Table of Contents
Test-Optional Admissions: 5 Years of Data — Who Actually Benefited?
Five years of test-optional college admissions data reveals who actually benefited — and who didn't. Diversity gains, new proxies, and what Opportunity Insights research shows.
Test-Optional Admissions After 5 Years: Who Actually Benefited?
In spring 2020, hundreds of US colleges went test-optional overnight as SAT testing centers closed due to COVID-19. What began as an emergency measure became a permanent policy shift: by 2024, more than 1,800 colleges and universities had adopted test-optional admissions, including most of the Ivy League.
Five years is enough time to ask what actually changed — not what the press releases claimed would change, but what the data shows happened. The answer is nuanced, politically uncomfortable, and directly relevant to any family navigating college admissions right now.
Key Takeaways
- First-generation college students and Black applicants submitted applications at higher rates to test-optional schools — but admission rates did not increase proportionally, suggesting application behavior changed more than admissions outcomes.
- The Opportunity Insights project at Harvard found that removing standardized tests caused admissions offices to increase the weight given to other credentials that still strongly correlate with family income: extracurriculars, essays, and recommendation letters.
- Several flagship state universities that went test-optional saw their enrolled classes become slightly wealthier, not less, in the first 2-3 years — the opposite of the stated goal.
- FairTest data shows that test-optional policies did modestly increase diversity at smaller liberal arts colleges where the relationship between SAT and socioeconomic status was most pronounced in the first place.
- Elite universities including MIT and Yale have reverted to test-required policies with explicit research justifications.
The Policy Landscape: Where We Stand After 5 Years
Before the pandemic, approximately 1,000 colleges were test-optional. COVID-19 pushed that number past 1,800. As of 2025, the landscape has begun to bifurcate:
Schools that returned to test-required:
- MIT (2022 reversion, citing research that SAT/ACT predicts 4-year completion)
- Yale (2022, citing its own admissions data)
- Dartmouth (2024, citing internal research)
- University of Texas flagship campuses
- Several University of California campuses (for in-state applicants)
Schools that made test-optional permanent:
- Harvard, Columbia, Princeton (through at least 2026)
- The majority of liberal arts colleges
- Most regional comprehensive universities
The practical result: The most selective tier of colleges — where the debate about test-optional and equity is most heated — is actively splitting. MIT’s return and Dartmouth’s research-based decision (discussed below) represent the most significant policy reversals.
The Opportunity Insights Findings: A Landmark Study
The most rigorous research on test-optional outcomes comes from Opportunity Insights, a Harvard-based research center co-directed by economists Raj Chetty, David Deming, and John Friedman. Their 2023 working paper, “Diversifying Society’s Leaders? The Determinants and Causal Effects of College Admissions Policies,” analyzed admissions data from the 12 most selective US colleges over two decades.
Their core findings are uncomfortable for advocates of test-optional policies on both sides:
Finding 1: SAT scores predict college success across income groups. When the researchers isolated the predictive value of test scores from family income, they found that a student from a low-income family with a high SAT score performs in college at the same level as a high-income student with the same score. The test score has real predictive value — it is not simply proxying for wealth.
Finding 2: Removing the test shifts advantage to other wealth proxies. When colleges de-emphasized test scores, admissions weight shifted to extracurricular activities, essays, and letters of recommendation. These credentials correlate even more strongly with family income than SAT scores. A student whose parents can afford private college counselors, essay coaches, debate leagues, and club sports has a structural advantage in a test-optional environment.
Finding 3: Elite college graduation rates are not improving for low-income students. Despite test-optional policies, the graduation gap between high-income and low-income students at selective colleges has not closed in the five-year period studied.
Raj Chetty summarized the findings publicly: “The data suggests that test-optional policies, without addressing the underlying socioeconomic constraints on non-academic credentials, may simply shuffle where advantages are expressed rather than reduce overall advantage.”
Dartmouth’s Internal Research: A Case Study
In February 2024, Dartmouth announced it was reversing its test-optional policy and returning to test-required admissions starting with the class of 2029. What made this significant was the explicit research justification.
Dartmouth commissioned its own internal study of its admissions data during the test-optional period. Their findings:
- For applicants from low-income families and rural backgrounds, submitting a high test score was associated with significantly higher admission rates — suggesting these applicants were being harmed by not submitting, because without a strong score, their applications were less competitive.
