Implementation of Whole School Restorative Approaches to Promote Positive Youth Development: Review of Relevant Literature and Practice Guidelines

: Positive youth development highlights the promotion of skills through engaging and caring settings and building opportunities for bidirectional and constructive relationships. Whole School Restorative Approaches (WSRA) promote school community relationships and social and emotional skills which are core components for positive youth development. To our knowledge, the literature reviews on the implementation of restorative practice approaches in schools do not focus speciﬁcally on WSRA and are not comprehensive. This study carries out a systematic review of the empirical evidence on the effectiveness of WSRA and develops some evidence-based practice guidelines on their effects. Thirteen studies met the inclusion criteria and we developed 10 evidence-based practice guidelines on the implementation of WSRA. There were many studies with WSRA showing positive results related to improving social and emotional skills and behavior. The highest level of scientiﬁc evidence is for secondary education where two WSRA are recommended. Although our review suggests that more rigorous methodological research is necessary, it also points out that the evidence available on WSRA implementation is encouraging. They could be used through all educational stages to deal with diversity and inclusion while promoting a school culture of peace based on the positive management of relationships and conﬂicts.


Introduction
The plasticity of individual development comes from the integration of contextual and biological systems [1]. One could take part in plasticity of developmental trajectories by promoting bidirectional and positive relationships between individuals and their context and, thus, supporting adaptive development [2]. Considering the above, positive youth development evolves as a framework that emphasizes the promotion of capacities by means of engaging and caring environments and creating opportunities for bidirectional and beneficial youth-context relationships rather than focusing on deficits [3][4][5]. Such an approach goes beyond single-level efforts and addresses a complex interaction of levels [6] including family, peer, school, and community environments. It is based on the importance of a successful attainment of milestones in terms of psychosocial development involving, for instance, social, emotional, and behavioral skills [7]. In particular, developmental outcomes in the acquisition and improvement of these skills can be influenced by the interaction

Materials and Methods
A systematic literature review is considered a compilation of research results carried out to previous criteria and in order to answer a research question. In order to be valid and of quality, a systematic literature review must be based on neutral criteria that enable the procedure to have clarity, transparency, and accuracy. We used a type of systematic review known as a scoping review which is generally used when a body of evidence has not yet been fully reviewed, or exhibits a great, compounded, or diverse nature not manageable to a more specific systematic review. The review was conducted considering the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [27].

Literature Search
Literature searches were designed to be as sensitive as possible in order to obtain as many papers as possible on the effects of WSRA in schools. In order to locate and identify significant studies meeting the inclusion criteria, we carried out focused searches in the general literature. Inclusion criteria were all types of study designs (i.e., meta-analyses, systematic reviews, randomized controlled trials, case-control studies, cohort studies, nonanalytic studies, and expert opinions) intended to study the implementation of WSRA in schools in primary and/or secondary education and considering any member of a school community (i.e., students, teachers, and/or families). One should note that designs with a high methodological level in the field of WSRA in schools are scarce; therefore, we considered all types of study designs in order to be able to reach all the scientific evidence available. We only considered studies published in peer-review journals, as in these journals papers are peer-reviewed, and thus, the evidence of the studies they publish is of high quality. We did not consider studies: (a) focused on pocket school restorative approaches because they aim to solve conflicts, and only when integrated with WSRA they promote positive youth-development; (b) on the effects of WSRA implementation in non-ordinary schools (e.g., special schools) since participants may have other needs beyond those of students in ordinary schools, which could make the integration of results difficult; and (c) published in journals with no peer-review. Therefore, dissertations, books, or magazines were not included. There were no time or language limits.
We conducted a computerized search to identify articles about the efficacy of WSRA in schools. Specifically, we conducted our computerized search in ProQuest (PsycINFO, Educational Resources Information Centre or ERIC, Social Services Abstracts, and Sociological Abstracts), PubMed, and Education Full-Text in November 2021. We used the following terms to conduct the search: "restorative", "justice", "practice", "circle", "conference", "mediation", and "school". The standard search strategy was as follows: restorati * AND justice OR practice * OR circle * OR conference * OR mediat * AND school *. In order to confirm the suitability of our search strategy, we compared the records retrieved in the databases with those published in high-impact journals where papers in this field may be presented. We also checked specialized websites such as National Educators for Restorative Practices, The International Institute for Restorative Practices, The Restorative Justice Council, Transforming Conflict, and The Schott Foundation. We additionally compared the records retrieved with others regarding systematic reviews close to the field of our study [14,[22][23][24][25][26]. We further reviewed the records of the identified articles. The search strategies may be redundant since they covered much of the same literature.

Analytic Method
This method has been used and considered valid for systematic reviews as an aid to determine the most highly rigorous evidence based on WSRA [28]. Specifically, it was intended to go through the internal validity of studies on the efficacy of the implementation of WSRA (see next section). To do so, we developed 15 research questions. We completed a cross tabulation of each selected study by 39 response categories in order to get frequencies and percentages of each of them.

