New research sheds light on why colleges choose test-optional admissions
Yale economist Navin Kartik and coauthors explore why admissions committees sometimes choose not to see test scores, offering a theory of how social pressure shapes university decisions.
Why would a university ever prefer less information about its applicants? That question lies at the heart of “The Test-Optional Puzzle” and “Test-Optional Admissions,” a pair of papers published in the AEA Papers and Proceedings and the American Economic Review, respectively, coauthored by Yale economist Navin Kartik with Wouter Dessein (Columbia University) and Alex Frankel (University of Chicago).
“From an economist’s perspective, it seems odd,” said Kartik, who joined Yale’s Department of Economics in Fall 2025. “Why would you ever want less information and not more?”
In recent years, there has been a vibrant debate about the role of standardized test scores in college admissions, with test-optional policies now widespread across U.S. higher education. Even prior to 2020, a notable fraction of four-year colleges and universities were already test-optional, meaning they did not require standardized test scores for undergraduate admissions. During the COVID-19 pandemic, when many students were unable to take the SAT or ACT, virtually all institutions across the country paused testing requirements. Yet even after the pandemic, notwithstanding a few prominent reversals (including Yale), most U.S. colleges have remained test-optional.
Kartik’s research contends that common claims about the benefits of test-optionality, such as a school being able to enroll a more diverse cohort of students, are not very compelling. “We were interested in understanding why a college would voluntarily give up information,” Kartik said. “Outside of a pandemic, or extreme costs of testing, the logic isn’t obvious. Wouldn’t a college be better off observing applicants’ scores and deciding how to use those scores to achieve whatever objective it has?” After all, standard economic theory suggests that decision-makers are better off with more information—even if that information is noisy, or affected by factors like differential resources for test preparation.
The authors then propose a novel theory to explain why institutions might voluntarily forgo information that could improve admissions decisions.
Journal Publication
Test-Optional Admissions
Authors: Wouter Dessein, Alex Frankel, Navin Kartik
A theory of social pressure
While advocates of test-optional policies often emphasize equity and access, Kartik and his coauthors suggest a different angle.
The researchers’ model points to how social pressure may be a key factor. Colleges may face reputational, legal, or political costs when their admissions decisions conflict with widely held views about merit. These pressures can come from outside observers, including alumni, courts, parents, or the broader public. When test scores are visible, observers may strongly disagree with decisions such as admitting low-scoring athletes or legacy applicants, while rejecting higher-scoring students.
“If the college doesn’t see the score, society infers something like an average,” Kartik explained. “That way, the school can admit more of the students it wants, without facing as much of a penalty from outside observers.” In the model, choosing not to require test scores allows colleges to reduce these “disagreement costs,” even though it also reduces the college’s ability to fine-tune its decisions to test scores.
Journal Publication
The Test-Optional Puzzle
Authors: Wouter Dessein, Alex Frankel, Navin Kartik
Beyond equity
The central contribution of the newest paper is showing how institutions may strategically choose not to know. Test-optional policies, the authors argue, can help colleges manage social disagreement while still pursuing their own institutional objectives.
To illustrate the idea, Kartik describes a stylized example involving athletic recruitment. A college may want to admit certain athletes even if their test scores fall below what outside observers would consider acceptable. If scores are visible, those decisions can trigger backlash. If scores are unseen, those same students appear closer to the average, reducing conflict without changing who gets admitted.
In earlier research published in the Journal of Political Economy, Kartik examined how differential test preparation resources can dilute what scores actually measure, shifting them away from academic ability and toward what he calls “gaming ability.”
“Test scores end up reflecting not just general preparedness, but also your ability to study to the test,” he said. “Whether you can afford tutoring, for example.”
Implications for admissions policy
For policymakers and admissions offices, the findings reframe the debate. The choice is not simply about whether standardized tests are fair or predictive. It is also about the political and social costs of using them.
Kartik emphasized that colleges face a trade-off between making the best intrinsic decisions and managing external pressure from observers. “By not seeing scores, you’re certainly worsening your ability to make the decisions you fundamentally want to,” he said. “But you may be able to mitigate external pressure.”
He also stressed that the research is theoretical rather than prescriptive. “We’re not saying this is the only explanation for test-optional policies,” he said. “But we think it’s a plausible one. We’re very agnostic about what the college is trying to maximize. Whatever your objective is, seeing information shouldn’t hurt you, external pressures notwithstanding.”
More broadly, the work connects to Kartik’s research agenda on how information is used, manipulated, or withheld in markets—from credit scores to recommendation algorithms. As he put it: “The fundamental question is why less information can sometimes look better than more. That’s what we’re trying to understand here.”
Meet Navin Kartik
Navin Kartik joined Yale’s Department of Economics in Fall 2025, after spending 17 years at Columbia University. An economic theorist, Kartik studies how information affects economics and politics.
His research examines situations where people and organizations face incentives to shape what others learn about them—from standardized tests and college admissions to credit scores, corporate disclosure, and political accountability. Across a range of settings, his work asks how information is communicated, when and how it improves or distorts decision-making, and how institutions can be designed to better use information.
At Yale, Kartik is excited to build connections across economics, political science, and data science, and to contribute to the department’s long tradition of theory that engages with real-world policy and institutional design. He is the first Director of the new Political Economy Program at the Cowles Foundation, which supports theoretical and empirical research on voting, elections, political institutions, lobbying and bargaining, international agreements, and the interaction of public policy with political and economic power.