The College Board has steered its signature test, the SAT, through celebrity bribery scandals, international cheating rings, rain-induced Scantron malfunctions, the rise of the rival ACT, and a steady onslaught of bad P.R. But the COVID-19 pandemic presented a new kind of crisis. Because of lockdowns, 2020 exam dates were cancelled across the world. Hundreds of universities suspended their testing requirements in response. The Washington Post declared “the beginning of the end of our obsession” with standardized exams. The College Board, which brought in more than a billion dollars in revenue in 2019, had considered designing a digital version of the SAT before, without much urgency. Now they scrambled to make it happen. In mid-April of 2020, they announced the development of an at-home version of the test that they internally referred to as the eSAT—“e” for both “electronic” and “emergency.”
By the summer, those plans had fizzled. One reason was the bumpy début of a computerized, at-home version of the Advanced Placement exams, in May and June of 2020, during which thousands of students experienced technical difficulties. Changing course, the College Board designed a kind of hybrid test. The digital SAT, which was announced this past January, will still be administered at testing centers, under proctored supervision, but students will use laptops or tablets—no No. 2 pencils necessary. It runs just two hours long. For a first trial run, last November, the College Board paid five hundred high schoolers from across the world a hundred dollars each to sit for the test. (“I only really did it for the money,” one participant told me.) A few days before Thanksgiving, a dozen senior staff members of the College Board gathered virtually to review the feedback.
The mood was optimistic. “This is my favorite slide,” a staff member said, as she pulled up a summary of the pilot results. “With only one exception”—a test-taker in São Paulo, Brazil, whose laptop had shut down and would not reboot—“every single student who started the exam was able to finish and successfully submit.” Another team member read aloud approving reviews from test-takers in New York City and Mumbai. The majority of students across every demographic had found the new test less stressful.
“We’ve been talking a lot with our comms team about the most persuasive voice for those who are nervous or potentially hostile to change,” Priscilla Rodriguez, the executive who oversees the SAT, said. A polished graduate of Harvard Business School, she wore a white turtleneck and spoke briskly. “They need to hear from their peers to go in open to the idea that this can be an easy and smooth experience,” she added. “They’re not gonna believe it from us.”
The paper SAT is notoriously vulnerable to security breaches and logistical snafus. During the review meeting, one staff member said he’d just learned that a sitting in rural Alaska had been cancelled because the booklets didn’t reach their destination in time. Since 2020, the test has been on hiatus in Egypt owing to leaked exam forms. For the digital SAT, which will roll out internationally next spring and in the U.S. in 2024, questions will be preloaded onto students’ computers and unencrypted on exam day. A teen-ager who shows up late—or needs to troubleshoot a faulty trackpad, as one pilot participant had—will be able to begin after the rest of the group instead of being sent home.
“It’s amazing to compare this to the straitjacket of what we used to do,” David Coleman, the College Board’s fifty-two-year-old C.E.O., said. He wore a bulky headset and had his face pressed enthusiastically toward his screen. He added, “I can smell the relief.”
Rodriguez smiled. “I’ll tell you, David, it’s ‘casual,’ ” she said. “And I put that in quotes, because it’s still an SAT. But it really did feel that way.”
The SAT has been overhauled so many times that one wonders whether it should still bear that name. When the test débuted, in 1926, the “A” stood for “aptitude,” and the questions were designed to gauge inborn intellect. Critics like to point out that Carl Campbell Brigham, the Princeton psychologist who developed the SAT from an Army I.Q. exam, was a prominent eugenicist, and that some élite colleges sought to use it to limit the admission of Jewish students, who were considered workhorses with “ambition but not brains,” as one Ivy League dean put it.
In recent decades, watchdog groups such as the National Center for Fair & Open Testing, or FairTest, have presented research suggesting that SAT questions historically favor affluent white men. One infamous example comes from an eighties test form: How many teen-agers outside the prep-school circuit would know that the analogue to “runner: marathon” is “oarsman: regatta”? Fifty-three per cent of white students and only twenty-two per cent of African Americans answered that question correctly. Since the nineties, the College Board has labored to reinvent the test as a more democratic metric of academic readiness and achievement. The verbal portion has done away with grammatical analogies, and the math portion now favors geometry and algebra over riddle-like quantitative reasoning. The “A” in “SAT,” which was changed in 1994 to denote “achievement,” no longer stands for anything.
