Responding to Evidence that Race-Based Law School Affirmative Action May Actually Hurt African-Americans:
By VIKRAM DAVID AMAR
|Thursday, Jan. 20, 2005|
What Professor Sander Argues About Race-Based Affirmative Action
Here is a capsule summary of Sander's argument:
Most law schools take race into account in a very substantial way in order to admit underrepresented minority students (Sander uses blacks as the paradigm). Accordingly, at each school, the admitted African-American students possess, as a group, numerical admissions credentials and academic skills that are considerably lower than those possessed by the non-minority law students.
As a result, when students matriculate, a sizable majority of blacks fare very poorly in law school classrooms, and receive grades that tend to place them near the bottom of their classes. After graduation, because their grades tend to be low, black law students are much more likely than are whites to fail the bar exam. And in the job market, low grades plague efforts by black law graduates to obtain the best law jobs.
Building on his statistical observations and analyses, Sander argues that many blacks with poor law school grades would have had better grades -- and would have ended up with a higher chance of passing the bar -- had they attended law schools more commensurate with their academic skills, as represented by their college grades and LSAT scores.
Indeed, Sander says that if race-based law school affirmative action were eliminated or reduced, the black bar passage rate would go up, and the number of blacks entering the bar would actually increase!
In the end, Sander argues that whether one focuses on passing the bar or getting a good job, there is a strong case that race-based affirmative action hurts, rather than helps, black law students.
Questions About Sander's Statistics and Methodologies
As might be expected, many academics are taking issue with Sander's analysis, and are writing responses that will be published in various law reviews in the coming months. Some of the criticisms go to Sander's statistical data sets -- asking, for example, whether the data he uses is broadly-based enough and whether it comes from years that are representative, rather than abnormal.
One particularly important critique goes to Sander's methodological assumption about how many blacks would - if affirmative action were ended -- actually attend less prestigious law schools after being turned down by the more elite schools they currently get into (by virtue of affirmative action). Some critics argue that Sander vastly overestimates how many blacks would still go to (lower-ranked) law schools, rather than eschew law school altogether in favor of business school, other grad school, or other career alternatives.
This is a powerful criticism. If Sander indeed understates the number of blacks who would leave the law school world altogether if affirmative action were ended, then his claim that the number of newly-minted black lawyers would rise in the absence of affirmative action seems much less plausible.
It is hard to know how off the mark Sander's assumptions on this score may be; determining exactly how many blacks would go to business school -- or do something else entirely -- if they were not getting into the more elite law schools is impossible. I will say, however, that Sander certainly should not assume that blacks who get turned down by the higher law schools would attend less competitive law schools at the same rate as do white law students who are denied admission at the higher law schools.
Why? Because black law school applicants who get turned down may have more, and better, options than their white counterparts. After all, even if race-based affirmative action were abolished at all law schools, affirmative action would remain for other graduate schools and in the business world.
But could Sander argue that affirmative action in those realms is similarly destructive to African-Americans, and thus should be reduced? It's possible, of course - but no one, to my knowledge, has done a Sander-like analysis for business schools or medical schools or entry-level jobs. If such analyses were done, they might well show different results than Sander's for law school: The effects of affirmative action in other areas may well be very different -- and the arguments in favor of reducing affirmative action in these other areas may thus be much weaker.
Another question has been raised about Sander's claim that if affirmative action were abolished, and black law students therefore attended law schools where their numerical admissions credentials quite closely paralleled those of their white law school colleagues, blacks' law school grades, bar pass rates, and employment prospects would all be the same as those of whites. Some social scientists argue that even among black and white students with comparable LSAT scores and grades, who are classmates at the same law school, the black students' grades and bar passage rates will tend to be lower.
The explanation for this, these scientists say, is that law school culture is not as comfortable for black students as it is for white students, for a variety of reasons. Always being a numerical minority may take a psychological toll, for instance. So may the lack of same-race role models on the faculty - as well as white professors' possible inclination to pick same-race protégés.
