RAND recently published its evaluation of a Pittsburgh Public Schools initiative, “Pursuing Equitable and Restorative Communities” (PERC), that used nonpunitive approaches, rather than traditional punishments like detention and suspension, to address student misbehavior. Researchers employed a randomized controlled trial to assess the effects of the program on student behavior, academic achievement, school climate, etc.
First, the good news…
We found strong evidence that PERC had positive effects on teachers’ perceptions of teaching and learning conditions. Teachers’ responses to the district’s Teaching and Learning Conditions Survey indicated significantly higher ratings of conduct management, teacher leadership, school leadership, and overall teaching and learning conditions in the PERC schools than in control schools. The impact of PERC on conduct management is driven by the positive and statistically significant impact on responses to items about whether faculty work in a safe environment and whether they understand policies regarding student conduct.
Although suspension rates have gone down in the district overall in the past few years, PERC further reduced both the number of days students were suspended and the number of suspensions. Not only were PERC students less likely to be suspended, but they were less likely to be suspended multiple times.
Now the bad news…
We also found negative impacts of PERC. Despite fewer suspensions, academic outcomes did not improve in PERC schools. At the middle grade level (grades 6–8), academic outcomes actually worsened in the treatment schools. Neither did we find fewer suspensions in middle grades. It could be that it is more challenging for restorative practices to positively affect middle grade students, at least within a two-year time frame.
We did not see fewer suspensions for male students, for students with individual education plans, or for incidents of violence or weapons violations. Neither did we see a reduction in arrests. This might be because teachers have more discretion to implement a restorative punishment for nonviolent behavior, whereas the district’s code of conduct requires a suspension for violent behavior. This, of course, raises the question of whether restorative practices can be effective in curbing the most violent behavior, at least within a two-year implementation period.
Two years may not be enough time to produce desired increases in student achievement, but a decline in performance is not a good sign.