Measuring Success: An Examination of the use of Standardized Tests in College Admissions

We conceived of this book in the spring of 2016 in response to the fragmented and incomplete state of the literature that informs debates and decisions related to college admission testing. For many students, teachers, parents, policymakers – frankly, nearly all of those immediately outside the testing industry and college admissions – the role of college admission tests remains a mystery. To supporters, standardized tests provide a neutral yardstick for measuring student potential and performance—particularly important given the varying levels of academic rigor across high schools. But detractors, including those who support test-optional policies, argue that college-entrance assessments are biased, misused, or overused.

With this in mind, we set out to assemble a comprehensive collection of new research on admissions testing from experts and practitioners on both sides of the debate, with an emphasis on methodological rigor that has too often been lacking from the discussion of such emerging practices as test-optional admissions. Beyond pure research, we also wanted to highlight the on-the-ground perspective of college enrollment officers who have changed or considered changing their testing policies.

In the 12-chapter volume that resulted, contributors provide detailed evidence that standardized test scores (especially when combined with grades) have significant predictive validity for college performance and completion across race, gender, or socioeconomic status. What’s more, several authors reveal significant flaws in the research that institutions frequently cite to justify their claim that going test optional will boost minority enrollment. A careful study by Andrew Belasco and his colleagues shows that test-optional policies generally enhance selectivity but do not in fact increase racial diversity.

Nevertheless, other chapters in Measuring Success show that certain admissions leaders continue to view standardized testing warily. Some believe testing requirements are likely to intimidate and discourage potential minority applicants. Others contend that even if tests provide useful predictive value, the prospect of increasing racial and economic diversity through test-optional policies presents a worthwhile trade-off for a modest decline in college-wide GPA performance. Perhaps the most striking claim of test-optional proponents, not always stated directly, is that the only way to prevent misuse of testing is in effect to tie the hands of admissions officers. Making admissions tests voluntary, this argument goes, forces enrollment managers to match their practice with their long-standing rhetoric about keeping the use of tests in perspective.

But this invites a question: what prevents admissions officers from practicing what they preach about holistic admissions without going test-optional? Given the well-documented evidence for the usefulness of tests as an admissions tool – along with well-established guidance that tests should be used together with grades and other criteria – why can’t enrollment managers continue to require SAT or ACT scores but ensure that their staffs are trained to use the tests appropriately? After all, colleges trust their staffs to implement complex need-blind admissions policies without removing information about students’ financial status from their deliberations. Conversely, if tests aren’t useful enough to be required of all students, why would admissions officers want any students to submit them?

Perhaps the biggest reason for continued interest in test-optional admissions is simply that some colleges see little downside – and a number of upsides – from implementing the policy. Their announcements about opening up access and bringing greater racial diversity to campus tend to garner positive publicity despite the lack of rigorous research supporting such claims. And removing testing requirements tends to increase application volume.

College admissions certainly isn’t the only field in which research doesn’t fully inform policy. But the findings of Michael Hurwitz and Jason Lee in their chapter on grade inflation should give pause even to those who aren’t persuaded by the evidence about the value of testing presented in other chapters. High school grade point averages continue to rise, with significant compression at the high end, making it particularly hard to distinguish applicants from one another. With these trends especially pronounced among students at affluent, mostly-white high schools, the notion that relying more heavily on grades will reduce inequities in the admissions process seems implausible.

As the college admissions debate continues, it has become clear that developing more useful, fine-tuned data for decision making will require more attention to rigorous research, not less. Test scores certainly need not be the only item in the admissions toolkit, but enrollment managers have ample reason to take them seriously as an important gauge of academic potential. As the research in this volume shows, there is no reason to believe that eliminating the information provided by test scores will help admissions officers meet their goal of opening doors for more aspiring students.

Jack Buckley is the senior vice president of research and evaluation at the American Institutes for Research and a research associate professor of applied statistics at New York University. He is the coauthor of Charter Schools: Hope or Hype? Lynn Letukas is an associate research scientist at the College Board. She is the author of Primetime Pundits: How Cable News Covers Social Issues. Ben Wildavsky is senior fellow and executive director at the College Board Policy Center. He is the author of The Great Brain Race: How Global Universities are Reshaping the World. They are coeditors of Measuring Success: Testing, Grades, and the Future of College Admissions