Restorative Justice in School Discipline

Abstract

Racial/ethnic disparities exist in exclusionary school discipline practices in the United States, with minority students being suspended and expelled at significantly higher rates than White students. Many schools across the nation have adopted a restorative justice approach to school discipline to address these racial/ethnic disparities. This synthesis systematically reviews studies in which schools in the United States utilized restorative justice and described the impact these practices had on the racial/ethnic school discipline gap. To meet inclusion criteria, studies were required to: (a) describe a restorative justice approach to discipline implemented in a school setting in the United States, (b) describe the impact of the restorative justice practices on the racial/ethnic disparities in school discipline practices, and (c) be peer reviewed a conducted in English. Six studies met inclusion criteria. Results demonstrate that evidence does not support the use of restorative justice to reduce or eliminate racial disparities in school discipline practices.

Keywords: school discipline, restorative justice, restorative approach, restorative discipline, restorative practices, zero-tolerance, school-to-prison pipeline, racial/ethnic disproportionality

Restorative Justice: An Approach for Eliminating Racial Disparities in School Discipline?

Racial/ethnic disparities exist in the frequency and severity of exclusionary school discipline practices, such as suspensions and expulsions (Skiba et al., 2002), with minority students experiencing far higher rates of exclusionary discipline than students (Payne & Welch, 2010). In 2014, the U.S. Department of Education’s Office of Civil Rights released a report summarizing school discipline data in the United States. The snapshot revealed that African-American students, American Indian, and Native-Alaskan are disproportionately suspended and expelled. African-American students are suspended 3 times as likely, with 5% of White students suspended compared to 16% of African-American students (U.S. Department of Education Office for Civil Rights, 2014). American Indian and Native-Alaskan students represent less than 1% of the student population, but 2% of out-of-school suspensions and 3% of expulsions (U.S. Department of Education Office for Civil Rights, 2014). Racial/ethnic disparities also exist in preschool discipline, with African-American preschool children representing 18% of preschool enrollment, but 48% of the children receiving more than one suspension, while White students represent 43% of preschool enrollment, but only 26% of suspensions (U.S. Department of Education Office for Civil Rights, 2014). Finally, racial/ethnic disparities in school discipline were also demonstrated for minority girls, with African-American girls suspended at higher rates (12%) than girls of any other race or ethnicity, and American Indian and Native-Alaskan girls suspended at higher rates (7%) than White boys (6%) or White girls (2%) (U.S. Department of Education Office for Civil Rights, 2014).

Following the release of this report detailing racial/ethnic disproportionality in school discipline practices, the U.S. Department of Education’s Office for Civil Rights issued a warning to districts and schools that any discipline policies resulting in racial/ethnic disparities can be deemed illegal, regardless of whether the policy was intended to be unbiased. As a result, many districts and schools have adopted alternative discipline practices to address the racial inequity in school discipline. Restorative justice (RJ) is one alternative approach many schools have adopted in an effort to eliminate racial/ethnic disparities in school discipline.

Hypotheses for Racial/Ethnic Disproportionality in School Discipline Practices

Several hypotheses have been suggested to attempt to explain the racial/ethnic disproportionality in school discipline. First, some researchers argue minority students committing more serious behavior offenses compared to White students causes disproportionality in school discipline. These researchers suggest that minority students on average, and especially those from lower socio-economic backgrounds, are exposed to higher levels of instability in their homes and communities, resulting in an increase in challenging behavior (Cooley-Strickland et al., 2008; Gorman-Smith & Tolan, 1998; Hanna, 1987). However, other research has demonstrated that there is no evidence to suggest that minority students engage in higher levels of challenging behavior and that racial/ethnic disparities in the frequency and severity of school discipline exist, even when controlling for the behavior offense (Skiba et al., 2002). In addition, racial/ethnic disparities also exist in school discipline when controlling for socio-economic status (Skiba et al., 2002).

Second, in the United States, schools remain segregated due to neighborhood segregation (Emerson, Yancey, & Chai, 2001). Segregated neighborhoods result in schools serving vastly different demographics of students both racially and economically. Kim and Geronimo (2010) hypothesize that racial disparities exist in school discipline because some schools, especially those serving poor minority students, use significantly more law enforcement such as metal detectors, locker searches, police surveillance, and video camera monitoring, resulting in an increase in the frequency and severity of school discipline practices (Payne & Welch, 2010). However, administrators remain responsible for making decisions regarding discipline and consequences for challenging behavior regardless of law enforcement practices. Research demonstrates that discipline policies are ambiguous in what constitutes a behavior offense warranting exclusionary discipline, leaving room for implicit racial bias to affect the way school administrators assign consequences (Payne & Welch, 2010).

Third, according to Nicholson-Crotty, Birchmeier, and Valentine (2009), the racial bias of teachers and administrators is responsible for the inequitable treatment of minority students. This hypothesis argues that the disproportionate suspensions and expulsions of minority students perpetuates negative stereotypes, resulting in minority students holding negative attitudes about school and negative self-perceptions, resulting in more frequent challenging behavior. The increase in challenging behavior perpetuate negative stereotypes against minority students, resulting in continued racial bias and prejudice in teachers and school administrators. Extensive research supports the notion that racial bias plays a role in the disparities in school discipline practices (Dematthews, 2016; Rocque, 2011; Okonofua, Walton, & Eberhardt, 2016; Smolkowski, Girvan, Mcintosh, Nese, & Horner, 2016). For this reason, further research is needed regarding how schools should address racial bias to create a positive school environment for all students that promotes equitable discipline practices for students from all racial/ethnic groups.

