July 28, 2026

How Colleges Actually Use Test Scores in 2026 Admissions

Elite university building with an empty desk and test paper symbolizing standardized testing in admissions

Seven of the eight Ivy League schools now require an SAT or ACT score. Five years ago, that number was zero. If you last paid attention to admissions testing during the pandemic, when "test-optional" felt like a permanent fixture that would outlast the crisis that created it, the ground has shifted under you, and it happened faster than almost anyone in higher education predicted. Understanding why requires looking past the headlines about individual schools and into the research, the legal backdrop, and the quieter mechanics of how a score actually moves through an admissions file.

The Great Reversal: Why Elite Schools Brought Testing Back

MIT was actually first, announcing in March 2022 that it would reinstate its SAT/ACT requirement starting with applicants for fall 2023 — a full two years before Dartmouth made headlines with the same move. At the time, MIT's dean of admissions argued that standardized tests helped identify talented students from under-resourced schools who might otherwise go unnoticed, a rationale that sounded countercultural in 2022 and became the standard justification by 2025.

Dartmouth followed in February 2024. Within eighteen months, Yale, Brown, Harvard, Cornell, Princeton, Penn, Stanford, Caltech, Georgetown, and Johns Hopkins had all reversed course too, along with public flagships like Ohio State, UT Austin, and Auburn. What started as an outlier decision at one tech-focused institution became the dominant position among the most selective schools in the country.

Columbia is the lone Ivy holdout, sticking with permanent test-optional status even as its peers reverse course, a distinction that's becoming increasingly conspicuous. Here's the current landscape at a glance:

School type	Testing policy for 2026-27	Examples
Test-required	SAT/ACT mandatory	Dartmouth, Yale, Harvard, MIT, Stanford, Ohio State
Test-flexible	Score required, but AP/IB accepted	Yale, some public flagships
Test-optional	Submission encouraged, not required	Chicago, Northwestern, Duke, WashU, Emory
Test-blind	Scores not considered even if sent	University of California system, CSU system

The reversal wasn't a vibe shift or a return to pre-pandemic nostalgia. It followed a specific research finding that admissions deans couldn't easily wave away, and it arrived at the same moment schools were under new legal pressure to justify however they evaluated applicants.

What the Research Actually Found

In January 2024, economists Raj Chetty and David Deming, working with a team including Sarah Friedman, Bruce Sacerdote, Douglas Staiger, and Michele Tine at Opportunity Insights, published an analysis of admissions and transcript data across the Ivy-Plus schools (the eight Ivies plus Chicago, Duke, MIT, and Stanford). The dataset was unusually large for admissions research: hundreds of thousands of student records spanning multiple entering classes, linked to actual college transcripts rather than self-reported outcomes.

The numbers were stark. Students entering with a perfect 1600 SAT or 36 ACT posted college GPAs 0.43 points higher than students who entered around 1200 SAT or 25 ACT — and those lower-scoring students were five times more likely to land a C+ or below in their first year. That gap held up even after controlling for the selectivity of the high school a student attended.

"Students admitted without submitting test scores under test-optional policies achieved GPAs equivalent to applicants who scored around 1300 on the SAT" — a full standardized-test gap between what admissions officers assumed about non-submitters and what actually showed up in the classroom once those students enrolled.

High school GPA, by comparison, barely moved the needle. The gap between a 4.0 and a 3.2 predicted only a 0.1-point difference in college GPA, meaning test scores turned out to be roughly four times more predictive of first-year performance than transcript grades. Part of the explanation is grade inflation: as more high schools cluster students near the top of the scale, GPA loses its ability to separate a genuinely exceptional student from a merely good one, while a standardized test, by design, still spreads scores across a wide range.

And critically, the study found no calibration bias: students from under-resourced high schools with the same test score performed just as well in college as their more affluent peers with that same score. That last finding is the one deans keep quoting, because it undercuts the core argument for dropping tests in the first place — the assumption that scores simply measure family income rather than academic preparation.

Why This Matters After the End of Affirmative Action

There's a second, less-discussed force behind the reversal: the Supreme Court's 2023 ruling in Students for Fair Admissions v. Harvard, which barred colleges from considering race directly in admissions decisions. That ruling didn't mention testing at all, but it reshaped the incentives around it.

