High school seniors vying for acceptance at UC Berkeley face daunting odds. Until about 50 years ago, Cal admitted any applicant with a B average in college-prep classes. As recently as 1985, the acceptance rate was above 50%. With a record 79,000 applicants in 2014, Berkeley now admits only 17%, placing it in a near-tie with UCLA for the title of most selective public university in the country.
How can Berkeley possibly choose wisely among all the overqualified teenagers eager to enroll? With little fanfare, Cal recently decided to give its overburdened admissions office yet more information to consider. When application forms become available Aug. 1, they'll look a little different: This year, for the first time, Cal will encourage seniors to submit letters of recommendation, one from a teacher and one from somebody else who knows them well.
The change sounds innocuous. Elite private colleges routinely ask for recommendations. Many public flagships do, too, including the University of Michigan, the University of Texas in Austin, and the University of North Carolina in Chapel Hill. It may seem eminently reasonable for colleges to gather information on, say, a student's ability to overcome personal challenges or lead others.
But when it comes to the UC Berkeley admissions process, long emblematic of the meritocratic ideal at elite public universities, nothing is uncontroversial.
For starters, there's the matter of race. For years, Berkeley sorted through its applicant pool by relying heavily, though not entirely, on an index combining grades and standardized test scores. Race-based affirmative action, however, gave Latino and African Americans applicants a significant boost. Then came 1996, and Proposition 209, which banned the use of race in admissions. Quite suddenly, black and Latino numbers dropped by half. "There was an atmosphere of grieving in our office," then-admissions director Bob Laird wrote after the fact.
Not coincidentally, two years later Berkeley adopted the "holistic" or "comprehensive review" approach used by selective private colleges. It weighs conventional academic qualifications together with extracurriculars, special talents, socioeconomic background, and personal essays. Since then, Latino numbers have bounced back, though African American enrollment remains below 3%.
By adding one more factor to the holistic mix, the introduction of recommendations will inevitably heighten complaints from affirmative action opponents. They routinely argue that comprehensive review at Berkeley lacks transparency and amounts to an end-run around Prop. 209 (a charge Cal disputes).
"They have no intention of conducting an honest evaluation of the recommendations," says Stuart Taylor, coauthor of "Mismatch: How Affirmative Action Hurts Students It's Intended to Help, and Why Universities Won't Admit It." The use of letters will discriminate against high-achieving Asian students in favor of less-prepared blacks and Latinos, he predicts.
This argument resembles criticism of anti-Jewish admissions policies at Ivy League schools in the first half of the 20th century. Elite universities began evaluating "character"—including through newly introduced letters of reference—to admit clubbable young men rather than what Yale's admissions director called "Old Testament" grinds who excelled on purely academic criteria.
There are objections on the opposite grounds, too. Despite his firm belief in affirmative action, Laird dislikes the new policy because he believes it will actually "perpetuate disadvantage." That's because large public high schools with many poor and minority students, whom teachers may or may not know well, are far less likely than elite private schools to know how to play the recommendations game. They may not present applicants in the sophisticated manner most likely to help them stand out among the masses knocking at Sather Gate.
It's not an improbable consideration. Bruce Poch, longtime dean of admissions at Pomona College and now head of admissions and counseling at the private Chadwick School, sees benefits to considering recommendations. But he also recalls that when he applied to Oberlin College from a New Jersey public high school, where only 10% of graduates went on to college, his teacher recommendation noted: "he writes good."
On a more practical level, the new policy will impose a heavy administrative burden. If all 79,000 Cal applicants submit two letters, that's 158,000 recommendations that teachers, counselors, coaches, or ministers will need to write, and admissions officers need to evaluate. "There's a pretty good chance there's going to be a revolution" among counselors and teachers, Laird contends.
It's too soon to know whether that dramatic prediction will pan out. There's already been a small-scale beta test. Each year the admissions office asks a few thousand particularly disadvantaged candidates to submit recommendations, which readers say they have found helpful in making judgments. Now, Cal is expanding the full-time admissions office staff from 32 to 40 to absorb some of the extra work (an army of poorly paid seasonal application readers already helps out).
And not all teachers will necessarily have to write more recommendations than they did before. Berkeley estimates that more than half of its in-state applicants apply to other selective colleges that require recommendations; teachers can just recycle the same letters.
But the biggest question mark hanging over Berkeley's large-scale experiment — so far unique among UC schools — is whether it can do anything to rationalize the subjectivity of the admissions process.
One of the architects of the plan, Panos Papadopoulos, is a mechanical engineering professor and outgoing chair of the academic senate. He wanted to see whether decisions under the old system were replicable. So he had the files of top applicants, who were automatically admitted based on a 1 out of 5 rating, completely rescored. The result? On second reading, 40% landed in the no-guarantees No. 2 and No. 3 piles.
Papadopoulos is convinced that letters will help address this problem by providing additional data points to distinguish between seemingly similar straight-A students, as well as borderline applicants much further down the ladder, who are appealing candidates but may or may not have what it takes to make it through Cal.
But it's equally plausible that recommendations, due to arrive en masse by Jan. 1, will have the reverse effect, and just make the subjectivity of the holistic system even worse.
No admissions system is perfect. Most of the world's universities use cut-and-dried admissions criteria: standardized tests and grades. Surely teenagers who have performed well educationally despite living in poverty or overcoming other challenges merit some degree of additional consideration. Then again, does adding factors such as "cross-cultural engagement" and "demonstrated concern for others" — two qualities letter-writers will be asked to address — depart too much from core notions of academic excellence?
These questions will inevitably persist as Cal and other universities keep trying to define — some would say redefine — merit. As state institutions experiment with ways to make consistent, multidimensional judgments with limited resources, they need to maintain public credibility. When Berkeley moves ahead with its new recommendations policy, it must be prepared to explain its admissions decisions to 65,000 or so disappointed applicants in ways they and the rest of the public can understand, and trust.
Ben Wildavsky is director of higher education studies at the Rockefeller Institute of Government, State University of New York, and policy professor at SUNY in Albany.