The Department of Education Rediscovers Differences in School Discipline Rates

The new data issued by the U.S. Department of Education’s Office of Civil Rights find, among other things, differences in suspension and expulsion rates among racial and ethnic groups. This has provoked reactions of shock by Education Department officials and other concerned parties. CNN quotes Education Secretary Arne Duncan proclaiming "Perhaps the most alarming findings involve the topic of discipline … The sad fact is that minority students across America face much harsher discipline than nonminorities, even within the same school. Some examples - African American students, particularly males, are far more likely to be suspended or expelled from school than their peers." The report has been discussed on television and in the news as if this were some new and alarming finding. In fact, though, the existence of these differences is not news at all. The variations in discipline are substantially the same as those that Steve Caldas and I reported in our 2005 book, Forced to Fail: The Paradox of School Desegregation (paperback edition, 2007), and in our discussion of the figure below Steve and I mainly laid out what was already obvious and widely-known: black students are disciplined more than any others, followed by Hispanics. Asian students are suspended and expelled less often than any others, including whites. Note that the data we used in the book come from a report in the 1990s. So, what’s the news here?

The best answer is that Secretary Duncan and others are trumpeting old news as an alarming discovery because they want to use it to push race-conscious policies in the schools. The Department of Education presents that data as "raising fundamental questions about fairness." In taking this preconceived perspective, the department betrays an implicit assumption that schools must be responsible for group differences in suspensions and expulsions. This may be because administrators and teachers, consciously or unconsciously, respond more severely to infractions by black or Hispanic students than to infractions by white or Asian students. Or it may be that behaviors really do differ because of varying situations outside the classrooms and schools should be responding in ways that will eliminate the differences.

The first possibility, that schools are engaging in racially discriminatory behavior in discipline, is theoretically possible, but not supported by the facts. The larger the number of black and Hispanic students in schools, according to long-standing evidence, the more disciplinary problems there are in those schools. As we reported in our book, one out of every five principals in schools with mostly minority students said that their schools had serious discipline problems. Half the principals in mostly minority schools reported having moderate problems. By contrast, the overwhelming preponderance of principals in schools with fewer than 5% minority students said that they had no discipline problems. The more minority students there are in a school, the more common the kinds of behaviors that lead to suspensions and expulsions.

But perhaps those principals’ reports are distorted by their own unconscious biases. That’s not likely. The students themselves much more often reported feeling unsafe, being threatened with weapons, and being in fights when there were more black and Hispanic students in the schools. In general, the evidence just doesn’t support the idea that discipline is being enforced in a discriminatory manner. It does support the idea that the groups vary in their actual rates of misbehavior and that punishment reflects this.

But isn’t it “racist” to suggest that orderliness is not evenly distributed across all groups? No, not only is it not morally wrong to present such a hypothesis, there are excellent, empirically based explanations for why behavioral variations by race and ethnicity exist.  One of these explanations comes from family background. An extensive social scientific literature, as well as common sense, tells us that behavior problems in school are clearly correlated with single-parent family structure. The demographic groups that are disciplined at higher rates than others also have much higher percentages of children living in single-parent families. In my own research, I’ve found that many of the educational disadvantages of minority children result from their being in schools containing large concentrations of children from single-parent families, a fact best explained by the behavioral problems in those schools. Beyond families, of course, there are also communities. Social order in schools is frequently a reflection of community social order.

But what about the other claim, that schools are failing to live up to their responsibilities to deal with misbehavior in a way that will erase all the family and community disadvantages that some students bring with them to the classroom? If Secretary Duncan and his associates think that is possible, I have to wonder what they’ve been smoking around the conference table. I hope that their answer would not be to lighten up on the suspensions and expulsions of black and Hispanic students. That would simply keep more problem students in the classrooms and further disrupt the learning of their classmates, who are most often other black and Hispanic students. More seriously, the shift of responsibility for individual actions from the individual to the institution can only worsen the problem by ingraining the view that anything pupils, or pupils in specific groups, do wrong is the fault of the school or the society.