1.
When were the studies conducted? The year when the study results were published was identified.

2.
Where were the studies conducted? The country where the study was conducted was identified. 3.
What was the range of the sample of target studies? The sample size across studies of target participants was identified. 4.
What was the setting of studies? Studies were assigned to one of the following categories: (a) primary education; (b) secondary education; and (c) both.

5.
Was there ethnic diversity in the context in which the WSRA was implemented? If so, how many ethnic groups were included in each study sample? Frequencies and percentages of studies conducted in contexts with ethnic diversity were calculated. The frequencies and percentages of the different ethnic groups in each study were categorized as follows: (a) two; (b) three; and (c) four or more. 6.
What type of research design was used? The frequencies and percentages of research design used were identified and categorized as follows: (a) systematic literature review; (b) randomized controlled trial; (c) quasi-experimental study with control group; (d) quasi-experimental study without control group; (e) qualitative study; and (f) case study. 7.
How long was the WSRA implemented for? Studies were assigned to one of the following categories: (a) one year or less; (b) from one year to two years; and (c) more than two years. The range of time of implementation of the studies was considered. 8.
What type of follow-up was used in the studies? Studies were assigned to one of the following categories: (a) no follow-up; (b) pre and post measurements; (c) pre and post measurement, and follow-ups. 9.
Which WSRA was implemented? The frequencies and percentages of each WSRA implemented were calculated. The WSRA implemented was identified across studies and essential elements were extracted. 10. What were the main outcomes of studies? Studies were assigned to one or more of the  [29]. Practice guidelines were assigned to one of the following grades: (1) A; (2) B; (3) C; (4) D.
To address questions 14 and 15, we used the SIGN method [29] as described in Table 1. This method is intended for the evaluation of the quality of the evidence of research studies, or in other words, it establishes the degree of confidence we can have in the results of selected studies. To do so, it considers the rigorousness of the study design as well as the risk of bias and the probability that there is a causal relationship between intervention and effect. Specifically, there are specific checklists according to study design to assess the internal validity of studies including items such as the clarity and appropriateness of the research question addressed, sample details (e.g., selection, size, baseline differences between groups, location, participants lost to follow-up, etc.) confounding variables, and statistical analyses. The checklists also allow for an overall assessment of studies. It establishes eight levels of evidence in a range from 1++ ("High-quality meta-analyses, systematic reviews of randomized controlled trials or randomized controlled trials with a very low risk of bias") to 4 ("Expert opinion"). The lower the score is, the higher the quality of evidence. Considering the level of evidence of the selected studies, the SIGN method [29] enables the preparation of evidence-based practice guidelines. It takes into account the applicability of results to the target population, the amount of evidence available and the consistency of results. It establishes 4 grades of practice guidelines in a range from A ("At least one meta-analysis, systematic review or randomized controlled trial rated as 1++ and directly applicable to the target population; or a body of evidence consisting principally of studies rated as 1+, directly applicable to the target population and demonstrating overall consistency of results") to D ("Evidence level 3 or 4; or extrapolated evidence from studies rated as 2+). To answer question 14, study quality was evaluated independently by pairs (VK and LME). Disagreements between evaluators were resolved by discussion and consensus by a third investigator (JAAC). To address question 15, we developed evidence-based practice guidelines following the described system. Table 1. Key to evidence statements and grades of recommendation by the Scottish Intercollegiate Guidelines Network (SIGN).

1++
High-quality meta-analyses, systematic reviews of RCTs a or RCTs with a very low risk of bias

1+
Well-conducted meta-analyses, systematic reviews or RCTs with a low risk of bias 1− Meta-analyses, systematic reviews or RCTs with a high risk of bias 2++ High-quality systematic reviews of case-control or cohort studies. High-quality case-control or cohort studies with a very low risk of confounding or bias and a high probability that the relationship is causal 2+ Well-conducted case-control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal 2− Case control or cohort studies with a high risk of confounding or bias and a significant risk that the relationship is not causal 3 Non-analytic studies, e.g., case reports, case series 4 Expert opinion

Grades of Recommendation
A At least one meta-analysis, systematic review or RCT rated as 1++ and directly applicable to the target population; or a body of evidence consisting principally of studies rated as 1+, directly applicable to the target population and demonstrating overall consistency of results B A body of evidence including studies rated as 2++, directly applicable to the target population and demonstrating overall consistency of results; or extrapolated evidence from studies rated as 1++ or 1+ C A body of evidence including studies rated as 2+, directly applicable to the target population and demonstrating overall consistency of results; or extrapolated evidence from studies rated as 2++ D Evidence level 3 or 4; or extrapolated evidence from studies rated as 2+ a RCT: Randomized controlled trial.
Inter-rater agreement was established between two members of the research team (VK and LME) in order to assess the proportion of agreement when determining levels of evidence and grades of practice guidelines. Both of them extracted data from the included articles and coded studies considering the parameters of our research questions. Agreement was considered acceptable if it met at least 80% [30]. Six studies (46%) were randomly selected and both investigators extracted data independently and added them to a codebook. Such data was assessed for consistency between the two evaluators by calculating the proportion of agreements and disagreements on the different codes. Interrater agreement was 97.4%. Cohen's Kappa was not estimated due to the scarce variability of agreement between both evaluators [31].

