Mounting empirical research shows that race-preferential admissions policies are doing more harm than good. Instead of increasing the numbers of African Americans entering high-status careers, these policies reduce those numbers relative to what we would have had if colleges and universities had followed race-neutral policies. We have fewer African-American scientists, physicians, and engineers and likely fewer lawyers and college professors. If, as the evidence indicates, the effects of race-preferential admissions policies are exactly the opposite of what was originally intended, it is difficult to understand why anyone would wish to support them rather than adhere to the principle of color blindness.
One of the consequences of widespread race-preferential admissions policies is that talented minority students end up distributed among colleges and universities in patterns that are very different from those of their white and Asian counterparts. When the schools that are highest on the academic ladder relax their admissions policies in order to admit more underrepresented minority students, schools one rung down must do likewise or they will have far fewer underrepresented minority students than they would have had under a general color-blind admissions policy. The problem is thus passed on to the schools another rung down, which respond similarly. As a result, students from underrepresented minorities today are concentrated at the bottom of the distribution of entering academic credentials at most selective colleges and universities.
The problem is not that no academically gifted African-American students are seeking admission to college and universities. The nation is fortunate to have many. But there are not enough at the very top tiers to satisfy the demand, and efforts to change that have had a pernicious effect on admissions up and down the academic pecking order, creating a serious credentials gap at every competitive level.
Unfortunately, a student whose entering academic credentials are well below those of the average student in a particular school will likely earn grades to match. Even former Princeton President William G. Bowen and former Harvard President Derek Bok, who were pioneers in formulating affirmative action policies, admit that the credentials gap has serious consequences. In The Shape of the River, they wrote, “College grades [for beneficiaries of affirmative action] present a … sobering picture…. The grades earned by African-American students at the [elite schools we studied] often reflect their struggles to succeed academically in highly competitive academic settings.”
Why is it not better to get bad grades at a top school than better grades at a school that is one or two rungs down from the top? Everyone knows that a good student can get in over his head if he is placed in a classroom with students whose level of academic preparation is much higher than his own. He can end up demoralized and learn less than he would have been capable of otherwise. Such a student, through no fault of his own, has been “mismatched.” He may give up on tough but rewarding majors in science and engineering and opt for soft majors that are less likely to lead to high-status careers.
The empirical evidence demonstrates that this is exactly what is happening: Beneficiaries of race-preferential admissions are, on average, less successful than similarly credentialed students who attend colleges and universities where those credentials put them in the middle or top of the class. Overall, race-preferential admissions policies as practiced today are hurting, not helping, when it comes to jump-starting the careers of preference recipients.
I have no doubt that those who originally conceived of race-preferential admissions policies nearly 50 years ago were acting in good faith. By lowering admissions standards for African-American and Hispanic students at elite colleges and universities, they hoped to increase the number of minority students on campus and ultimately to promote their integration into high-status careers.
The real conflict over race-preferential admissions policies has not been about good or bad faith or whether we should aspire to be a society in which members of racial minorities are fully integrated into the mainstream. There is no question that we should. The conflict is about the means. Should we allow the principle of color blindness under the law to be sacrificed in the hope that in the long run, it will help us become a society of equal opportunity?
No less a liberal icon than California Supreme Court Justice Stanley Mosk warned of the risks associated with such temporary compromises with principle when, writing for the California Supreme Court in Bakke v. UC Regents (1976), he held racially discriminatory admissions policies to be unconstitutional:
To uphold the [argument for race-preferential admissions] would call for the sacrifice of principle for the sake of dubious expediency and would represent a retreat in the struggle to assure that each man and woman shall be judged on the basis of individual merit alone, a struggle which has only lately achieved success in removing legal barriers to racial equality.
Justice Mosk understood something basic about race discrimination. Throughout history, the temptation to engage in it has almost always come packaged with a justification that many found appealing at the time. When the country has succumbed to that temptation, however, it has almost always come to regret it. By 1976, when Mosk was writing, we owed it to ourselves to be more skeptical.
The Bakke case was brought by Allan Bakke, the son of a mailman and a schoolteacher—hardly a scion of wealth and privilege. After serving in Vietnam as a medic, he had twice applied to the University of California at Davis Medical School for admission. “More than anything else in the world, I want to study medicine,” he wrote in his personal statement, and his actions proved it was true. He was rejected both times, however, under circumstances that pointed strongly to his race.
At the time, UC–Davis Medical School had a two-track admissions system. The first 84 out of 100 seats in the class were given to the most qualified applicants regardless of their race, ethnicity, or other disadvantage. Bakke just missed making the cut. The remaining 16 seats were reserved for the disadvantaged, but in practice, “disadvantaged” always meant members of racial minorities. Bakke could have been the fatherless son of an illiterate washerwoman and it would not have mattered: Because he was white, he did not qualify.
It is worth pointing out that at UC–Davis Medical School, race was no mere tiebreaker in otherwise close cases. Bakke had a college grade point average (GPA) of 3.46 and an undergraduate science GPA of 3.44 (back before grade inflation’s long march through the academy) as well as a commendable record of volunteer emergency room service at a local hospital, where he frequently worked late into the night with victims of car accidents and street fights. By contrast, the average “disadvantaged track” admittee in 1973 had a college GPA of 2.88 and an undergraduate science GPA of 2.62.
The gap in standardized test scores was also wide. Bakke’s Medical College Admission Test (MCAT) scores put him at the 97th percentile (Science); 96th percentile (Verbal); 94th percentile (Quantitative); and 72nd percentile (General Information). On the other hand, the average “disadvantaged track” admittee in 1973 had MCAT scores in the 35th percentile (Science); 46th percentile (Verbal); 24th percentile (Quantitative); and 33rd percentile (General Information).
Mosk’s decision for the California Supreme Court was instantly denounced by supporters of race preferences. Hundreds of demonstrators gathered beneath his office window to demand its reversal. Thousands rallied elsewhere. When visiting local campuses, Mosk would routinely find himself greeted by picketers and hecklers.
When UC–Davis, in a conciliatory gesture, invited Mosk to give the law school commencement address in 1978, students insisted that he decline the honor. When he accepted anyway, a quarter of the graduating students walked out. But Mosk was undaunted. “Judges in California cannot be intimidated,” he stated firmly. “Lawsuits are won and lost in courtrooms, not on the streets.”
At the time, Mosk would probably have laughed to hear his insistence on color blindness characterized as “conservative.” Up to then, he had been accused far more frequently of leaning too far to the left. But whatever his political persuasion, Mosk had been a stalwart ally of the civil rights movement from its beginning. At a time when it was not yet fashionable to do so, he had quit fraternal organizations to protest their refusal to admit African-American members. As California’s Attorney General, he established the office’s civil rights division and banned the Professional Golfers Association from using state golf courses until such time as it agreed to admit African-American members.
Far from seeing a contradiction between his support for the civil rights movement and his opposition to the “minority friendly” admissions policies in Bakke, Mosk viewed them as one and the same. His opposition to racial discrimination was a matter of principle, and he was unwilling to sacrifice that principle for what he called the “dubious” practical gains promised by supporters of race-based preferences.
Mosk’s vision of civil rights did not prevail. Two years later, his opinion was superseded by the U.S. Supreme Court’s decision in Regents of the University of California v. Bakke. In that fractured decision, four justices would have affirmed Mosk’s decision, while four would have accorded public universities and other governmental bodies virtually unlimited power to discriminate in favor of racial groups that they perceived to be disadvantaged.
Justice Lewis Powell, Jr., whose lone opinion is now much admired by advocates of race-preferential admissions policies, purported to take a middle path. His opinion took the position that race preferences are permissible, but only as a method of capturing the benefits of a diverse student body for all students. At the same time, Powell’s opinion decided the case for Mr. Bakke apparently on the ground that the inflexibility of the admissions policy disproved its claim to be so motivated.
In the years following Bakke, race-preferential admissions policies mushroomed on college and university campuses. As they grew, millions of Americans, correctly or incorrectly, came to view themselves as beneficiaries. More important, a thriving diversity and remedial education bureaucracy was established to administer the policies and deal with the consequences. Rightly or wrongly, many of those who are so employed believe their jobs to be in jeopardy whenever race-preferential admissions policies are debated. By 2003, when the Supreme Court had the opportunity to address the issue again, it essentially adopted Powell’s position.
Thus, if Mosk was right that the benefits promised by race-preferential admissions were “dubious,” the mistake will be difficult to correct at this late date. It is not just the iron rule of bureaucracy at work today—that first and foremost, bureaucracies work to preserve themselves. Many distinguished citizens—university presidents, philanthropists, judges, and legislators—have built their reputations on their support for race-based admissions. Their jobs may not be at stake, but their sense of accomplishment almost certainly is. Overcoming that is not easy.
But if anything can cause good-faith supporters to stop and reconsider, it is the mounting empirical research showing that race-preferential admissions policies are doing more harm than good even for their intended beneficiaries. If this research is right, we now have fewer, not more, African-American physicians, scientists, and engineers than we would have had if colleges and universities had followed race-neutral policies. We have fewer college professors too and likely fewer lawyers. Ironically, preferential treatment has made it more difficult for talented African-American and Hispanic students to enter high-prestige careers.
No one should want to support race-preferential admissions policies if their effects are exactly the opposite of what was hoped for. Even if the consequences of race preferences turned out to be simply a wash—neither increasing nor decreasing the number of African-American and Hispanic professionals—it is difficult to understand why anyone would wish to support them rather than adhere to the principle of color blindness.
How can it be that affirmative action actually reduces the number of minority professionals graduating from colleges and universities? One of the consequences of widespread race-preferential admissions policies is that talented minority students end up distributed among colleges and universities in patterns that are very different from those of their white and Asian counterparts. When the schools that are highest on the academic ladder relax their admissions policies in order to admit more underrepresented minority students, schools one rung down must do likewise or they will have far fewer underrepresented minority students than they would have had under a general color-blind admissions policy. The problem is thus passed on to the schools another rung down, which respond similarly. As a result, students from underrepresented minorities today are concentrated at the bottom of the distribution of entering academic credentials at most selective colleges and universities.
The problem is not that no academically gifted African-American or Hispanic students are seeking admission to colleges and universities. The nation is fortunate to have many. But there are not enough at the very top tiers to satisfy the demand, and efforts to change that have had a pernicious effect on admissions up and down the academic pecking order, creating a serious credentials gap at every competitive level.
Unfortunately, a student whose entering academic credentials are well below those of the average student in a particular school will likely earn grades to match. The reason is simple: Entering academic credentials matter. While some students will outperform their academic credentials, just as some students will underperform theirs, most students perform in the range that their entering credentials suggest. Anyone who claims differently is engaging in wishful thinking at students’ expense.
Some of the best available data on this point are for law schools. In elite law schools, 51.6 percent of African-American law students had first-year GPAs in the bottom 10 percent of their class as opposed to only 5.6 percent of white students. Nearly identical gaps existed at law schools at all levels (with the exception of historically minority schools). At mid-range public law schools, the median African-American student’s first-year grades corresponded to the 5th percentile among white students; for mid-range private schools, the median for African-American students corresponded to the 7th percentile among white students.
Overall, with disappointingly few exceptions, African-American students were grouped toward the bottom of their classes. Moreover, contrary to popular belief, the gap in grades did not close as students continued through law school. Instead, by graduation, it became wider.