Through each iteration of the SAT, one constant has been vehement pushback. “I think I’ve been through half a dozen overhauls,” Robert Schaeffer, of FairTest, told me. A tawny Florida resident, Schaeffer has worked for FairTest since its founding, in 1985, and is now the executive director. Retooling the SAT is “largely rearranging deck chairs, or putting lipstick on a pig,” he said. “In my experience, it’s never made the test a better or fairer predictor, which is what it’s supposed to do.”
In 1969, Bowdoin College became a pioneer in test-optional admissions. “Achievement tests cannot escape cultural bias,” the school’s admissions director said at the time. In the following decades, more than a thousand accredited four-year institutions adopted similar policies. Until the pandemic, though, few of the nation’s most selective schools were willing to disrupt their norms. COVID-19 “forced an experiment,” Schaeffer said. Many of the schools that went test-optional received a surge of applications. Ivy League universities were so inundated that they took an extra week to release 2021 admissions decisions. Acceptance rates, once a bragging right for exclusive colleges, have plummeted so low that some have stopped reporting them.
Jonathan Burdick, the vice-provost for enrollment at Cornell, told the Times that none of the university’s efforts to diversify its applicant pool had been as effective as waiving the testing requirement. It’s less clear, however, whether more diverse applicant pools translate to more diverse enrollment numbers. Burdick told me that thirty per cent of Cornell’s entering freshman class in 2022 chose not to submit scores. The same class included fifty per cent more first-generation students than in the previous year, but only a “modest increase” in students of color. At Amherst College, where roughly a third of admits didn’t submit scores, the proportion of domestic students of color also “increased modestly,” according to the school’s dean of admission and financial aid. Several schools that I contacted declined to provide me with similar data. One admissions officer told me, “We believe the forces at play are complicated and are not easily captured by simple descriptive statistics.”
What’s certain is that test-optional policies have influenced the College Board’s latest rebranding. When the organization announced the digital SAT, in January, it described the exam as a “lower-stakes test” in a “largely test-optional world.” “Simply put, the debate in our society is no longer ‘Should every student be required to supply the SAT?’ ” Coleman told me, one morning last winter. “That debate is over.” We were in his office at the College Board’s New York headquarters, on the seventeenth floor of a high-rise near the World Trade Center. A framed portrait of Martin Luther King, Jr., hung over his desk. With the lighter-weight digital SAT, Coleman said, his goal is to provide “the most widely available, inexpensive way” to help students stand out. He added, dryly, “There’s a part of me, if you want to know, that likes the fact that people will no longer be able to blame the SAT.”
A Rhodes Scholar with degrees from Yale, Oxford, and Cambridge, Coleman made his name in the early two-thousands as an architect of the Common Core standards, the controversial set of K-12 academic guidelines now implemented in more than forty states. When he started at the College Board, in 2012, the SAT was largely administered on weekends, and most students paid a fifty-dollar admission fee. Today, the majority of students sit for the exam free of charge during school hours. But opponents of the College Board believe that even a more accessible test contributes to educational inequality. An article published last year by the National Education Association argued that efforts to root out biased material “often do not detect underlying bias in the test’s form or content.” In Coleman’s view, the SAT merely exposes inequities that already exist. “I wish that banning the test could fundamentally alter inequality in America,” he told me, noting, “We can waive the test, but will we be fulfilling our promise to students?” In a recent essay for The Atlantic, the behavior geneticist Kathryn Paige Harden made a similar point, calling it “ironic” that COVID-19 accelerated the trend toward dropping the SAT, given that “the course of the U.S. pandemic offers a clear lesson: Without tests, the problem is harder to see and harder to solve.”
