The Disparities of Disparate Impact

Is there a racist behind every tree in the American education forest? That’s the spin a lot of people have given to last week’s massive trove of federal data on school discipline and sundry other topics. “Black students face more harsh discipline” headlined the New York Times. “Minority students face harsher punishments,” quoth the Associated Press. “An educational caste system” stormed the head of the country’s largest coalition of civil-rights groups.

Is there a racist behind every tree in the American education forest? Photo by Stuart

The federal data (from 2009-10) cover a multitude of issues but what caught most eyes was the finding that black and Latino students are suspended or expelled from school in numbers greater than their shares of the overall pupil population. “The undeniable truth,” declared Education Secretary Arne Duncan, “is that the everyday educational experience for many students of color violates the principle of equity at the heart of the American promise.” Declaring that the new data paint “a very disturbing picture,” Assistant Secretary (for Civil Rights) Russlynn Ali proudly informed the media that her office has “launched 14 large-scale investigations into disparate discipline rates across the country.”

Ponder the phrase: disparate discipline rates. This arises from the doctrine of “disparate impact,” a sly phrase coined as a means of boosting civil rights in the realm of employment law. It means, in effect, that discrimination may be afoot—and enforcement called for—whenever a seemingly neutral or universal policy gives rise to disparities (by race, gender, etc.) in whatever benefit or harm that policy leads to. But it’s by no means limited to employment any longer.

At the Education Department’s Office of Civil Rights (OCR), the enforcers hunt for disparities in sundry realms of education from college admissions to Advanced Placement course access, as well as discipline and more. If they find that something good or bad isn’t getting bestowed across the entire eligible population in proportion to the basic demographics of that population, they sense “disparate impact” at work, which is invariably accompanied by at least a hint that discrimination must be the cause of it.

Such hints swiftly get picked up, echoed, and amplified. That’s what Wade Henderson, CEO of the Leadership Conference on Civil and Human Rights, was about when he  thundered that the OCR data “points [sic] to mass and systemic discrimination in our public education system” and it’s the Education Department’s duty to “investigate school districts…and take appropriate enforcement action.”

Ms. Ali, one senses, also sees that as her duty.

The primordial problem with this whole line of analysis, of course, is that an infinity of good and bad things get unevenly distributed across populations for reasons that have absolutely nothing to do with the kinds of discrimination that are banned in our laws and Constitution. People who aren’t very smart are disproportionately rejected by the Princeton admissions office. People who aren’t very tall seldom make it onto varsity basketball teams. (It often appears that white and Asian students—pace Jeremy Lin—don’t either.) Those who can’t hear very well seldom play violin in the school orchestra. And on and on.

As for school discipline, there’s a reason for it. It’s to make naughty, disruptive, or disorderly kids behave or exit, both for their own good and for the good of the school as an educational institution. Enforcement ranges from keeping kids safe (from weapons and fighting, for example) to creating a calm, respectful atmosphere in which those who are serious about learning can study without disturbance. Discipline, in other words, isn’t only about those being disciplined. It’s even more important for everybody else.

The Wall Street Journal’s invaluable editorial writer, Jason Riley, picked up on this in a perceptive March 10 columntitled “What about the kids who behave?” “The Obama administration’s sympathies,” he wrote, “are with the knuckleheads who are disrupting class, not with the kids who are trying to get an education. But is racial parity in disciplinary outcomes more important than school safety? Going easy on the students who behave badly—especially in inner-city schools where the problem is pronounced—is an odd way of advancing black education and closing the learning gap. Black kids already tend to be stuck in dropout factories with the most inexperienced teachers. Must they be consigned to the most violent schools as well?”

Riley correctly added that data such as these create an even stronger argument for school choice—charters, vouchers, and more—to enable low-income families whose kids are serious students to escape intolerable schools for better ones. (Let’s hope Ms. Ali’s enforcers don’t bully the charter and private schools into disciplinary submission, too.) On some parts of the choice agenda—charters in particular—the Obama-Duncan administration has been positive. It’s been death on vouchers, though, even in inner-city Washington D.C., due in no small part to its pals in the teacher unions.

Apropos of which, another much-discussed pattern in the new OCR data is the presence of lesser-paid teachers in heavily minority schools. Again, the civil rights folk imply that this signals discrimination against black and brown kids. (“Give ‘em the cheap teachers.”) But in the real world it almost certainly stems from the fact that—as Riley noted—inner-city schools have, on average, less experiencedteachers, hence teachers with relatively lower salaries. Why? Many reasons, of course, including the disruptive and insufficiently disciplined atmosphere in some such schools, but also because—thanks, once more, to the teacher unions—veteran instructors enjoy contractual provisions that allow them to choose their schools and for some reason (badly behaved students, perchance?) more than a few shun inner-city postings. We shall see whether Mr. Duncan and Ms. Ali manage to move teachers around against their will to overcome this particular “disparate impact.” (There’s a big loophole in the spending-comparability requirements of the Title I program which, if closed, would prod such re-assignments.) Or if they push hard against discipline policies aimed at keeping inner-city (and other) schools orderly enough that those who do teach in them will come back.

They’ll be under plenty of pressure to do that. At the UCLA-based Civil Rights Project, for example, long-time activist Gary Orfield is calling for “stepped-up enforcement actions by the Office for Civil Rights to respond to the stark disparities in discipline, not to mention the many other indicators of injustice and inequity. For example, the number of “disparate impact” interventions has been disappointing… OCR should actively investigate the pronounced disparities revealed by the data. Where unjustifiable policies are to blame, OCR should use its enforcement authority as well as technical assistance resources to spur schools and districts to replace the ineffective policies with less discriminatory ones.”

Expect much more of this sort of thing—and don’t be surprised if Ms. Ali is quietly soliciting it!

One more point. The OCR data themselves emerged from an overhauled version of a long-standing biennial survey of schools and districts serving about 85 percent of U.S. students. They’re self-reported, however, and susceptible to error and misinformation at every level. Examine closely the results for any given district or school and you’re apt to find stuff that doesn’t make sense on its face. Consider, for example, the absurdity of the Seattle Public Schools reporting that they spent just $323.53 per pupil on instructional-staff salaries in 2009—about $2,623.14 less than the neighboring Shoreline School District. It was probably a misplaced decimal point—but it made it into the national data set, the averages and, presumably, the “disparities.”

Is such information robust enough to sustain enforcement actions? It obviously is in the eyes of Orfield, Henderson and others. The prior question, however, is whether “disparate impact” is a reasonable basis for such actions in the first place. What if it is simply true—regrettable, but true—that some kids or groups of kids break school rules more often than others?

-Chester E. Finn Jr.

This post originally appeared on the Fordham Institute’s Flypaper Blog

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428

For subscription service to the printed journal
Phone (617) 496-5488

Copyright © 2024 President & Fellows of Harvard College