Before the ruling, test-optional policies gave admissions offices flexibility to build a diverse class using race as one input among many. After it, schools needed criteria that could withstand legal scrutiny while still identifying talented students from a wide range of backgrounds — and the Opportunity Insights finding that scores predict success consistently across income and demographic groups gave them exactly that. A test score is, in a sense, race-neutral by construction, which makes it legally safer ground than it was in 2019.

This is a genuine tradeoff, not a clean win for either side of the debate. Critics point out that access to test prep, retakes, and even the ability to travel to a testing center still correlates with family resources, so a "race-neutral" number isn't automatically a "resource-neutral" one. Schools that reinstated testing requirements have generally paired the move with expanded fee waivers and free official test-prep partnerships to blunt that criticism, with mixed success depending on the institution.

Test-Optional Doesn't Mean Test-Blind — And That Matters

Here's the part that trips students up every cycle. Even at schools that say "optional," most applicants who get in still sent a score.

During the height of test-optional admissions, roughly 90% of admitted students at selective schools submitted scores anyway. Test-optional, in practice, mostly meant "optional for people whose scores would hurt them" — which is a very different policy than the marketing suggests, and one that created a quiet two-tier system within test-optional applicant pools.

The National Bureau of Economic Research's 2025 digest on this topic flagged a specific irony: test-optional policies ended up hurting high-achieving, disadvantaged students the most, because that group submitted scores at unusually low rates, effectively hiding evidence that could have helped their applications. A first-generation student with a 1450 SAT, for example, might assume a strong score matters less at a test-optional school and withhold it, not realizing that number would have meaningfully strengthened their file relative to their GPA alone.

A simple decision framework, if you're applying somewhere test-optional this cycle:

Pull the school's mid-50% range from its Common Data Set or admissions site.
Compare your score to the 50th percentile, not the 25th. If you're at or above it, submit.
If you're below the 25th percentile, withholding usually helps — but only if your GPA, rigor, and essays are doing real work elsewhere.
Reach schools are the exception: a slightly below-median score can still add context a reach school wants, especially paired with an upward grade trend.
Recruited athletes, legacy applicants, and homeschooled students face different calculus — homeschoolers in particular often benefit from submitting scores even when below-median, since a test result is one of the few standardized data points a school has to compare them against a traditional transcript.

Don't confuse "optional" with "irrelevant." That's the mistake that costs applicants admits they were actually qualified for.

How a Score Actually Gets Read: Superscoring and Flexibility

Not every 1450 is treated the same way, and that surprises a lot of families who assume a score is a score everywhere.

More than 60% of selective colleges now superscore the SAT — taking your highest section scores across multiple sittings and combining them into a new composite. A growing number extend the same courtesy to the ACT. Harvard is the notable exception here: it evaluates scores from a single sitting only, so gaming the system with a retake-heavy strategy doesn't help there the way it might at, say, Vanderbilt, where a student could take the test three times and have the admissions office assemble their best math score from one sitting and their best reading score from another.

Test-flexible policies add another wrinkle. Yale, for instance, lets applicants submit AP or IB exam results instead of an SAT or ACT score, provided they clear a minimum threshold across multiple subject tests. It's testing, technically, just not the testing you were expecting, and it opens a path for students at schools with strong AP programs but limited access to SAT test dates.

None of this means the score is the whole application. Admissions officers still read essays, weigh course rigor, and consider recommendation letters — a score just sets the range they're reading everything else against. Think of it less as a gate and more as the frame around the picture: it doesn't determine what the rest of the file looks like, but it shapes how generously that file gets read.

Beyond Admissions: Money and Placement

Scores don't stop mattering once you're admitted. Two uses get overlooked constantly, and both can have a bigger dollar impact on a family than the admission decision itself.

Merit scholarships frequently run on published score thresholds, especially at public flagships and regional privates. The University of Alabama publishes tiered SAT/GPA cutoffs that scale from partial tuition to full tuition plus a stipend, and it's not alone — the University of Kentucky, University of Arizona, and dozens of other publics run similar automatic-award systems that a family can calculate before ever submitting an application. Roughly speaking:

Scores in the low 1200s start unlocking entry-level institutional merit awards, particularly at public universities.
The 1350–1450 range opens larger award tiers at many mid-selective privates.
Scores above 1500 frequently trigger a school's top merit tier, sometimes stacked with honors-college admission.
The most reliable target is your target school's 75th percentile among admitted students — that's where the bigger money tends to live.