I am not aware of anyone who disputes these figures. Even strong supporters of racial preferences have conceded that the problem is “real and serious” and that “the average black law student’s grades are startlingly low.”
William G. Bowen and Derek Bok, authors of The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions, are long-time advocates of race-based admissions policies. As former presidents of Princeton and Harvard, respectively, they can fairly be said to have been among those who invented race-based admissions. Nevertheless, they candidly admit that the credentials gap has serious consequences: “College grades [for beneficiaries of affirmative action] present a … sobering picture,” they wrote. “The grades earned by African-American students at the [elite schools we studied] often reflect their struggles to succeed academically in highly competitive academic settings.”
Why are poor grades a problem? Why is it not better to get bad grades at a top school than better grades at a school that is one or two rungs down from the top? Everyone knows that a good student can get in over his head if he is placed in a classroom with students whose level of academic preparation is much higher than his own. He can end up learning less than he would have been capable of otherwise. Such a student, through no fault of his own, has been “mismatched.”
I have every confidence, for example, that I could learn basic physics, despite the fact that I have never taken a course in it and my mathematics skills are a little rusty. On the other hand, if you were to throw me into the Basic Physics course at the California Institute of Technology with many of the very best science students in the world, I would be lost and likely to learn little if anything. I would be mismatched. On a good day I might make a few lame jokes about how law professors just are not geeky enough to do physics; on a bad day, I might even get a little prickly about it. But it is unlikely that I would come out of that class as competent in the basic principles of physics as I would have in a less high-powered setting.
What is remarkable is that anyone thought that recipients of affirmative action would somehow be immune to this phenomenon, which subtly or not so subtly affects us all at one time or another.
Sometimes the problem is that a less than stellar performance leads to a loss of enthusiasm for academic pursuits and causes students to put their energies into other endeavors like athletics, social life, or campus politics. Everyone needs a niche. If, for the beneficiaries of racial preferences, excelling at academics in their peer group seems out of reach, they will look to stand out in other ways. Alas, the other ways available to them are usually less effective at promoting their integration into high-prestige careers and thus into mainstream society.
Whatever the mechanism, as I will detail below, the evidence is getting very nearly overwhelming at this point: Race-preferential admissions policies as practiced today are hurting, not helping, when it comes to jump-starting the careers of preference recipients. But let’s take the evidence a step at a time.
Why Minority Students Flee Science and Engineering
When Americans want to make the point that something is not overwhelmingly difficult, we say, “Hey, this isn’t rocket science.” In doing so, we acknowledge that mastering real rocket science—and by extension other hard sciences and engineering—really is tough. As one Yale University student complained to The Wall Street Journal, “In other classes, if you do the work, you’ll get an A…. In science, it just doesn’t work that way.”
Many students who start out majoring in science or engineering eventually switch to something easier. Some drop out of school altogether, and a few even flunk out. It should surprise no one that those who fail to attain their goal of a science or engineering degree are disproportionately students whose entering academic credentials put them toward the bottom of their college class. Not all stereotypes about science and engineering students are accurate, but the notion that they tend to be highly credentialed and hardworking is largely on target. They have to be.
What does surprise some, however, is this: Several unrebutted empirical studies have now demonstrated that part of the effect is relative. An aspiring science or engineering major who attends a school where her entering academic credentials put her in the middle or at the top of her class is more likely to succeed than an otherwise identical student attending a more elite school where those same credentials place her toward the bottom of the class. Put differently, an aspiring science or engineering major would be smart to attend a school where her entering credentials compare favorably with those of her classmates.
Part of the reason may be that science and engineering are ruthlessly cumulative. A student who has difficulty with the first chapter in the calculus textbook is apt to have difficulty with later chapters and with subsequent courses in the mathematics curriculum. By contrast, an English literature student who simply fails to read the Chaucer assignment is not necessarily at a serious disadvantage when it comes to reading and understanding Virginia Woolf. Since quitting science and engineering is easy—ordinarily, all one has to do is switch majors—the attrition rate is quite high. By senior year, there are significantly fewer science and engineering majors than there were freshmen initially interested in those majors. The pool of potential physicians, engineers, and scientists thus shrinks.
Beneficiaries of affirmative action are by no means the only ones who need to take note. This should be of serious concern to college athletes, children of alumni, and other special admittees. But as a nation concerned with integrating more African Americans and Hispanics into leadership positions, we need to be especially attentive to the implications for affirmative action.
A few years ago, a friend of mine took his very talented high school senior daughter to visit some of the nation’s top science-specialty colleges and universities, including the California Institute of Technology, the Massachusetts Institute of Technology, and Rice University. Several of these institutions tried to show off their commitment to diversity by having an attractive and personable minority student conduct the campus tour. But a pattern emerged among the tour guides: In the course of the tour, these students would reveal that while they had started out intending to major in science or engineering, they had switched to a softer major.
These are the schools attended by the nation’s rocket scientists—both literally and figuratively. The ones that get in or almost get in, whether as beneficiaries of affirmative action or not, are all remarkable talents. But there are differences in the level of academic preparation between those who were admitted strictly on their academic credentials and those who needed a special preference. The former, no matter what their race, often have been living and breathing science or engineering since they were in elementary school. Those who need a preference to get in are very talented, but they are not quite the caliber of those who did not.
These beneficiaries of preferential treatment nevertheless had as their first choice the study of science and engineering. That is why they chose one of these schools in the first place. All of them could have succeeded in that ambition just about anywhere else, but misguided admissions policies caused them unwittingly to attend the only schools on Planet Earth where they would instead come out communications majors. This waste of scientific talent did not have to be.
African-American college graduates age 35 and under are only 36 percent as likely as their white counterparts to hold a bachelor’s degree in science or engineering. They are 15 percent as likely to have a PhD in those areas. The figures for Hispanics are not quite as lopsided, but they are close. Hispanic college graduates age 35 and under are only 41 percent as likely as their white counterparts to hold a bachelor’s degree in science or engineering. The corresponding figure for PhDs is 26 percent.
All other things being equal, one might expect a minority whose members are less likely to be native English speakers to major in science and engineering more commonly than native English speakers, but this does not seem to be the case with Hispanics. On the other hand, Asian college graduates are more than four and a half times more likely to have a bachelor’s degree in science or engineering and a little over seven times more likely to have a PhD in those areas than white graduates.
If we were discussing fairly narrow fields—like acoustical engineering or Russian literature or golf—this lack of representation would be of no great moment. There is no segment of the labor force that proportionately reflects the nation’s demographic profile. Physicians are disproportionately ethnic Chinese, Cuban, Subcontinent Indian, or Jewish. Hispanic jockeys dominate horse racing; and as a result of the efforts of Hollywood actress Tippi Hedren, ethnic Vietnamese dominate the manicure business. The wine industry has more than its proportional share of Italian Americans. Even within professions, disproportionality is the rule, not the exception. Among lawyers, litigators are disproportionately Irish American. Among physicians, radiologists are disproportionately Subcontinent Indian.
Science and engineering, however, are not narrow fields. Obtaining an initial degree in science or engineering is the gateway to a large number of respected professions and occupations, from aviation inspector to zoologist. These fields represent a significant portion of the world’s most rewarding jobs. If African Americans and Hispanics are facing significant impediments to entering these growing fields, that is a situation that calls for attention.
Lack of investment in a solution does not appear to be the problem. Concern over underrepresentation in science and engineering is not new. On November 13, 1992, more than two decades ago, the popular magazine Science issued a special news report entitled “Minorities in Science.” In it, the editors lamented:
For 20 years, science has been wrestling with “the pipeline problem”: how to keep minorities from turning off the obstacle-strewn path to careers in science, mathematics, and engineering. Thousands of programs have been started since the late 1960s to bring diversity to the scientific work force. But their results have been dismal….
By 1992, the National Science Foundation had already spent over $1.5 billion on programs designed to increase the number of minorities in science or engineering. Officials at the National Institutes of Health estimated that they had pumped an additional $675 million into the system. Uncounted state, local, foundation, and industry programs contributed much more.
Nevertheless, the consensus of opinion was that much of the money had been spent unwisely. In the earliest days of affirmative action, “colleges took any person of color who wanted to become an engineer, regardless of their background,” said Mary Perry Smith, a former Oakland schoolteacher and founder of California’s Mathematics, Engineering, Science Achievement (MESA) program, which promotes minority student participation in those fields. “They tried to turn students who barely knew algebra into engineers and it was a total failure.”
Things have gotten better. According to its website, MESA, founded in 1970, serves “educationally disadvantaged students” and “to the extent possible by law, emphasizes participation” “from groups with low rates of eligibility for four-year colleges.” By targeting students in high school and middle school, MESA tries to cultivate student interest in science and engineering and make sure that students get the experience they need to succeed in these disciplines in college. Programs like MESA require one-on-one contact with students, and they appear to have enjoyed a measure of success.
Even considering the modest success of programs like MESA, however, most of the news of the past 40 years has been disappointing. In 1992, Dr. Luther Williams, then the assistant director of education and human resources at the National Science Foundation, was exasperated enough to declare, “The country cannot repeat the experiment of the last 20 years.” Williams, who later went on to become provost of Tuskegee University, a historically black university with a reputation for emphasizing a science and engineering curriculum, was blunt: Those vast expenditures were “an incredible waste of financial and human resources.”
Williams may have sounded more pessimistic than he intended. Billions of dollars had been spent, and much of it produced nothing, but some of the programs that have encouraged minorities in science and engineering—like MESA—may well have been worth it. There is no reason to abandon successful efforts. The real question is whether we can do better, and the answer to that question is an optimistic yes. All we have to do is stop shooting ourselves in the collective foot.
First of all, the problem with the underrepresentation of African Americans, Hispanics, and American Indians in science and engineering is not a lack of interest on the part of the minority students. Some have assumed the opposite: that a lack of role models would cause a lack of interest among African Americans and Hispanics. It is true that minority students are less likely to have a friend or family member in science or engineering, but that does not seem to be the root of the problem. Study after study has found, if anything, that African-American and Hispanic students are slightly more interested in pursuing science and engineering degrees than white students are.
For example, Professors Alexander W. Astin and Helen S. Astin of UCLA’s Higher Education Research Institute examined a sample of 27,065 students enrolling as freshmen at 388 four-year colleges in 1985. They found that the rate of initial interest in majoring in a biological science, a physical science, or engineering was, in descending order, 52.6 percent for Asians, 35.7 percent for Chicanos, 34.5 percent for American Indians, 34.2 percent for African Americans, and 27.3 percent for whites.
These findings are consistent with numerous other efforts to study the issue. Every time, it comes out the same: African Americans and other racial minorities are more interested in science and engineering than whites are. There may well be a lack-of-interest problem among Americans in science and engineering that needs fixing, but if there is, it is more with college-bound whites than with African Americans and Hispanics. If one wants to understand the root of the underrepresentation problem, one must look elsewhere.
Some researchers have done just that. They have shown that the problem comes a little further down the pipeline, with attrition. While African Americans and Hispanics have higher rates of initial interest than whites, they are less likely to follow through with that interest. Somewhere in college, the intention to graduate with a degree in science or engineering withers and dies. In one study of elite colleges and universities, for example, 70 percent of Asians persisted in their ambition, while 61 percent of whites, 55 percent of Hispanics, and 34 percent of blacks did.
Some observers have been inclined to ask, “What accounts for disproportionate minority attrition?” But the right question is, “What accounts for student attrition from science and engineering in general?” Once that question is answered, the question about disproportionate minority attrition essentially answers itself.