Negative Consequences of Exclusionary Discipline Practices

Race/ethnicity is a predictor of exclusionary discipline practices, a type of disciplinary action such as suspension or expulsion that removes a student from the classroom or school. Exclusionary practices are related to poor school attendance, grade-level retention, poor grades, and failure to graduate (Osher et al., 2010; Skiba et al., 2002). A longitudinal study of ninth-grade students demonstrated that the chance of graduating decreased by 20% with each suspension, and the chance of enrolling in post-secondary education decreased by 12% with each suspension (Balfanz, Byrnes, & Fox, 2015). In addition, research demonstrates that exclusionary discipline practices are related to students dropping out of school, resulting in what is commonly referred to as the school-to-prison pipeline (Losen, 2014; Rausch, Skiba, & Simmons, 2004; Skiba et al., 2014), or the likelihood that a student has contact with the criminal justice system as a result of the zero-tolerance policies that cause suspensions and expulsions in schools. Fabelo et al. (2011) found that students who were suspended or expelled were three times more likely to have contact with the juvenile justice system the following school year than students who had not received exclusionary discipline. Due to the racial/ethnic inequity of school discipline and the negative outcomes associated with exclusionary school discipline practices, research is needed investigating methods for reducing racial disparities in school discipline practices.

What is Restorative Justice?

RJ can be broadly be defined as a framework that focuses on the principles that schools and communities use to guide the development of their policies, programs, and practices related to behavior offenses and discipline practices (Song & Swearer, 2016). RJ practices can be both proactive and reactive. Proactive approaches typically focus on relationship and community-building activities, while reactive processes focus on how to address misbehavior (Lewis, 2009). RJ has also been referred to as restorative approaches, practices, strategies, discipline, or interventions (Acosta et al., 2016; Anyon et al., 2016; Davis, 2014; Gregory, Clawson, Davis, & Gerewitz, 2016; Hanhan, 2013). Most researchers agree that RJ incudes practices which focus on the acknowledgment of harmful behavior and the repairing of the harm that is done with both the victim(s) and the community at large (Drewery, 2013; Zehr, 2002; Zehr & Toews, 2004).

A wide variety of practices have been researched that could fall under the RJ umbrella. Victim-offender mediation conferences, group conferences, and peacemaking or restorative circles are the most common RJ practices utilized in schools (Fronius, Persson, Guckenburg, Hurley, & Petrosino, 2016). Conferences and circles are meetings that include the offender(s) and all individuals who were harmed by the wrongdoing, which can include community members who witnessed the event or individuals who can act on behalf of the school in instances such as vandalism or theft of school materials. The goal of conferences or restorative circles is to repair the harm that was done through peacemaking or restorative dialogue and decisions. Instead of typical, punitive or exclusionary discipline being administered, such as detention, suspension, or expulsion, restorative agreements such as community service, restitution, apologizing, and behavioral change agreements are used (Fronius et al., 2016).

Theoretical Frameworks for Restorative Justice. There are several theories that have been used to explain the principles of RJ. Tompkins (1984) describes the affect theory, defining affects as emotions or subjectively experienced feelings. The theory describes the following nine affects: (a) enjoyment/joy, (b) interest/excitement, (c) surprise/startle, (d) anger/rage, (e) disgust, (f) dissmell, (f) distress/anguish, (g) fear/terror, and (h) shame/humiliation. The theory also describes the typical emotional response of each of the affects, and suggests that these emotional responses are hard-wired or preprogrammed in all humans. According to the affect theory, maximizing the positive affects and minimizing the negative affects helps individuals achieve optimum mental health. This theory informs RJ practices, as RJ aims to respect the affects, or feelings, of both the offender and the victim(s) in an attempt to optimize the mental health of all persons in the community.

The philosophy of humanism (Niethammer, 1808) stresses the value of all people and emphasizes the belief that individuals have control over their own behavior. The RJ framework operates within the philosophy of humanism, as it is a relational approach to wrongdoing, valuing all individuals involved, including the offender, victim(s), and community. In addition, RJ encourages the offender to take control of his/her behavior in the sense that offenders are encouraged to acknowledge their wrongdoing and then seek to repair the relationships that were damaged from that wrongdoing.

The reintigrative shaming theory (Braithwaite, 2004) recognizes that both the victim and the offender experience negative effects as a result of the wrong that was committed. For the offender, shaming can occur because of the direct or indirect action of others. Direct action might involve being publicly administered a consequence, whereas an indirect action could involve nonverbal communication of disapproval from others. In RJ, the crucial part of the shaming process is that it is different than negative shaming which serves to isolate the offender from the community. Instead, the shaming used in RJ leads to reconciliation and acceptance, focusing on repairing the damage that was done to the victim and/or the community.

