.webp)
During the last several months, we’ve watched test scores return to center stage at some of America’s top colleges. Admissions offices shared the results of their internal analyses, revealing the powerful predictive value of test scores. Some admissions offices retracted test-optional policies, because test-submitters had such significant advantages that the phrase “optional” became misleading. At the same time, we’ve seen other colleges play down internal analyses that validated the predictive value of testing and commit to test-optional policies for the long haul. Nothing is ever simple in the realm of college admissions.
Under the paradigm of test-optional admissions, there’s a great deal of ambiguity regarding the relative weight of test scores in the admissions process. Colleges have a natural incentive to encourage more students to apply, to yield the biggest possible pool of applicants from which to build a class. Most colleges with test-optional policies have advised students that they will be judged equally with or without test scores.
What we learned from Yale is that once applications arrive inside the admissions office, the treatment of students with and without test scores was not remotely similar. In fact, students without test scores had a 2% chance of gaining admission, while students with test scores had a 6% chance of admission—triple the odds. The Dean of Undergraduate Admissions, Jeremiah Quinlan, understood that “students who were not submitting their test score were inadvertently hurting their chances of admission to Yale.” He admitted, “I became more and more convinced that we weren’t being honest about the reality of our admissions process to students and parents.” Calling the policy test-optional felt disingenuous, and the admissions office reversed course and reinstated testing requirements.
Yale came clean and corrected the misperception that students had an equal shot of getting in with or without tests, but how many other admissions offices have yet to make that correction? This is something that none of us can know. And that’s troubling.
We learned from the University of Texas at Austin that test scores are nearly essential to ensure the right students are admitted to the right academic programs. Students without test scores were achieving Freshman grades nearly a full point (0.86) below their matched peers who submitted test scores, and were 55% more likely to be on academic probation (below a 2.0 GPA). Students performing at lower levels require more resources and academic support and pose a greater risk of dropping out. This incredible discrepancy in academic performance drove the Texas flagship to reinstate test requirements, and other schools will surely follow.
The Ivy Plus study conducted by Raj Chetty and other researchers from the Opportunity Insights group found that for the Ivies plus Chicago, Stanford, Duke, and MIT, standardized test scores were much better than high school grades at predicting collegiate academic performance. The results of their analysis were striking.
In their internal study of variables predicting performance at Dartmouth, researchers Bruce Sacerdote and colleagues found that high school grades predict a mere 9% of the variance in college grades, while the SAT, by itself, explains 22% of the variance. Combining the two variables explains 25% of the variance in grades at Dartmouth. The SAT is the single best predictor of academic achievement for all demographic groups, which helped fuel Dartmouth’s decision to reinstate testing requirements.
It’s not surprising that UT Austin was the first public school to reinstate testing following announcements from Dartmouth, Yale and Brown, as the UT acceptance rate for out-of-state students, a mere 8%, approaches the admit rate for this highly select group of colleges. UNC Chapel Hill posted an 8% admit rate for out-of-state students for the class of 2026, and we’ve heard that this rate has fallen to 6% for the class of 2028. There’s a very good chance that additional public Ivies such as UNC Chapel Hill will take a page out of the Ivy playbook by reinstating testing requirements.
Test-optional admissions provides many benefits to colleges:
Given these benefits, some schools will be unwilling to revert to test-required admissions, even if their own institutional data suggests the value of that approach. Duke was part of the Opportunity Insights study, which revealed the unique predictive power of standardized tests, but decided to commit to a test-optional policy. Duke is now driving in 54,000 plus applicants (up from 36,000 in 2019) for an acceptance rate of 5.1%, and it may be loathe to give up those gains in applicants and selectivity.
In the discussion regarding reinstating test-requirements, board members of the University System of North Carolinaexpressed concerns that requiring test scores might drive students away from the UNC system, especially at less selective schools.
One colleague framed the discussion in terms of state demographics. States that are growing in population, like Texas, may be less worried about hitting future admissions targets and more likely to reinstate testing requirements. On the other hand, states that are stagnating or losing population may be much less likely to reinstate testing requirements, for fear of losing potential applicants.
Admissions officers at Yale were worried about SAT and ACT testing access for the kids coming from California. Given that the UC and Cal State systems are test-blind, fewer testing sites are opening their doors on weekends, and kids are struggling to find available seats. Some of the independent college counselors with whom we work have shared stories of their California students going to great lengths to take the March SAT: driving 3 hours round-trip, or even having to fly to Washington or Idaho to find an open seat. Due in part to this burden of access to testing, Yale decided to adopt a test-flexible position and allow students to submit APs or IBs in lieu of SATs or ACTs. If testing access remains so challenging in California, this could impact the decision of some colleges to reinstate testing requirements.
The window is closing for colleges to make announcements that will affect the current crop of juniors, but more colleges will announce the return to testing for future classes. Many selective schools across the country have yet to take a post-pandemic position on testing. If the data from MIT, Dartmouth, Texas and other selective schools reflects broader trends, we can only expect more colleges will follow in the footsteps of these institutions.
