The Senate hearing at which I testified yesterday, chaired by Sen. Dick Durbin, seemed designed to raise support for legislation imposing federal mandates on states or districts to curtail the use of out-of-school suspensions, to make suspension policies uniform across schools, or both.

The motivations for such legislation are understandable. Out-of-school suspensions do little to help the suspended students educationally and they are correlated with arrests and incarceration. The Judiciary Subcommittee on the Constitution, Civil Rights, and Human Rights, which Sen. Durbin chairs, is particularly interested in these facts because African American students are much more likely to be suspended than whites.

But the facts do not support the kind of legislation that seems to be under consideration. Two recent and highly sophisticated studies by Rochester University professor Joshua Kinsler shed new light on the well-established trends noted above. For the first time, Kinsler factored-in between school variations in discipline policy when looking at the racial disparity in out-of-school suspensions. He discovered that, within any given school, black and white students sent to the principal’s office for a given reason are issued the same suspensions at the same rates. The disparity is all between schools.

Schools with predominantly black student bodies are more likely to issue suspensions, and to issue longer ones, for a given offense. White students at those schools get the same treatment, but most white students are in predominantly white schools that are less severe in their discipline policies. Black students at mostly white schools also get less severe punishments.

Kinsler did find that African American students were more likely to be referred to the principal’s office, which has long been seen as evidence of systemic racism. To investigate that explanation, Kinsler looked for any relationship between teachers’ referral rates to the principal’s office and the race of those teachers and of the students they refer. He found none. This does not mean that racism plays no role, but it calls into question the view that racism is a dominant factor in referrals to the principal’s office.

In a subsequent empirical study, Kinsler investigated what would happen if all schools were compelled to observe a more lenient suspension policy, to close the black/​white discipline gap. He found that this would disproportionately hurt the achievement of African American students, widening the black/​white achievement gap. The reason for this, according to Kinsler’s findings, is that serious suspensions do in fact discourage misbehavior, and that removing disruptive students from the class does improve the achievement of the other students.

Kinsler’s methodology, which jointly models discipline policy, student behavior and student academic achievement, is the most advanced I’ve seen used in this field. Unless and until his findings are found to be in error, or are contradicted by similarly sophisticated research, it would be unconscionable and counterproductive to impose a blanket reduction in suspensions on the nation’s schools.

None of this is meant to defend the cavalier use of out-of-school suspensions. As I explain in my written testimony to the Committee, there are much better alternatives and there are policies that will systematically encourage the use of those superior alternatives. I sincerely hope that Senator Durbin and his Committee do not leap before they look at these alternatives and at professor Kinsler’s findings.