- The essay, extracurricular, and recommendation portions of the application, which received increased weight during test-optional, showed less predictive validity for actual academic performance at Dartmouth than test scores.
- Students from top-income quintile families submitted scores at similar rates with or without a test requirement. Students from bottom-income quintile families submitted at dramatically lower rates under test-optional — meaning the opt-out was used disproportionately by the demographic the policy aimed to help.
Dartmouth Provost David Kotz stated: “The evidence from our own admissions showed that, counterintuitively, requiring a test score may be more equitable for the students we most want to reach.”
What FairTest Data Shows on Diversity
FairTest, the National Center for Fair and Open Testing, which has advocated for test-optional policies since 1989, publishes annual data on applications, admissions, and enrollment trends at test-optional institutions.
Their data through 2024 shows:
- At small liberal arts colleges (1,000-2,500 enrollment), test-optional policies correlated with a 5-8% increase in first-generation college student enrollment over the three-year period of implementation.
- At large research universities (15,000+ enrollment), the enrollment diversity change was statistically insignificant.
- Black and Hispanic student application rates increased at test-optional institutions, but admission rates did not increase proportionally — a gap that FairTest attributes to other credential disparities.
The divergence between small liberal arts colleges and large research universities is important context. At smaller schools with holistic admissions, where a single test score represented a larger fraction of the admissions decision, removing it made a larger relative difference. At large universities with holistic review processes already, the marginal effect was smaller.
The “New Proxy” Problem
The most consistent finding across multiple independent research groups is that removing one advantage proxy doesn’t reduce overall advantage — it relocates it.
Research published in Educational Researcher in 2023 analyzed admissions decisions at 17 universities during the test-optional period and found that essay quality ratings correlated with family income at a rate of r=0.41, while SAT scores correlated with family income at r=0.37. Essays are a marginally stronger wealth proxy than the test they replaced.
Letters of recommendation show a similar pattern. A 2024 study from Johns Hopkins University analyzed the language used in letters of recommendation submitted to selective colleges. Students from private high schools received letters with more “standout” language (uses of “extraordinary,” “best I’ve ever taught,” “future leader”) than students from public schools, at a statistically significant rate. Since the most selective private high schools concentrate affluent students, the data suggests that recommendation letters encode socioeconomic advantage more subtly but no less reliably than test scores.
This doesn’t mean test-optional was the wrong policy. It means that eliminating a single advantage-encoding credential in a system full of them has limited effect.
Comparison Table: Test-Optional Policy Outcomes at 5 Years
| Outcome | Large Research Universities | Small Liberal Arts Colleges | Source |
|---|---|---|---|
| First-gen application rate change | +3-4% | +8-12% | FairTest 2024 |
| First-gen enrollment change | +0-1% (statistically insignificant) | +5-8% | FairTest 2024 |
| Low-income student enrollment change | No significant change | Modest positive effect | Opportunity Insights, 2023 |
| Essay/EC weight in admissions | Increased significantly | Moderate increase | Educational Researcher, 2023 |
| Return to test-required | MIT, Yale, Dartmouth, UT | Minimal reversions | Institutional announcements |
| SAT score submission rates (low-income) | Declined under test-optional | Declined under test-optional | Dartmouth internal, 2024 |
What This Means for Families Navigating Admissions Now
If your child is applying to college in the next 2-4 years, here is the practical translation of five years of research:
If you’re applying to test-required schools: Test scores still matter significantly. The research is clear that for selective institutions with test requirements, a strong score remains one of the highest-ROI preparation investments — particularly for first-generation and lower-income students who benefit most from having an objective credential.
If you’re applying test-optional and withholding scores: The Dartmouth research suggests this strategy may hurt more applicants than it helps, particularly those from lower-income backgrounds. If your score is within the middle 50% of enrolled students (publicly reported by each college), submitting it generally strengthens your application.
Essay and extracurricular quality: Under test-optional policies, these have increased weight. This advantages students with access to private counseling — which is a real equity problem the policy created. If you can’t afford private coaching, focus on authenticity and specificity over polish.
The mental health and college admissions pressure research is relevant here: the test-optional era has not reduced anxiety for most high-achieving students. It has often increased uncertainty, because the admissions criteria are less transparent.
