A new study that's raising controversy in law-school circles questions whether admissions preferences for black students help them or, ironically, set them back in their careers.
Research by a respected law professor at the University of California, Los Angeles, asserts that blacks who benefit from affirmative action are being admitted to law schools where they find themselves in over their heads, achieving lower grades and failing the bar exam in higher numbers than they would have without the preferences.
The research by Prof. Richard H. Sander, scheduled for publication in this month's Stanford Law Review, turns traditional critiques of affirmative action on their heads. It already is under assault.
Some critics say the study dramatically understates the positive impact of affirmative action on black law students. Based on the same data the study used, Richard Lempert, a professor of law and sociology at the University of Michigan, argues that eliminating affirmative action in law-school admissions would reduce the number of black attorneys by at least a quarter.
"I and other people who looked closely at it absolutely despair at the quality of the research," says Prof. Lempert, an architect of the University of Michigan law school's affirmative-action program, which the U.S. Supreme Court upheld in a pivotal ruling last year. "His conclusions are just dead wrong."
Usually, social conservatives decry preferences because of perceived unfairness to white applicants. Although critics have talked before of a "stigma" that damages black recipients, the new analysis stands out for its detailed focus on alleged harms to the careers of black students.
"We need to take seriously the idea that there are potential costs to minorities who benefit from racial preferences," Prof. Sander says in an interview.
Prof. Sander, who describes himself as a lifelong Democrat sympathetic to the goals of affirmative action, claims that abolishing preferences wouldn't reduce the number of black lawyers. In fact, he estimates it would likely increase the cohort of black attorneys emerging from the Class of 2004 by 8% and the number of those passing the bar the first time by 22%.
The study comes after the Supreme Court last year in the Michigan case narrowly endorsed the use of race as a factor in undergraduate and law-school admissions. The court ruled that diversity in higher education was necessary to cultivate "a set of leaders with legitimacy in the eyes of the citizenry."
Drafts of the study, which hasn't been made public, have circulated among experts, and Prof. Sander discussed it at a recent academic conference. Critics, including Prof. Lempert, are drafting a harsh critique to submit to the Stanford Law Review.
Prof. Sander relied primarily on data that the Law School Admission Council collected on 27,000 students who entered 160 U.S. law schools in 1991, including their grades in college, test scores and bar-exam results.
The study found a stark achievement gap between blacks and whites throughout the nation's law schools. Close to half of the black law students ended up in the bottom tenth of their class. African-Americans were more than twice as likely as whites to drop out -- and more than six times as likely to fail state bar exams after multiple tries.
Prof. Sander argues that the reason for this outcome stems from a "mismatch" between the credentials of the black students and the institutions they attend. Because they have weaker credentials, he says, the students achieve lower grades. And since grades are strongly correlated to success on the bar exam, he argues, these students failed the bar in higher numbers.
He argues that students who perform at the bottom of their classes at more selective colleges often are confused by tougher material taught at speeds that challenge higher-achieving classmates. At less selective colleges, the material tends to be simpler, so these students can pull into the middle of their class and pick up the baseline information needed to pass the bar exam. And he says there is a "cascade effect" on every tier of law school, from Harvard and Yale down the ranks, ensuring that, at each level, blacks perform worse and are less likely to become lawyers.
By the study's tally, 86% of blacks currently admitted to law schools would still gain admission without preferences. But they would attend less competitive schools, where they would compile stronger records. The remaining 14% -- 500 to 600 a year -- would likely drop out or fail the bar.
To preserve diversity, Prof. Sander recommends setting modest goals for racial preferences -- about 4% in law school classes -- instead of aiming for twice that figure, which he says is typical. Less selective schools would be able to meet that figure without affirmative action, he argues.
But University of Michigan's Prof. Lempert says the study makes a number of unreasonable assumptions. Without affirmative action, many African-Americans wouldn't attend law school at all, he says. The study assumes that black applicants would merely go to a less selective school, but Prof. Lempert says many would be unable or unwilling to go to such schools because they might be so far away that the students wouldn't even consider them.
He also notes that, although there is a correlation between grades and bar passage, many other reasons explain blacks' poor performance on the test. These, he said, include a documented "stereotype threat," the tendency of minority groups to conform to negative stereotypes about their abilities.
Prof. Sander says he found no data to support Prof. Lempert's critique. James Lindgren, a law professor at Northwestern University, who is reviewing the same data, also found nothing suspect in the study. Now that the Supreme Court has accepted the legality of affirmative action, Prof. Lindgren says, the study might help "the debate move into a more fruitful and nuanced discussion about whom it helps and whom it hurts."
The new study's conclusions contrast sharply with a prominent study of affirmative action chronicled in the 1998 book "Shape of the River," by former Harvard University President Derek Bok and former Princeton University President William Bowen. Their research, based on voluminous data from selective colleges, concluded that racial preferences were enormously beneficial to African-Americans, who went on to earn unusually high numbers of professional and graduate degrees and achieve success in business and other endeavors.
Christopher Edley Jr., dean of the law school at the University of California, Berkeley, says that, even if Prof. Sander's findings are correct, he would suggest taking measures to improve African-American student performance, rather than scrap affirmative action. Prof. Edley also says the study gave insufficient weight to the academic benefits of diversity, for which there is "universal celebration" on his campus.
At UCLA, Prof. Sander has been at the center of the debate over diversity. In 1997, after a voter initiative banned affirmative action in California, Prof. Sander helped design and implement a preferential formula to help socioeconomically disadvantaged applicants. But the school turned to other methods after that system failed to achieve enough racial diversity to satisfy some faculty.