Jerome Gorin / Getty

When the Trump administration released its school-safety report last month, it landed with a thud—and only partly because it’s a clunky 180 pages. Many of the recommendations in the report, authored by the Federal Commission on School Safety, are aimed at fostering a better school climate—how a school feels to the students who attend it—whether that’s through improved access to counseling and mental-health services or a greater emphasis on social-emotional learning. But other recommendations were met with derision, such as a proposal to rescind an Obama-era rule urging schools to be mindful of whether they might be punishing minority students at a higher rate than white students.

Study after study has shown that black students are unevenly suspended or expelled from schools nationwide. The 2014 school-discipline guideline was the Obama administration’s attempt to remedy that. The Trump commission, however, argued that deciding how students should be disciplined should not be the federal government’s job, but the teachers’. Both administrations, at least, agreed that discipline was also a matter of school climate—something educational leaders have been trying desperately to improve.

A new study by the Rand Corporation, a nonpartisan think tank, shows just how crucial improving the climate at school can be to helping decrease suspensions. In 2013, Pittsburgh’s public schools were trying to figure out how to remedy racial disparities in discipline. At the time, they had mandatory diversity training for staff that sought to address implicit bias and discrimination in the classroom, but they wanted to do more. Restorative practices, which are nonpunitive ways of responding to conflicts, had been gaining momentum among school leaders as a way to help curb suspensions.

So the district got a grant to try out restorative practices in their schools, randomly selecting 22 of them to receive the restorative treatment, while 22 others went about business as usual. The basic goal of restorative practices is to build relationships between teachers and students, so that students will be less likely to act out. Teachers start off the school year by asking students innocuous questions such as what the students did that summer. As the year goes on, the questions grow more personal and introspective, and students build trust with the adults and classmates around them. Of course, formal times for such events can be time-consuming, so it is often recommended that the practices are woven into the day. As much as restorative practices aim to change how students are disciplined, they also seek to change the behavior that might require discipline, improving the overall climate of the school.

The researchers examined the schools—elementary, middle, and high schools—over two years and found that restorative practices greatly reduced the number of school days lost to suspension, particularly among elementary schoolers. The dip was most acute among black, low-income, and female students, and nonviolent offenses drove the decline. “It seems to be the case that restorative practices were providing an alternative that the staff felt they could use to enforce discipline, [especially] for offenses that weren’t extremely serious in the sense of endangering people’s safety,” John Engberg, a researcher at Rand, told me.

On top of that, the report found no negative impact on the test scores of students in the schools that had restorative treatments. “That seems to indicate that keeping kids in school is not leading to a deteriorating learning environment,” Engberg said. And, for their part, teachers who worked at schools with the restorative treatments rated their climate as comparatively more positive.

There were some things that restorative practices couldn’t change, though. Sure, academic outcomes, such as test scores, didn’t drop, but they didn’t improve either. The decline in suspension rates was most stark for elementary-school students rather than middle- or high-school students, where the effects were more muted, suggesting that early intervention is important.

Changing a school’s climate is a long process, Catherine Augustine, a senior researcher at Rand, told me.“This isn’t Let’s go to a one-day workshop and we’ll all be restorative,” she said. It takes work from teachers, faculty, staff, and students. And the researchers themselves still have a lot of work to do in terms of understanding how restorative practices work and whether the gains made by the elementary schoolers will carry forward through middle and high school. Still, the bipartisan goal of improving school climate may not be as elusive as it seems.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.