The usual culprit is lagging entering science credentials like math SAT score, the grades received for high school courses in mathematics and science, and the number of such courses taken, all of which are strongly correlated with persistence in science. The better one’s entering science credentials, both in the absolute sense and in comparison to one’s classmates, the more likely it is that a student with an initial interest in science and engineering will persevere and ultimately succeed in earning a degree.
This is a double whammy for African-American and Hispanic students as a group. While there are many individual students of all races with excellent entering science credentials, the average for African Americans and Hispanics as groups is below that for the nation as a whole. Until this situation changes, students from those groups are all but certain to have higher than average attrition rates from science and engineering. That is bad enough by itself, but as a result of race-preferential admissions policies, as well intentioned as these policies are, those same groups are concentrated toward the bottom of selective colleges and universities. That constitutes a completely unnecessary second hit.
Improving the absolute entering science credentials of underrepresented minority students is imperative, but the painstaking work and progress has been slow. On the other hand, improving those credentials in comparison to those of other students at the same school is easy and would significantly increase the number of underrepresented minority students who graduate with science and engineering degrees. All we have to do is stop encouraging students to enroll in schools at which they needed a preference to be admitted.
The earliest of the studies showing mismatch to be part of the problem, The Role of Ethnicity in Choosing and Leaving Science in Highly Selective Institutions, was published in 1996 by a team of scholars led by Dartmouth psychologist Rogers Elliott. It found that the single most important cause of black attrition from science at the selective institutions studied was the “relatively low preparation of black aspirants to science in these schools.” The authors were careful to use the word “relatively.” For successful students, it was not just entering credentials demonstrating highly developed ability at science that mattered, but comparatively high credentials. A student who attended a school at which his math SAT score was in the top third of his class was much more likely to follow through with an ambition to earn a degree in science or engineering than was a student with the same score who attended a school at which that score was in the bottom third of the class. The problem for minority students was that, as a result of affirmative action, being in the top third of the class was relatively rare.
For some, this is counterintuitive. The more prestigious the school, they believe, the more adept it should be at graduating future physicians, scientists, and engineers, no matter what their entering credentials. But instructors everywhere must pitch the material they teach to a particular level. They can pitch to the top of the class, to the middle, or to the bottom, but they cannot do all three at the same time.
At elite colleges and universities, pitching to the bottom of the class is uncommon, especially in the science and engineering departments. The whole point of these institutions is to teach to the top. That is the reason that highly credentialed students, who may have been underchallenged in high school, are willing to travel thousands of miles and incur significant debt to attend them. If these institutions were to abandon that practice and resolve to teach to the bottom of the class, there would be no good reason for them to exist.
Elliott and his co-authors cited the extraordinary record of historically black colleges and universities (HBCUs), which graduate far more than their share of black engineering and science majors, as further support for their findings. With only 20 percent of total African-American enrollment, these schools had been producing 40 percent of the African-American students who graduate with natural science degrees. Those students frequently went on to earn PhDs from mainstream universities. The National Science Foundation has reported, for example, that of the approximately 700 African Americans who earned a doctorate in science or engineering between 1986 and 1988, 29 percent earned their undergraduate degree from an HBCU. For biologists, the figure was 42 percent, and for engineers, it was 36 percent. This success has continued into more recent years.
There is nothing magical about HBCUs. It is just that unlike at other colleges and universities, credentials gaps are not an issue at the historically black institutions. One faculty member at a historically black school—North Carolina Central University’s Walter Pattillo, Jr.—told Science magazine in 1992: “The way we see it, the majority schools are wasting large numbers of good students. They have black students with admissions statistics [that are] very high, tops. But these students wind up majoring in sociology or recreation or get wiped out altogether.”
A more recent study by University of Virginia psychologists Frederick Smyth and John McArdle (now at the University of Southern California) confirmed Elliott’s findings, and the effects were not subtle. In Ethnic and Gender Differences in Science Graduation at Selective Colleges with Implications for Admissions Policy and College Choice, Smyth and McArdle found that among a sample of underrepresented minority students at 23 universities who intended to major in science, mathematics, or engineering, 45 percent more of the women and 35 percent more of the men would have succeeded in attaining their goals if they had attended schools where their entering credentials had been about average.
Those figures, upon reflection, are stunning. If, in just the time it would take a high school senior to graduate from college with a degree in science or engineering, we could add as success stories 45 percent more women and 35 percent more men at these schools, that would be a breakthrough of epic proportions. Half of that would still be huge. A tenth would at least be a step in the right direction.
Another study—this one by UCLA law professor Richard Sander and UCLA statistician Roger Bolus—pulled data from nine University of California campuses. The authors came to a similar conclusion. “Minority attrition in science is a very real problem,” they wrote, “and the evidence in this paper suggests that ‘negative mismatch’ probably plays a role in it.” Their multiple approaches to the data yielded consistent results:
[S]tudents with credentials more than one standard deviation below their science peers at college are about half as likely to end up with science bachelor degrees, compared with similar students attending schools where their credentials are much closer to, or above, the mean credentials of their peers.
All three of these studies were highlighted in a report entitled Encouraging Minority Students to Pursue Science, Technology, Engineering and Math Careers, issued by the U.S. Commission on Civil Rights in 2010. The researchers’ conclusions have not been challenged, much less successfully rebutted. Nevertheless, they have been ignored by colleges and universities.
In fact, they have been worse than ignored. In the most recent Supreme Court case to challenge the constitutionality of race-preferential admissions policies—Fisher v. University of Texas (2013)—the University of Texas argued that it needed to use race-preferential admissions in order to ensure racial diversity not only in the student body as a whole, but also in science classes (as well as other classes and programs). But UT’s preferential treatment of minority students was likely aggravating rather than alleviating the problem of underrepresentation in science. The more colleges and universities engage in race preference, the fewer the African Americans, Hispanics, and American Indians who will graduate with such degrees.
The evidence continues to accumulate. Recently, Duke University economists Peter Arcidiacono and Esteban Aucejo and Duke University sociologist Ken Spenner found evidence supporting the mismatch thesis when researching the major choices of undergraduates enrolled at Duke. In their article What Happens After Enrollment? An Analysis of the Time Path of Racial Differences in GPA and Major Choice, they found that black undergraduates were much less likely to persist with an entering goal of majoring in engineering, the natural sciences, or economics than white students were. Approximately 54 percent of black males switched out of these majors, while only 8 percent of white males did.
Once again, the problem was not lack of interest in science and engineering among black students: Before starting at Duke, more black students than whites indicated an initial interest in majoring in these subjects. Instead, the differences in attrition were best explained by entering academic credentials.
These authors also helped to dispel the common belief that beneficiaries of affirmative action catch up after their freshman years with their better-credentialed fellow students. What happens instead is that many transfer to majors where the academic competition is less intense and where students are graded on a more lenient curve. Their GPAs increase, but their standing relative to other students taking the same courses does not.
Again, the authors show that this effect is by no means confined to beneficiaries of affirmative action. White children and grandchildren of alumni who receive legacy preferences have the same experience, earning lower grades than white non-legacies at the end of their first year. While the gap narrows over time, it is only because legacy students also shift away from the natural sciences, engineering, and economics and toward the humanities and social sciences.
This helped the authors to respond to the argument that underrepresented minority students abandon science and engineering because they have no role models there or because they are somehow made to feel unwelcome. It is exceedingly unlikely that anti-legacy bias, lack of legacy role models on the faculty, or any other argument commonly advanced to explain racial disparities in science explains the legacies’ collective drift toward softer majors. If it is the wrong explanation for legacies, it is overwhelmingly likely to be the wrong explanation for underrepresented minorities as well.
The study created a firestorm at Duke, where the administration, instead of taking the research to heart, focused on pacifying indignant students, alumni, and faculty members who felt insulted by the results. In an open letter to the campus responding to demands that the university condemn the study, Duke provost Peter Lange and other administrators stated that they “understand how the conclusions of the research paper can be interpreted in ways that reinforce negative stereotypes.” They assured students that there are no easy fields of study at Duke and took the position that insofar as the mammoth problem identified in the study exists, it could easily be solved through student counseling and a few tweaks to the science curriculum.
Evidently, business will remain as usual at Duke. Potential affirmative-action recruits with an interest in science and engineering will continue to be told that Duke is the school for them. They will not be told that their chances of success in their chosen fields would be greater at Ohio State or, for that matter, at the University of North Carolina at Chapel Hill. Nor will they be told that if they switch majors to disciplines like African and African-American studies, art history, English, sociology, and women’s studies, they are less likely to enjoy lucrative careers or indeed to get jobs at all. In securities law, this would qualify as actionable fraud. In higher education, it is considered virtuous.
Should Colleges and Universities Have Predicted the Mismatch Problem?
Everyone seems to want to enroll in the most highly ranked college or university that will take them—with U.S. News & World Report as the most important, if not the sole, arbiter of rank. This kind of fierce competition for admission can be a good thing. At many institutions, it has helped to create a rich intellectual climate. Even students who are ultimately denied admission to their first-choice school often benefit from the academic skills they developed in making the attempt.
In some cases, however, the competition is over the top. Parents fight to enroll their children in the “right” kindergarten in the belief that it will be a pipeline into the “right” college or university. Business Week has reported that some parents are willing to fork over as much as $40,000 for advice from so-called experts on getting their son or daughter into an Ivy League school.
It is unlikely that these anxious parents and their nervous college-bound offspring know about the research conducted a few decades ago by University of Chicago sociologist James A. Davis. Outside the field of sociology, it seems largely to have escaped notice. Published in 1966 under the title The Campus as a Frog Pond: An Application of the Theory of Relative Deprivation to Career Decisions of College Men, it might have given them reason for pause. It is not obvious that a student who enrolls in a school that he got into only because of some special dispensation thereby gains a career advantage.
Davis set about to study which students are most likely to enter high-prestige careers like law and medicine. More specifically, he wanted to know who is more likely to pursue a high-prestige career, the student who earns low grades at a highly competitive school or the otherwise identical student who attends a somewhat less competitive school and hence gets higher marks. Looking at data collected by the National Opinion Research Center on 35,000 students from 135 colleges and universities, Davis found that college grades were more strongly correlated with the decision to enter a high-prestige career than was the quality of institution.
Davis was not saying that the best route to a high-prestige career is to graduate at the top of the least competitive school one can find. The local community college is unlikely to be the best option for the most academically talented students. Both the quality of the institution a student attends and his grades are positively correlated with the likelihood that the student will enter a high-prestige career. Davis simply found that grades mattered somewhat more. All other things being equal, students were less likely to pursue a high-prestige profession if they had attended a highly competitive school at which they received low grades than they were if they had attended a somewhat less competitive school and received correspondingly better grades.
As Davis himself recognized, these findings tend to “challenge the notion that getting into the ‘best possible’ school is the most efficient route to occupational mobility.” In some cases at least, the added self-confidence one enjoys as a result of being the big frog in a small frog pond may outweigh whatever advantages an elite education in a more glamorous frog pond can offer. Davis therefore offered the following words of advice: “Counselors and parents might well consider the drawbacks as well as the advantages of sending a boy to a ‘fine’ college, if, when doing so, it is fairly certain he will end up in the bottom ranks of his graduating class.”