The SAT will always be an imperfect tool. The pertinent question may be whether it’s less imperfect than other metrics used to measure college readiness. A 2015 report published in The Annals of the American Academy of Political and Social Science noted that, since the seventies, participation in school clubs and sports teams has risen among upper-middle-class students and declined among their working-class peers. New research from a team at Stanford suggests that students’ personal essays are more predictive of family income than their SAT scores are. When I asked Coleman about the Varsity Blues college-admissions scandal, in which wealthy parents paid to have their children’s test scores doctored, he said, “Add up all the cheating scandals any way you want—they don’t compare to the number of students who get substantial help on their essays.”
Universal testing can theoretically provide a more level playing field. In Michigan, the economist Joshua Hyman found that mandating the SAT or ACT increased the number of low-income high-school students who went on to four-year universities. Similar findings have come out of Maine, Illinois, and Colorado. Rodriguez, the College Board executive, often shares her own story by way of example. The daughter of Colombian immigrants, she said that after she took the SAT, in her junior year of high school, she began getting mail from “colleges I’d never heard of, in states I’d never been to,” and earned a number of scholarships because of her scores. (She ultimately matriculated at the University of Virginia, in her home state.) “I believe that the SAT gives students the ability to walk through a door that they may not have even known was there,” she told me.
In 2019, the College Board announced the so-called adversity score, which attempted to contextualize students’ SAT results by calculating their level of disadvantage on a scale from one to a hundred. After a backlash—it was folly to “expect an appended metric to fix an inequality,” my colleague Jiayang Fan wrote—the score was abandoned. Coleman admitted at the time that the College Board had “perhaps overstepped,” but the organization has retained a version of the same tool, called Landscape, which compares an applicant’s test scores with those of others from the same high school. Underlying the College Board’s strategies is a conviction that quantifying students’ records can help promote social justice. Coleman told me he worries that, in a test-free world, disadvantaged kids will get “sorted into futures” that don’t require the skills measured by the SAT. “It’s a terrifying thought in a way, because it almost hints at the idea that we would dare say to any community in this country, ‘You cannot master the math,’ ” he said. In a Times Op-Ed from March, John McWhorter denounced the test-optional movement for imposing “low expectations” on some students of color. For administrators eager to increase diversity, he added, the pandemic seems to have provided “cover for a policy change that some schools wanted to make anyway.”
Universities that have experimented with policy change are now deciding whether to stick with it for the long term. Last month, M.I.T. restored its testing requirement, explaining in a press release that making admissions decisions without scores “tends to raise socioeconomic barriers.” Stu Schmill, the dean of admissions, told me that the SAT can be a “good pathway” for students who lack access to advanced school coursework. In contrast, Harvard announced in December that it would suspend its requirement for four more years. The University of California system has gone even further. In 2019, a largely Black and Hispanic school district in Compton worked with a coalition of students and advocacy groups to sue the system; their lawsuit argued that the use of standardized testing in admissions amounted to discrimination. Last year, as part of a settlement, the U.C. system accelerated plans to eliminate even optional testing from admissions, becoming the highest-profile American institution to go “test blind.”
One winter afternoon, I returned to the College Board’s offices to experience the digital SAT myself. The Omicron variant was surging, and most staff members were still working remotely, but an open MacBook Pro awaited me on a conference table in the “First Generation Room,” which features an enormous mural of Howard Schultz, the C.E.O. of Starbucks, who grew up in public housing in New York City before attending Northern Michigan University. I sat down and followed prompts on the screen to accept the test’s terms and conditions. Then the computer entered “lockdown mode.” A message warned that if I tried to access the Internet or open other applications, my scores would be disqualified.
The digital SAT felt far less gruelling than any other standardized test I can recall taking. A matinée instead of a marathon, it lasted only half as long as the four-hour exam that I sat through as a high-school junior a decade ago. Back then, the test contained ten sections, including one “experimental” portion—you didn’t know which—that wouldn’t count toward your score. Similar to the more recent iteration of the paper SAT, the new test contains a two-part reading-and-writing section followed by a two-part math one (now with a built-in calculator). On the digital exam, the reading-and-writing portions contain roughly fifty short passages, each followed by a single question. The pressure to speed-read page-long articles is gone. So is the sense that failing to crack any individual passage could cost you many points. Mercifully, no questions asked whether a paragraph’s tone could be described as “speculative” or “incisive” or “sardonic.”
