Association for Public Policy Analysis & Management

Schools across the country are seeking out alternative ways to address student behavioral issues rather than suspending students, which isolates the student from the school community and may not address underlying issues. On the one hand, providing students with emotional supports may have a positive effect on school culture if students are developing new strategies to deal with problems they face. At the same time, though, relying less on suspensions means that disruptive students who otherwise would have been suspended are in the school building more, which may have a negative effect on other students and school climate.

Identifying the effect of suspensions on student and school climate is an important but challenging task due to questions of the direction of causality and endogenous responses of schools. In order to understand the effect that reducing suspensions can have on student learning and the school climate, we focus on a policy change in Chicago Public Schools (CPS) that dramatically reduced the extent to which schools could use suspensions. Just prior to the beginning of the 2012-13 school year, CPS made changes to the Student Code of Conduct. The amendment mandated that principals acquire central office approval if they wanted to suspend students for more than five days. The district also eliminated mandatory 10-day suspensions and expulsions for the most severe offenses such as illegal behaviors around weapons possession, narcotics trafficking, and sexual assault. Note that these major changes to the SCC should affect the average length of out-of-schools suspensions. We start by exploring how the policy changed suspension practices in Chicago high schools, and then we use that exogenous variation to identify the effect of suspension reductions on student and school outcomes.

We compare suspension usage pre- and post-policy using a school fixed effects framework as well as student-level control variables to show the policy resulted in a reduction in the length of out-of-school suspensions in CPS high schools. After the policy change, these suspensions were on average about 0.5 days shorter, which is a substantial drop from the pre-policy trend in suspension length. We also show a marked decline in the proportion of CPS high school students that received a suspension longer than five days.

Because we see precipitous drops in these measures of suspensions that were targeted by the policy and this was the only policy change at the time that would affect the length of suspensions, we use this decline in the length of suspension to look at changes in student and school outcomes. Overall we find that after the policy students were in school more on average, and we find that reading test scores remain similar to pre-policy years with slight increases in math scores. However, both student and teacher reports of safety and order in the school building declined after the policy. Reducing the extent to which schools can use suspensions to address behavioral issues, then, has complex implications for principals and teachers in terms of the resources they need to help disruptive students.