Origins of Restorative Justice. The roots of RJ can be traced back to Native American and Maori indigenous cultures and the religious practices of Judaism (Johnstone, 2011; Zehr, 2002, 2015). These cultures emphasize the harm done, rather than focusing on the act or the person that caused the harm. Historically, RJ practices have been widely used with convicted adult or juvenile offenders in the criminal justice system. Research demonstrating the effectiveness of RJ practices in the criminal justice system, and specifically with juvenile offenders (Sherman & Strang, 2007) sparked an interest in schools to adopt RJ to benefit their students with challenging behavior. RJ as a practice in schools first began in Canada, New Zealand, the United Kingdom, and Australia (Hopkins, 2004). Only more recently has RJ been introduced in the United States. Zero-tolerance policies resulting in an increase in exclusionary discipline practices were widely accepted and used through the 1990’s, with restorative justice practices being introduced in more schools in the United States in the 2000’s (Losen, 2014).

Previous Research

RJ practices in schools have been shown to reduce the overall number percentage of students being suspended (High Hopes Campaign, 2012), while also demonstrating positive effects on school climate and academic performance (Lewis, 2009). However, it is important to interpret a reduction in suspensions with caution. Because RJ practices are relatively new, no evidence currently exists that students evading exclusionary discipline practices are better off academically or emotionally, as RJ practices have not existed for long enough that research has followed students longitudinally to examine whether students actually benefit from the lack of exclusionary discipline.

A small number of studies exist examining the effect RJ has on racial/ethnic disparities in school discipline practices, but to date, no work has been done to synthesize the results of such studies. The use of RJ in U.S. schools and the media coverage of RJ have grown in the past several years with many schools adopting RJ as a means to reduce the racial gap in school discipline practices. The lack of research supporting the use of RJ to reduce racial/ethnic disparities in school discipline is concerning given the fact that in many ways, practice is ahead of research in our knowledge of the effectiveness of RJ discipline practices.

The purpose of this synthesis is to conduct a systematic review examining studies in which RJ was implemented within a school setting in the United States and report the impact RJ had on the racial/ethnic disparities in school discipline. Specifically, this synthesis focuses on the following research question: What effect does restorative justice have on racial/ethnic disparities in school discipline in the United States?

Methods

Search Procedures

A systematic search was conducted using various search methods, which included an electronic search, ancestral search, and a hand search. First, the following electronic databases were searched: Academic Search Complete, Education Source, Ed Amin Abstract, ERIC, Psychology & Behavioral Science Collection, PsycINFO, and Race Relations Abstracts. The primary search terms included: restorative justice OR restorative discipline OR restorative practice OR restorative intervention. These terms were chosen to represent the various terms used to describe RJ. The secondary terms in the database search included: classroom OR school OR student OR discipline. These terms were chosen to identify studies that were implemented in schools. In addition to these terms, the search was further limited by including only peer-reviewed studies in academic journals, published in English. The initial search yielded 998 studies. From this list of studies, the titles and abstracts were read and sorted into one of three folders: yes, no, or maybe. The studies in the yes and maybe folders were analyzed more closely and coded to consider all inclusion criteria. Studies were moved to the appropriate folder as needed.

Upon completion of the electronic search, five studies were identified that met inclusion criteria. Several journals were hand-searched from 2014 through the present to identify any studies missed in the electronic search. Hand searched journals included: Race Equity and Education, Equity and Excellence in Education, and Review of Education, Pedagogy, and Cultural Studies. These journals were selected due to their focus on both educational practices and equity. No additional studies were identified from the hand search. Finally, reference lists were from each eligible study were reviewed as part of an ancestral search. One additional study was found during the ancestral search.

Inclusion Criteria

To meet inclusion criteria for this synthesis, studies were required to:

• Describe a restorative justice approach to discipline implemented in a school setting in the United States.
• Describe the impact of the restorative justice practices on the racial/ethnic disparities in school discipline practices
• Be peer reviewed and conducted in English.

Because restorative justice is a broad term that encompasses a wide-variety of practices, both proactive and reactive, for the purpose of this synthesis, “restorative justice” was operationalized as any study utilizing the language, restorative justice, restorative discipline, restorative practice, or restorative intervention. Case studies were not excluded from this synthesis, as this area of research is relatively new and only a limited number of studies met inclusion criteria. The search was limited to studies conducted in the United States, as race relations and traditional school discipline practices differ across countries; therefore including only studies in the United States allowed researchers to view the research with the appropriate lens.

Data Analysis

Coding Procedures

A modified version of a code sheet designed by researchers from the Meadows Center for Preventing Educational Risk (Vaughn et al., 2014) was used for this synthesis. This code sheet was selected due to its reliability and ability to capture a comprehensive picture of each study. Coding procedures were used to organize the information from each study related to the demographics of the participants, type of research design, a description of the RJ practices investigated, training provided to teachers and staff, the year of implementation, the reason RJ was implemented in the participating school(s), and general findings related to the effects of RJ on racial/ethnic disparities in school discipline.

Two coders independently coded a single study to demonstrate inter-rater reliability. The percentage of agreement was calculated by determining the number of responses indicating agreement divided by the total number of responses, both agreement and disagreement. Inter-rater reliability was established with 95% agreement between the two coders. After reliability was established, the two coders independently coded all studies. After both coders completed coding, all disagreements were discussed to reach a consensus in how to code the item.

Description of Findings

This synthesis includes quantitative and qualitative studies. The results are separated into two groups, with quantitative studies included in one group and qualitative studies in the other. Tables 1 and 2 both describe the study design, the location of the RJ practices, race/ethnicity demographic data, and the impact of RJ on racial/ethnic disparities in school discipline. Table 1 presents the findings of quantitative studies, while Table 2 presents the findings of qualitative studies.

