In all the talk about the Affirmative Action cases at University of Michigan, no one is looking at any real data. For example, MIT’s amicus curiae brief says the same old tired things about promoting diversity. I say tired because this argument is no longer persuasive to the public or the courts, even though it is right.

What would be persuasive would be evidence that shows that students admitted under affirmative action policies which consider race as part of the process do just as well at university as other students. Now either this statement is true or it isn’t. Let’s consider both cases.

Case A) Yes it is true: those students do just as well as ( or better than ) other students:

I remember a similar discussion from my years at MIT, when complaints were made that male Asian students were being discriminated against in the admissions process, because “less qualified” female Asian students were given preferential admission. The Admissions Office released some data looking at the actual performance of students which showed that the female Asian students actually performed better than “similarly qualified” male Asian students, thus justifying the preference given to those students during the admissions process.

Confronted with such evidence, how could a court strike down this process?

Case B) No it isn’t true: those students don’t perform as well as other students:

One would think that if a University truly had “promoting diversity” as its goal, then it would put resources into making sure that once a minority student gets to university, that the student actually succeeds. After all, it really doesn’t do any good for anybody if more minority admissions simply translates to more minority dropouts.

Logically then, if the students don’t perform just as well as other students, shouldn’t we be blaming the university for failing to follow though with the promise of diversity in education? We have seen this pattern in hiring of women and minority faculty, where although hiring patterns have changed tenure patterns have not. In either case, the conclusion should be clear: just changing the admissions policies isn’t enough: universities have a duty to make sure minority students succeed. If they don’t, it is the education policy which is broken, not the admissions policy.

Epilogue: I hope at some point in the court cases, someone will pick up on the idea that admissions policies are testable, and that looking at the performance of students after they are admitted can be valuable when evaluating the admissions policies.