To raise students’ test scores, schools should stop pushing out troublesome students, argues a new report.
A recent report published in December’s issue of the American Sociological Review finds that students in schools with high rates of suspensions suffer academically -– even if they are not being suspended themselves. The report, compiled by University of Indiana professor Brea Perry and University of Kentucky professor Edward Morris, concludes that high rates of suspensions can have a negative impact on the test scores of students who have not been suspended, and that schools may be better served by only suspending students in moderation.
Evidence shows that high incarceration rates can have a devastating impact on offenders, as well as on surrounding families and communities. The researchers set out to see if a similar tactic would yield the same results in schools -– if high rates of suspension could negatively impact other students' academic achievement.
To glean their results, researchers followed about 16,000 middle school and high school students in a Kentucky district over the course of six semesters. They tracked the students’ scores on a statewide test administered three times a year, and compared it to their schools’ suspension rates during the time of test taking. Researchers controlled for factors that are correlated with a school’s suspension rate, like poverty. They also adjusted for factors like the annual number of drug infractions in a school, violent infractions and incidents of disruptive behavior, so that a school’s normal level of crime would not confound “the association between suspension and achievement,” says the report.
Perry told The Huffington Post the researchers adjusted for these factors so that “any effect suspension has is over and above those events occurring. We wanted to find if something about the punishment itself ... is harmful to students above and beyond the disruption of being in a high violent context.”
Researchers found that high rates of school suspensions had a substantial negative impact on individual students' test scores –- especially in schools with typically low levels of violence.
“Low and moderate levels of suspension are benign … there is no benefit or harm,” said Perry. But when the suspension levels become "excessive," it starts to affect "the achievement of non-suspended students.”
She continued, “[Excessive suspension caused] negative consequences in all schools, but it was especially harmful when the level of violence in school is low. … Suspension is most detrimental when it is probably perceived by students as illegitimate, overused or used inappropriate. Kids are looking around saying, ‘This is happening way too often, things in school are great, and you’re using suspension.’”
According to the report, when a school with typically low levels of violence experiences high rates of suspension, “the predicted percentile score in reading achievement decreases from about 54th at the mean level of suspension to 28th at very high levels of suspension.” Overall, researchers conclude that high suspension rates "can create a heightened sense of anxiety" for students and that "turnover of suspended students in and out of classrooms creates unstable, socially fragmented environments."
Suspension rate disparities in the studied district mirrored national suspension disparities –- meaning that African-American students were disproportionately suspended at the same rates reflected in national data.
The research demonstrates that suspension should be used as a last resort, limited to situations where the safety of students is threatened, said Perry. She noted that a school used in the study recently implemented its own suspension diversion program, in which misbehaving students were asked to sit and do homework rather than being sent away. She says the school saw an increase in test scores.
“There’s nothing good about [suspension] for suspended kids, but also nothing good about it for non-suspended kids,” said Perry.