Whatever Davis was thinking about when he published The Campus as a Frog Pond, it was unlikely to have been affirmative action. The year was 1966, and though there were already a few who advocated extra efforts to recruit African-American students, the term “affirmative action” was not yet a household word, much less the euphemism for the preferential treatment for minority applicants that became so thoroughly institutionalized by the late 1970s. Still, the connection between his work and race-preferential admissions policies should have been obvious.
It is unclear why Davis’s conclusions received so little attention outside the world of sociology. Perhaps it was because it was in no one’s clear interest to publicize them. Top schools had every reason to ignore them. They need students to form the bottom of their classes, and if hearing about Davis’s frog-pond effect should discourage some students from enrolling, these schools would just have to reach deeper into their applicant pool to find students who were not so discouraged.
Schools one rung down faced the same problem—on both ends of their recruitment efforts. If they tried to recruit students away from a top school by arguing the frog-pond effect, they would risk getting the word out to their potential recruits for the bottom of their classes. For those students, the message would be: You are better off at a school somewhere on the third rung. In the end, second-rung schools would have succeeded only in making it clear that they are indeed second-rung. For the most part, only schools at the very bottom of the academic ladder have an incentive to argue the frog-pond effect, and these unselective schools (of which there are many) rarely invest heavily in recruitment. They let students come to them.
Perhaps a more important reason that colleges and universities did not draw a connection between Davis’s findings and affirmative action was simply the temper of the times. It is difficult to overstate the sense of urgency that some of the earliest advocates of race-preferential affirmative action programs had. They did not view themselves as initiating an admissions policy that would remain in place for nearly half a century. Their race-preferential admissions were a frantic response to the civil unrest that had swept through Watts in 1965; Atlanta, Baltimore, Chicago, Cleveland, and New York in 1966; Detroit and Newark in 1967; and especially cities across the country in 1968. Whatever else the late 1960s might have been, they were not a time of slow and deliberate reflection.
Very early college and university affirmative action programs, several of which were financed by the Rockefeller Foundation, focused on recruiting African-American students from the inner city. Criminal records were not a barrier to admission; in fact, leadership in a street gang was sometimes considered a plus. Academic standards were not just relaxed, they were sometimes ignored altogether.
In at least one documented case, a good academic record was counted against an African-American applicant for admission. Thomas Sowell recounted a case in which a young African-American woman, who had apparently just missed being admitted to Cornell University’s regular program, was denied entry to Cornell on the ground that despite her humble background as the daughter of laundry workers, her academic record was too good for Cornell’s special program. The internal report on her read:
[H]er cultural and educational background does not indicate deprivation to the extent necessary for qualification as a disadvantaged … student. In spite of the fact that both her parents are laundry workers, she has been adequately motivated by them to a point where she has achieved academic success and some degree of cultural sophistication.
At the time, many elites thought—in retrospect, too optimistically—that students from the most disadvantaged circumstances had overcome so much to get where they were that once admitted to a selective college or university, they would outperform better-credentialed students born to more prosperous parents. They were thus unlike the legacy students that Davis had been studying. Despite the lack of early emphasis on academic achievement in their lives, they would flourish in high-powered academic settings and go on to become pillars of the community—or so it was hoped.
It is easy to lampoon the radical chic notion that elite colleges and universities can come together with street gang members to improve society, but I will leave that task to others. Suffice it to say that while the prediction of academic success was made in good faith and was not clearly and provably wrong given what was known at the time, it turned out to be wishful thinking. There were, of course, a few success stories, but the most significant consequence of this naïve form of affirmative action was to make the schools most dedicated to it—from UCLA to Cornell—even more chaotic than other American campuses of the era.
A reassessment of the policy was clearly in order. Given the fanfare with which the programs had been launched, however, it would have been difficult to abandon them entirely. The alternative was to reconfigure them in hopes of making them effective. Two changes were made.
First, there was a consensus that rather than scour the inner cities looking for students who did not fit their usual academic profile, colleges and universities would scour the high schools for African-American students who did. If they turned out to be from disadvantaged circumstances, that would be an added bonus. On the other hand, if they turned out to be middle-class or even upper-middle-class suburbanites from Bethesda, New Rochelle, or Wilmette, so be it. The goal was no longer broadly to prevent unrest in the inner city—something that elite colleges and universities had always been ill-equipped to do. The focus was simply on the slow, deliberate expansion of the African-American middle and upper middle classes by graduating more African-American students ready to enter high-status careers.
Second, there was agreement that to make race-preferential admissions work, minority outreach and recruitment efforts would need to be stepped up, remedial education would need to be significantly expanded, and administrators would have to be hired whose job it was to oversee these efforts.
The new race-preferential admissions policies were thus not the starry-eyed policies of the late 1960s. Diehard advocates of the more radical approach were sorely disappointed, and they were not entirely off base: When an African-American son of a brain surgeon can get a preference over the Vietnamese-American daughter of a dishwasher, it is hard to be sure that one is really making the world better.
At no point, however, did college and university administrators think seriously about whether their policies could be counterproductive, as Davis’s Frog Pond research indicated. A few commentators like UCLA economics professor Thomas Sowell and Yale law professor Clyde Summers specifically warned them. The reaction of colleges and universities to these warnings was not, however, to soberly assess the evidence, but rather to keep any evidence under wraps. Scholars could not study the issue if they did not have access to data.
Law Schools Engage in Secrecy
Secrecy has always been an important element of any race-preferential admissions policy. Consider, for example, the case of Timothy Maguire. In early 1991, this third-year law student and former Peace Corps volunteer took a job as a part-time file clerk in the Georgetown University Law Center admissions office, figuring it would be an easy way to earn some extra money without cutting too much into his study time. He never dreamed of the trouble he was getting into: In a few weeks, he would be vilified by the Georgetown University faculty, his fellow students, and the editorial pages of some of the nation’s major newspapers. Over the course of the next year and a half, he would be fighting for his right to practice law.
Maguire had heard many times that affirmative action was all about putting a gentle thumb on the scale in favor of minority applicants. It was a tiebreaker, nothing more. But the admissions files he was seeing told a different story. The gap in academic credentials between white and African-American admittees was very stark.
Maguire therefore decided to do a test. Taking what he regarded as a random sample of the files, he did a few back-of-the-envelope calculations and found that the average white student accepted to the law school had an LSAT score of 43 and an undergraduate grade point average of 3.7. The average accepted black student, on the other hand, had an LSAT score of 36 and an undergraduate grade point average of 3.2. To put those figures in perspective, an LSAT score of 43 was just shy of the top 5 percent among those who took the exam. A score of 36, on the other hand, was only in the top 30 percent.
To be sure, students in the top 30 percent are good students, all fully capable of becoming, in one form or another, successful lawyers. They are more typical, however, of law schools like Drexel University, the University of Louisville, and the University of Nevada than they are of highly competitive, academically oriented schools like Georgetown, which holds itself out as a law school for outstanding students, not just for good students. Rightly or wrongly, the admissions office would not have given a white student with an LSAT score of 36 and an undergraduate GPA of 3.2 a second glance: Of over 100 white admittees sampled, not a single one had an LSAT score under 39.
Maguire published his findings in the student-run newspaper, the Georgetown Law Weekly, along with an essay critical of the school’s separate and less-demanding admissions standards for African Americans. He called the credentials of white and African-American admittees “dramatically unequal” and argued that Georgetown was being dishonest in failing to inform its students about the gap. The failure to disclose the credentials problem made enrolling a racially diverse class seem easy, he wrote. It made it appear that if the school did not have enough African-American students, it was because the school just did not care enough to reach out and encourage them to attend.
The campus erupted. Within days, the Black Law Students Association had filed a formal complaint demanding that Maguire be expelled. The next week, 600 Georgetown law students crammed into a lecture hall for an emotionally charged “town meeting,” while others spilled over into another classroom to watch the event on closed-circuit television.
Maguire had obviously touched a raw nerve. “The central issue is racism,” said a white female second-year law student. “I think the article is assaultive. People were injured. I think this kind of speech is outrageous.” Another student called the article an attempt to chill Georgetown’s “commitment to legal education for African Americans.”
While they were not numerous, a few students defended Maguire, who had decided not to attend the meeting himself. Third-year student Adam Magazine told the crowd that affirmative action had been “swept under the rug before, because white people were afraid to say anything and black people felt threatened.” Maguire was thus performing a service of sorts by getting the issue into the open. “Are we really going to say that because we don’t like what [Maguire] said we are going to throw him out of school?” he asked.
Magazine’s question may have been intended as rhetorical, but it was not treated as such. Shouts of “Yes!” rang up from some in the crowd. Tempers were flaring.
All during the meeting, Dean Judith Areen blandly assured the crowd that Maguire had gotten his facts wrong. The gap was not what he suggested it is—or so she implied. She steadfastly refused, however, to provide the actual figures or to provide any details whatsoever as to Georgetown’s actual affirmative action policy. This caused even supporters some unease. “Affirmative action is a good thing,” said a male first-year student. “But so much of what we’re saying today is in the dark, because the administration won’t give us the facts.”
As the story leaked out to the mainstream press, Dean Areen kept up her assurances that Maguire had gotten it wrong and that his “random sample” was not random at all. Maguire had evidently taken a large pile of admissions files representing students who had been admitted, but he had no way of knowing whether that pile was really a random sample.
Following her lead, The New York Times editorialized that Maguire was “without the scarcest hint that he knows what a random sample is.” Even if true, this was harsher than necessary for the Times editors to make their point. This was a law student they were attacking, a law student writing in a school newspaper who was no doubt astonished to find himself suddenly thrust onto the national stage. Surely, the editors of one of the world’s leading newspapers could pick on someone their own size.
But the editors showed no mercy. They went on to ridicule Maguire, accusing him of writing “pretentiously” and stating that “he has learned very little” and “hasn’t a clue about the broad purpose of a great law school.” The raw nerve that Maguire had touched evidently had dendrites reaching into the editorial offices of The New York Times. If anyone had ever had any doubt about whether it was safe to talk critically about affirmative action in public, that doubt would be erased by this incident.
As it turned out, Maguire was more likely understating the gap than overstating it. Shortly after the controversy exploded, an internal memorandum surfaced, the authenticity of which was confirmed to The Washington Post by university sources. Authored by Georgetown admissions director Andrew Cornblatt, it stated that the median LSAT score for full-time African-American students at Georgetown in 1989 “increased to 33, up from 32 last year and 30 two years ago.” According to the memorandum, the median for the entering class as a whole in 1989 was 42, which would make Maguire’s figure of 43 for white students in all likelihood on target.
A score of 33 was not quite in the 56th percentile, and a score of 32 was in the 52nd percentile. Both scores were thus quite ordinary among test takers nationally. Among actual law students, however, they were below average, since low scorers on the LSAT frequently do not attend any law school. A score of 30 was in the 41st percentile—below the average for test takers and much below the average for actual law students.
In the midst of the fury over his article, Maguire was charged by the school with violating confidentiality (although he had disclosed no individual information and had published only the kind of information found in the Cornblatt memorandum and reported to the American Bar Association and U.S. News & World Report). The case was to be tried by a panel of two professors and one student, and possible sanctions included expulsion. Fortunately for Maguire, his lawyers were able to negotiate a settlement with Georgetown. Rather than being expelled, he was issued a letter of reprimand and allowed to graduate.
This did not sit well with everyone. An unhappy faculty group accused the administration of “a panicked reaction” that failed to “celebrate and vigorously defend” affirmative action. On graduation day, approximately a month and a half after publication of the offending article, a number of students wore green ribbons to protest the settlement. A few carried placards like the one that read, “Ethics … A Meaningless Word.” A group of about 10 African-American alumni, calling themselves the Concerned Black Law Alumni of the Georgetown University Law Center, staged their own muted protest.
