The Dartmouth Report

I've been following recent developments at Dartmouth College closely for a couple of different reasons. Their current president, Sian Beilock, was ours for six years. Their Middle East Dialogues initiative, launched through a collaborative effort by faculty in the Jewish Studies and Middle Eastern Studies programs, has fostered constructive engagement to a degree not seen on peer campuses. And their recent decision to reinstate a standardized testing requirement for all undergraduate applicants was based on a detailed and very interesting empirical analysis that has been made public.

The most striking claim in the Dartmouth report is that the test-optional policy adopted during the pandemic made it more difficult for the admissions office “to identify high-achieving less-advantaged applicants.” The strongest evidence for this claim comes from a pool of about five thousand applicants who submitted scores but then requested that they be disregarded in the evaluation process.

Like many of its peer institutions, Dartmouth interprets scores in context—within any given range of scores, applicants with identifiably disadvantaged backgrounds are more likely to be given offers of admission. Advantage is assessed based on first-generation status, as well as neighborhood and school characteristics. As the authors state, “a score of 1400 for an applicant from a high school in a lower-income community with lower school-wide test scores is a more significant achievement than a score of 1400 for an applicant from a high school in a higher-income community with higher school-wide test scores.”

The idea is that students who have faced significant hardships prior to application have exhibited greater latent ability at any given score. The logic is simlar to that underlying percentage plans at public universities that guarantee admission based on class rank rather than absolute performance.

Given this approach to admissions, an applicant’s decision to submit scores under a test-optional policy should depend on their individual, neighborhood, and school characteristics. But the authors of the report find that this is not the case in practice:

All groups (e.g. students from higher- versus lower-income neighborhoods or students from more- versus less-advantaged backgrounds) submit scores at roughly the same rate for a given SAT score. This similarity actually hinders the admission chances for less-advantaged applicants, who would raise their admissions probabilities by revealing their scores at some range…

Consider students with a score of 1450-1490 from less-advantaged backgrounds. These students increased their admission probability by a factor of 3.7x (from .02 to .074) by revealing their score. The test-optional policy thus led to Admissions not identifying these high-achieving applicants as highly prepared… There are hundreds of less-advantaged applicants with scores in the 1400 range who should be submitting scores to identify themselves to Admissions, but do not under test-optional policies.”

In effect, a certain group of applicants from less advantageous backgrounds were sabotaging their own prospects of admission by witholding scores. In doing so, they were also making it harder for the university itself to identify the applicants it most wanted to admit. Dartmouth has responded by removing the discretion of applicants to harm themselves and the institution in this way.

There is no doubt that moving to test-optional admissions expanded the applicant pool. At Dartmouth, for example, the pool was almost thirty percent larger in the test optional cohorts (2021-2022) than in the immediately preceding test-required cohorts (2017-2019). However, the larger pool was not much different in composition—the two sets of cohorts had roughly the same first-generation share and comparable neighborhood income distributions. It is for this reason that the College is willing to accept a return to smaller applicant pools.

Still, the reasoning behind the reinstatement will strike some people as paternalistic—the university is trying to prevent people from damaging their own prospects, and doing so by limiting their discretion. While this policy (and similar decisions at MIT and Yale) are perfectly defensible, there is an alternative approach that might solve the problem in a less paternalistic manner. A university could require the submission of scores from all applicants, but also offer them the option of having these scores disregarded in the evaluation process. This would be a test-optional policy in effect, but would generate exactly the kind of data that could help an institution identify applicants who were hurting their own prospects of admission, so that it may increase outreach efforts and better target the dissemination of information.

I don’t know of any university contemplating such a policy, but for those who want to remain test-optional without losing highly qualified applicants from less advantaged backgrounds, it does seem worth considering.