Harvard’s cult of personality

Later this month, the Supreme Court will hear oral arguments in the case of Students for Fair Admissions Inc. v. President and Fellows of Harvard College, which may decide the future of affirmative action. But when considering the narrower question of whether Harvard discriminates against Asians in admissions, it is just as important to examine the facts of the case in the court of public opinion as in the court of law. 

After all, this case has brought to light facts about college admissions at Harvard (an elite institution if there ever was one) that would otherwise be hidden from the public.

The evidence in the case implies that at least one of two disturbing conclusions is true. Harvard either discriminates against Asian applicants or, consistent with a pernicious stereotype, Asian applicants have substantially worse personalities than those of any other racial group.

Harvard’s admissions process assigns applicants ratings on five criteria that, according to Harvard, do not consider race: academics, extracurriculars, athletics, recommendations and personality. These individual ratings are combined into an overall rating that can consider race.

At trial, Harvard showed that while Asians scored higher than whites on academics and extracurriculars, and similar to whites on recommendations, they scored lower on personality and much lower on athletics. Whites were also more likely to have high ratings across multiple criteria. This combination of relative strengths and weaknesses across applicants resulted in parity. Asians and whites were admitted to Harvard at roughly equal rates. Therefore, if the individual ratings were unbiased, there was no evidence of discrimination against Asians.

Students for Fair Admissions (SFFA) contended that the personality ratings were in fact biased against Asians because Asians had consistently lower personality ratings than whites, Blacks and Latinos. Harvard, in response, argued, and the lower courts agreed, that SFFA could not prove that Asians’ low personality ratings were the result of discrimination.

Harvard and the lower courts are legally correct, but not because Harvard’s personality ratings are demonstrably unbiased. Rather, they are difficult to objectively assess in a way that could prove bias. Harvard’s admissions officers determine personality ratings from applicant essays, teacher recommendations, alumni interviews and any other pertinent information in the application. According to Harvard, personality ratings do not directly consider race, but experiences tied to an applicant’s race can be a factor.

Still, a closer examination of Harvard’s personality scores reveals two unsettling facts. First, when comparing applicants with the same level of academic achievement, Asians always had the worst personality scores of any group. The SFFA constructed an index based on Harvard’s academic rating and used it to divide applicants into 10 evenly-sized groups (deciles) based on the strength of their academic performance. Within each of the 10 academic deciles, Asians had the lowest personality scores across all of the racial groups. A perfect 10 for 10.  

Second, within each racial group, high academic performance strongly predicted high personality ratings, but Asians had the lowest average personality rating even though they had the highest average academic rating. For applicants in the bottom decile of the academic index, the chances of scoring a high personality rating were between 8-9 percent for all groups. But for applicants in the top academic decile, these numbers were 21 percent, 29 percent, 33 percent and 43 percent for Asians, whites, Latinos and Blacks, respectively. Whites in the top academic decile had a nearly 40 percent greater chance of scoring a high personality rating than Asians, and this percentage was even greater for the other groups.  

In sum, Asians not only had the lowest personality scores for any given academic level, but higher academic performance didn’t translate into a winning personality for Asians to nearly the same degree that it did for other groups. In fact, Asian applicants in the top academic decile received personality scores that were comparable to Black, Latino and white applicants in the third, fourth and fifth academic deciles.

In contrast, Harvard alumni rated Asians similar to whites on personality and better, on average, than Latinos and blacks. But it is the admissions office, not alumni, that ultimately determines Harvard’s personality ratings.

It is not possible to discern the truth about discrimination from these facts alone and their conflicting interpretations. But someone must be right. Either Harvard needs to take a hard look at cultural bias in its admissions process, or Asian applicants to Harvard need to spend less of their scarce time on academics and more on personality development.

Prasad Krishnamurthy is a professor of law at U.C. Berkeley School of Law.