Unfortunately for Maguire, the story did not end there. After graduation, Maguire took and passed the New Jersey bar examination. But he was found to be unfit for the practice of law by a panel of the Committee on Character. Only after protracted litigation was he finally able to practice the profession for which he had trained.
After such a story, no one should be surprised to learn that only a small number of intrepid souls are willing to bring the facts about affirmative action to public light or even to debate it. As far as I have been able to determine, Maguire himself never commented again on the subject in public.
Large Gaps in Credentials Continue Today
Twenty-five years after Bakke, twin cases against the University of Michigan—one against its College of Literature, Science, and the Arts and the other against its law school—came before the Supreme Court, challenging the constitutionality of race-preferential admissions policies. The resulting decisions more or less followed Justice Powell’s reasoning in Bakke.
Of the two, Grutter v. Bollinger (2003) was by far the more significant. In that 5–4 decision, diversity was held to be a compelling purpose (or, more accurately, the majority deferred to the University’s judgment that diversity was a compelling purpose). The law school was thus given a green light to continue with the very large preferences it had been according underrepresented minority students.
The 6–3 decision in Gratz v. Bollinger (2003) was at least superficially in plaintiff Jennifer Gratz’s favor. The echo of Bakke was clear. The Court held that like UC–Davis Medical School’s special admissions program in Bakke, the college’s point-driven admissions program was too inflexible. The strong implication was that this inflexibility evidenced a motivation other than the desire to capture the benefits of a diverse student body for all students.
Under that point system, the college added 20 points to the academic index of all African-American applicants for admission—the equivalent of an entire letter grade in high school GPA, all other things being equal. An African-American applicant with a 3.0 GPA (straight Bs) would be admitted before a white or Asian student with a 3.99 (just shy of straight As) if their records were otherwise the same. By contrast, the children and grandchildren of alumni received only one point—still one point too many for a state institution, but at least a small enough advantage to make mismatch an unlikely problem. The legacy preference was thus a true tie-breaker.
Just as in Bakke, colleges and universities interpreted the Court’s objection to the admissions procedures in Gratz to be essentially cosmetic. A few changes were necessary, but the college itself was not hindered from granting large preferences to underrepresented minority groups. Instead, the size of those preferences grew in the aftermath of the Grutter and Gratz decisions. By 2005, the distribution of high school GPAs among admittees to the college was as follows:
For combined SAT (Math + Verbal), it was as follows:
In 2005, an Asian applicant stood essentially a zero percent chance of being admitted with a GPA and SAT scores equal to the median GPA and SAT of black admittees. A white student stood only a 1 percent chance.
This is consistent with the practices at other colleges and universities for which firm evidence is available, including schools that are somewhat less selective than the University of Michigan. The Center for Equal Opportunity is an organization that studies this issue. Its research shows, for example, that in 2006, for an in-state black male applicant with a GPA and ACT scores equal to those in the 25th percentile for black admittees, the probability of being accepted was 83 percent at Miami University of Ohio. In contrast, the probability of an in-state white male applicant with the same GPA and ACT scores being accepted was only 32 percent. In 2006, the median high school GPA for black admittees at MU was 3.45; for whites, it was 3.73. The median ACT score was 23 for blacks and 27 for whites. For those students who submitted SATs, the median combined (Math + Verbal) score was 1060 for blacks and 1220 for whites.
Some of the best-documented evidence of race-based admissions policies is again at law schools, nearly all of which give preferential treatment in admissions to members of underrepresented minorities, in part because the schools’ accreditation can depend on it. As a result, the average black student has an academic index that is more than two standard deviations below that of his average white classmate.
As the Center for Equal Opportunity’s data indicate, this is by no means confined to the most selective law schools:
- At the University of Nebraska College of Law, a white resident with credentials like those of the average black admittee in 2006 had just a 1 percent chance of admission. By contrast, a black resident with those credentials had a 79 percent chance and an otherwise similar black non-resident had a 35 percent chance. The probability that a Hispanic resident with the credentials of the average black admittee would be admitted was 43 percent. A similar Hispanic non-resident had a 10 percent chance. As a result, the average African-American admittee had an LSAT score in the 29th percentile of test takers nationally, with the average white applicant in the 75th percentile and the average Hispanic somewhere in between.
- At Arizona State University, a white resident with the same credentials as the average black admittee in 2006 stood only a 2 percent chance of admission, while his black resident counterpart had a 96 percent chance and his black non-resident counterpart had a 92 percent chance. The corresponding figures for Hispanics were 64 percent (resident) and 46 percent (non-resident). African-American admittees thus had an LSAT median in the 68th percentile of test takers nationally, with Hispanics at the 72nd percentile and whites at the 86th percentile.
Similarly, the Center for Equal Opportunity studied the admissions policies for six public medical schools: the Medical College of Georgia, SUNY–Brooklyn College of Medicine, University of Michigan Medical School, Michigan State College of Human Medicine, University of Oklahoma College of Medicine, and University of Washington School of Medicine. It found that all of them engage in significant preferential treatment for African-American students in admission. All but the Medical College of Georgia give preferential treatment to Hispanics, and the University of Oklahoma College of Medicine also gives preferential treatment to American–Indian applicants.
The Shape of Misinformation
“Wait!” one might ask. “Didn’t that famous book by William G. Bowen and Derek Bok—The Shape of the River—disprove the mismatch hypothesis back in the 1990s?”
No, it didn’t.
In September 1998, when The Shape of the River made it into print, race-preferential admissions policies were perceived by their supporters to be under siege. California’s Proposition 209, which prohibited such policies, had been adopted in 1996. Washington State had a similar initiative coming up in a few weeks on the November ballot. Those who backed these initiatives hoped to bring similar initiatives to other states soon.
The Shape of the River was the Establishment’s response. Generously funded by the Andrew W. Mellon Foundation and other foundations with long records of support for race-based admissions, it was authored by two former Ivy League university presidents who viewed themselves as the cavalry arriving just in the nick of time to rescue affirmative action.
As it turned out, the Washington State initiative passed anyway, but there is a good chance The Shape of the River did slow the progress of similar initiatives in other states in the years that followed. The book was cited by Justice Sandra Day O’Connor as support for the Court’s opinion in Grutter v. Bollinger. Despite its failure to stop the Washington State initiative, it has been very influential.
Judging from the attention the book received in the media, the Mellon Foundation publicists must have been working around the clock. Few books have been published with as much fanfare.
- According to a fawning editorial in The New York Times, the book’s “findings provide a strong rationale for opposing current efforts to demolish race-sensitive policies in colleges.” The Shape of the River “flatly refutes” the arguments of its opponents.
- Newsweek’s Ellis Cose commented that the book was the “most ambitious study to date of the effects of affirmative action in higher education” and “an important corrective to conservative propaganda.”
- The Atlanta Journal-Constitution’s Cynthia Tucker, casting aside the usually prudent presumption of good faith on the part of one’s fellow Americans, wrote that “the Bok–Bowen report will not convince the die-hard opponents of affirmative action—those who cling to a narrow view of ‘merit’ or who, for whatever reason, do not wish to see the pool of opportunity expanded.” “Nevertheless,” she added, “the report is a welcome antidote to those constant critics who cannot see evidence of achievement beyond mere test scores and grade point averages.”
Some of the commentary was aimed specifically at the issue of mismatch. Harvard University sociologist Nathan Glazer argued in The Washington Post that it was now “clear” that worries over mismatch were misplaced. The Pittsburgh Post-Gazette editorialized that the notion that race-based admissions policies have hurt African-American students “is one that can be dismissed.” But were they right?
In some ways, the authors of The Shape of the River were in an enviable position. Their team of researchers had unrestricted access to the Mellon Foundation’s massive “College and Beyond” database, which contains records for about 80,000 students, all of whom had enrolled at one of 28 colleges and universities mostly in 1976 or 1989. Any social scientist studying education issues would consider himself lucky to have such a resource (the Mellon Foundation, however, denied access to Harvard University professor Stephan Thernstrom and former Vice Chair of the U.S. Commission on Civil Rights Abigail Thernstrom, whose view of affirmative action is more skeptical than that of Bowen and Bok). It includes information about each student’s academic credentials and performance as well as his subsequent career. It thus provides an invaluable opportunity to study many aspects of affirmative action. As a result, the book positively bristles with charts and graphs.
The database is perhaps especially useful for studying the mismatch hypothesis. Bowen and Bok were able to group the participating schools into three tiers—Tier 1 (e.g., Princeton and Yale); Tier 2 (e.g., Northwestern and Vanderbilt); and Tier 3 (e.g., Pennsylvania State and the University of North Carolina)—and attempted to draw conclusions about the effects of attending one tier as opposed to another.
Most of the book is devoted to demonstrating uncontroversial facts. The authors effectively show, for example, that there is indeed a large gap between the academic credentials of African-American and white college applicants. If top-tier schools were to terminate racial preferences for African Americans but otherwise maintain their current admissions policies, the result would be entering classes that have significantly fewer African-American students at highly selective schools.
The authors also show that substituting class-based preferences for race-based preferences would not lead to classes of the same racial composition. Since only 17 percent of all low-income students graduating from high school who scored in the top 10 percent on standardized tests are black or Hispanic, the primary beneficiaries of class-based affirmative action would be poor Asians or poor whites. (This latter argument is, of course, a double-edged sword for the authors. It explains why preferences for low-income students are not a substitute for preferences for underrepresented racial minority students, but it also reveals that many of the beneficiaries of those preferences are members of the middle class and sometimes higher.)
Then they get to the more controversial parts. After conceding—as they must—that race-based admissions policies lead to poor grades for beneficiaries of preference, the authors assert that African-American students who attend schools at which they are substantially mismatched are nevertheless better off with lower grades at a more elite school than they would have been with the higher grades at the less elite school.
In attempting to prove this, Bowen and Bok show that the mean earnings of black men with SAT scores of less than 1000 who attend what they identify as a Tier 1 school are higher than the earnings achieved by their counterparts at Tiers 2 and 3. It is on this basis that they claim the right to pooh-pooh the notion that black students “pay a penalty in life after college” for having attended schools at which they were mismatched. “On the contrary,” they conclude, “the black (and white) matriculants with academic credentials that were modest by the standards of these schools appear to have been well-advised to go to the most selective schools in which they were admitted.”
If the authors intended to disprove the mismatch hypothesis, however, they made at least two serious errors in the way they conducted their study.
First, they took account only of students’ SAT scores and not of other academic credentials such as high school rank. One cannot assume that a student with a combined SAT score of 1200 at Princeton is the equivalent of a student with the same score at Pennsylvania State University. There is an excellent chance that the first student has a substantially better high school GPA as well as other distinctions in his favor. That is why he is at Princeton and the other student is at Penn State.
Comparing students with the same SAT scores and finding that students at the more elite school have higher graduation rates and higher post-graduation earnings, even though they appear to be mismatched at the more elite school, is an unfair comparison. It is overwhelmingly likely that the typical such student attending the less elite school has a less elite high school record to match.
This is not the kind of error that two former Ivy League presidents should have made. Much of their universities’ public relations strategy is built around the theme that they routinely reject applicants with perfect or near-perfect SATs who do not otherwise measure up to their standards. Prior to publication of The Shape of the River, they appeared genuinely to believe that students with identical SAT scores are not necessarily equivalent (and they are quite right about that). Suddenly, they were asking readers to forget their earlier position.
