I teach calculus at Berkeley. Some of my students can’t do middle school math.

In March, two California families opened college admissions decisions that should make anyone who cares about higher education stop and think.

One student — I’ll call him Elias — was admitted to a top University of California engineering program. His family was thrilled though surprised. Math had been his weakest subject in high school, but they assumed the university had seen something they hadn’t.

Another student — Nadia — watched rejection after rejection arrive from UC campuses. One turned her down even though she had earned A’s there in Calculus II and other college-level STEM courses through concurrent enrollment.

These are real students. I’ve changed their names, and withheld their schools, to protect their privacy.

Seen side by side, their outcomes don’t make sense. They make perfect sense only if you imagine admissions working less like careful evaluation and more like chance.

I don’t use that comparison lightly.

I teach mathematics at UC Berkeley. Three days a week, I stand in front of lecture halls with 500 to 1,400 students. By the second week of the semester, I already know who is in trouble. Some students are five to eight years behind in mathematics. Office hours that should be spent discussing integrals instead become lessons on fractions and basic algebra you would expect students to learn in middle school.

That is why this spring and summer, thousands of faculty members across all 10 UC campuses — including more than two-thirds of UC’s mathematics faculty — signed open letters (opens in new tab) urging the system to restore the SAT and ACT in admissions. Five Nobel laureates joined us, along with professors from across the university, including Beatriz Manz, Berkeley’s first Latina full professor.

We did so because what an admission letter is supposed to mean — that a student is ready to attend the school — no longer matches what we see in our classrooms. 

This isn’t a new debate.

In January 2020, a UC task force (opens in new tab) studied standardized testing and recommended keeping it. Then-President Janet Napolitano nevertheless urged the Board of Regents to phase out the SAT and ACT, citing pandemic disruptions and long-standing equity concerns. The regents voted that May to make testing optional, and a court ruling (opens in new tab) later that year, focused on access for students with disabilities, required UC to become completely test-blind through 2025. 

Since then, admissions officers have been forbidden from considering the one standardized measure that provided a common academic benchmark across California’s thousands of high schools.

Many peer universities have since reconsidered. MIT, Stanford, Harvard, and Yale all suspended testing requirements during the pandemic. All have reinstated them. Across the Ivy League, similar reversals are underway.

UC has moved in the opposite direction.

Faculty asked the university to revisit the policy quickly, but the timeline has repeatedly slipped. The Academic Senate now plans a review that will conclude in June 2027, meaning no new admissions policy could take effect before fall 2028.

Students don’t have that kind of time: They attend college once!

The reason so many faculty have spoken out isn’t philosophical. It’s because of what the data — and our classrooms — show.

Before 2020, Berkeley administered a diagnostic exam to incoming Calculus I students. Between 2018 and 2020, 71% of roughly 2,200 students tested as ready or nearly ready for calculus. At the bottom of the distribution there was almost nobody. Just 0.14% tested below basic algebra.

After UC adopted test-blind admissions, the picture changed dramatically.

Among approximately 2,800 Calculus I students tested between 2021 and 2023 with a different diagnostic that measured proficiency in eight mathematical topics, the share who were ready or nearly ready for calculus (meaning, they had proficiency in five to eight subjects) fell to 51%, dropping to 44% by 2023. Most troubling, the bottom of the distribution exploded. Seventeen percent of students answered none of the eight diagnostic topics correctly. By 2022-23, more than one-third had severe math deficits, and the single most common score on the exam was zero.

The classroom itself changed shape.

Before 2020, most students clustered toward the “ready” end of the spectrum, with only a tiny tail at the bottom. Afterward, the middle hollowed out, and the weakest students became the largest single group. We didn’t end up with one somewhat weaker calculus class. We ended up teaching two different classes in one lecture hall: one for students who were prepared for college calculus and one for students who needed remedial mathematics that nobody had scheduled.

That distinction matters because it has been misunderstood publicly. An Aug. 12 article (opens in new tab) in the San Francisco Chronicle said that in the pre-2020 readiness assessment, 48% of students who took the exam were prepared for Calculus I. The article said the faculty reported higher readiness for students admitted under the test-blind policy. But that is incorrect. Our analysis (opens in new tab) defined calculus-readiness as students passing seven or eight out of eight topics on the diagnostic exam. By 2023, that figure had fallen to 26%. More important than any percentage is the distribution itself. The crisis isn’t simply that readiness declined; it’s that the middle collapsed while the bottom became the largest group in the room. The Chronicle didn’t mention that. 

Nor is Berkeley an isolated case. Although Berkeley’s before-and-after diagnostics were not identical exams, UC San Diego (opens in new tab) — which used the same placement test before and after 2020 — has documented the same broad pattern. This is not one professor’s anecdote or one campus’ problem. 

Those of us teaching such courses don’t experience the consequences as abstractions. We watch students struggle through material they were expected to master years earlier. We pause lectures to reteach linear equations instead of expanding students’ understanding of differential equations.

The people arguing this test-blind policy on television, in committee meetings, and in politics do not stand in front of 700 students three times a week or grade a first midterm with a pre-curved D/F average.

We do.

In July, two dozen professors traveled to San Francisco for the UC Regents meeting. Some came from as far as San Diego. Each had one minute to describe what was happening in their classrooms.

