One of the by-products of writing a weekly column is that I am always on the lookout for topics to write about or fresh angles to cover. I self-servingly argue that makes me more effective in my day job as a college counselor, but I also know that I face the temptation of viewing every news story and conversation through a lens of possible column material.
Apparently that’s true of scholars as well. One of the dueling expert witnesses in last year’s court case dealing with Harvard University’s treatment of Asian American applicants, Students For Fair Admission v. Harvard, was Peter Arcidiacono, an economics professor at Duke University.
Now Arcidiacono, who testified on behalf of the plaintiffs, and two other economics professors, Josh Kinsler from the University of Georgia and Tyler Ransom from the University of Oklahoma, have collaborated on two working papers for the National Bureau of Economic Research, both of them based on the data Arcidiacono reviewed and analyzed for the Harvard case. The Duke economist’s ability to turn his courtroom gig into two scholarly papers is not only an example of social science research at its best, but also pays tribute to the Seinfeld episode where Jerry and George pitch their idea for a show about nothing to NBC executives. Stand in line at a Chinese restaurant? “That’s a show.” Testify in court? That’s a paper.
The Harvard case provided a public glimpse not seen before into how the selective college admissions process functions. Lawyers for Harvard tried to argue that details about institutional policies and procedures should be kept out of the public eye on grounds that they were proprietary. That argument seems akin to claims of executive privilege and national security as justifications for government secrets. The danger is not that releasing information will be damaging but rather embarrassing.
The first of the papers produced by Arcidiacono et al. examines preferences given to legacies and athletes by Harvard. Their research shows that 43 percent of white admits at Harvard are “ALDCs,” a designation that includes athletes, legacies, “dean’s list” (donors) and children of faculty members.
Those who fall into the ALDC designation are “disproportionately white and come from higher income households.” The paper finds that LDC applicants (those other than athletes) are on average stronger than non-ALDC applicants, but the average LDC admit is weaker than the average non-ALDC admit, and from that fact draws a conclusion that there is an advantage in the admissions process for applicants in that group.
Arcidiacono and his co-authors conclude that only 25 percent of white applicants admitted in the ALDC groups would have been admitted without the benefit of membership in that category. Their analysis suggests that a 10 percent chance of admission rises to 50 percent for a legacy, 70 percent for children of donors on the dean’s list, and “near certainty” of admission as a recruited athlete.
The information about the role of athletics in Arcidiacono’s paper prompted Derek Thompson to write an article for The Atlantic with the title “The Cult of Rich-Kid Sports.” Thompson argued that Harvard’s nonscholarship Division I athletic program “functions as affirmative action for white affluence.”
Harvard offers more varsity sports (42) than any other Division I program. The need to recruit athletes capable of competing becomes a priority in the admissions process, and according to Thompson, Harvard’s student body contains more student athletes than Ohio State’s. Arcidiacono’s paper found that in the six admission cycles between 2014 and 2019, recruited athletes made up more than 10 percent of those admitted despite being less than 1 percent of the applicant pool. The research states that recruited athletes have an admit rate of 86 percent, which would seem to be better odds than the overall 5 percent rate. Being a recruited athlete is by far the best admission “hook.”
The natural presumption would be that the athletic program is an engine for bringing racial and socioeconomic diversity to Harvard, but Thompson’s article argues that is not the case. He describes a number of the sports offered by Harvard, such as squash, crew, water polo and lacrosse, as “true rich-kid sports,” and cites National Collegiate Athletic Association data from 2017-18 that fewer than 30 of the Ivy Leaguers competing in crew and lacrosse were black. Arcidiacono’s research cites a Harvard Crimson survey estimating that a quarter of recruited athletes in the Class of 2019 came from families earning more than $500,000 per year. That’s not as high as legacies (40.7 percent) but substantially higher than the student body as a whole (15.4 percent). The working paper finds that just over 3 percent of white admitted athletes come from economic disadvantage.
But is any of that shocking to anyone who follows higher education? And is Harvard doing anything different than numerous other highly selective institutions in using the admissions process to achieve a variety of institutional goals? I’m sure that there are researchers who would challenge Arcidiacono and his colleagues’ analysis of the data, as David Card from the University of California, Berkeley, did in the trial itself, but having data available publicly to analyze and debate is a good thing for those of us who believe that transparency should be a guiding principle for college admission.
The larger question is whether the outsize preferences given to ALDCs and other groups in admission is particularly onerous at a place like Harvard. Selective admission is a zero-sum game, with good news for one applicant removing opportunity for another applicant. When an institution is admitting only one in 20 applicants, it is forced to make fine distinctions among applicants, all of whom are superbly qualified, and the consequences of setting aside large segments of the student body to special interests are greater for applicants and for the institution. I suspect that the Harvard case will lead to a broader discussion about preferences of all kinds in college admission. That’s a discussion worth having.
The second working paper by Arcidiacono, Kinsler and Ransom raises more interesting philosophical questions. I will address them in a future column, perhaps as early as next week.