Backstage at the Admissions Office

We often hear officials at America’s elite colleges and universities crowing about their “rich and interesting” student bodies, and how they assemble them by using “holistic” procedures to evaluate the huge numbers of applicants. The implicit idea is that the holistic approach is much better than a meritocratic system that admits the best qualified students.

An article published in the New York Times recently, “Confessions of an Application Reader,” does considerable damage to that idea. It was authored by Ruth Starkman, who teaches ethics at both the University of San Francisco and Stanford. She served as an “external reader” for the 2011 admissions year at nearby UC-Berkeley and found the experience depressing.

Along with about 70 other “readers” she helped to rank the applications of nearly 53,000 students to the famous university. Of those students, only about 21 percent were accepted. Professor Starkman’s piece sheds a great deal of light on the process by which Berkeley decided to accept some applicants and reject the rest.

Berkeley, we learn, is not content to simply take the advice of veteran scholars like Starkman on the merits of applying students. No, the readers first had to undergo a training program intended, she writes, “to ensure our decisions conformed with the criteria outlined by the admissions office.”

Ostensibly, the goal was to give each applicant “as close to equal treatment as possible,” but Starkman’s point is that the officials actually wanted something very different. Berkeley was looking for ways to guarantee that there would be enough students from “underrepresented minority groups” in a state with a constitutional amendment (Proposition 209, passed in 1996) against using race, gender or ethnicity in admissions at public institutions.

UC officials have never made a secret of their opposition to Proposition 209. Former Berkeley chancellor Robert Berdahl, for example, lamented that the law got in the way of the school’s efforts at recruiting more “underrepresented” students and said, “We need to think about new strategies…” The new strategy is to use “holisitic” evaluations where other factors serve as proxies for race and ethnicity.

Starkman found that Berkeley’s system was marked by “fundamental unevenness” and was “confusingly subjective.” Of course! That’s how the school tips the scales in favor of students who supposedly produce the desired “diversity” on campus, but without acknowledging that it desperately wants to fill quotas of them.

When she asked an official after one of the training sessions how to consider race, Starkman was told that doing that would be illegal, but that the school wanted to get “the bigger picture” about the student’s life. It became evident to her that “the bigger picture” was designed to improve the rankings of certain minorities, but not of students who just had outstanding academic records.

In one instance, Starkman was told that a low-income student who had excellent test scores should only be ranked a “3” (that is, middle-of-the-pack). The smudge on his “bigger picture” was having served in the Israeli army.

Readers were told not to downgrade students because they hadn’t taken any Advanced Placement courses in high school, but should instead pay the most attention to unweighted grade point averages. Those considerations work in favor of students who went to high school in lower-income areas, a proxy for race.

The same was true of the emphasis readers were supposed to put on “stressors.” If the reader detected anything to indicate that the student might have struggled against socioeconomic disadvantage, that was a reason to raise the applicant’s ranking. Conversely, readers were instructed not to elevate applicants whose essays showed anything that “smacks of privilege.”

Once the training was over and the readers were at work on their rankings, Berkeley’s agenda became more obvious to Starkman when she received emails from an admissions official saying that she wasn’t evaluating correctly. She had given too many “outlier” scores that were lower than they should have been.

Berkeley did not really want her professional judgment about the applicants. It wanted her to go along with the university’s admissions agenda.

Another reason why Starkman’s evaluations were so often “wrong” was probably that she got tired of essays that expressed “a torrent of woe.” Many students know that Berkeley (and other top schools using “holistic” evaluations) look for those “stressors” and therefore pile on their hardships. Starkman became numb to it but that meant ignoring reasons for elevating some applications.

Enlightening as the article is, it suggests but never explicitly raises a crucial question: Can “holistic” admissions truly improve a college or university?

Holistic evaluation and admissions involves a trade-off: some “diverse” students with relatively low academic ability are admitted on speculation that they will do well enough in college, while an equal number of students who are not diverse (at least in the ways that count) but who have demonstrated their high academic ability are rejected. Is that wise?

One skeptic is Thomas Sowell, who had a long teaching career at schools that played the game of admissions preferences. In his 1993 book Inside American Education, he quoted Harvard’s dean of admissions, “We want to serve the best students from all backgrounds and we’re trying to choose people who will be leaders later on….If we’re driven exclusively by academic qualities, we would have a much less rich and interesting student body….”

To that notion, Sowell made this sharp reply.

“What will look ‘rich and interesting’ to superficial people can of course differ greatly from what scholars who are masters of their respective intellectual disciplines will find to be students able to plumb the depths of what they have to offer. Dull-looking nerds can revolutionize the intellectual landscape and produce marvels of science, even if their life stories would never make a good movie or television mini-series.”

In short, using “holistic” evaluations involves a bad trade-off, not a good one.

Twenty years have gone by since the publication of Sowell’s book, but nothing has changed. Berkeley, Harvard, and other prestige schools continue to say that they’re improving their campuses (and indeed the nation) by using preferential admission polices that are supposed to build those “rich and interesting” student bodies.

But is a university really better off with one fewer exceptionally smart student with near-perfect SAT scores and one more student who will have a hard time with English 101? Starkman suggests not, writing, “Could this student with a story full of stressors and remedial-level writing skills survive in a college writing course?”

I imagine that Ruth Starkman will take a lot of heat for her confession that “holistic” evaluations is another instance where a nice-sounding idea has poor results. But I say “Bravo” for daring to question a myth.