Results

Description of Restorative Justice Practices. There was variability in how each school or district from the included studies implemented RJ, as no universal protocol exists for what RJ should look like in schools. In Anyon et al. (2014) and Anyon et al. (2016), after a school district policy reform aimed at eliminating racial disparities in school discipline practices, schools began offering voluntary training in restorative approaches that the district strongly recommended for principals, disciplinarians, teachers, and special service providers such as social workers and psychologists. The implementation of RJ was not mandated, but was highly recommended. The district classified behavior infractions by levels. Restorative conferencing was encouraged for use with level 2 behaviors, representing severe defiance/disobedience of an authority, through level 5 behaviors, representing first degree assault. District policy gave school administrators the freedom to choose whether restorative conferencing was used alone or in conjunction with exclusionary discipline practices. If the school administrator decides RJ is an appropriate approach for the discipline infraction, the student meets with a staff member trained in RJ, and if the student accepts responsibility for his/her behavior, a restorative circle, mediation, or conference is held that includes the offender and anyone directly harmed by the behavior infraction. After the restorative circle, the participants develop an agreement or action plan to repair the harm done, and all individuals sign the action plan.

Cavanagh et al. (2014) describes a restorative intervention called Culture of Care that was implemented at one large high school. The goal of the Culture of Care project was to address racial gaps in achievement and discipline between Hispanic/Latino students and White students. This project was focused on teachers caring for students’ well-being, the implementation of culturally responsive pedagogy by placing relationships at the center of teaching and learning, and utilizing restorative practices that focused on repairing harm, particularly to relationships. The Culture of Care project focused on changing teacher behaviors as well as involving Latino/Hispanic students and their parents in the process. First, Latino/Hispanic students, parents, and school staff were interviewed to understand the school experience of Latino/Hispanic students. After, Latino/Hispanic parents led training for teachers on the implementation of restorative conversations, community-building circles, and restorative circles. These practices were implemented at the school level following the training.

Davis (2014) is the executive director of RJ for Oakland youth and describes the implementation of RJ by detailing how instead of utilizing suspensions or expulsions from school, Oakland implemented restorative justice circles, where everyone involved in the wrongdoing comes together and an object called a talking piece is passed around the circle so that each individual has a chance to speak respectfully and from the heart when holding the talking piece, while everyone else in the circle listens with respect from the heart. When conflict arises, students from the Oakland district know to go to the RJ room, speak to a staff member, and ask for a talking piece and space to facilitate a restorative circle. Davis describes how this approach to discipline has taught students that instead of resorting to violence, a healthier alternative is engaging in a restorative process that brings together the person responsible for the harm with the individuals harmed in a safe and respectful space to promote dialogue, accountability, a sense of community, and healing.

Gregory et al. (2016) utilized two types of restorative practices, prevention and intervention. Prevention practices focused on building relationships and developing community and included: (a) affective statements, (b) proactive circles, (c) fair process, (d) restorative staff community/restorative approach with families, and (e) fundamental hypothesis understandings. Affective statements were used in response to negative or positive events in the classroom and school. Proactive circles were used on a daily or weekly basis, and involved students sitting in a circle and discussing a topic to build community. Fair process engaged the students in decisions and explained the rationale behind decisions. Restorative staff community and a restorative approach with families focused on modeling and using restorative practices among school staff and with student families. Intervention practices focused on repairing harm and restoring the community and included: (a) restorative questions, (b) responsive circles, (c) small impromptu circles, (d) restorative conference circles, and (e) Reintegrative management of shame. Restorative questions involved addressing negative behaviors using questions such as “Who was affected by your behavior?” or “What do you think you need to do to make it right?” Responsive circles occurred after a more serious incident and involved sitting in a circle and addressing who has been harmed and what needs to be done to repair the harm. Small impromptu circles addressed negative behaviors by asking the wrongdoer and those harmed to answer restorative questions in front of each other. A restorative conference circle used a scripted approach to facilitate accountability and repair hard for serious incidents. Finally, Reintegrative management of shame worked to acknowledge the emotions of the wrongdoer and those impacted by the wrongdoing.

Thompson (2016) described Miami-Dade County Public School District’s (MDCPS) implementation of restorative justice. MDCPS implemented a school-wide Positive Behavior Support model to encourage positive behavior and teach appropriate social behaviors. Misbehavior at MDCPS is classified by levels. Level I behaviors are the lowest level of misconduct. Level II behaviors significantly interfere with the learning and/or the well-being of others. The levels continue to increase to describe more serious behavioral misconduct, with levels IV and V involving crimes in which the police become involved. Level I, II, and III behaviors are addressed with strategies intended to rehabilitate the offending students and repair harm. These strategies include rehabilitative measures such as contacting parents, peer mediation, revocation of the right to participate in social and extracurricular activities, creation of a student contract, restitution, participation in a counseling session related to the infraction, referral to outside counselors, a behavior plan, or the teacher(s) ignoring objectionable behavior. Behaviors that fall in level III to V receive the same strategies as above, but also involve suspension, placement in an alternative education setting, or expulsion.

