For several decades, taking the ACT or SAT1 has been a rite of passage for high school students dreaming about going to college. According to ACT, these tests “motivate students to preform to their best ability” and “provide colleges and universities with excellent information for recruiting, advising, placement and retention.” The folks at College Board, which owns SAT, agree, claiming that their test helps students “demonstrate knowledge or skills in areas that current research tells us are most critical for college readiness and success.”
Whether or not you believe that ACT/SAT help predict success in college (and we’ll get to why that is an open debate soon) there is no denying that these tests have also been sources of dread and anxiety for students and parents, who worried that a "bad" score could torch a kid's chance to get into the college of their dreams or who hope that high scores will be rewarded with scholarships. This anxiety has help turn taking the ACT/SAT from a rite of passage to a billion dollar industry complete with test prep courses, private coaches, and countless students taking the tests multiple times in the hopes of increasing their score and thus their chances to get into a highly selective institution.
(A quick reminder: the majority of colleges admit the majority of their applicants, the majority of the time. Test scores matter quite a bit more in the admissions decision for highly selective schools than they may for a student applying to a less selective college)
While there have always been some schools that are test optional2 (especially in the community and technical college sector, which may use test scores for course placement but rarely require them for admissions), the vast majority of bachelor’s degree granting institutions accepted and/or required ACT or SAT scores for admissions3.
And then the pandemic hit.
The Abrupt Shift to Test Optional
Suddenly, in a matter of weeks, the slow debate happening in some states and at some liberal arts colleges about possibly becoming test optional was abruptly and definitely answered. With students unable to take the test due to Covid restrictions, schools were essentially forced to adopt test optional or test flexible policies. By the 2022-2023 academic year, the percentage of colleges requiring test scores dropped to single digits. That year, only 43% of college applicants submitted a score, a drop of more than 30% since pre-pandemic levels.
By fall 2023, some 80+ schools (including the entire University of California system) went a step further and became “test blind”, meaning that students aren’t even given the option to submit test scores.
At the time, many predicted that test optional admissions policies would become a permanent fixture in higher ed. In 2022, Janet Godwin, the CEO of ACT, shared that while she preferred colleges require tests, she was also “not surprised by the test-optional movement. It’s the new normal. It’s here to stay.”
While test optional still continues to be the dominant model at most colleges and universities, some very big name schools have announced plans to return to a requiring standardized test scores from all their applicants. So far, this list includes schools like MIT, University of Texas Austin, Yale, Harvard, Brown, and Dartmouth. While some schools in the “Ivy plus” cohort, like Stanford have committed to test optional for the 2024-2025 admissions cycle, I will be very surprised if we don’t see more highly selective schools reinstating test requirements in the next two years4.
Why return to required test scores?
In announcing the return of required test scores, some of the schools have framed this as an equity issue. According to Hopi Hoekstra, Harvard’s Dean of Dean of the Faculty of Arts and Sciences:
Standardized tests are a means for all students, regardless of their background and life experience, to provide information that is predictive of success in college and beyond. Indeed, when students have the option of not submitting their test scores, they may choose to withhold information that, when interpreted by the admissions committee in the context of the local norms of their school, could have potentially helped their application. In short, more information, especially such strongly predictive information, is valuable for identifying talent from across the socioeconomic range… Fundamentally, we know that talent is universal, but opportunity is not. With this change, we hope to strengthen our ability to identify these promising students and to give Harvard the opportunity to support their development as thinkers and leaders who will contribute to shaping our world
Other schools have echoed this idea that standardized tests can help reveal “diamonds in the rough” and have suggested that this move is, in some ways, a response to the Supreme Court decision making race-conscious admissions practices unlawful.
While I want to assume good intent and that the decision makers at the most high profile colleges and universities are truly trying to figure out how to make college admissions more equitable, I’m also aware that Harvard’s own research has found that the best predictor of how well a student will do on the SAT is to ask for their parent's income information. Given the unequal wealth distribution along racial lines in the US, it’s hard to see how requiring standardized test scores is likely to help shift admissions patterns. Children of wealthy parents are 13 times more likely to score over a 1300 hundred on the SAT than children of lower income parents and are already significantly more likely to be admitted to highly competitive colleges than their peers.