Overall Results
We identified a total of 7775 records by means of the computerized search. We did not identify more records in the manual search (See Figure 1). After eliminating duplicates, we excluded 7658 that did not meet inclusion criteria. Most of the studies excluded were out of scope since they cover literature on engineering, dentistry, palliative care, health care services, eating disorders, LGBT student experiences, divorce, forensic evaluation, prisons, criminal behavior, consumer satisfaction or consumer attitudes, among others. We ultimately included 13 studies.

D
Evidence level 3 or 4; or extrapolated evidence from studies rated as 2+ a RCT: Randomized controlled trial.
Inter-rater agreement was established between two members of the research team (VK and LME) in order to assess the proportion of agreement when determining levels of evidence and grades of practice guidelines. Both of them extracted data from the included articles and coded studies considering the parameters of our research questions. Agreement was considered acceptable if it met at least 80% [30]. Six studies (46%) were randomly selected and both investigators extracted data independently and added them to a codebook. Such data was assessed for consistency between the two evaluators by calculating the proportion of agreements and disagreements on the different codes. Inter-rater agreement was 97.4%. Cohen's Kappa was not estimated due to the scarce variability of agreement between both evaluators [31].

Overall Results
We identified a total of 7775 records by means of the computerized search. We did not identify more records in the manual search (See Figure 1). After eliminating duplicates, we excluded 7658 that did not meet inclusion criteria. Most of the studies excluded were out of scope since they cover literature on engineering, dentistry, palliative care, health care services, eating disorders, LGBT student experiences, divorce, forensic evaluation, prisons, criminal behavior, consumer satisfaction or consumer attitudes, among others. We ultimately included 13 studies.

Analytic Method: Research Questions and Responses
Appendix A shows the main characteristics of the papers identified. Most of the studies were published between 2015 and 2020 (69.2%) and were conducted in the United States of America (38.5%), United Kingdom (30.8%), and Australia (15.4%).
As for the sample characteristics, five (38.5%) studies were conducted with students with a sample range from 40 to 2771, two (15.4%) with school staff with a sample range from 6 to 95, two studies (15.4%) included both students and school staff with a sample range from 14 to 412, three (23.1%) included both students and school staff but did not specify the sample size and 1 (7.7%) included students but the sample size was not specified either. Eight studies (61.5%) were conducted in secondary education, two studies (15.4%) in primary education and the rest (3; 23.1%) in primary and secondary education.
Eight studies (61.5%) reported ethnic diversity in the setting where the WSRA was implemented, while the rest of studies (38.5%) did not report data on this dimension. Of those reporting data on ethnic diversity, four studies (50%) included more than four ethnic groups and four studies (50%) did not report data on the number of ethnic groups included.
They included a total of seven quasi-experimental studies (53.8%), four qualitative studies (30.8%), and two randomized controlled trials (15.4%). With regard to a follow-up, seven studies (53.8%) considered post-intervention measures, five (38.5%) pre-and postmeasures, and one (7.7%) pre-, post-, and follow-up measures at 24 and 36 months. As for the time of implementation of the different WSRA across studies, six studies (46.2%) implemented the WSRA for more than two years, five (38.5%) within one year and two years, and two (15.4%) less than one year. The time of implementation of the different approaches ranged from 6 months to 5 years.
Four studies (30.8%) included WSRA based on the essential elements stated by the International Institute of Restorative Practices model, and two studies (15.4%) the Learning Together Intervention. The remaining seven studies (53.8%) incorporated different approaches (i.e., the Whole School Restorative Approach, the Restorative Practice Approach, Restorative Approach based on the European Forum for Restorative Justice, the Traditional Whole School approach and the Proactive-only Whole School Approach at the same time, the Restorative Justice approach as part of the Safe Schools initiative, the Pilot Community Conferencing and Restorative Practices Program, and The Restorative Whole-School Approach).
Taking into account the results of our analytic method, there was enough scientific evidence to develop evidence-based practice guidelines on the effects of the implementation of WSRA in schools. Out of the thirteen studies, twelve (92.3%) were considered for the preparation of practice guidelines since one study (7.7%) did not show positive changes associated with the WSRA implementation. The grade of practice guidelines ranged from D to A: D, 8 practice guidelines (80%); B, 1 practice guideline (10%), and A, 1 practice guideline (10%). Out of the ten practice guidelines, 2 (20%) were related to primary education, 5 (50%) to secondary, and 3 (30%) to both school settings. In addition, 2 (20%) practice guidelines were related to the implementation of the Learning Together Intervention and 8 (80%) to other specific WSRA. See Appendix A for more details of studies.