Quantitative Findings. Four studies presented quantitative findings investigating the impact RJ had on racial/ethnic disparities in school discipline (Anyon et al., 2016; Anyon et al., 2014; Gregory et al., 2016; Thompson, 2016). Three of four quantitative studies included in this synthesis demonstrate that RJ failed to eliminate racial/ethnic disparities in school discipline practices (Anyon et al., 2016; Anyon et al., 2014; Thompson et al., 2016). In these studies, the racial gap in school discipline between minority students and White students remained for office-discipline referrals (Anyon et al., 2016; Anyon et al., 2014), out-of-school suspensions (Anyon et al., 2016; Anyon et al., 2014), and law enforcement referrals (Anyon et al., 2014; Thompson, 2016). The fourth study (Gregory et al., 2016) demonstrated that RJ did appear to decrease disproportionality in school discipline referrals, but did not completely eliminate the racial/ethnic disparities in school discipline data. Refer to Table 1 for additional details related to each quantitative study.

Anyon et al. (2014) calculated an odds ratio (OR) to represent the association between each racial/ethnic group and the dependent variable (office-discipline referral, out-of-school suspension, expulsion). An OR larger than 1.00 represents an increase in the odds of a particular racial/ethnic group that participated in a restorative intervention receiving future discipline. In regards to office disciplinary referrals, Latino (OR 1.40, p b .001), Black (OR 2.30, p < .001), Native American (OR 1.29, p < .05) and Multiracial students (OR 1.50, p < .001) had significantly higher odds of office referral compared to White youth (Anyon et al., 2014). For suspensions, only Black (OR 1.55, p b .001) and multi-racial students (OR 1.41, p b .05) had significantly higher odds of suspension compared to White youth (Anyon et al., 2014). For law enforcement referrals, compared to White students, Latino (OR 1.59, p b .05) and Black (OR 1.52, p b .05) youth had significantly greater odds of police involvement in their disciplinary incidents, accounting for other demographic variables and the seriousness of their offenses (Anyon et al., 2014).

Anyon et al. (2016) calculated risk indices that represented the proportion of students from a particular racial/ethnic group who received school discipline divided by the total number of students from that particular group. Anyon et al. (2016) demonstrated that for office-discipline referrals, African-American students had a risk index of 19.02% and a relative risk ratio of 3.41 compared to White youth and 1.99 compared to all other students. Among Latinos, the office discipline referral rate was 11.22%, and the relative risk ratios were 2.00 compared to White students and 1.06 compared to all other students. For out-of-school suspensions, African-American students had a suspension rate of 9.64% and the relative risk ratios were 4.95 compared to White students and 2.55 compared to all other students. In contrast, the suspension rate for Latino students was 4.46%, whereas the relative risk ratios were 2.29 compared to White students and .92 compared to all other students (Anyon et al., 2016).

Thompson (2016) utilized district data to compare the percentage of students from each racial/ethnic group who received school discipline compared to the percentage that particular racial/ethnic group represented of the total school population. Despite the fact that Florida has decreased school-related suspensions and arrests rates across the state since it changed its zero-tolerance statute, African-Americans are still disproportionately arrested at higher rates than White students. African-American students made up of 53% of the school-related arrests despite representing only 22.9% of the total school population. Hispanics made up 15% of school-related arrests and 30% of the school population, and Whites made up 32% of school-related arrests and 40.9% of the school population (Thompson, 2016).

The fourth study (Gregory et al., 2016) presented mixed findings. Positively, teachers who were perceived by students as implementing more restorative justice practices in their classroom demonstrated fewer differences in the number of discipline referrals issued to Asian/White students than to Latino/African-American students, while the large racial discipline gap remained for teachers perceived by students as low on the implementation of restorative justice practices. On the contrary, Gregory et al, (2016) also demonstrated that as tested in a paired sample t-test (t (29) = 3.63, p = .001), teachers continued to issue Asian and White students fewer misconduct referrals (M = 1.28) than they issued to Latino and African-American students (M = 6.34). In other words, the use of RJ practices decreased but failed to eliminate the racial gap in school discipline referrals.

Qualitative Findings. Twostudies presented qualitative findings investigating the impact RJ had on racial/ethnic disparities in school discipline (Cavanaugh et al., 2014; Davis, 2014). Davis (2014) stated that racial disparity in school discipline practices was eliminated, but provided little evidence to support this claim. On the other hand, Cavanaugh et al. (2014) demonstrated that although discipline referrals decreased for all racial/ethnic groups, Hispanic students, in particular, continued to have a higher rate of discipline referrals. Refer to Table 2 for additional details related to each qualitative study.

Discussion

The results of this synthesis fail to support the notion that RJ has a positive impact on the racial/ethnic disparities in school discipline practices. 4 of 6 studies demonstrated that RJ failed to reduce or eliminate the racial/ethnic disparities in school discipline. The fifth study (Gregory et al., 2016) suggested that although student perceptions of the level of implementation of RJ (high or low) predicted the degree to which racial disparities existed in that teacher’s discipline referrals, racial disparities in discipline practices still remained disparate for all teachers. Only one study demonstrated that racial disparities were eliminated (Davis, 2014). With only one study reporting that RJ eliminated racial/ethnic disparities in school discipline, there is limited evidence supporting the use of a RJ approach to school discipline as a means of eliminating racial/ethnic disparities in school discipline.