There are lots of reasons for this, of course, but as with MANY things related to highly selective college admissions, it's generally better to be born wealthy than smart.
There are other rationales that have been given for returning to required test scores, including that test scores help identify the students who are most likely to succeed academically. This argument in favor of tests actually points to one of the most essential questions about the role of ACT and SAT in college admissions: do they actually help predict which students are more likely to be successful at a particular college or university?
The answer is pretty simple: it depends who you ask.
According to the folks in charge of the ACT, their research finds a correlation between ACT score and college performance about 75% of the time. For the other 25% of students, they either do better or worse than their scores would predict5. The College Board also argues that SAT scores offer a higher level of predictive ability, with higher test scores leading to improved student retention and likelihood of graduation.
Others, including institutional researchers for higher education institutions and state data systems, argue that unweighted high school GPA is the most predictive tool for measuring the likelihood that a student will succeed academically.
During a 2022 meeting of the Iowa Board of Regents regarding the future of testing requirements, their Chief Academic Officer, Rachel Boon, noted “Our findings continued to indicate that the tests do have some value on predicting first year GPA (grade-point average), but ultimately had a limited relationship to the likelihood of graduation.” Her sentiment echoes findings from researchers at the University of Chicago who found that, for Chicago Public School students, students’ high-school grade point averages are five times stronger than their ACT scores at predicting college graduation.
Some colleges, potentially including the University of North Carolina system, are considering an admissions model that tries to split the difference and will require test scores if students have lower GPAs, so a student with 2.8 would be required to submit but a student with a 3.8 would not.
The truth is that both GPA and high school GPA have some basic predictive abilities: generally speaking, the higher either measure is, the more likely it is that students will succeed academically in college. Of course, some students will be high in test scores and GPA, so if they do well in college, is it their test score or their GPA that predicted it?
Neither measure can fully predict who’ll drop out or not make it to graduation because there are a whole host of other factors that go into that which neither measure can account for (for example: how much a student parties, mental health challenges that might show up during college, new motivation inspired by finding a major that a student is passionate about, etc)
I will say that, in my experience in admissions at a moderately selective public institution, I tended to find test scores the most useful when they were wildly different than I might have predicted from a student’s GPA. I think of the student with a 2.4 GPA and a very solid 34 ACT who might not have been admitted if we’d only looked at the GPA, but whose 34 signaled that there was some significant academic potential that this kid hadn’t explored yet. But outside of the extremes, test scores told me little about a student. There just isn’t that much difference in outcomes for a student with a 21 ACT versus a 24. A student with 30 ACT and one with a 36 are so statistically similar in terms of likely college outcomes that either of them could be admitted to a highly selective college and do just fine.
Grade Inflation and the Return to Test Scores
I suspect that another significant driver for the return to required test scores is the well documented trend of high school grade inflation. Please note that I’m not talking about the absolutely ridiculous trends in weighted GPAs. I’m taking about unweighted GPAs, which have been steadily rising over the the last 15 years and rose noticeably during and after the pandemic. These GPA increase may reflect some impact of schools using pass/no pass grades during the pandemic, as well as more “courtesy As” for students who managed to show up and try in virtual classes.
But it isn’t just the pandemic influencing this trend, as the average high school GPA has increased from a 3.0 in 2009 to a 3.22 in 2021, even while standardized test scores on both statewide assessments and tests like ACT fell during this time. We are now at a point where, across the country, more students get “As” than “Bs”, which is leading to some researchers affiliated with ACT to argue that “As average high school GPA continues to increase, more students are receiving A grades and fewer students are receiving B and C grades. This makes it more difficult to use GPA to understand students’ academic achievement and preparation for college.”
Given that there is also declining trust in the authenticity of other elements of a student’s application (did they write their own essay? did they pay someone else to do it or to edit it? did they use generative AI?), it isn’t surprising that some schools want to lean back into a measure like ACT or SAT scores that feels more objective to them.
The Upside of Requiring Test Scores
While I am personally skeptical that ACT or SAT scores are better predictors of college success than high school GPA (though I do think grade inflation is real), I think there might be some upsides about the return of required test scores at highly selective institutions.
