The world of elite college admissions experienced a major shake-up last year. Many of the nation’s top schools started requiring applicants to submit test scores again, after failed experiments with “test-optional” systems inspired by the pandemic and activist pressure.

In announcing the change, one school after another touted how well standardized tests predicted applicants’ college performance. They also noted that, under the previous test-optional regime, many poor students chose not to submit scores that would have boosted their admission chances. Elite schools often assess applicants’ test scores on a curve—a score that would hurt a rich kid from an elite private school might help a poor kid from a low-performing public one—but outsiders had no way of knowing how this would have applied to their particular applications.

A new working paper from Bruce Sacerdote, Douglas O. Staiger, and Michele Tine examines the admissions practices at Dartmouth, which will again require test scores, starting with the class of 2029. The researchers found that at Dartmouth, “less-advantaged” students—including first-generation college students, those from neighborhoods with below-median incomes, and those whose high schools ranked high on the College Board’s Challenge Index—boosted their chances by submitting SAT scores at or above 1420, even though the school’s median score is 1520 and its 25th percentile score is 1440.

As Sacerdote and his coauthors demonstrate, the SAT is highly predictive of students’ performance at Dartmouth. SAT scores explained about 22 percent of the variation in Dartmouth students’ first-year GPAs; high-school GPAs, by contrast, explained only 9 percent. “[T]he relationship between academic performance in college and test scores,” the authors note, “is quite similar across subgroups including subgroups by income, gender, and more versus less advantaged high school.”

The researchers then examined what happens when applicants have the option of submitting this valuable, unbiased data point—which, under Dartmouth’s test-optional policy, a striking 46 percent declined to do. Fortunately, the researchers were able to access some of these kids’ SAT scores, even though those scores weren’t used in the admissions process, allowing them to estimate the effects on an applicant’s chances had he submitted his scores.

Sacerdote and his coauthors found that less advantaged students who didn’t use their scores had admission rates of around 5 percent—and these rates didn’t correlate much with their (hidden) SAT results, suggesting that the rest of those students' applications failed to reveal the talents captured by the test. Such students who did send their SATs were penalized for low scores, especially those below 1300, but were aided by scores above 1400.

For better-off students, however, the researchers found that submitting test scores was less consequential. These applicants’ chances of admission were positively correlated with their test results even if they didn’t submit the scores. This could suggest that wealthier applicants who declined to submit their scores demonstrated their skills elsewhere in their application, or, as the authors suggest, that admissions offices were simply more familiar with wealthy high schools’ transcripts and could better infer their students’ abilities from their grades.

What about the other alleged benefits of going test-optional? At Dartmouth, the researchers found that the policy led to a higher volume of applications, though it didn’t make the applicant pool much more demographically or academically diverse.

The policy also favored applicants who hid low test scores, were admitted to the school, and ultimately failed to meet Dartmouth’s academic demands. By contrast, it hurt high performers—especially high-performing poor kids who didn’t know that their scores would have boosted their chances of admission. It’s welcome news that many elite institutions are changing course.

Photo by: Education Images/Universal Images Group via Getty Images

Donate

City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next