Before the pandemic, a growing number of colleges stopped requiring applicants to submit SAT or ACT scores, as a way to increase diversity on their campuses. But researchers are finding that the test-optional policy isn’t substantially raising the share of low-income students or students of color at colleges that have tried it.
The latest study, published in the peer-reviewed American Educational Research Journal in April 2021, found that test-optional admissions increased the share of Black, Latino and Native American students by only 1 percentage point at about 100 colleges and universities that adopted the policy between 2005-06 and 2015-16. The share of low-income students, as measured by those who qualify for federal Pell Grants, also increased by only 1 percentage point on these campuses, compared to similar schools that continued to require SAT and ACT scores.
“We’re moving the needle a little bit but it’s a drop in the bucket for what we need,” said Kelly Rosinger, an assistant professor of education at Penn State and a former admissions officer at the University of Georgia, in reaction to this study. Her earlier research on a group of selective liberal arts colleges, which had adopted test-optional policies before 2011, didn’t find any diversity improvements on those campuses.
Rosinger said not to expect “dramatic gains” in diversity from eliminating testing requirements because the other qualifications that admissions departments weigh, such as extracurricular activities and advanced high school courses, “tend to privilege the same students who are privileged by test scores.” Well-to-do families can pay for extras like sports and music lessons and high schools in wealthier neighborhoods are more likely to offer advanced coursework.
Going test-optional did increase the share of women, who already made up a majority of students on these campuses, by 4 percentage points. Test-optional policies especially benefited women because they’ve historically had lower test scores than men but higher grades.
In 1970, Bowdoin College was the first college in the country to go test-optional. A small number of liberal arts colleges followed suit not only to foster diversity but also because it appealed to an anti-testing sensibility. The test-optional movement gathered steam in the 2000s as concerns mounted over the fact that wealthier students could hire tutors, take the tests multiple times and post higher scores.
The latest study, conducted by Christopher Bennett when he was a doctoral student at Vanderbilt University, homed in on 99 private colleges and universities that opted to waive testing requirements between 2005-6 and 2015-16. Some were small, highly selective liberal arts colleges, such as Bennington College, Smith College and Wesleyan University. Others were less selective and some were larger research universities with graduate schools, such as American University in Washington, D.C., and DePaul University in Chicago. No public universities were included in this study because not enough state schools with competitive admissions had gone test-optional before the pandemic.
Bennett made an effort to capture how much more diverse the 99 colleges became as a result of the test-optional policy. This kind of analysis is tricky since many colleges were becoming more diverse in recent decades, even if they didn’t adopt a test-optional policy. That’s because the percentage of Latino high school students was soaring and more low-income Americans were applying to college. To isolate how much student diversity could be attributed to test-optional policies, Bennett first calculated that the share of students who were Black, Latino or Native American increased from 13 percent to 17 percent at the 99 schools after adopting test-optional admissions. That’s a 4 percentage-point increase.
He then compared these diversity gains to what happened at a similar group of colleges that didn’t adopt test-optional policies until after 2015-16. Their racial and ethnic diversity also increased by about 3 three percentage points even as they continued to require SAT or ACT scores. The difference between the two was a slight 1 percentage point. Similarly, the share of low-income students who were Pell Grant recipients rose from 25 percent to 28 percent at the test-optional schools. But that 3 percentage point gain whittled down to 1 once it was compared against similar colleges that were still requiring test scores. That’s why Bennett determined that the test-optional colleges increased their diversity only slightly more than similar colleges that were still requiring applicants to submit test scores.
Many colleges that went test-optional have boasted that their Black and Latino student population increased by double digits. Indeed, that did happen. Bennett found a 3 to 4 percent increase in the enrollment of Pell Grant recipients from low-income families and a 10 to 12 percent increase in enrollment of Black, Latino and Native American students. But there were so few of these students to start that even these large increases didn’t change the demographics of the campus all that much. And that’s why the total share of students of color only shifted by 1 percentage point. My best analogy is low-fat milk. If you increase the amount of fat by 50 percent, that sounds like a lot more fat. But your total milk fat might rise from 1 percent to only 1.5 percent. It’s still skimmed milk.
The starting line also helps explain why the share of women went up by a lot more. Women already made up more than 50 percent of college students and so even a relatively small increase could raise their total share by a noticeable amount.
Another surprising finding is that only 20 to 30 percent of students took advantage of test-optional admissions, according to Bennett’s rough estimate. The majority of applicants continued to submit test scores. Bennett said the low uptake also helps explain why it didn’t result in a big increase in diversity. “We shouldn’t really see effects that are enormously large if very few students are actually taking advantage of the test-optional option,” Bennett said
During the pandemic most colleges and universities went test-optional because students couldn’t sit for exams in person. Florida’s public universities were one of the few big exceptions and still required SAT or ACT scores. It will be interesting to see how the diversity of the 2021-22 freshman classes change when all the data are in. If colleges remain bastions of privilege, that could slow the momentum of test-optional admissions. Colleges had to hire many more admissions staffers and application readers to sift through applications without test scores. Test scores are an efficient way to reduce the applicant pool. Based on this pre-pandemic research, it may seem that the small diversity gains from test-optional admissions are not worth the cost.
Akil Bello, senior director of advocacy at FairTest, which believes that standardized testing is flawed and misused, argues that colleges shouldn’t reinstate SAT and ACT requirements after the pandemic even if test-optional admissions prove not to raise the diversity of the nation’s college campuses very much. “You’re removing a barrier, and it does no harm to students,” said Bello. “What’s the point of keeping this barrier?”
I believe that many college leaders sincerely want to improve diversity. If small fixes like test-optional admissions don’t work well, and more radical solutions like affirmative action are under legal attack, there are no obvious answers for college admissions departments.
This story about test-optional admissions was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for the Hechinger newsletter.
The interface is already broken as a result of the “name buy” system created by the testing companies, who reap millions of dollars in revenues selling names back to university admissions departments in one year cycles. Universities are wedded to purchasing names from their same feeders schools and do not have access to a more diverse student pool because they are naturally risk averse; the one year cycle is institutional behavior that inhibits universities from expanding their feeder pool. Also, students do not necessarily know which school has a program that best fits their gifting because there exists a 700 to 1 student counselor ratio (in California), which is entirely unsustainable to help students find the “right fit.” Counselors generally counsel “top of mind” and for the best students as opposed to the B- or C+ student who could match well to the small liberal arts college. What is transpiring is a groundswell of students wanting more choices and colleges seeking students outside of institutional methods. This trend combined with the research that cumulative GPA is a better indicator of student success we are seeing movement to a holistic approach better matching colleges with students beginning much earlier. The interface to higher-ed has been dominated by tests producing the “name buy” system, a one-year recruiting cycle and an unsustainable counseling ratio. This system has created a pathology that universities cannot recruit before the SAT, and students cannot find the right college program that fits their gifting. Because of the oncoming enrollment cliff going universities must now re-think how they recruit students and students are desiring more choices. The pandemic has accelerated a review of these practices. Albion College in Michigan is an example of university admissions departments adjusting to the new reality.