At one “content briefing” that I attended, a College Board employee explained that the new test maintains the same “level of complexity” and that “the same number of students are getting the questions right and wrong”; what’s changed is the “perception of the passages being accessible.” Coleman told me, “Intimidation offers no value to us.” Including many short reading passages also gives the digital test’s writers more room to showcase “different cultures, different perspectives, different authorial voices,” as another executive put it. My exam featured passages about the Haitian American novelist Edwidge Danticat, the Chinese film director Chloé Zhao, and the Yankton Sioux writer Zitkála-Šá. A pie chart summarizing the racial and ethnic breakdown of a sample reading-and-writing section showed only eighteen per cent “White American or European Descent.” During the same briefing, one staffer explained that the grammar questions on the test were designed to accommodate speakers of vernacular or nonstandard English and to avoid “matters of contested usage,” including the singular “they.”
Some critics have dismissed the digital SAT as the College Board’s ploy to remain relevant in the face of financial losses. An investigation in Forbes estimated that cancelled testing during the spring and fall of 2020 may have cost the College Board as much as two hundred million dollars in lost revenue. Last January, in what many interpreted as a sign of downsizing, the organization announced that it would scrap all of its subject tests, as well as the SAT’s optional essay section. Neither Coleman nor Rodriguez would quantify the losses that the College Board has suffered as a result of the pandemic, but Coleman cited the “pleasure of being nonprofit.” (Public disclosures from 2020 show that the College Board was operating with more than three hundred million dollars in cash and savings, plus more than eight hundred million dollars in investments.) I asked whether the organization would go out of business if every university in the country ditched its testing requirements. “Not even,” Coleman said. Rodriguez pointed out that, in a recent survey, the College Board found that more than eighty per cent of students still wanted the option to submit test scores to colleges.
Schaeffer, the FairTest director, told me that in theory he’s open to the idea of a purely optional SAT. “If a teen-ager seriously believes that filling in bubbles on a Thursday or Saturday morning adds something to their portfolio, they can send that in,” he said. But he worries that the spectre of standardized-testing requirements will be hard to shake. “For teen-agers whose parents, teachers, older siblings, and guidance counsellors grew up in a world in which the SAT and ACT were mandatory, switching to test-optional is a sea change,” Schaeffer said. He criticized the multibillion-dollar test-prep industry for “scaring kids about the importance of test scores.” (“Don’t cut out the ACT or the SAT,” one instructor from Kaplan warned in a recent Facebook video, pointing out that access to scholarship money, even at test-optional schools, can still depend upon one’s standardized-test scores.) The SAT has dominated the college-admissions process for so long. It has absorbed so much collective anxiety about who gets ahead in our so-called meritocracy. One struggles to imagine that taking the test could feel like anything less than a make-or-break endeavor.
Geovanna Toribio, a senior from Brooklyn, participated in the digital pilot last November, after taking the paper SAT three times. “It was more to the point,” she told me, of the new test. “I wish they did this two years before.” The child of a Dominican Italian father and an Ecuadorian mother, Toribio will attend SUNY Albany in the fall. She hopes to become a police detective. But she told me that her friends who dream of careers in academically cutthroat fields such as medicine and law had been “absolutely terrified of the SAT.” Test-optional policies ended up helping them, she said: “They’re first-generation students like me, and they’re able to go to the colleges of their dreams because they don’t have to submit their SATs.” Yet Toribio also recalled that, when she applied to colleges, she felt pressure to submit her SAT scores to the most competitive schools on her list, even if they didn’t require them. “I was, like, ‘If I don’t turn in my SAT scores, it’s just somebody else that’s gonna get the spot,’ ” she said, adding, “I wanted to show off that, Hey, I took the SAT. I’m not doing test-optional.”