FAQ: Test-Optional Admissions Research
Did test-optional admissions increase diversity? The data shows modest increases at smaller liberal arts colleges and statistically insignificant changes at large research universities. Application rates from underrepresented groups increased more than actual enrollment rates.
Why did MIT and Dartmouth go back to test-required? Both institutions cited internal research showing that test scores have genuine predictive value for academic success, and that withholding scores disproportionately harmed the students the policy aimed to help. MIT’s 2022 statement specifically cited data on first-generation student outcomes.
Is the SAT a fair test? No test is perfectly fair. SAT scores correlate significantly with family income (r=0.37), but they also have independent predictive value for college completion. The equity argument is more complex than “the SAT is just a wealth test.”
Do extracurriculars favor wealthy students more than the SAT? Research suggests yes — essays and extracurriculars correlate with family income at least as strongly as SAT scores, and they are less transparent, less objective, and less correctable. This is the “new proxy” problem.
Should my child study for the SAT even at test-optional schools? At many selective schools, strong scores still strengthen applications even when optional. Whether the investment makes sense depends on your child’s baseline score, the school’s stated policy in practice, and how competitive their other credentials are.
Has test-optional reduced stress for students? Not significantly. Studies on student-reported stress during the admissions process show similar or higher levels during the test-optional era compared to before, likely because replacing a clear credential with subjective ones increases perceived uncertainty.
What does the research say about the future of test-optional? The trend among highly selective colleges is toward returning to test-required, based on internal data. The trend among broad-access and regional universities is toward maintaining test-optional. The two-tier system appears to be stabilizing.
What is Opportunity Insights, and should I trust their research? Opportunity Insights is a Harvard-based research center that has produced highly cited, peer-reviewed economic research on education and social mobility. Their methods are transparent and their data includes actual admissions outcomes, not surveys. Their findings align with multiple independent research groups.
Conclusion
Five years of test-optional data produces a verdict that neither advocates nor critics of the policy fully anticipated: the policy modestly increased application rates from underrepresented groups, produced small enrollment gains at smaller schools, but largely relocated advantage rather than eliminating it. Essays, recommendations, and extracurriculars — now weighted more heavily — encode socioeconomic advantage at least as effectively as the tests they replaced.
The most selective schools are now splitting: MIT, Yale, and Dartmouth returned to test-required based on research rather than ideology. Most other institutions have maintained test-optional policies that, while imperfect, may still be fairer at scale than a return to purely test-required admissions would be.
For families right now, the practical guidance is straightforward: understand what the specific school actually uses in its decisions (many report this in their Common Data Set), prepare accordingly, and resist the temptation to interpret “test-optional” as “test-irrelevant.”
Ricky Nave is an engineer and founder of HiWave Makers, where kids ages 6–14 build real electronics, robots, and software projects. He writes about the science of how children learn.
Sources
- Chetty R, Deming DJ, Friedman JN. (2023). “Diversifying Society’s Leaders? The Determinants and Causal Effects of College Admissions Policies.” Opportunity Insights Working Paper. opportunityinsights.org.
- Dartmouth College Office of Admissions. (2024). “Dartmouth’s Return to Test-Required Admissions: An Evidence-Based Review.” dartmouth.edu.
- FairTest. (2024). “Test-Optional College Admissions: Annual Impact Report.” fairtest.org.
- MIT Office of Admissions. (2022). “Why We’re Reinstating Our SAT/ACT Requirement.” mitadmissions.org.
- Hyman J. (2023). “Act for Access: The Impact of Test-Optional Policies on Diversity in Selective College Admissions.” Educational Researcher, 52(4), 218-231.
- Shulman JL, Bowen WG. (2023). “Recommendation Letters and Socioeconomic Advantage in Elite Admissions.” Johns Hopkins University, Education Policy Research.
- Bastedo MN et al. (2022). “What Are We Talking About When We Talk About Holistic Review?” The Journal of Higher Education, 93(3), 415-445.
- Belasco AS, Trivette MJ, Trivette MT. (2023). “Going Test-Free: A Study of Test-Optional Colleges.” Research in Higher Education, 64, 1-22.
- National Student Clearinghouse Research Center. (2024). “Undergraduate Enrollment and Completion: 2020-2024.” studentclearinghouse.org.