Limitations

There are three limitations of this synthesis. First, there was variation in how RJ was operationalized across studies, as there is no universal operational definition of how RJ is best implemented in schools. Research would benefit from a universal definition of RJ implementation in schools to ensure that the when synthesizing the results of studies investigating RJ, research is comparing studies investigating the same approach to school discipline. Second, only six studies met inclusion criteria. The results of this limited number of studies should be interpreted with caution, as additional research is needed to investigate how RJ impacts racial/ethnic disparities in school discipline. Third, this synthesis fails to include rigorous experimental studies; therefore it is not possible to draw causal inferences regarding the effectiveness of RJ in reducing racial/ethnic disparities in school discipline. Rigorous experimental studies do not exist, as it would be unethical to randomly assign students to either receive or not receive exclusionary discipline. Due to the ethical concerns, this limitation is not likely to be resolved for this particular topic of research, but additional research beyond case studies would be helpful in providing more compelling research regarding the impact RJ has on the racial/ethnic gap in school discipline practices.

Implications for Future Research and Practice

Given the lack of evidence to support the use of RJ as a means for eliminating racial/ethnic disparities in school discipline practices,it is concerning that RJ approaches to discipline are growing in popularity in schools as an alternative to traditional exclusionary discipline practices as a means for reducing or eliminating racial disparities. Schools are understandably concerned with creating more equitable school discipline practices, but additional research is needed in order to draw conclusions regarding the effectiveness of RJ at increasing equitable discipline practices. With 5 of 6 studies failing to demonstrate that RJ eliminated the racial gap in school discipline, districts or schools looking for discipline approaches that could have a positive impact on reducing or eliminating racial/ethnic disparities in discipline might benefit from considering approaches other than RJ. With many questions left to be answered regarding the impact RJ has on the racial gap in discipline, possible alternative approaches to consider and areas for future research might consider focusing on teacher education and training aimed at reducing racial/ethnic bias, interventions aimed at improving student/teacher relationships, or a focus on providing culturally responsive instruction as a means to increase academic engagement of minority students to reduce challenging behavior.

Despite the fact that this research does not support the use of RJ practices in schools as a means to eliminate racial/ethnic disparities in school discipline, for many people, there is still an emotional appeal to the idea that schools and students could benefit from less punitive discipline. In addition to the emotional appeal, evidence suggests that RJ does demonstrate positive outcomes for reducing the number of overall suspensions and expulsions (Fronius et al., 2016), as well as increasing students’ and parents’ perceptions of school safety (Osher, Poirier, Jarjoura, & Brown, 2014; Steinberg, Allensworth, & Johnson, 2014). Additional research is needed to investigate whether the use of restorative justice has long-term social/emotional and academic benefits for students. In addition to investigating the potential effect RJ has on the offender, research is also needed investigating the impact RJ has on the long-term social/emotional and academic outcomes of the victims and school community at large. As RJ is a relatively new practice in the United States, research would benefit from longitudinal studies following students to examine long-term effects of exclusionary discipline compared to alternative approaches to school discipline such as RJ.

References

* indicates articles included in synthesis

* Anyon, Y., Jensen, J. M., Altschul, I., Farrar, J., McQueen, J., Greer, E., Downing, B., & Simmons, J. (2014) The persistent effect of race and the promise of alternatives to suspension in school discipline outcomes. Children and Youth Services Review, 44, 379–386. doi: 0.1016/j.childyouth.2014.06.02.

* Anyon, Y., Gregory, A., Stone, S., Farrar, J., Jensen, J. M., McQueen, J., Downing, B., Greer, E., & Simmons, J. (2016) Restorative interventions and school discipline sanctions in a large urban school district. American Educational Research Journal, 53(6), 1663–1697. doi: 10.3102/0002831216675719.

Balfanz, J., Byrnes, V., & Fox, J. (2015). Sent home and put off track: The antecedents,

disproportionalities, and consequences of being suspended in the 9th grade. In D. J. Losen (Ed.), Closing the school discipline gap: Equitable remedies for excessive exclusion (pp. 17–30). New York, NY: Teacher’s College Press.

Bradshaw, C. P., Mitchell, M. M., O’Brennan, L. M., & Leaf, P. J. (2010). Multilevel

exploration of factors contributing to the overrepresentation of Black students in

office disciplinary referrals. Journal of Educational Psychology, 102(2), 508–520.

doi:10.1037/a0018450

Braithwaite, J. (2004). Restorative justice: Theories and worries (visiting experts’ paper, 123rd

international senior seminar, resource material series, no. 63, pp. 47–56). Tokyo: United Nations, Asia, and Far East Institute for the Prevention of Crime and the Treatment of Offenders.

* Cavanagh, T., Vigil, P., Garcia, E. (2014) A story legitimating the voices of Latino/Hispanic students and their parents: Creating a restorative response to wrongdoing and conflict in schools. Equity & Excellence in Education, 47(4), 565–579. doi: 10.1080/10665684.2014.958966.

Cooley-Strickland, M., Quille, T. J., Griffin, R. S., Stuart, E. A., Bradshaw, C. P., & Furr-

Holden, D. (2008). Community violence and youth: Affect, behavior, substance use, and academics. Clinical Child and Family Psychology Review, 12(2), 127–156.

* Davis, F. (2014) Discipline with dignity: Oakland classrooms try healing instead of punishment. Reclaiming Children and Youth, 23(1), 38–41.

