediation session." This shifts responsibility for minor disputes from staff to learners, freeing up staff time and building learner agency.
Measure cultural change. Conduct a school climate survey: "Do you feel safe here? Are relationships respectful? Does the school care about repair, not just punishment?" Improved scores indicate culture shift.
Address safeguarding boundaries explicitly. By Year 3, staff should be very clear: restorative approaches are not appropriate for allegations of abuse, serious violence, or criminal behaviour. These must go through formal discipline and child protection procedures immediately. Restorative conferences may follow much later, if both parties consent.
Implementation looks different in primary and secondary schools.
Primary learners have less developed theory of mind and emotional regulation. Restorative practices work best when simple, frequent, and embedded in day-to-day relationships.
Restorative conversations are the primary tool. Teachers ask: "What happened? What were they feeling when you did that?" or "How do you think that made them feel?" Young learners struggle with abstract concepts, so stay concrete: "When you pushed, they fell and got hurt." Always end with action: "What will you do different next time?"
Circles in primary are shorter (5-10 minutes) and use physical objects (talking piece) to keep attention. Prompts should relate to their world: "Who helped you this week?" or "What's something you're proud of?"
Parental involvement is critical. Primary learners' behaviour is still very shaped by home. Restorative conversations are more powerful when parents reinforce at home. A note home: "Your child and another had a conflict today. We've talked about how the other felt. Can you ask them about it and help them think of a way to repair things?"
Secondary learners have stronger theory of mind, peer allegiances, and autonomy drive. Restorative approaches must respect their growing independence and peer relationships.
Restorative conversations work well if framed as dialogue, not interrogation. Learners will shut down if they feel blamed. Frame it: "I want to understand your perspective. What happened from your view?" Acknowledge their autonomy: "I'm not here to punish you; I want you to think through impact and decide what repair looks like."
Circles and conferences are powerful in secondary because peer opinion matters. When a learner hears from peers how their behaviour affected them, it carries more weight than a teacher saying so. Peer mediators are also highly effective in secondary—learners often talk to peers before adults.
Boundaries matter. Secondary learners push back if they perceive unfairness. Be very clear: restorative justice is voluntary, and if they refuse, formal consequences apply. This clarity often makes learners more willing to engage.
Most schools encounter resistance when implementing restorative justice. Here are the big ones.
The objection: "This is soft. Learners won't take it seriously. We'll lose authority."
The reality: Restorative approaches hold learners more accountable, not less. A learner in detention has 1 hour to sit and think. A learner in a conference must face the person affected, hear their pain, and collaboratively solve the problem. This is harder than detention.
How to address it: Invite sceptics to observe a restorative conversation or circle. Let them see the learner engaged, thinking critically, and taking responsibility. One observation converts more staff than ten conversations about theory.
The objection: "We don't have time for this. Teachers are already stretched."
The reality: Restorative conversations take 5-15 minutes. They replace punishment conversations that take 5-15 minutes anyway. Over time, fewer repeat incidents mean less overall time spent on behaviour.
How to address it: Calculate the time saved. If exclusions drop from 40 to 20 per year, your school saves approx. 40 hours of reintegration meetings, SEN review meetings, and alternative provision coordination. Use that evidence to justify training time upfront.
The objection: "Learners won't engage. They'll refuse to apologise or they'll be insincere."
The reality: Some learners will be reluctant initially. This is normal. The key is offering choice: "You can have a conversation with me, or we can go straight to a consequence. What would you prefer?" Many learners choose dialogue once they realise it's fair and focuses on understanding, not blame.
How to address it: Never force a learner into a restorative process. If they refuse, apply the standard consequence. However, be clear: "If you'd like a chance to repair things, I'm open to that." Often, once the emotion cools, learners come back requesting restorative dialogue.
The objection: "Some staff facilitate conversations well. Others make it worse."
The reality: Quality varies. A poorly facilitated conversation where the adult does all the talking or doesn't listen to the learner's perspective defeats the purpose.
How to address it: Invest in ongoing training, not one-off induction. Annual refresher sessions, peer observations, and reflective practice (staff discussing "how did that conversation go?") build consistency. Have one lead facilitator who coaches others.
The objection: "Is restorative justice safe for allegations of abuse or assault?"
The reality: No. Restorative justice is never appropriate as a first response to abuse or serious violence. Child protection and formal discipline must come first. However, restorative approaches may be offered much later (months or years) if both parties consent, as part of healing and reintegration.
How to address it: Build this distinction into your policy explicitly. Train staff on red lines: "If there is evidence of abuse, violence, or criminal behaviour, you go to the designated safeguarding lead and behaviour lead immediately. Restorative processes come later, if at all." This clarity protects both learners and staff.
