Study: Racial disparities in middle school discipline escalate early in the year and dip ahead of breaks

A new study finds that daily discipline rates in middle school escalate quickly in the first days of school and fluctuate throughout the year in predictable ways — often dipping ahead of breaks.

These escalations are most acutely felt by Black students, the data shows. By November, the Black student discipline rate is 10 times higher than it was at the beginning of the year and 50 times higher than the white student rate at the beginning of the year.

The study, published in the Proceedings of the National Academy of Sciences, demonstrated that discipline dipped ahead of breaks, including Thanksgiving, winter break and spring break, and then it escalated rapidly in the days after the break. Researchers also translated the data affecting Black students into sound to convey how the constant barrage of discipline might feel to them.

There was no sustained “cool-down” effect in the days after breaks, which surprised author Sean Darling-Hammond, assistant professor of biostatistics, community health sciences and education policy at UCLA.

“It indicated just how pernicious, pervasive and stable this escalation that occurs is,” he said. “Once you get to November, this is your new baseline of discipline.”

Racial disparities that show up in the first 10 or 20 days can predict the disparities that will persist all year. Researchers considered factors such as income and gender, and it looked at both suspensions and lower-level disciplinary rates. But the data consistently showed that Black students were far more likely to be disciplined than any other group.

“Teachers’ perception of students change very quickly at the beginning of the year, and that’s very different for students of different races,” said Darling-Hammond.

He said the findings point to the need for professional development on discipline to come early in the year.

“Teachers are coming into the year with every intention of being fair and equitable in how they respond to students, but something is hijacking their ability to do that,” he said. “If we can start to understand exactly what that is when it happens, we can design interventions that are more responsive.”

One trend didn’t show up in the data: annual standardized testing didn’t have any effect on discipline rates.

Researchers chose to focus on middle school-aged children because there is a “giant jump” in discipline between elementary school and middle school, said Darling-Hammond. Focusing on this precarious time could potentially help students avoid trajectories that are an “on-ramp to the school-to-prison pipeline.”

California has been trying to reduce suspensions and expulsions for several years by hiring more counselors, addressing the underlying cause of students’ misbehavior and encouraging restorative justice programs. In 2019, California banned suspensions in elementary and middle schools for willful defiance, defined as disrupting school activities or defying school authorities. Until the pandemic, schools were showing progress in reducing suspensions, which tend to primarily involve Black students and students with disabilities.

The study looked at four years of data from a single school district whose name and location were not revealed by researchers, due to a data-sharing agreement with the district. Darling-Hammond noted that is one of the top 30 largest school districts in the country. This data set from the 2015-16 school year to the 2018-19 included the daily discipline rates for 46,964 students in 61 middle schools that served sixth through eighth graders.

The school district was chosen for study because of the “extraordinarily rare” granular data set that the school district collected on discipline, said Darling-Hammond. It included the demographics of students involved, including race, gender, economic status and whether the student was an English learner or special education student. It included the reason for the disciplinary action and the outcome, which ranged from a verbal warning to suspension.

But researchers said that the school district in question was also a useful one to study because it included demographics similar to many other American school districts. It included both urban and suburban schools, and its racial demographics are close to the demographics of the United States: 5% Asian, 18% Black, 17% Hispanic and 55% White. The median income is $54,000 and 58% are eligible for free and reduced-price lunch.

The district began to implement “restorative practices” and “positive behavioral interventions and supports” practices during the 2015-16 school year. This showed up in the data that showed an overall reduction in discipline, but it showed consistently wide racial disparities.

“This isn’t a story of ‘nothing works,'” said Darling-Hammond. “This is a story about how what we’re doing could work a lot better if we are attending to the dynamics of what’s happening in school and when it’s happening.”

The study was co-authored by Jennifer L. Eberhardt, professor of organizational behavior and psychology at Stanford, Jason A. Okonofua, assistant professor of psychology at UC Berkeley and Michael Ruiz, researcher at UC Berkeley.