From Retributive to Restorative: An Alternative Approach to Justice

• In Event: Disrupting Discipline: An In-Depth Examination of Restorative Practices’ Impacts and Implementation

Abstract

Classroom management and discipline represent one the hardest parts of school officials' jobs. Over the last five decades, educational authorities have increasingly turned to using exclusionary discipline, with the rate of school suspensions more than doubling for Black and Latino children. In SY2012, 3.5 million students were suspended, losing nearly 18M days of instruction due to “zero-tolerance” policies. Being in a stricter school can lead to long-term negative consequences.

One justification for policies encouraging more punitive discipline is the desire to prevent negative spillover effects from disruptive students that make it hard for other students to learn. This concern has led districts to approach conflict resolution from a zero-sum perspective: suspend students whom they view as disruptive in an effort to hold them accountable and potentially harm those students, or avoid suspensions and harm their classmates whose learning now may be disrupted. While educators are increasingly aware of the potential harms of suspensions, they seek concrete responses to undesirable behavior. A growing movement within education has sought to find a solution that avoids this tradeoff by deterring undesirable behavior without imparting harm.

We investigate restorative justice practices (RP), which emphasize community building and restoration, as an alternative to the traditional punitive approach. We leverage the rollout of RP across Chicago public high schools (CPS) beginning in SY2014. CPS provided training to school staff that emphasized less punitive and more reparative strategies when engaging with students. Using a difference-in-differences-style design, we examine how student educational and behavioral outcomes, school climate perceptions, and criminal legal system engagement respond to RP exposure.

We find RP decreased out-of-school suspensions by 17%. We find no evidence of corresponding increases in in-school suspensions, suggesting that students are receiving more in-school instruction time in response to policy adoption. We identify an 18% decline in child arrests, with significant decreases both during school hours and on school grounds and outside of school, and for both violent and non-violent offenses. This suggests that the introduction of RP generated meaningful changes in underlying student conduct and demonstrates that school practices may meaningfully shape socializing behaviors.

We find an improvement in student-reported perceptions of school climate, which include measures of belonging and safety. Exploring the importance of implementation, we show our results are driven by CPS' intensive coaching program that involved ongoing professional development of staff throughout the year. We do not identify increased classroom disruption as measured through GPA or test score changes. Finally, we find that Black males experience the largest declines in out-of-school suspension days and arrests, and separately increases in attendance and in math test scores.

Our results provide evidence that an alternative approach exists that helps solve the perceived tradeoff between educators feeling the need to suspend students or else face learning disruptions within schools. Our findings suggest that no such tradeoff exists. RP has the potential to improve student perceptions of school climate and reduce behavioral incidents inside and outside of school without harming academic performance, potentially improving the daily experiences of all students.

Authors

• Anjali Adukia, University of Chicago

• Benjamin Feigenberg, University of Illinois at Chicago

• Fatemeh Momeni, University of Chicago