Learners with social, emotional, and mental health needs (SEMH) are disproportionately excluded from school. Restorative approaches can support these learners, though implementation requires adjustments.
Reduces anxiety. Learners with anxiety often avoid conflict because they fear adult anger. Restorative conversations, framed as understanding rather than interrogation, reduce anxiety. A learner thinks: "They want to understand, not blame me" rather than "I'm in trouble."
Restorative practice builds belonging. SEMH learners often feel disconnected. Community circles foster relational safety (Wachtel, 2016). Regular sharing creates psychological safety, preventing behavioural problems (Cameron & Thorsborne, 2001).
Teaches emotion regulation. Restorative conversations model how to name emotions, reflect on impact, and problem-solve. Learners with poor emotion regulation begin to internalise these skills through repeated, low-pressure dialogue.
Increases agency. Many excluded learners have internalised a narrative: "I'm bad. I always mess up." Restorative approaches shift this: "This behaviour caused harm, and you have the power to repair it." This builds agency and hope.
Adapt facilitation for emotional intensity. An emotionally dysregulated learner may not be ready for a dialogue immediately after an incident. Wait 30-60 minutes for emotions to cool. Start with a brief conversation with a trusted adult (key worker, learning mentor) before introducing peers.
Use shorter, simpler language. Learners with lower processing speed or communication difficulties need shorter prompts: "What happened?" (pause for answer) "How did that affect them?" (pause) rather than multi-part questions.
Involve parents or carers proactively. SEMH learners often benefit from parental support. Ring home immediately: "Your child had a conflict today. We're working on repair. Can you support at home?" This consistency improves outcomes.
Link to trauma-informed practice. Many SEMH learners have trauma histories. Restorative approaches align with trauma-informed teaching principles: safety, choice, collaboration, trustworthiness. A conversation framed as "I'm here to help you think this through" feels safer than "You broke the rule."
Use restorative circles carefully. Full circles may be too intense for a learner in crisis. Instead, use smaller circles (4-6 people) or one-to-one conversations. Build toward whole-group circles as the learner's regulation improves.
Restorative justice is powerful, but it is not a panacea. Be realistic about what it can and cannot achieve.
If a learner is being abused at home, a restorative conversation will not stop the abuse. Restorative justice addresses school-based harm and relationships. Serious safeguarding concerns require child protection procedures, not restorative dialogue.
Restorative justice is slower than traditional consequences. A learner can receive detention instantly; a restorative conference requires scheduling, facilitator training, and willingness from both parties. In the UK, where schools are under pressure to show quick behaviour improvements, this can feel frustrating. However, the research suggests that slower repair leads to fewer repeat incidents—so it is faster overall.
A learner can be forced into detention. They cannot be forced into a genuine restorative dialogue—they will sit silent and the process fails. If a learner refuses, you apply the standard consequence and move on. This means restorative justice works best for about 70-80% of incidents; the remainder require formal discipline.
If a learner is missing school because of transport insecurity, hunger, or untreated ADHD, a restorative conversation will not solve those problems. Restorative justice must be paired with whole-family support: referrals to counselling, breakfast clubs, SEND assessment, family liaison work.
The evidence base for restorative justice is strong, but most studies are in secondary schools and majority-white communities. Research on primary, special schools, and diverse communities is limited. Implementation quality varies widely—some schools have seen dramatic reductions in exclusions, others have seen minimal change. The variable findings likely reflect differences in staff training, whole-school commitment, and baseline culture.
Restorative justice is a powerful, evidence-based approach to behaviour management that repairs harm and rebuilds relationships. It is not a quick fix; schools that commit to it over 3 years see marked improvements in school climate, reduced exclusions, and better outcomes for vulnerable learners. Here are the essentials:
Schools like those in Barnet have demonstrated that restorative justice works in the real world, not just research. The investment is substantial—time, training, facilitator support. But the payoff—fewer excluded learners, better relationships, healthier school culture—is worth it.
Restorative justice shows promise, but limitations exist. Most research comes from secondary schools (mostly white) in North America and Australia. Little research exists for primary schools or diverse, deprived schools. Implementation varies; some schools cut exclusions by 70%, while others see little change. Training, commitment, and school culture likely affect results, (Hopkins, 2011; Morrison, 2006). We need more research to find what ensures success. Studies often measure exclusion rates and school climate, (Cameron & Thorsborne, 2001); adult outcomes need study. Schools must have realistic expectations; restorative justice is a helpful approach but not a cure-all, (Reimer, 2011).