Whole-School Restorative Approaches: Evidence Available, Its Quality and Development of Evidence-Based Guidelines
In this section, we present in detail the scientific evidence identified and show that, according to the SIGN methodology [29], its quality is high enough to develop evidencebased guidelines.
With regard to primary education, two studies were selected. Goldys (2016) [32] prepared a report on the effects of the Whole School Restorative Approach after four years of implementation. It was based in the USA and the entire school community was considered in the sample although its size was not stated. It included a diverse population of students. By means of post-intervention measures, the results showed that the WSRA was associated with improvement and enduring changes in behavior (level of evidence: 3). In Canada, Reimer (2011) [33] conducted a qualitative study at a primary school with a diverse community of students on the effects of the Restorative Justice Approach as part of the Safe Schools Initiative after 5 years of its implementation. The study sample included one school administrator, one board administrator and four teachers. There was no followup. The results showed that the intervention seemed to be associated with benefits for students in providing character education, valuing students as individuals, listening to their voices and giving them ownership of the issue and assisting in healing relationships. An improvement in classroom climate and the whole school climate was also observed (level of evidence: 3). Both studies are non-analytic and have a risk of bias. They allow the preparation of two low grade practice guidelines as follows:

•
The Whole School Restorative Approach as the behavior plan to change school culture is recommended in primary education to improve behavior in settings with a diverse population of students and in the long term [32] (Grade of recommendation: D).

•
The Restorative Justice Approach as part of the Safe Schools Initiative is recommended in primary schools characterized by a diverse community of students to provide character education; valuing students as individuals, listening to their voices and giving them ownership of the issue and assisting in healing relationships in the long term. It is also recommended for the improvement of classroom and whole school climate in the long term [33] (Grade of recommendation: D).
Regarding secondary education, we identified eight studies. Two of them were related to the so-called Learning Together Intervention. Bonell et al. (2018) [34] conducted a study on the Learning Together Intervention that included 7121 students in secondary schools with a high degree of diversity in terms of ethnicity and a three-year follow-up. The intervention had a small but significant effect on bullying but no effect on aggression at 24 months' follow-up. At 36 months' follow-up, the intervention group showed higher quality of life and psychological well-being and lower psychological difficulties than the control group. The intervention was also associated with reductions in police contact, smoking and alcohol and drug use (level of evidence: 1++). Warren et al. (2020) [35] conducted a qualitative study on the Learning Together Intervention within the aforementioned controlled trial. No details were reported on the sample participating which included students and staff. By means of post-intervention measures, the results showed that The Learning Together Intervention helped to create a more inclusive and cohesive school environment (level of evidence: 3). While the first study is a randomized control trial design with low risk of bias; the second study includes a non-analytic design and, thus, has a risk of bias. Considering the different level of evidence of both studies, two practice guidelines are prepared since an integrated recommendation on the Learning Together Intervention cannot be prepared. The practice guidelines are as follows:

•
The Learning Together Intervention is recommended, in the long term, in order to decrease bullying while improving quality of life, psychological well-being and psychological difficulties of students in secondary schools with high levels of student diversity. It is also recommended in order to reduce police contact, smoking and alcohol and drug use in the long term [34] (Grade of recommendation: A).

•
The Learning Together Intervention is recommended in secondary schools with high levels of student diversity to create a more inclusive and cohesive school environment in the long term [35] (Grade of recommendation: D).
Another study conducted in secondary education is that of Wong et al. (2011) [36]. It used a quasi-experimental design to study the effectiveness of The Restorative Whole-School Approach after two years of implementation. It was based in Hong Kong and included a control and a partial group, pre-and post-measures and a sample size of about 1200 students. The pre-and post-measures showed a significant reduction in bullying in the intervention group and higher social and emotional skills in comparison with both control groups (level of evidence: 2++). This quasi-experimental study shows a very low risk of bias and a high probability that the relationship is causal. One practice guideline is prepared on The Restorative Whole-School Approach model of Wong et al. (2011) [36]:

•
The Restorative Whole-School Approach model is recommended for students in secondary schools in order to reduce bullying and increase empathy and self-esteem in the medium term [36] (Grade of recommendation: B).
Four studies on the effectiveness of WSRA including the essential elements as described by The International Institute of Restorative Practices Model were identified. Acosta et al. (2019) [6] conducted a randomized controlled trial on the efficacy of Restorative Practices Intervention that included 2771 students in secondary schools, with high levels of diversity in terms of ethnicity, and pre-and post-intervention measures after two years of implementation. There were no differences between the study groups in school connectedness, positive developmental outcomes and bullying victimization (level of evidence: 1+). Gregory et al. (2015) [37] conducted a quasi-experimental study in the USA one year after the WSRA was implemented. It included a control group, a sample size of 412 ethnically diverse students from secondary schools, and post-intervention measures. Their results showed that higher implementation levels of the WSRA were associated with better teacher-student relationships. Students perceived teachers as more respectful, and they issued fewer discipline referrals compared with low intervention implementation (level of evidence: 2+). Mansfield et al. (2018) [13] prepared a report on a WSRA, after two years of implementation, in the USA in a school that enrolls a diverse student body. It included pre-and post-measures but no control group. The number of participants was not stated in the paper. The results showed a decline in in-school and out-of-school suspensions and a decrease in the discipline gaps across race/ethnicity, gender and special education (level of evidence: 2−). Mirsky (2007) [38] prepared another report in the USA within a range from two to four years after the WSRA was implemented. It included a large sample size (n = 2148 students) at three secondary schools. The results found the WSRA to be associated with a decrease in disciplinary referrals, detentions, incidents of disruptive behavior, and out-of-school suspensions in post-intervention measures. There was also an establishment of a culture of collaboration among staff and a significant increase in students reporting other students for behavioral problems, students self-reporting, and parents reporting their children (level of evidence: 3).
There is a well-conducted cohort study with a low risk of confounding or bias and a moderate probability that the relationship is causal [37]. The remaining studies selected are non-analytic studies. The evidence available on the effectiveness of WSRA including the essential elements of the International Institute of Restorative Practices Model allows for an integrated practice guideline. Therefore, one specific practice guideline is prepared according to study outcomes:

•
The WSRA including the essential elements of the International Institute of Restorative Practices Model is recommended in ethnically diverse secondary schools, to improve teacher-student relationships while decreasing exclusionary discipline referrals in the medium term; to reduce in-school and out-of-school suspension rates as well as the number of infractions leading to suspension and the discipline gaps, to decrease exclusionary discipline referrals and detentions, to improve teacher-student relationships, and to improve collaboration among staff members in the medium and long term [13,37,38] (Grade of recommendation: D) The last of the studies regarding secondary education is that conducted by Norris (2019) [12] on the effectiveness of the Traditional Whole School Approach versus the Proactive-only Whole School Approach after 18 months of implementation by means of a quasi-experimental design. This study was conducted in the United Kingdom with a comparison of the aforementioned WSRA and a sample of about 600 students. The Proactiveonly Whole School Approach showed higher levels of happiness and school engagement in comparison with the Traditional Whole School Approach (level of evidence: 2−). This is a case-control study with a high risk of confounding or bias and a significant risk that the relationship is not causal. The resulting practice guideline is formulated as follows: • The Proactive-only Whole School Approach is recommended in secondary schools to increase levels of happiness and school engagement in the medium term [12] (Grade of recommendation: D).
As for studies conducted in primary and secondary education at the same time, we identified three studies including three different WSRA. Moir and MacLeod (2018) [39] conducted a quasi-experimental study on the effectiveness of the Restorative Approach based on the model of Education Scotland (2015) after 6 months of implementation and included 18 multiagency partners. Post-intervention measures showed that Educational Psychology Services that implemented the aforementioned WSRA were associated with an improvement in relationships and a reduction in exclusions/referrals to alternative schools (level of evidence: 3). Kehoe et al. (2008) [40] conducted a qualitative study in Australia on the effectiveness of the Restorative Practice Approach after at least 5 years of implementation in schools that had diverse ethnic populations. It included a sample size of 54 participants together with teachers and students. Their post-intervention measures showed that the implementation of the WSRA was associated with better student behavior as well as enhanced social and emotional skills (level of evidence: 3). Shaw (2007) [41] carried out a qualitative study on the effectiveness of the Pilot Community Conferencing and Restorative Practices Program in Australia, after two years of its implementation, that included 18 primary and secondary schools and post-intervention measures. The sample size was not stated in the study. The results showed that such a program can be effective for repairing relationships, acknowledging consequences of behavior, and resolving disputes. For teachers and administrators, the program represented a shift in thinking with regard to justice and discipline (level of evidence: 3).
The evidence identified on those three specific WSRA for primary and secondary education had a high risk of confounding or bias and the relationship studied may not be causal. The following three specific practice guidelines are prepared:

•
Restorative Approaches of Education Scotland is recommended in primary and secondary education to improve relationships and reduce exclusions/referrals to alternative schools in the short term [39] (Grade of recommendation: D).

•
The Restorative Practice Approach is recommended in primary and secondary schools with diverse ethnic populations to improve student behavior and social and emotional skills in the long term [40] (Grade of recommendation: D).

