Does Harvard Discriminate?

Students for Fair Admissions is suing Harvard University for discriminating against Asian Americans in its admissions process.

Let’s be clear: Harvard does discriminate. When the admissions office must select approximately 2,000 admits from about 40,000 applicants, it had better discriminate! The question is: How does it do so? Harvard employs a “holistic admissions” process, which I describe below, to yield a class that will produce, as it explicitly states, “leaders in many disciplines who make a difference globally.” SFFA raises a fundamental question about whether the process is fair. Only using a very narrow conception of merit could the process be considered unfair.

Most of the attention to race issues in college admissions began during the civil rights movement and after President Lyndon Johnson issued his executive order on affirmative action. In this context, Asian Americans’ academic and occupational success led them to be labeled a “model minority” -- they were positioned as “proof” that nonwhite status was no impediment to academic brilliance and used as evidence that the U.S. stratification system was not aligned against all minorities. Needless to say, systematic variations of success among Asian Americans (Korean immigrants earn bachelor’s degrees at twice the rate of Vietnamese immigrants, for instance) puts a lie to those claims.

Nevertheless, in the late 1980s, responding to publicity that there was a quota for Asian Americans that allowed elite institutions to increase its overall ethnic diversity, the U.S. Department of Education’s Office of Civil Rights investigated whether Harvard (and UCLA) were discriminating against Asian Americans in its admissions process.

By claiming that Asian Americans are being evaluated differently for admission than are white applicants, opponents of affirmative action attempted to highlight that it’s not minority students who aren’t meritorious -- it’s those minorities, primarily African Americans and Latinx. It’s those minority students who don’t perform well on standardized tests, implicitly held as the only legitimate benchmark of merit. In now launching a similar suit, SFFA’s goal is the same: to ensure that groups that have experiences, skills, talents and dispositions that don’t align with these traditional conceptions of merit should not be admitted to Harvard and, by extension, to other selective institutions.

To the extent that an institution recognizes such diverse claims to merit, SFFA argues, it is engaging in illegal practices: it is discriminating, in this case, against Asian Americans. Somehow, SFFA believes that Harvard’s “fairness” should be focusing on traditional, narrow notions of merit, not on its mission of developing global leaders, let alone increasing opportunity for all. SFFA hasn’t recognized that “history is a graveyard of classes which have preferred caste privileges to leadership,” as sociologist Digby Baltzell famously wrote.

A major claim of the plaintiffs in the current case against Harvard is that Asian Americans are assigned lower “personal ratings” (one of four subjective overall ratings used in admissions decisions) than other groups and that this is why otherwise-qualified Asian Americans are denied admission. Yet this reflects a fundamental misunderstanding of Harvard’s (and many other selective institutions') practice of “holistic admissions.”

Based on my observations of admissions meetings (admittedly, many years ago, but the process is largely the same), I can say with confidence that no particular rating for a given applicant determines admission. In subcommittee meetings organized by geographical categories called dockets, the primary reader of a folder has the most say -- in part by assigning the academic, athletic, personal and extracurricular ratings based on their personal assessment rather than a formula. But a second reader’s comments, along with messages from athletic coaches, music professors and others, are equally important in framing the subcommittees’ conversations.

Not only is the preliminary decision to recommend for or against admission based on these conversations but, before the final decisions are made, the full admissions committee meets and applicants again move back and forth across the admit/reject line. The idea that a slightly lower personal rating -- even a systematically lower rating across applicants from the same group -- has a significant effect on these much more consequential processes is not defensible. There are many applicants who are admitted with a relatively low rating on one of the four subjective ratings.

In this process, the impact on admission of any given rating or attribute of an applicant varies based on everything else that is in the folder, all the applicants within the docket and, eventually, all the applicants in the pool. While these contextual elements should be clear from the various analyses (used as evidence in the lawsuit) of the current Harvard admissions process, it was certainly apparent from my analyses of the Harvard Class of 1984 applicants. For example, among those who averaged over 700 on standardized tests, over 70 percent of applicants from the two local dockets (Boston and Cambridge) were admitted compared to 43 percent from urban/suburban dockets near New York City and Philadelphia.

The admissions office makes decisions not only on the basis of the strengths of individual applicants, but just as importantly on how these strengths articulate with institutional needs: Does it need a left tackle to protect the quarterback’s blind side? Does it need a French horn player for the orchestra? Did the field hockey team graduate a large number of players last year? Is the dance or physics program particularly underpopulated? So, context is critical both in considering a given applicant and in making decisions about the overall composition of the class. And nonacademic criteria play an important role in this process.

Using nonacademic criteria in admission decisions was helpful to the powers that be at Harvard in slowing the “Jewish invasion” in the 1920s but is now critical in broadening and democratizing access for current students. We know from excellent data (from 1976 onward) that students of color who were admitted to highly selective institutions by taking race into account performed very well and graduated in high numbers, even when admitted with lower test scores. Further, they succeeded in graduate and professional schools and in the labor market. So, using broader conceptions of merit has been shown to be an effective way to build new leadership.

Were Harvard’s entering class selected by an SFFA admissions committee, would it be comprised of all valedictorians? All perfect scorers on the SAT? If it were, would the university be able to fulfill its mission? Would it be a place that would attract 40,000 applicants and yield four of five admits in pursuing its goal of developing global leaders?

The larger question is whether an approach narrowly focused on test scores would be fair to the students in our racially and socioeconomically segregated K-12 system. Harvard’s approach, which attempts to include diverse notions of excellence among its admits, has done quite well in increasing the number of students of color in its entering classes. Even with its efforts to reduce economic barriers to access (full grants are given to those deemed financially eligible), the Harvard entering classes are extremely well-off compared to the households of their high school mates. Perhaps Harvard can examine more closely the 3 percent of its class that comes from the top 0.1 percent of the income distribution (a 3,000 percent overrepresentation). Perhaps Harvard can also encourage more applications and admits from the bottom fifth of the income distribution and increase the percentage of its entering class from this income segment from its current less than one in 20 to, say, one in 10.

Surely, to attain both fairness and its goal of developing global leaders, Harvard should continue broadening the range of talents it privileges and diversifying the backgrounds of its students. Ignoring the history and current state of race/ethnic inequality and focusing instead on implementing snapshot notions of colorblindness -- SFFA’s goal -- is not the route to fairness. Fixing the inequality in K-12 education and broadening our conceptions of merit and diversity are a good place start on the journey to fair and equitable opportunity.