Sander does not really delve into the possibility that it is an uncomfortable law school culture, rather than the "mismatch effect" of students attending schools other than the ones their grades and test scores would suggest, that leads to African-American law students' comparative underperformance.
A Question About Sander's Framing of the Inquiry
My own reactions to Sander's interesting article focus less on the statistical rigor of his assumptions and calculations, and more on the way he has styled his conclusions.
Let us assume that Sander is correct that law students with incoming admissions credentials significantly lower than those of most of their classmates will likely be better off - in terms of bar passage and job options - at less competitive law schools. Even so, that same insight would likely be true for white law students whose admissions numbers are lower than those of their law school classmates as well.
In other words, Sander's point should not really be about race-based affirmative action in itself, but rather about any admissions policies and decisions by a law school that result in the admittance of students whose college grades and LSAT scores are significantly lower than the law school's medians.
So, if Sander is correct, any law student whose admission was based less on his numerical credentials, and more on intangible diversity factors - whether resulting from his race, age, alumni connections, past work experience, geography, or another factor - would be well-advised to think about attending another, less competitive, school.
The question becomes, then, why are we focusing only on the racial dimension of the problem? Earlier opponents of affirmative action who criticized it as being non-meritocratic failed to expand their critique to include non-meritocratic admissions criteria that have nothing to do with race (like alumni legacy status). So, too, Professor Sander's discussion of the counterproductive aspects of affirmative action fails to encompass how other admissions criteria are (or at least may be) similarly counterproductive.
Some Questions About Sander's Prescription
Moreover, even if Sander's analysis and focus are correct, his primary prescription as to what ought to be done in response might not be. If black law students might be better off attending less competitive schools, perhaps the answer is not to abolish or pare back race-based affirmative action, but simply to provide full disclosure about the effects of affirmative action to black applicants, letting each admittee decide for herself whether she thinks she can buck the statistical trends Sander documents.
If this system were followed, students could take their own individual experiences into account. An admittee whose strong college performance was not predicted by her SATs and high school grades may believe that, once again, she will "overachieve" in law school - and enter the best school that accepts her. In contrast, an admittee who is highly sensitive to, and demoralized by, getting poor grades may realize she would prefer being a student with a better grade point average at a lower-tier school. An admittee who knows his charm and public speaking finesse will help him prevail in Moot Court may not worry as much about grades; but a shy admittee who knows he does not interview especially well may want to go to a school where his grades will likely be better.
Providing full disclosure -- rather than ending (or reducing) the affirmative action programs themselves -- might be particularly desirable if one believes that the advantages of affirmative action transcend the particular law students who get into law school by virtue of them. In other words, if one widens the focus from the students themselves (on whom Sander focuses) to society at large, the benefits of affirmative action persist even if a substantial number of students who are admitted under such programs struggle.
When minority students attend law school, the educational experience for all students (not just those of color) is enhanced by diversity. When even a small number of minority lawyers succeed, their success is not only valuable in itself, but also inspiring; they may now serve as role models for other aspiring lawyers of color. Given these benefits, one might argue against Sander's bottom line even if his numbers are unassailable.
In other words, what Sander's numbers reveal, and what we do with those revelations, are separate questions.
The Important Service Sander Has Provided
Even with all the questions and quarrels about Sander's article that can be raised, Sander has performed a brave and enormously helpful service by putting his data and his analysis on the table. Most important, from my perspective, is that Sander's article demonstrates how much law schools have been failing to serve the minority students they have admitted.
The data on low grades and low bar passage rates for black law students that Sander compiles are startling. And they clearly point up the need for law schools to devise, fund, and implement more and better programs for helping students who come into law school not as academically prepared as their classmates.
The big message I take from Sander's piece is not necessarily that affirmative action at the admissions door is pernicious. Rather, the message might be that affirmative action at the admissions door without more is not very useful; it must be followed up with programs that make the law school experience - and not just the admissions experience - successful for students of color.