•
Pilot Community Conferencing and Restorative Practices Program is recommended in primary and secondary schools in order to repair relationships, acknowledge consequences of behavior and resolve conflicts in the medium term [41] (Grade of recommendation: D). Table 2 includes the details of the approaches identifies and their corresponding practice guidelines. Other approaches: -All teachers at the School had been trained in the Tribes Learning Communities program. Tribes is a program that began in the 1970s and, according to the Tribes Website, is a research-based process that creates a culture that maximizes learning and human development. It is a step-by-step process that teachers implement in their classrooms to teach students skills to enable them to work collaboratively in long-term groups called Tribes It is recommended in primary schools to provide character education; valuing students as individuals, listening to their voices and giving them ownership of the issue and assisting in healing relationships. It is also recommended for the improvement of classroom and whole school climate [33].

D Secondary education
The Learning Together Intervention -Primary restorative practices using respectful language to challenge or support behavior and circle time to build relationships. -Secondary restorative practices involved implementing restorative conferences to address more serious behavior problems. -Lesson plans and slides to guide teachers' delivery of 5-10 h per year of lessons on social and emotional skills for students. -Action group meetings of at least six staff and six students, held twice per term, to review relevant school policies and coordinate the intervention.
It is recommended in order to decrease bullying while improving quality of life, psychological well-being and psychological difficulties of students in secondary schools. It is also recommended in order to reduce police contact, smoking and alcohol and drug use [34].

Name of Approach Essential Elements Practice Guideline Grade of Recommendation
The Learning Together Intervention -Primary restorative practices using respectful language to challenge or support behavior and circle time to build relationships. -Secondary restorative practices involved implementing restorative conferences to address more serious behavior problems. -Lesson plans and slides to guide teachers' delivery of 5-10h per year of lessons on social and emotional skills for students. -Action group meetings of at least six staff and six students, held twice per term, to review relevant school policies and coordinate the intervention.
It is recommended in secondary schools to create a more inclusive and cohesive school environment [35]. It is recommended in ethnically diverse secondary schools, to improve teacher-student relationships while decreasing exclusionary discipline referrals in the medium term; to reduce in-school and out-of-school suspension rates as well as the number of infractions leading to suspension and the discipline gaps, to decrease exclusionary discipline referrals and detentions, to improve teacher-student relationships, and to improve collaboration among staff members in the medium and long term [13,37,38]. Once all members of staff were trained to this level, principally during school in-service days, subsequent evening sessions were made available to those who wished to pursue further instruction on restorative practices -Implementation of positive communication strategies between staff and students within the everyday discourse of school interactions, utilizing enquiring questions, and student-led participation strategies.
The Proactive-only Whole School Approach is recommended in secondary schools to increase levels of happiness and school engagement [12].

D Primary and secondary education
Restorative Practice Approach -The approach aims to deal with student behavior in a holistic way through increasing social and emotional skills in students and staff as well as handling behavioral incidents as they arise -Holistic approach that uses modeling pro-social behaviors and a common language that avoids focusing on blame or excuses -A continuum incorporating both reactive and proactive approaches This approach is recommended in primary and secondary education to improve student behavior and social and emotional skills [40]. D Table 2. Cont.

-
The reactive approach involves formally bringing together the students in a conference, after an incident has occurred, to discuss what happened and how to resolve the issue. Teachers ask students a series of affective questions when reacting to these situations -Reactive approaches are enhanced through proactive, holistic and relational style of language to communicate with students. This is achieved through the use of affective language. Affective language involves using statements about how someone was impacted by the other person's behavior with the aim of eliciting feelings.