(Please note that I also believe that high selective admissions is NEVER going to be a fully fair or equitable process. It is a process designed to replicate privilege and wealthy students will always have a better shot than other students do)
The first potential upside is that reinstating test scores might reduce the number of students applying to highly selective institutions. When schools like Harvard went test optional, most of them saw a marked increase in their number of applicants. Harvard, for example, saw an almost 10% increase in applicants from 2021 to 2022, a spike that many believe came from students who may have previously self-selected out due to their test scores. For example, a student with an unweighted GPA of 3.9 and an ACT score of 30 or SAT of 1300 might have felt like they had a shot once they didn’t have to submit their very good, but below Harvard average, scores. Given that Harvard didn’t admit more students, this spike in applications drove the already low acceptance rate to under 5%.
Returning to required test scores may thin the applicant pool a bit, which is probably not a bad thing.
The other reason I’m not mad about a return to test scores is that I don’t think selective schools were actually as neutral about scores as they claimed to be. In fact, there is pretty solid evidence that students who didn’t submit test scores were often less likely to be admitted than students who did submit scores. According to one report, “the 2022 acceptance rate at Fordham University was 63 percent among students who submitted scores, compared with 49 percent among those who did not. Similarly, Boston College’s 2022 incoming class recorded an acceptance rate of 25 percent among those who submitted scores and 10 percent among students who did not. This admittance discrepancy holds true for other big name schools, including Barnard, the University of Virginia, Georgia Tech, Amherst, and many more.”
This is partially due to the fact that the students who were most likely to submit scores were submitting high scores, so that strengthened their applications. Dr.Kelly Slay, an assistant professor at Vanderbilt University, has been studying the impact of test optional admissions and noted in an interview with The Hechinger Report that admissions reps are trained to look for and make meaning out of test scores, so when a student doesn’t submit them, the absence is taken as a signal, an indication that the test scores must have been low. When comparing to candidates with similar grades, that absence counts against them. As one admissions rep noted “I think the students that do have the strong test scores still do have that advantage, especially when you have a student that has strong test scores versus a student who doesn’t have test scores and everything else on the academics is more or less the same.”
Finally, while it is still early days in terms of understanding the impact of test optional policies in terms of increasing racial and socio-economic diversity, the early findings are not encouraging. From The Hechinger Report:
Earlier quantitative studies found that the test-optional movement, which has spread to over 1,700 colleges, failed to substantially raise the share of low-income students or students of color. For example, one study published in 2021 found that the share of Black, Latino and Native American students increased by only 1 percentage point at about 100 colleges and universities that adopted the policy between 2005-06 and 2015-16. A separate study of a group of selective liberal arts colleges that adopted test-optional policies before 2011 didn’t find any didn’t find any diversity improvements on those campuses.
Ultimately, I would prefer that more states and schools would either go test blind like California or be more transparent about the fact that even applying without test scores may not actually increase a students chances of getting admitted. As with all things in the admissions world, more transparency is always a good idea.
ACT and SAT are both accepted at every college that accepts test scores that I’m aware of. Whether a student takes ACT or SAT is largely a reflection of geography and students who take both tests generally get comparable scores (e.g. a kid who is in the 90th percentile of ACT scores is likely to be at or near the 90th percentile on the SAT, so there’s not usually a huge advantage to taking both.
This includes schools that were test optional or “test flexible” for students with certain minimum unweighted high school GPAs, which range from as low as 2.5 (University of the Ozarks) to as high as 3.5 (Washington State University)
Many non-highly selective schools (both public institutions and smaller privates) may have requested scores but rarely used them as a reason to deny students admissions, if the student had a satisfactory high school GPA. At a prior college of mine, I know that we admitted students who had ACT scores ranging from 12-15 as long as their unweighted GPA was over the 3.0 mark.
I’m curious about Columbia, however, as they announced in 2023 that they had made a “permanent” decision to stay test optional
It’s worth noting that I don’t think ANYTHING will ever accurately predict for all students. There are too many life variables that can happen in college that can derail success or make some students do better than expected.