Outside the university, school districts have begun acting on their own. On Aug. 6, the San Dieguito Union High School District voted 4-0 (opens in new tab) to urge UC to reinstate standardized math testing for STEM admissions and encouraged other California school districts to consider similar resolutions.

Jennifer Doudna, the Nobel Prize-winning Berkeley scientist known for pioneering gene editing, summarized the concern in a letter to the regents this summer.

“The test-blind policy was a well-intentioned experiment,” she wrote, “but the reality in our classrooms shows that it has failed.” Admitting students who are unprepared, she warned, “sets them up for discouraging struggle. … We cannot afford another cycle of administrative delay.”

That last sentence is at the heart of this debate.

Defenders of scrapping the SATs and ACTs blame COVID or AI for a broad decline in K-12 education. They say those are the only reasons UC students aren’t as ready for math — not the lack of standardized testing in admissions. 

Both those factors surely contributed. California’s K-12 math program isn’t a success story. In 2025, only 37% of students in the state performed at grade level (opens in new tab). 

But neither fully explains what faculty are seeing.

General learning loss would lower the entire distribution. It would not hollow out the middle while piling students up at exactly zero. Something else appears to be compounding the problem. Whatever the mix of causes, the problems at UC accelerated precisely when the system discarded the one standardized baseline it once had. Without standardized tests, admissions officers are flying blind, especially as grade inflation has made transcripts a weaker indicator of readiness, and artificial intelligence has made admissions essays a weaker signal of students’ work. 

It’s also worth remembering that students entering in fall 2027 were in fifth grade when the pandemic began. If, seven years later, many cannot reliably solve a basic inequality such as x² > 4 — a four-answer multiple-choice question that fewer than half answered correctly — then “pandemic learning loss” is no longer doing all the explanatory work.

UC has twice tried to build a replacement for the SAT and ACT. Twice it has failed.

The 2020 task force estimated that a UC-specific admissions test would take nine years to develop. Berkeley Senate leaders projected a cost of $100 million to $150 million (opens in new tab). A second effort examined California’s Smarter Balanced Assessment Consortium (SBAC) — a standardized test that evaluates public school students’ mastery of English and math. In 2021, the Academic Senate concluded (opens in new tab) it was “not appropriate as an admissions test” and would likely cost about as much as the SAT while reproducing many of the same inequities critics hoped to solve.

Yet even after those conclusions, BOARS — the UC faculty committee that oversees admissions — spent much of spring 2026 revisiting SBAC before internal opposition forced that effort to be abandoned and the regents imposed an accelerated review. 

Meanwhile, another admissions cycle passes.

Which brings us back to Nadia and Elias.

Both took California’s standardized math assessment as juniors in 2024. Both later applied to UC Berkeley.

Two real Bay Area high schools put numbers to this pattern. At (opens in new tab)Dougherty Valley High School (opens in new tab) in San Ramon, 432 of 756 juniors exceeded California’s math standard in 2024, as measured by SBAC. Of the school’s 542 Berkeley applicants that fall, 44 were admitted (opens in new tab) — an acceptance rate of 8%.

At (opens in new tab)Mission High School (opens in new tab) in San Francisco, 193 of 202 juniors did not meet the state math standard. Yet Berkeley admitted 34 of the school’s 76 applicants — a 45% acceptance rate.

Based on those figures, at least 174 Dougherty Valley students who exceeded the state benchmark were rejected, and a student with Nadia’s academic record would have been among them.

At least 25 Mission High students who failed to meet the benchmark were admitted, and a student with Elias’ academic record could have been among them.

Even accounting for students who did not take the state assessment, the disparity barely changes.

Today, Nadia is thriving at an out-of-state university and is on track to graduate in three years.

UCB’s own data (opens in new tab) suggest Elias has roughly a coin-flip chance of failing his first calculus course.

I don’t believe either student was well served.

Perhaps Elias would have been better off beginning at a community college before transferring to UC engineering once his foundation was stronger. Perhaps Nadia should have had the opportunity to build her career in the California university system she had already shown she could succeed in.

Instead, both lost something.

The irony is difficult to ignore. UC has tried to make SBAC a replacement for the SAT, yet its admissions decisions often run contrary to those state standardized tests, letting in more students who fail the SBAC. Without SAT or ACT scores — and without faculty input into the admissions criteria — UC’s process struggled to distinguish a student on the edge of failing calculus from one who had already excelled in college-level mathematics.

On the current schedule, no policy change could affect admissions before fall 2028.

That means every student applying until then will pass through the same system that produced Nadia’s and Elias’ outcomes.

Faculty have presented years of classroom evidence. Nobel laureates have spoken. School districts have begun passing resolutions. Yet the university’s response remains another study, another committee, and another delay.

The question is no longer whether more evidence is needed.

It’s how many more students we are willing to gamble with before acting.

Zvezdelina Stankova is a teaching professor of mathematics at UC Berkeley and recipient of the Mathematical Association of America’s Deborah and Franklin Tepper Haimo Award — the nation’s top honor for college math teaching. She is also one of the five original authors of the UC STEM faculty open letter urging the University of California to restore the SAT and ACT.