Second, The Shape of the River did not even compare students with the same SAT scores. It compared students within broad bands. While banding cannot always be avoided, it is what statisticians do when they set out to muddy the waters, and it seems to have done so in this case. The authors divided students into five categories: those with SATs below 1000, those with scores from 1000 to 1099, those with scores from 1100 to 1199, those with scores from 1200 to 1299, and those with scores at 1300 and above. They then compared post-graduation earnings of minority students in each of these broad bands at each of the three tiers.
Such an approach has superficial appeal until one understands that the composition of each band differs markedly depending upon which tier is being examined.
- At Princeton and Yale, students with SAT scores below 1000 are a rarity. Those that exist are not just likely to have particularly fine high school grades; they are likely to have SAT scores that are only a tiny bit below 1000. At Penn State or the University of North Carolina, however, such students are below the median but not so far below it as to be a rarity. Consequently, the median student in the “below 1000 band” for the Tier 1 schools will almost certainly have an SAT score quite a bit above that of the median student in the “below 1000” band in the Tier 3 schools.
- At the opposite end of the spectrum, the same is true. The median student in the “1300 and above” band attending a school like Princeton and Yale almost certainly has a higher SAT score than the median student in that band attending Pennsylvania State or the University of North Carolina.
- In the intermediate categories—1000 to 1099, 1100 to 1199, and 1200 to 1299—results will be skewed in one direction or another depending on the level of selectivity of the tier being examined.
One cannot compare bands of this kind without building in a bias that makes the earnings enjoyed by the graduates of the more highly ranked schools seem more attributable to the school than they really are. It is easy for schools to have higher graduation rates and higher earnings when they simply have more academically gifted students as their raw material. If they did not have higher graduation rates and earnings, it would mean that something was wrong.
Given those two serious methodological flaws, The Shape of the River’s results seem preordained. One would have to expect the data to show that the policies that these two university presidents had been following all along were justified and that mismatch was not a problem.
But lo and behold, even given these flaws, evidence of mismatch comes peeking out from behind the charts.
For example, while Bowen and Bok are careful not to draw attention to it, their own figures show that one category of black men—those with SAT scores between 1000 and 1099—earn more if they attend Tier 2 schools rather than Tier 1 schools. Similarly, their figures show that black women with SAT scores between 1100 and 1199 earn more if they stay away from Tier 1 schools and attend Tier 2 schools instead. These students’ results show just the opposite of what Bowen and Bok claim they have proven.
Meanwhile, buried in Appendix D.5.4 and Appendix D.5.5 is a bombshell to which the authors seem wholly oblivious. The charts in these appendices contain a more sophisticated analysis that is barely mentioned in the text.
These charts attempt to tease out how various factors influence the subsequent earnings of African Americans who attended one of the 28 colleges or universities in the Mellon Foundation database. Included among these factors are several pre-college considerations: the socioeconomic status of the student’s family, SAT scores, and whether the student was in the top 10 percent of his high school class. Also included are several factors from the student’s college days: the selectivity of the college or university he attended; his major; whether his grades put him in the top third, middle third, or bottom third of the class; and whether he went on to earn a graduate or professional degree. The effect of each of these factors was measured.
The authors purport to show that, on average, attending a Tier 1 school rather than a Tier 3 school contributes to the income levels of both African-American men and African-American women. But something important appears just a few rows down: College grades generally contribute more. Again and again, through the different permutations of the analysis they conduct, the authors’ own figures show it.
Imagine two African-American males with the same SAT scores. Both were in the top 10 percent of their high school classes, and both come from middle-class families. Only their colleges are different. Bowen and Bok demonstrate convincingly that if the two have the same major in college and similar grades, the one who attended a Tier 1 school will earn about $17,365 more than the one who attended a Tier 3 school.
But what if they don’t have similar college grades? Bowen and Bok also show that if one of those students is in the top third of his college class and the other is in the bottom third of his, the former will earn an average of $34,089 more.
By the authors’ own calculations, therefore, it is better to be an African-American male at Penn State in the top third of the class than to be an African-American male at Princeton in the bottom third of the class. The increased earnings he gets from high grades are worth almost twice the increased earnings he gets from attending a Tier 1 school. And there’s more: The boost in earnings he would get for majoring in natural science as opposed to the humanities would be a whopping $49,537.
If one’s class rank and major were unrelated to the level of selectivity of one’s college, then it would be perfectly sensible for Bowen and Bok to celebrate the finding that, on average, black male students get an earnings boost from attending a Tier 1 school instead of a Tier 3 school. But they are not unrelated. For students who would not have been admitted but for racial preferences, the chances of earning grades in the top third of the class are exceedingly remote, and the chances of graduating with a degree in natural science are greatly reduced.
The only question is whether an African-American student who attends a Tier 1 school and winds up in the bottom third of the graduating class would likely have been in the top third of a Tier 3 school. The answer to that question, at least in many cases, is yes.
Consider, for example, an African-American male student with SAT scores of 1300 who just missed being in the top 10 percent of his high school class. He is a talented student by any ordinary measure. If he attends Pennsylvania State University, his SAT scores will put him exactly at the 75th percentile in the entering class of 2011 according to U.S. News & World Report. That would give him a very strong shot at earning grades in the top third. If he starts out intending to major in a natural science, there is an excellent chance that he will stick to it. If he enrolls at Princeton instead, his SAT scores will put him 90 points below the 25th percentile for that school, making it much more likely that his grades will be in the bottom third, possibly even the bottom third of the bottom third.
Similarly, figures in Appendix D.5.5 predict that the average African-American female with a combined SAT score of 1400 will earn more than $3,800 extra by attending the University of North Carolina if, as her SAT scores suggest she will, she graduates in the top third of her class rather than by attending Yale University if, as her scores suggest, she graduates in the bottom third. As an added bonus, if she has a desire to major in engineering, graduating with such a degree will earn her, on average, an additional $17,894 according to Bowen and Bok’s calculations. And she is much more likely to do that at UNC than at Yale.
Have Bowen and Bok by themselves proven that mismatch is hurting minority students? No, they have not, but they have gone further in that direction than they have in proving that is it not hurting them. When one combines their results with those of other researchers, a picture begins to emerge.
Another Mellon Foundation Project Finds that Affirmative Action–Induced Low Grades Hurt Minorities
In 2003, five years after The Shape of the River was published, another project of the Andrew W. Mellon Foundation came to fruition. This one, however, was not released with the same fanfare. In fact, the Mellon Foundation did everything but disown it.
The book was called Increasing Faculty Diversity: The Occupational Choices of High-Achieving Minority Students, and it was authored by Stephen Cole, professor of sociology at the State University of New York at Stony Brook, and Elinor Barber, a research associate in the provost’s office at Columbia University. Its purpose was to advise the Ivy League and other colleges and universities on how to increase the racial and ethnic diversity of their faculties. The original research for the project—which consisted of individual interviews with students, focus groups, and many thousands of questionnaires—was heavily underwritten by Mellon.
Cole and Barber’s study easily could have become just one more among the large number of reports on diversity funded by some hapless foundation with more money than it knows how to spend. It could have recited the usual clichés: Colleges and universities should make greater efforts to make minorities feel welcome; they should celebrate diversity; they should go the extra mile. Universities could wallpaper their administrative offices with such reports.
Instead, the authors actually tried to find an answer to the question of why so few high-achieving African Americans and Hispanics choose to go to graduate school with an eye to becoming college and university professors. Unlike many of their predecessors who have weighed in on that question, they declined to ignore the problem of affirmative action–induced low grades. As the authors put it:
[Many of the best-prepared African-American students] are admitted to schools where, on average, white students’ scores are substantially higher, exceeding those of African Americans by about 200 points or more. Not surprisingly, in this kind of competitive situation, African Americans get relatively low grades. It is a fact that in virtually all selective schools (colleges, law schools, medical schools, etc.) where racial preferences in admission is practiced, the majority of African American students end up in the lower quarter of their class….
As Cole and Barber acknowledge, that leads to problems:
It is not at all surprising that academic performance in college should turn out to be an important influence on the decision to select academia as a career. If a student is not academically successful and has not received rewards for his or her academic performance, it would make little sense for that student to think of spending the rest of his or her life in a job where “being good in school” is a prerequisite.
No surprise there: Young people tend to go into fields in which they perceive they will do well. Weak swimmers do not sign up for training as lifeguards, and the kind of person who perpetually burns the toast does not seek the training necessary to become a professional chef. At college, students get a sense of how good they would likely be as college professors from their grades. Students who are not toward the top of the class are less likely to think of academia as the best place to apply their talents. Cole and Barber, unlike their predecessors, were willing to say so.
One approach to that problem might be to try to tell minority students not to worry about their lackluster grades, that colleges and universities will rush to hire them anyway. Telling students, “We don’t really care how good you will be at this job; we want you anyway” is not, however, a good hiring strategy. Nobody with the long term in mind wants to enter a career when his comparative advantage is his skin color.
Cole and Barber put the matter more delicately, but they agreed. They cautioned against “highly visible affirmative action programs” that “may send a positive message” that “elite colleges and universities care strongly about having a diverse student body,” but that also may telegraph to minority students “that they were admitted to a selective school because of affirmative action programs and that they ‘don’t really belong.’”
Rather than recommend staying the course on affirmative action, Cole and Barber argued for a change in direction. Finding that “African American students who attend less selective schools are more likely than those who attend selective schools to persist with a freshman interest in academia,” they advised that:
Instead of recommending that minority students go to the most prestigious school they can get into, high school guidance counselors should recommend that each student go to a school where he or she is likely to do well academically. An HBCU may be such a school. Guidance counselors, in short, should try to reduce some of the lack of fit between the level of academic preparation of minority students and the schools where they enroll.
In some ways, The Shape of the River and Increasing Faculty Diversity are twins. They both contain a wealth of data, and in both books the data showed that affirmative action–induced low grades were creating a serious obstacle to minority achievement. The difference is that Cole and Barber understood what they had uncovered and reported it in the text of the book forthrightly. In The Shape of the River, the data were ignored or misinterpreted.
Unlike The Shape of the River, Increasing Faculty Diversity was barely mentioned at all in the press. The hardworking publicists for the Mellon Foundation were apparently on holiday. An exception to the media blackout was an article entitled “The Unintended Consequences of Affirmative Action” in the Chronicle of Higher Education, which reported on the efforts of the Mellon Foundation to distance itself from Stephen Cole (Dr. Barber had passed away while the book was in preparation).
“Researchers report the findings as they see them, and they may not be consistent with what we’d like to see or what we think are there,” Harriet Zuckerman, senior vice president of the Mellon Foundation, told the Chronicle. “The Mellon Foundation is not on this, just as it is not on other research we support.” Readers should be “cautious about putting much weight on certain findings,” she said.
Jeffrey H. Orleans, executive director of the Council for Ivy Group Presidents, which had furnished seed money for the project before the Mellon Foundation stepped in with major funding, reacted similarly: “There are a whole lot of data in here, and if one started out with an ideological position—whatever it was—you could find a whole lot to support that.”