Discussion
The objective of this study was to conduct a comprehensive and systematic literature review on the effects of WSRA in schools and develop evidence-based practice guidelines following the SIGN system [29]. The final aim of the study is to provide guidance on the effects of WSRA implementation on outcomes related to positive youth development.
Thirteen studies met the inclusion criteria and we developed 10 evidence-based practice guidelines on the implementation of WSRA considering the level of evidence: study design and intervention results. Most studies showed positive results associated with its implementation and, especially, on improving social and emotional skills and behavior. Those are key components of approaches directed at promoting positive youth development [20]. The results suggest that the WSRA provide a valuable framework that allows the development of successful restorative youth programs that improve conflict resolution skills, reduce discipline and disruptive problems, and enhance school attendance. Additionally, the WSRA approach improves communication competences and emotional literacy in teachers, other school staff, families, and students, which boosts both social-emotional development and long-term mental well-being in school-age children and adolescents. It is noteworthy that the WSRA have been used successfully in different cultural contexts, enhancing a school climate of equity, inclusion, and sensitivity to diversity where respect to the traditional ethnic culture of each school is fundamental. Even so, the evidence available does not allow for the preparation of general practice guidelines since we identified specific WSRA with specific components. This makes it difficult to integrate evidence on general practice guidelines.
When looking into specific WSRA, the highest level of scientific evidence is for secondary education. On the one hand, the Learning Together Intervention [34] is associated with improvements in quality of life and psychological wellbeing. It is also efficacious in reducing police contacts and substance use. There is also scientific evidence supporting another specific approach in secondary education, The Restorative Whole-School Approach [36] to decrease bullying while improving social and emotional skills. Again, outcomes considered key components of approaches aimed at promoting positive youth development [20]. These two WSRA are strongly recommended considering their grade of recommendation. As for the remaining approaches identified, further evaluation of their implementation is suggested in order to have more data to increase the grade of their evidence-based guidelines.
It is worth highlighting that only two randomized controlled trials have been conducted on WSRA. They are both based on secondary education. The first one evaluated the efficacy of the Learning Together Intervention [34] as already mentioned. This intervention includes not only restorative practice elements but also methods to modify school policies and systems and another for social and emotional education. Since some WSRA also include such methodology, this study was also considered in our review. The second study found that the Restorative Practices Intervention has no benefits over the control group [6]. This is probably because the control group had implemented restorative practices at some degree. Despite the above-mentioned limitations, this is not to say that such initiatives have no value. Firstly, they are the first randomized controlled trials conducted on WSRA. This is quite a challenge considering they both include more than 2000 students in secondary schools. The implementation of such designs in ordinary school settings is also a challenge given that daily life in schools is quite complex. Secondly, the experience of conducting such studies and their results establishes the conditions for more rigorous research and for a theoretical basis that supports the association of a positive school climate with positive youth development [6] and with the improvement of risk and health outcomes [34]. When looking into the body of evidence, the approaches including the essential elements of the International Institute of Restorative Practices Model are those that have more evidence on their effectiveness [6,13,37,38]. Most of the results indicate the effectiveness of such approaches in secondary schools in improving relationships while decreasing disciplinary measures. This is consistent with the need of school approaches to build positive rela-tionships and to resolve conflicts in a more proactive way [42]. An integrated practice guideline was prepared although its grade of recommendation suggests further research on its effectiveness.
Our review suggests the need for more rigorous methodological research concerning such approaches, but also the other identified WSRA and in primary and secondary education. On the one hand, more rigorous quasi-experimental studies and randomized controls are needed [8]. In this sense, it is also necessary to use evaluation measures within a psychometric analysis framework (i.e., validity, reliability, and responsiveness) [43] to evaluate the effects of WSRA and the use of blind raters to evaluate their effects to minimize biases. We also recommend a high degree of fidelity to the implementation of WSRA controlling the challenges that are involved (e.g., lack of staff buy-in) [8], as only a few studies did so since poor implementation may yield mixed outcomes [44]. The impact of the implementation of WSRA on outcomes such as academic achievement and satisfaction of participants is also suggested. They should address the limitations of those that have already been carried out [6] and provide a higher level of evidence of those with a low grade of recommendation. The lowest grade of recommendations does not support the use of a specific educational practice. Even so, it points out scientific evidence on such practice and future lines of research. In addition, more studies in primary education are needed. The sooner approaches that promote positive youth development are implemented, the better for prevention of relational and emotional difficulties.
It should be noted that research in the field is complex because WSRA may vary in their conceptual basis, principles, and methods according to their origins, and they may also have to adapt to a specific school's way of doing things, the cultural setting and related factors [10]. Specifically, the implementation of WSRA in schools involves careful awareness of the setting (e.g., clinical, state, private, etc.), participants (e.g., mixed ethnic groups, staff reticence, expectations of students, etc.), school culture (e.g., secular, religious, level of authoritarianism, etc.), and the history of the school community (e.g., school policy and commitment to move to a new educational approach, etc.) [45,46]. As for the setting, most of the studies were conducted in educational centers with a high degree of student diversity and situated in vulnerable areas at a social and economic level, which may make it difficult to generalize the results to other contexts. We suggest carrying out further studies in other settings to have more scientific data available to support the implementation of WSRA in broader international contexts. Further research should also consider the effect of context, type of school, and developmental stage on WSRA implementation. The sustainability of the WSRA implementation is relevant. It depends on the variability of financial support and the guarantee of stability in the training processes that support staff rotation. This makes study comparisons difficult, and in addition, each approach may have its own specificity. For instance, the Restorative Practices Intervention [6] and the Learning Together Intervention [34] may seem comparable but drilling down into the models shows they use different processes and practices. This should not be seen as a drawback but rather as the potential of WSRA to adapt to school reality by changing or creating processes to fit the setting. We propose the implementation of WSRA as a long-term and continuous process rather than a time-limited intervention [47]. Although most studies yielded significant results in the medium and long term, one should note that promising results were also shown in the very short term [38]. The implementation of the WSRA by means of educational psychology services might have generated a structure that optimized the processes required for proper WSRA implementation. More studies are needed to corroborate our hypothesis and, thus, be able to prepare more robust practice guidelines.
Although there are a few literature reviews on the implementation of restorative approaches in schools, none of them seem to focus specifically on the effects of WSRA. As already mentioned, they show some limitations [14,[22][23][24][25][26] that we address by means of a less-restricted literature search. Based on the scientific evidence available, the present work is a very first attempt to develop evidence-based practice guidelines on the effects of WSRA. To our knowledge, such an initiative has not been conducted to date. To do so, we used the SIGN system [29]. Compared to other systems used to evaluate evidence, this classification system assigns greater importance to the quality of the evidence behind each practice guidelines. It also highlights that the body of evidence should be regarded as a whole. It does not depend on single studies to support each practice guideline. It highlights the importance of well-conducted non-analytical studies to sustain practice guidelines when more rigorous designs such as randomized controlled trials cannot be carried out for practical or ethical reasons as it usually happens in the school setting [48]. A systematic review is considered the most appropriate methodology for evaluating the evidence available on a methodologically varied group of studies and procedures. Such diversity made it difficult to conduct a meta-analysis. Even when evaluating the same or similar WSRAs, there was a high level of diversity, and we were not able to combine the results into a general estimate of effect. Our study only included peer-reviewed studies. So, doctoral dissertations, master's theses, books or book chapters, meeting papers, and grey literature articles were out of our review, which may have decreased the body of evidence for our analyses and left aside important evaluation efforts [49]. It is worth noting that peer-review articles always underwent an evaluation process, which may, in turn, increase the level of evidence available for our analyses and resulting practice guidelines.
Our study offers evidence on the different approaches of WSRA used in schools and may generate leads about the links among WSRA and positive youth development, since as suggested by Acosta et al. (2021), restorative principles could be an effective means to boost positive youth development in school settings. That involves selecting an evidence-based approach that promotes positive youth development through supporting safe and protective school climates that may improve social, emotional, behavioral, and mental health outcomes for students. The research support for the implementation of WSRA approaches is encouraging especially for secondary education, adapting restorative principles to the heterogeneity of the cultural background in schooling context. They could be used to deal with diversity in terms of ethnicity and socio-economic adversity. The implementation of WSRA seems to generate a school culture of peace that promotes skills for the positive management of relationships and conflicts. Those skills are essential for daily-life functioning. Thus, their implementation may be promoting mental health and nurturing the resilience of the whole school community. Funding: This study has been funded by Instituto de Salud Carlos III through the project "Evaluation of the effectiveness of the program for the education of social, emotional and mental health of adolescents living in residential care centers" with reference PI18/01859 (Co-funded by European Regional Development Fund/European Social Fund "A way to make Europe"/"Investing in your future").