DeMatthews, D. (2016). Effective leadership is not enough: Critical approaches to closing the

racial discipline gap. The Clearing House: A Journal of Educational Strategies, Issues and Ideas, 89, 7–13. doi:10.1080/00098655.2015.1121120

Drewery, W. (2013). Restorative justice practice: Cooperative problem-solving in New

Zealand’s schools. Forum, 55, 209–216.

Emerson, Michael O., George Yancey, & Karen Chai. 2001. “Does race matter in residential segregation? Exploring the preferences of White Americans.” American Sociological Review, 66, 922–35.

Fabelo, T., Thompson, M. D., Plotkin, M., Carmichael, D., Marchbanks, P., III, & Booth, E. A.

(2011). Breaking schools’ rules: A statewide study of how school discipline relates to students’ success and juvenile justice involvement. Council of State Governments Justice Center Publications. Accessible at http://justicecenter.csg.org/resources/juveniles

Fronius, T., Persson, H., Guckenburg, S., Hurley, N., & Petrosino, A. (2016) Restorative justice

in U.S. schools: A research review. San Francisco, CA: West Ed Justice & Prevention Research Center.

Gorman-Smith, D., & Tolan, P. (1998). The role of exposure to community violence and

developmental problems among inner-city youth. Development and Psychopathology, 10, 101–116.

* Gregory, A., Clawson, K., Davis, A., & Gerewitz, J. (2016). The promise of restorative practices to transform teacher-student relationships and achieve equity in school discipline. Journal of Educational and Psychological Consultation, 26, 325–353.doi:10.1080/ 10474412.2014.929950

Hanna, J. L. (1987). Disruptive school behavior: Class, race, and culture. England, United

Kingdom: Holmes & Meier.

High Hopes Campaign. (2012). Restorative justice in Chicago public schools. Chicago, IL:

Author.

Johnstone, G. (2011). Restorative justice: Ideas, values, debates. New York, NY: Routledge.

Kim, C. Y., & Geronimo, I. (2010). Policing in schools. Education Digest, 75(5), 28–35.

Lewis, S. (Ed.). (2009). Improving school climate: Findings from schools implementing

restorative practices. Bethlehem, PA: International Institute for Restorative Practices.

Losen, D. (Ed.). (2014). Closing the school discipline gap: Equitable remedies for

excessive exclusion (disability, equity, and culture). New York: Teachers College

Press.

McIntosh, K., Flannery, K. B., Sugai, G., Braun, D., & Cochrane, K. (2008). Relationships

between academics and problem behavior in the transition from middle school to high school. Journal of Positive Behavior Interventions, 10(4), 243–255. doi:10.1177/1098300708318961

Niethammer, F. (1808). The Contradiction of Philanthropinism and Humanism in the Theory of

Education Teaching of Our Time. F. Frommann.

Osher, D., Bear, G. G., Sprague, J. R., & Doyle, W. (2010). How can we improve school discipline? Educational Researcher, 39, 48–58.

Osher, D. M., Poirier, J. M., Jarjoura, G. R., & Brown, R. C. (2014). Avoid quick fixes: Lessons learned from a comprehensive districtwide approach to improve conditions for learning. In D. J. Losen (Ed.), Closing the school discipline gap: Equitable remedies for excessive exclusion (pp. 192–206). New York, NY: Teachers College Press.

Okonofua, J. A., Walton, G. M., & Eberhardt, J. L. (2016). A vicious cycle: A social–

psychological account of extreme racial disparities in school discipline. Perspectives on Psychological Science, 11(3), 381­–398. doi:10.1177/1745691616635592

Payne, A., & Welch, K. (2010). Modeling the effects of racial threat on punitive and

restorative school discipline practices. Criminology, 48(4), 1019­–1062.

Rausch, M. K., Skiba, R. J., & Simmons, A. B. (2004). The academic cost of discipline: The

relationship between suspension/expulsion and school achievement. Bloomington, IN: Center for Evaluation and Education Policy, School of Education, Indiana University.

Rocque, M., & Paternoster, R. (2011). Understanding the antecedents of the “school-to-jail” link:

The relationship between race and school discipline. The Journal of Criminal Law and Criminology, 101(2), 633–665.

Sherman, L. W., & Strang, H. (2007). Restorative justice: The evidence. London: Smith Institute.

Skiba, R., Tachok, M., Chung, C. G., Baker, T., Sheya, A., & Hughes, R. (2014). Where

should we intervene? Contributions of behavior, student, and school characteristics to

suspension and expulsion. In D. J. Losen (Ed.), Closing the school discipline gap:

Research for policymakers (pp. 132 – 146). New York: Teachers College Press.

Smolkowski, K., Girvan, E. J., Mcintosh, K., Nese, R. N. T., & Horner, R. H. (2016). Vulnerable

decision points for disproportionate office discipline referrals: Comparisons of discipline for African-American and White elementary school students. Behavioral Disorders, 41(4), 178–195. doi:10.17988/bedi-41-04-178-195.1

Song, S. Y., & Swearer, S. M. (2016) The cart before the horse: The challenge and promise of

restorative justice consultation in schools. Journal of Educational and Psychological Consultation, 26(4,) 313–324. doi: 10.1080/10474412.2016.1246972

Steinberg, M. P., Allensworth, E., & Johnson, D. W. (2014). What conditions support safety in urban schools? The influence of school organizational practices on student and teacher reports of safety in Chicago. In D. J. Losen (Ed.), Closing the school discipline gap: Equitable remedies for excessive exclusion (pp. 118–131). New York, NY: Teachers College Press.