Cole was obviously furious about the treatment he received. He told the Chronicle that he “wouldn’t get up and make a blanket statement that I’m opposed to affirmative action. But if you’re looking at this issue, affirmative action is contributing to the number of minority students getting lower grades, which seems to contribute to them selecting non–high achievement careers.” As for comments like Zuckerman’s and Orleans’s, he said, “I was trained at a time before social science became so politicized…. I believe that social science should be objective and value-free, and you should design a study to answer a question and what the answer is, that’s what it is.”
According to the Chronicle, Cole figured “that there is ‘no chance’ he’ll receive money again from the Mellon Foundation. ‘And I don’t care.’”
Where Are the Missing Black Lawyers?
At around the same time, UCLA law professor Richard Sander conducted his first study of mismatch (prior to the science and engineering study already discussed above). Unlike his work in science and engineering, his law school work received quite a lot of attention, in part because the student editors of the Stanford Law Review, which published the piece, solicited responses.
Evidently anticipating that some of the same accusations of racial bias hurled at Stanley Mosk, Timothy Maguire, and others might be thrown at him, Sander took the unusual (and somewhat awkward) step of including biographical information in his otherwise dry and scholarly article:
No writer can come to the subject of affirmative action without any biases, so let me disclose my own peculiar mix. I am white and I grew up in the conservative rural Midwest. But much of my adult career has revolved around issues of racial justice. Immediately after college, I worked as a community organizer on Chicago’s South Side. As a graduate student, I studied housing segregation and concluded that selective race-conscious strategies were critical, in most cities, to breaking up patterns of housing resegregation. In the 1990s, I cofounded a civil rights group that evolved into the principal enforcer (through litigation) of fair housing rights in Southern California. My son is bi-racial, part black and part white, and so the question of how nonwhites are treated and how they fare in higher education gives rise in me all the doubt and worries of a parent. As a young member of the UCLA School of Law faculty, I was deeply impressed by the remarkable diversity and sense of community the school fostered, and one of my first research efforts was an extensive and sympathetic analysis of academic support as a method of helping the beneficiaries of affirmative action succeed in law school.
Of course, even if Sander had been a lifelong opponent of race-based admissions, it is not clear why those facts would in themselves be sufficient to dismiss his findings out of hand. William G. Bowen, as president of Princeton University from 1972 to 1988, and Derek Bok, who presided over Harvard from 1971 to 1991, were pioneers in formulating affirmative action policies and therefore had every motivation to want to find that those policies had been successful. This is useful to know in evaluating their strongly pro-preference The Shape of the River, but it is not by itself reason to ignore their findings. Ultimately those findings must be addressed on their own terms. Unfortunately, however, Sander turned out to be right in believing that he would be viewed suspiciously for his work.
Sander’s work brought to public attention two undisputed facts:
- During the period for which data were available, the average African-American law student had an academic index that was more than two standard deviations below that of his average white classmate, and
- A majority of African-American law students during that period were in the bottom 10 percent of their law classes.
More important, he demonstrated that this disappointing law school performance was almost entirely the result of race-preferential admissions. When African-American and white law students with similar entering credentials competed against each other, they performed very nearly identically. Race-based admissions, therefore, were creating the illusion that African-American students were destined to do poorly in law schools at every level. The real problem was far less daunting: Fewer African-American students than anyone would prefer had the entering academic credentials necessary for admission on a color-blind basis to the most elite law schools, but many more would likely do well at schools a little farther down the pecking order—if they were attending those schools.
Just like supporters of race-based admissions at the undergraduate level, supporters of race-based admissions in law school often asserted that despite the increased likelihood of poor grades, minority students were better off accepting the benefit of a preference and graduating from the most prestigious law school willing to take them. Someone had to be at the bottom of every class; there was no real harm in its being minority students.
Up until Bowen and Bok’s misdirected effort in The Shape of the River, few advocates of race-preferential admissions at the undergraduate level made any effort to demonstrate that beneficiaries of preference were in fact better off than they would have been had they attended a school in which their entering credentials matched those of their peers. It was largely taken on faith. That is why the Sander study was so important. Unfortunately, the news that he was able to provide bore out the fear that mismatch is hurting African-American law students’ career prospects.
Sander noted two important effects of race-based admissions policies.
First, African-American students attending law schools failed or dropped out at much higher rates than white students: 19.3 percent vs. 8.2 percent. Overwhelmingly, this phenomenon was associated with poor performance and not financial hardship. Since many of these students who left law school would likely have performed better at a less competitive law school, they were in a very real sense victims of race-preferential admissions.
Second, among African Americans who graduated and took the bar, the proportion of those who passed on the first attempt was not just lower than that for whites; it was lower even when one controlled for academic index (a combination of LSAT and college GPA). For example, 71 percent of African Americans with an index of 400–460 failed the bar on their first effort, while only 52 percent of whites did. Similarly, 26 percent of African Americans with an index between 640 and 700 failed their first time, while only 13 percent of whites did.
Ultimately, only 45 percent of African Americans who entered law school passed the bar on their first attempt as opposed to over 78 percent of whites. Even after multiple attempts, only 57 percent of African Americans succeeded. The gap was thus never closed.
Something was clearly wrong. When African-American and white law students with similar academic credentials competed against each other at the same school, they earned about the same grades, and when African-American and white students with the same grades from the same tier school took the bar examination, they passed at the same rate. Yet African-American students as a whole had dramatically lower bar passage rates than white students with similar credentials. What could explain this?
As Sander pointed out, the most plausible answer was that they were not attending the same law schools. The white and Asian–American students were likely to be attending a school that takes things a little more slowly and spends more time on matters that are covered on the bar exam. They were learning while their minority peers were struggling at more elite schools, where the bar exam and doctrine are often mocked as things of little importance by faculty members.
Sander calculated that if law schools were to use color-blind admissions policies, fewer African-American law students would be admitted to law schools (3,182 vs. 3,706), but since those who were admitted would be attending schools where they had a substantial likelihood of doing well, fewer would fail or drop out (403 vs. 672). In the end, more would pass the bar on their first try (1,896 vs. 1,567), and more would eventually pass the bar (2,150 vs. 1,981) than under the current system of race-based admissions.
Sander’s findings had to be deeply troubling to any fair-minded supporter of race-preferential admissions, but unlike the mismatch literature in science and engineering, this was just a single study using a single database. For that reason, the U.S. Commission on Civil Rights, while impressed with the study, recommended in its 2007 report, Affirmative Action in American Law Schools, that more research be undertaken.
The stakes were obviously high. If Sander was right or even partly right, it may fairly be said that decades of race-preferential admissions at law schools were for nothing—or indeed worse than nothing. Just as Justice Mosk had warned, the principle of nondiscrimination would have been sacrificed “for the sake of dubious expediency,” a practical gain that never materialized.
It would have been nice if someone besides Sander had set out to find a database that would allow a second look at the issue of law school mismatch, but given the treatment that Sander and Timothy Maguire before him received, it is not surprising that researchers did not at first step forward. So Sander took the lead. At a minimum, what he needed was a state bar that kept records on the race of each person who had sat for the bar examination and how that test taker had performed on the examination along with his LSAT score and college GPA and the law school he attended. The state would also need to be a large one with a wide range of law schools.
Sander found such a state in California. Alone among the 50 states, California had all of the data he would need to retest the mismatch theory.
Sander assembled a team of researchers with diverse opinions about the mismatch issue and approached the State Bar of California for permission to examine the records. In a better world, this request might have been pro forma. The whole purpose of keeping such records is to allow useful research to be conducted on them, and the State Bar of California had allowed them to be used for education research in the past. The privacy of individuals would have to be protected, but that was hardly an insurmountable obstacle. Sander and his colleagues were not interested in identifying individuals. They just needed to work out a way to turn over the data that would ensure that everyone’s needs were accommodated.
At first, it seemed more than likely that he would have their cooperation. The professional staff with whom he was dealing seemed to understand the importance of the project and appeared prepared to recommend to the Board of Bar Examiners of the State Bar, whose final approval would be necessary, that the project be undertaken.
Then, slowly, a door began to close.
The first sign of trouble came from the Long Island–based Society of American Law Teachers. Calling itself “the largest membership organization of law teachers in the United States,” SALT sent a letter urging the State Bar of California not to cooperate with Sander.
SALT’s self-description, however, was a bit misleading. In fact, the only reason SALT can claim to be the largest membership organization of law professors in the country is that there is no general-interest association of law professors. The Association of American Law Schools, which functions in effect as the largest association of law professors, is in fact, as its name implies, an association of law schools. SALT is an activist group that leans so far to the left that it is in danger of falling over on its side. It is in no way representative of law professors, who, while tending very much toward the left side of the political spectrum, are not usually in SALT territory.
The SALT letter subtly threatened the State Bar of California with future litigation if it turned over the data. It was clear that not all of Sander’s critics had an interest in ensuring that he had access to newer data with which to test the mismatch theory.
Another letter came from an individual who had complained in the past that the Sander study should be discounted because it relied on data that were several years old. Now he was seeking to prevent Sander and his co-investigators from getting newer data, arguing, among other things, that bar examination scores are a poor “proxy for ‘student learning’” and that their disclosure “risks stigmatizing African-American attorneys regardless of how successful they may be in legal practice.” (Sander, of course, had not requested names or any information that would have allowed him to identify particular persons.)
Then the law school deans started to weigh in. They had heard—presumably from SALT—that research into mismatch was going to take place and that it could make them look bad.
Sander enlisted Gerald Reynolds, chairman of the U.S. Commission on Civil Rights, to fly to San Francisco and plead with the Committee of Bar Examiners on behalf of the project. He also got letters from several of the nation’s more iconoclastic former law school deans who happened to reside in California. But these efforts were unavailing.
It is something of an honor to be appointed to the Committee of Bar Examiners of the State Bar of California. It is not something lawyers undertake because they are looking to do battle with activist organizations like SALT or with law school deans. I find it unlikely that too many committee members were out to protect affirmative action even at the expense of its supposed beneficiaries, but perhaps the situation reached a point where cooperating with Sander would have been unpleasant for them, and it was certainly easier to say no than to say yes. So that is what they did.
Sander then requested the records pursuant to the California Public Records Act, California’s version of the Freedom of Information Act. When they were not forthcoming, he sued in California Superior Court. The case eventually reached the California Supreme Court, where it languished until December 19, 2013, when the Court ruled that the State Bar of California must cooperate with Sander and remanded the matter for a determination of the details. The story is still developing.
What Has Happened in States in Which Race-Preferential Admissions Have Been Outlawed?
We already have a good idea about what happens when race-preferential admissions are outlawed, since several jurisdictions, beginning with California in 1996, have adopted popular initiatives doing exactly that. The operative clause of California’s Proposition 209, which amended the state constitution, reads: “The State shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin.”
Similarly worded initiatives have been passed in Washington State (1998); Michigan (2006); Nebraska (2008); Arizona (2010); and Oklahoma (2012). Most recently, in 2014, the U.S. Supreme Court, in the curiously titled Schuette v. Coalition to Defend Affirmative, Integration and Immigrant Rights and Fight for Equality by Any Means Necessary, held these initiatives to be constitutional—a point that never should have been subject to doubt.
In California, the state for which we have the most data, Proposition 209 meant more underrepresented minority honor students and fewer underrepresented minority students with unacceptably low grades. It also meant higher graduation rates for underrepresented minority students.
Consider the case of the University of California at San Diego, a highly selective institution but not quite as selective as the University of California’s flagship campus at Berkeley. In 1997, prior to Proposition 209’s implementation, only one African-American student at UC–San Diego had a freshman-year GPA of 3.5 or better: a single African-American honor student in a freshman class of 3,268. In contrast, 20 percent of the white students on campus had such a GPA.