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. Pre-and postmeasures Inventory school climate (school climate)/Self-report/Internal consistency (omega) between 0.70-0.77 National Adolescent Health Study (school Connectedness)/Self-report/Omega = 0.81 Peer Attachment Scale (A three-item scale developed by Acosta (2003) measuring encouragement from peers to do well in school, confiding in peers, emulating peers, and considering peers' reactions before acting was used)/Self-report/Reliability not reported Improvement System-Rating Scale (Social Skills) /Self-report/Omega 0.74-0.87 Three items from the Communities That Care Survey (bullying)/Self report/Reliability not reported Other measurements: Student report of restorative practices/Self-report/Reliability not reported Intervention had a small but significant effect on bullying but no effect on aggression at 24 months' follow-up. At 36 months' follow-up, the intervention group showed higher quality of life and psychological well-being and lower psychological difficulties than the control group. Intervention was associated with reductions in police contact, smoking and alcohol and drug use.

1++
Goldys (2016) [32] Describes and examines the effectiveness of restorative practices to behavior change and management, and the promotion of a school culture of peace Greater implementation levels associated with better teacher-student relationships.
Students perceived teachers as more respectful, and they issued fewer discipline referrals compared with low intervention implementation.

2+
Kehoe (2018) [40] Explores the impact of restorative practices on student behavior Intervention can be an effective process for repairing relationships, acknowledging consequences of behavior and resolving disputes. Intervention as an opportunity to teach for transformation. For teachers and administrators, the intervention represented a shift in thinking with regard to justice and discipline.
3 Warren (2020) [35] Examines the participants accounts of 'Learning Together' whole-school intervention-related processes in the prevention of bullying and aggression as well as the improvement in student health in England secondary schools Pre-and postintervention measures Self-report: Chinese Adolescent Self-Esteem Scales Self-esteem (self-esteem).
Cronbach Within-subject comparison analysis, paired t test and treatment effect size estimates The intervention group showed a significant reduction in bullying, higher empathic attitudes and higher self-esteem in comparison to both control groups