Boston University’s Andrew Bacher-Hicks discusses causal evidence that suspensions lead to negative later-term outcomes.
Descriptive research has highlighted a relationship between students who are suspended and negative outcomes later in life. In 2014, findings like these prompted the Obama Administration to urge schools toward less strict disciplinary approaches than the zero-tolerance policies popular at the time. This guidance has since been rescinded by Education Secretary Betsy DeVos based on the claim that it made schools less safe for students. And while many past studies looked at the correlation between suspended students and later involvement with the justice system, causal evidence has been scarce.
While suspensions had previously been correlated with negative life outcomes, Andrew Bacher-Hicks, Assistant Professor of Education at Boston University, and his colleagues at the University of Colorado Boulder and Harvard University devised a study to tease out whether strict school discipline was actually causing these events. Leveraging data from a school zone boundary change in 2002-2003, a “natural experiment,” Bacher-Hicks’ research examined the effect of attending a strict school on kids.
“There’s this commonly-held belief that suspensions can lead to what’s known as the school-to-prison pipeline,” explained Bacher-Hicks. “But there hasn’t been a lot of causal evidence of this. Our study set out to estimate the causal impact of attending a school that is strict on discipline.”
From the 1970s until 2002, schools in Charlotte Mecklenburg operated under a court-ordered desegregation plan that involved busing students from different neighborhoods into schools. Yet a legal battle that ended in 2002 prompted school zone boundaries to be redrawn based on school capacity and the number of students living around a school. These new boundaries often bisected neighborhoods.
“There are a number of studies that have leveraged this exact same boundary change as a natural experiment,” offered Bacher-Hicks. “The thing that was most interesting to us is that schools in this district had substantial differences in their discipline policies.” This change in boundary allowed them to track a number of variables for students assigned to relatively strict schools, compared to students assigned to less strict schools.
“There’s important nuance to this study,” Bacher-Hicks continued. “Rather than tracking the effect of suspensions themselves, we really looked at the effect of what we deemed ‘strict schools.’ This designation includes the overall package of disciplinary policies, which notably includes suspension rates.” Thus, the study doesn’t link specific instances of suspension to specific outcomes for an individual later on. “We’re not able to determine with this study what would happen if two students are placed in the same school and you suspend one and not the other. This study looks more about the differences of sending a student to a strict school vs. a less strict school.”
The study found that, on net, the effects of attending a strict school are negative across all students. “Schools that suspend more students see a host of negative outcomes later in life,” he explained. These negative outcomes included lower educational achievement, lower graduation rates, lower college enrollment rates, and higher involvement in the juvenile and adult criminal justice systems.
These negative effects are, however, concentrated among Black males. “The gap between Black male students and all other students was considerable,” said Bacher-Hicks. “Across just about every outcome, the harmful effects of suspensions were greatest for Black males.”
A more surprising finding of the study with sweeping implications is the lack of positive spillover of suspensions for any student subgroup. “We looked at whether attending a strict school had harmful effects for one subgroup but positive effects—such as increased test scores or a reduction in likelihood of dropping out—in another, but we found no evidence of that.” No positive benefits of strict school discipline that many have claimed for decades were born out in the data of this study.
Findings like these have obvious and lasting implications on the conversation around both discipline and race in the US. When viewed in concert with research examining the effects of police violence on students, for example, this study adds affirmation to claims that discriminatory policies and structures start early and have lasting effects. To engage with findings like these on a practical level, districts and schools can work with states to track long-term outcomes of discipline policies and practices in their communities. Additionally, districts can design mechanisms that check for “strictness” disparities among schools and determine the best way forward.