Failure rates were similarly skewed. Fully 15 percent of African-American students and 17 percent of American Indian students at UC–San Diego were in academic jeopardy (defined as a GPA of less than 2.0), while only 4 percent of white students were. Other underrepresented minority students hovered close to the line. Since UC–San Diego did not keep separate statistics for those minority students who needed a preference in order to be admitted and those who would have been admitted regardless of race, it is impossible to say exactly how high the failure rate was for preference beneficiaries in particular. Suffice it to say that it was high.
This was not because there were no other African-American students capable of doing honors work at UC–San Diego. The problem was that such students were often at Harvard, Stanford, or Berkeley, where often they were not receiving honors. Similarly, white and Asian students were not magically immune from failure. Those who were at high risk for failure, however, had not been admitted there in the first place. Instead, they were at less competitive campuses of the University of California—Davis, Irvine, Santa Cruz, or Riverside—where their performance was more likely to be acceptable or even better than acceptable.
Being toward the bottom of the class can be demoralizing for anyone, just as being toward the top of the class can be uplifting, but bad grades created by preferential admissions policies can be especially troubling. When a class consists of students with roughly the same level of academic preparation, the student at the bottom of the class ordinarily will be the one who did not try hard enough. At the end of the term, he may decide to change his priorities and study more. Or, wisely or foolishly, he may decide that his priorities were right and that he would rather excel at football, work part-time, or spend time with his friends and family rather than improve his academic performance.
In contrast, when a student is having difficulty not because he is not trying, but rather because his academic credentials are significantly lower than those of his classmates, the lesson learned is not that hard work pays off, but rather that it doesn’t. You can knock yourself out, but even when you do, you may only eke by. Sometimes, students in that position find themselves dependent on the kindness of a teacher to allow them to get by—another unfortunate lesson in a nation that prizes self-reliance and an independent spirit.
Race-preferential admissions policies add an additional wrinkle. If an individual student is doing poorly at a selective college, that does not necessarily mean that his sisters or brothers or cousins are likely to be doing poorly too. It does not mean that his friends at church are doing poorly. But it is one thing for an individual student to find himself toward the bottom of the class. It is quite another for an African-American student to find himself toward the bottom of the class and to find half of his African-American friends and acquaintances there too.
It is easy to develop a sour-grapes attitude under those circumstances: “It’s all politics,” or “getting good grades isn’t really a black thing.” Culture comes from shared experiences, and affirmative action had been giving too many California minority students the shared experience of feeling unsuccessful at academic pursuits.
Then came Proposition 209. Although initially held up in litigation, it went into effect in time to affect the undergraduate admissions decisions for the entering class of 1998.
The media coverage of the implementation seemed calculated to create the impression of an educational crisis. “Acceptance of Blacks, Latinos to UC Plunges,” one Los Angeles Times headline bellowed. For a few days, UC’s color-blind admissions policies, which had caused the share of Berkeley’s offers of admission to African Americans, Hispanics, and American Indians to go from 23.1 percent of the total offers to 10.4 percent, permeated the national airwaves.
The claim that Berkeley had become “lily white” was unfounded. In 1997, 58.6 percent of its freshman admissions went to students who had checked minority boxes—primarily blacks, American Indians, Asian Americans, and Hispanics. When Proposition 209 went into effect, the figure declined to 48.7 percent, but only a bare majority of seats went to whites. Asian Americans accounted for about 38 percent of the total.
Moreover, the non-Asian minority students who would have attended Berkeley in the past had not simply vanished. They had been accepted to somewhat less highly ranked campuses—often UCLA and UC–San Diego—based on their own academic records rather than their skin color. In turn, students who previously would have been admitted to UCLA or UC–San Diego on a preference were now being admitted to schools like Davis, Irvine, Santa Cruz, or Riverside that, while somewhat less competitive, were nevertheless part of the prestigious UC system, which (in theory at least) caters only to the top 12.5 percent of California’s high school graduates.
In fact, both UC–Riverside and UC–Santa Cruz posted impressive gains in minority admissions. At Riverside, for example, black and Latino student admissions shot up by 42 percent and 31 percent, respectively. Santa Cruz’s increases were less dazzling but nevertheless notable. UC–San Diego reported mixed results. Black enrollment there was down 19 percent, but the enrollment of some other “underrepresented” groups like Filipinos and Latinos was actually up by 10 percent and 23 percent, respectively.
At UC–San Diego, the performance of black students improved dramatically. No longer were African-American honor students a rarity. Instead, a full 20 percent of the African-American freshmen were able to boast a GPA of 3.5 or better after their first year. That was higher than the rate for Asians (16 percent) and extremely close to the rate for whites in the same year (22 percent). Suddenly, African-American students at UC–San Diego found themselves on a campus where achieving academic success could be considered normal. In contrast to the experience of African-American students in other places, nobody could accuse them of “acting white” or “acting Asian.” Striving for and attaining academic excellence has lost any hint of association with race.
The sudden collapse in the minority failure rate was perhaps even more impressive. Once racial preferences were eliminated, the difference between racial groups all but evaporated at UC–San Diego, with black and American Indian failure rates falling to 6 percent.
UC–San Diego’s internal academic performance report announced that while overall performance had remained static, “underrepresented students admitted to UCSD in 1998 substantially outperformed their 1997 counterparts,” and “the majority/minority performance gap observed in past studies was narrowed considerably.” But “narrowed” was an understatement. The report found that for the first time, there were “no substantial GPA differences based on race/ethnicity.” A discreet footnote makes it clear that the report’s author knew exactly how this happened: 1998 was the first year of color-blind admissions.
Granted, UC–San Diego had 12 fewer African-American freshmen in the first year of Proposition 209’s implementation, forced as it was to reject students who did not meet the academic standards of the rest of the class. But it also had seven fewer African-American students in academic jeopardy at the end of the first year. Meanwhile, those 12 students probably attended a school where their chances of success were greater.
The next few years were good ones. Proposition 209’s critics have been loath to admit it, but the big news following its implementation was skyrocketing minority graduation rates. As Richard Sander and Stuart Taylor, Jr., reported in their 2012 book Mismatch: How Affirmative Action Hurts Students It’s Intended to Help, and Why Universities Won’t Admit It:
Minority graduation rates rose rapidly in the years after Prop 209, and on-time (four-year) graduation rates rose even faster. For the six classes of black freshmen who entered UC schools in the years before race-neutrality (i.e., the freshman classes of 1992 through 1997), the overall four-year graduation rate was 22 percent. For the six years after Prop 209’s implementation the black four-year graduation rate was 38 percent. Thus, even though the number of black freshmen in the UC system fell almost 20 percent from 1997 to 1998, the number of black freshman who obtained their degrees in four years barely dipped for this class, and the entering class of 2000, four years later, produced a record number of blacks graduating on time. The increase in black six-year graduation was less dramatic (63 percent before and 71 percent after Prop 209) but still substantial.
Not all of this astonishing increase can be traced to Proposition 209, but Duke University researchers have found that about 20 percent of the overall increases in graduation rates of UC minority students can be. If Proposition 209 had been implemented with greater rigor, it would likely have contributed even more. In a world in which steps forward in education are rare and incremental when they occur at all, that is a stunning victory.
Moreover, the gains have not been limited to grade point averages and graduation rates. Between 1997 and 2003, the number of African-American and Hispanic students graduating with a degree in science or engineering rose by about 50 percent. Not unrelatedly, the number of African-American and Hispanic students majoring in ethnic studies and communications fell by 20 percent. Academic self-confidence was growing among minority students.
Note the Triple Crown: (1) College grade point averages of underrepresented minority students and (2) graduation rates of such students were improving at the same time that (3) they were increasingly majoring in science and engineering. Ordinarily, these three goals would be difficult to achieve at the same time. For example, grading curves are traditionally lower in science and engineering departments than they are in the rest of the university, so it is remarkable that grade point averages would be going up alongside increases in the number of science and engineering majors.
Now combine those victories with an increase in graduation rates. When graduation rates increase, it is generally because some weaker students, who might have dropped out in an earlier time, are managing to make it to the end. One would thus expect increasing graduation rates to have a depressive effect on grade point averages and/or on the proportion of students majoring in science and engineering. Instead, improvements were made in all three areas. It is as if Ford had come up with a new automobile that was both more luxurious and better on gasoline mileage—and cheaper too.
Sadly, the effect has probably not lasted—at least not entirely. The University of California at Berkeley, for example, has been developing techniques that allow it to get around Proposition 209 to the greatest extent possible while still enabling it to argue publicly that it is in compliance. Some of these techniques are more legally defensible than others. As a result, the performance gap may have begun to return. At least for a while, however, California was leading the way in demonstrating that the gap can be closed and that the campus culture of racial separation can be changed, and Proposition 209 is still likely to be having a positive effect.
The Road Not Taken
I began this essay with California Supreme Court Justice Stanley Mosk. When asked to abandon the principle of color-blindness in exchange for “dubious expediency,” he declined. The indications are that his doubts were prescient.
Mosk was never forgiven—not even in death—for his deviation from what had become liberal orthodoxy. In his 2001 New York Times obituary, he was accused of having a knack for anticipating and bending with political currents. For good or ill, nothing could be further from the truth. When it came to standing up for what he believed, Stanley Mosk was unbendable.
Despite Mosk’s earlier history of support for civil rights, it was Justice Lewis Powell, a mild-mannered Nixon appointee, who ended up beloved by the civil rights establishment. Unlike the irascible Mosk, however, the gentlemanly and accommodating Powell was an unlikely civil rights hero.
As chairman of the Richmond School Board between 1953 and 1961 and a member of the Virginia Board of Education between 1961 and 1969—the crucial years following the Supreme Court’s decision in Brown v. Board of Education (1954)—Powell was in a position to take a leading role in dismantling Jim Crow. But Powell, who later went on to be president of the American Bar Association, did not distinguish himself as an advocate of desegregation “with all deliberate speed.” As Jerome Karabel put it:
His own carefully worded assessment of his service in these positions was that it had taken place when the pace of desegregation had been “necessarily more measured than civil rights leaders would have liked.” But this was a rather generous interpretation of his role in the years after the Brown decision, for when Powell stepped down as chairman of the Richmond School Board in 1961, after eight years of service, only 2 of the city’s 23,000 black children attended school with white children. And during his two terms with the state Board of Education, Powell’s sympathetic but fair-minded biographer reports that “he never did any more than was necessary to facilitate desegregation … [and] never spoke out against foot dragging and gradualism.”
I mean no disrespect to Lewis Powell, who I believe was on balance a good jurist, but when the right thing to do was stand on principle in the face of demands for expedience, Stanley Mosk was the one to call on, not Powell. Powell’s reputation as a resolute supporter of civil rights is undeserved. A former ABA president, he could be described as a conciliator, or he could be described as a man who was disinclined to rock the boat, but the two descriptions of Powell’s temperament are one and the same. It is a virtue or vice depending upon the situation.
History does not disclose its alternatives. Still, it is difficult not to wonder about what might have been if Stanley Mosk had sat on the U.S. Supreme Court instead of Lewis Powell.
Gail Heriot is a Professor of Law at the University of San Diego and a member of the United States Commission on Civil Rights. The views and opinions expressed herein do not necessarily reflect those of the United States Commission on Civil Rights.