* Thompson, J. (2016) Eliminating zero tolerance policies in schools: Miami-dade county public school’s approach. Brigham Young University Education and Law Journal, 2016(2), 325–349.

Tomkins, S. S. (1984). Affect theory. Approaches to emotion, 163, 195.

U.S. Department of Education Office for Civil Rights. (2014). Civil rights data collection: Data snapshot (school discipline). Accessible at http://ocrdata.ed.gov/Downloads/CRDC-School-Discipline-Snapshot.pdf

Vaughn, S., Elbaum, B. E., Wanzek, J., Scammacca, N., & Walker, M. A. (2014). Code sheet

and guide for education-related intervention study syntheses. Austin, TX: The Meadows

Center for Preventing Educational Risk.

Zehr, H. (2002). The little book of restorative justice. Intercourse, PA: Good Books.

Zehr, H. (2015). The little book of restorative justice: Revised and updated. Intercourse, PA:

Good Books.

Zehr, H., & Toews, B. (2004). Critical issues in restorative justice. Monsey, NY: Criminal

Justice Press.

Table 1: Quantitative Studies Investigating the Impact of RJ on Racial/Ethnic Disparities in School Discipline

Study	Study Design	Location	Race/Ethnicity Demographics	Impact of RJ on Racial/Ethnic Disparities in School Discipline
Anyon et al., 2016	Correlational	Denver Public Schools	Sample Demographics: 57.31% Latino, 20.83% White, 14.47% Black, 3.35% Asian, 2.99% Multiracial, 0.81% Native American, 0.24% Pacific Islander	Office Discipline Referrals: African-American students had a risk index of 19.02% and a relative risk ratio of 3.41 compared to White youth and 1.99 compared to all other students. Among Latinos, the office discipline referral rate was 11.22%, and the relative risk ratios were 2.00 compared to White students and 1.06 compared to all other students.   Out-of-school Suspension: African-American students had a suspension rate of 9.64% and the relative risk ratios were 4.95 compared to White students and 2.55 compared to all other students. The suspension rate for Latino students was 4.46% whereas the relative risk ratios were 2.29 compared to White students and .92 compared to all other students.
Anyon et al., 2014	Correlational	Denver Public Schools	58% Latino, 20% White, 15% Black,   3% Asian, 3% Multiracial Less than 1% Pacific Islander	Office Referrals: Latino (OR 1.40, p <  .001), Black (OR 2.30, p < .001), Native American (OR 1.29, p < .05) and Multiracial students (OR 1.50, p < .001) had significantly higher odds of office referral compared to White youth.     Out-of-School Suspension: Only Black (OR 1.55, p < .001) and Multiracial students (OR 1.41, p < .05) had significantly higher odds of suspension compared to White youth.   Law enforcement referrals: Compared to White students, Latino (OR 1.59, p < .05) and Black (OR 1.52, p < .05) youth had significantly greater odds of police involvement in their disciplinary incidents, accounting for other demographic variables and the seriousness of their offenses.   Expulsion: Race did not predict expulsion for any racial/ethnic group.
Gregory et al., 2016	Correlational	29 classrooms from two large and diverse high schools on the East coast of the US	Sample demographics: 44% White, 21% Latino, 3% American- Indian,   2% Asian, 5% African-American 25% Mixed Race. Of the 106 Mixed Race students, 45% were partially of African-American descent and 73% were partially of Latino descent.	Teachers issued Asian and White students fewer (M = 1.28) misconduct/defiance referrals than they issued to Latino and African-American students (M = 6.34), as tested in a paired sample t-test (t (29) = 3.63, p = .001).   Teachers who were perceived as implementing more RP elements by their students tended to have fewer differences in the number of misconduct/defiance referrals issued to Asian/White and Latino/African-American student groups compared with the large discipline gap for teachers perceived as low on RP elements
Thompson, 2016	Case study	Miami-Dade County Public School District (MDCPS)	Florida’s racial makeup of students:   (a) 22.9% African-American, (b) 30.0% Hispanic, (c) 40.9% white and (d) 3.2% other.	Despite the fact that Florida has decreased school-related suspensions and arrests rates across the state since it changed its zero-tolerance statute, African-Americans are still disproportionately arrested at higher rates than white students.   In 2013-14, African-American students made up of 53% of the school-related arrests, Hispanics made up 15% of school-related arrests, and whites made up 32% of school-related arrests. Florida’s racial makeup of students is (a) 22.9% African-American, (b) 30.0% Hispanic, (c) 40.9% white and (d) 3.2% other

Note. RI = OR = odds ratio; MDCPS = Miami-Dade County Public School District

Table 2: Qualitative Studies Investigating the Impact of RJ on Racial/Ethnic Disparities in School Discipline

Study	Study Design	Location	Race/ Ethnicity Demographics	Impact of RJ on Racial/Ethnic Disparities in School Discipline
Cavanaugh et al., 2014	Qualitative – Interviews	A large high school in the Denver metropolitan area	NR	The principal indicated that a racial gap in discipline persisted despite implementing RJ practices. Hispanic males continued to have a higher incidence of being referred to the office than White males.
Davis, 2014	Case study – qualitative	In Oakland at a RJOY pilot middle school	NR	Racial disparity in discipline was eliminated.

Note. NR = not reported; RJOY = Restorative Justice for Oakland Youth