Course placement is the quieter use case. Community colleges, which are mostly open-enrollment and don't touch test scores for admission, often still use SAT or ACT results to place students directly into credit-bearing math and English courses, skipping costly non-credit developmental coursework entirely. A student who never planned to use their SAT score for admission at all can still save a semester of tuition by having it on file for placement purposes.

The Public University Divide

If there's one thing that separates 2026 from 2020, it's a genuine, durable split among public systems, not just private schools quietly disagreeing with each other.

The University of California and California State University systems remain test-blind — they won't consider a score even if you send one, a policy the UC system locked in after years of litigation and internal review, including a faculty task force that ultimately recommended against reinstating testing despite the same predictive-validity data other schools cited to go the opposite direction. Meanwhile, Southern public flagships went the opposite direction: Florida, Georgia, Alabama, and Ohio State all reinstated requirements, citing the same Opportunity Insights research driving the Ivy reversals.

This isn't a strictly red-state, blue-state story, but the geographic pattern is hard to miss. It also means a student applying to both UC Berkeley and Georgia Tech in the same cycle is navigating two entirely different philosophies of what a test score is even for — one system treats it as a source of bias to be engineered out, the other treats it as the single clearest antidote to bias in the file.

Bottom Line

Check each school's specific policy every cycle — don't assume last year's list still applies, since dozens of schools flipped their stance in the past eighteen months alone.
If a score sits at or above a school's median, send it, even where optional. The data shows admissions officers reward transparency, not omission.
Look past the admission letter. Scores can unlock merit money and skip remedial coursework, so check scholarship thresholds and placement policies separately from admission requirements.
Understand superscoring and single-sitting rules before you register for a retake — a strategy that helps at one school can be wasted effort at another.
The single biggest shift for 2026: elite schools now treat test scores as a fairness tool for under-resourced applicants, not a barrier to them. That reframing, backed by hard transcript data rather than intuition, is why the pendulum swung back so fast.

Frequently Asked Questions

Is it true that test-optional colleges don't care about scores at all?

No, and this is the most common misconception in the entire topic. During peak test-optional years, around 90% of admitted students at selective schools still submitted scores, and research shows students who withheld strong scores may have hurt their own chances. Test-optional means the school won't reject you solely for not submitting, not that scores are ignored when present.

How do I decide whether to submit my SAT or ACT score to a test-optional school?

Compare your score to the school's mid-50% range, usually published on its Common Data Set or admissions website. If you're at or above the median, submit; if you're well below the 25th percentile, withholding is usually the safer call, provided the rest of your application — GPA, rigor, essays, recommendations — is strong enough to carry the file on its own.

Will more colleges reinstate testing requirements after 2026?

Likely, though not universally. Momentum from the Opportunity Insights research and the Ivy League's near-unanimous reversal is pushing more selective private and public flagships toward requirements, but schools like Chicago, Northwestern, and the entire UC system have signaled they're staying test-optional or test-blind for the foreseeable future, in some cases citing entirely different research conclusions from the same underlying data.

Do test scores matter for financial aid, or just admission?

Both. Many schools, especially public flagships and mid-tier privates, publish explicit SAT/ACT thresholds tied to merit scholarship tiers, separate from whatever policy governs admission itself. A strong score can unlock money even at a school where submitting it for admission was optional.

What's the difference between test-optional and test-blind?

Test-optional means you can choose whether to submit a score, and admissions officers will consider it if you do. Test-blind, the policy at UC and CSU schools, means officers won't look at a score even if you send one — it simply isn't part of the file they review, no matter how strong it is.

Does superscoring mean I should take the SAT multiple times?

It depends entirely on the school. Over 60% of selective colleges superscore, combining your best section scores across sittings, which does reward multiple attempts. But schools like Harvard evaluate only a single sitting, so check each target school's specific policy before planning a retake strategy around it — two or three attempts at a superscoring school can meaningfully raise your composite, while the same effort at a single-sitting school mostly just adds stress.

Why did the Supreme Court's affirmative action ruling affect testing policy?

The 2023 SFFA v. Harvard ruling barred colleges from considering race directly in admissions, pushing schools toward criteria that predict success consistently across demographic groups without explicitly weighing race. Since the Opportunity Insights research found test scores work the same way regardless of a student's background, reinstating testing became an attractive, legally defensible way to keep identifying talented students from a wide range of backgrounds.