The SATs are: a) dying; b) already dead; c) alive and well; d) here forever

A student uses a Princeton Review SAT Preparation book to study for the test. | Joe Raedle/Getty Images

The confusing future of standardized testing, explained.

On March 1, Columbia became the first Ivy League university to permanently suspend its longstanding requirement that applicants submit their scores on the SAT. It was the latest in a series of setbacks for the college testing industry.

Between 2000 and 2018, some 200 colleges and universities adopted similar policies. It was hardly a groundswell — there are about 2,300 public and private four-year colleges and universities in the US — but it cracked the door to a different future for standardized testing.

Covid-19 pushed that door wide open. The pandemic scrambled the logistics of test administration and caused most colleges to go “test-optional.” Covid had the same effect on mandatory admissions testing that it had on the practice of requiring white-collar workers to go to the office five days a week: It transformed a growing but not-yet-mainstream trend into a sudden sea change.

The number of test-takers plummeted during the pandemic and has only partially rebounded. Moreover, a sizable number of those who do take the exams aren’t submitting their scores, as policies like Columbia’s become the norm.

Meanwhile, an expected Supreme Court decision outlawing affirmative action admissions policies may give top colleges another reason to pull back from tests that have long played a key role in defining American meritocracy. Going test-optional or test-blind — that is, not even submitting test scores as an option — could be seen by colleges and universities as a way to continue their commitment to diversifying their student bodies in a post-affirmative action world.

Despite these developments, it’s too early to declare the death of college testing. Even as the SAT and its chief rival, the ACT, have become less important in admissions, they are becoming more universal for a different purpose: as a measure of high school achievement. More and more high schools have turned to the SATs and ACTs as their standard assessment tool for their students’ progress, entirely separate from the college admissions process.

The result is a standardized testing landscape that has been shaken up, and whose future looks murky, at best.

The different uses of standardized tests for America’s colleges, explained

While news outlets trumpeted that Columbia had “dropped,” “dumped,” or “ditched” the SAT, those depictions elided a more nuanced truth: Test-optional, Columbia’s policy, does not mean no tests at all.

Indeed, it’s likely that many Columbia applicants will continue to voluntarily submit scores. The only major institutions to go “test-blind” — meaning they refuse to consider tests in any way — are California’s public universities, which opted to do so in 2021 and 2022.

Given their sheer size — California State universities enroll nearly 500,000 students, along with another 280,000 in the University of California system — such a move by itself is a significant blow to admissions testing. But other public systems haven’t followed suit. Students took around 3 million SAT and ACT tests last year, up from 2.8 million in 2021, but down from 4 million in 2019.

The key question is whether test-optional is the new normal or a transition state to test-blind. According to a database maintained by FairTest, an anti-testing organization, fewer than 10 of the colleges that stopped requiring tests in the 2000s went fully test-blind.

According to the College Board, which administers the SAT, the first results of mass test optionality were roughly: 20 percent of students skipped the test, 30 percent took the test but didn’t submit their scores, and 50 percent took the test and submitted their scores. That means that the raw-number drop in the number of tests taken understates the true decline of testing, because it includes a lot of scores that weren’t submitted.

But it’s hard to predict what will happen next, because different colleges use admissions tests in very different ways.

Their stated reasons are often similar — they say they want to make sure students are prepared to succeed in college. While research shows that college success can be mostly predicted by high school grade point averages — unsurprisingly, doing well in school is a good indicator that you’re going to do well in school — grade point averages and tests together are more predictive of college success than GPAs alone. The difference isn’t huge, but it’s real.

But at super-elite institutions, there are probably 10 or 20 students in the applicant pool smart enough to succeed for every one who is admitted. Predicting success is not the issue. The Harvards and Princetons use the SAT more like an IQ test — they want an exam that reliably distinguishes the 99th percentile of smart from the 95th. That’s why the SAT deliberately includes questions that almost everyone gets wrong — and why high SAT scores are still the most widely accepted currency of undergraduate prestige.

For large, mid-tier public universities like the University of Tennessee and the University of Central Florida, standardized test scores serve a different purpose. They remain very useful as a first-order sorting mechanism for qualified applicants. These schools process tens of thousands of applications and typically don’t have the financial resources necessary to give each one a thorough “holistic” review. SAT and ACT scores come in handy in that context.

Then there are the hundreds of less selective public and private colleges — typically institutions facing a sharp, looming demographic decline in the number of new college students. They come closest to using admissions tests for the official purpose of predicting success, because it costs them money when students drop out.

SATs are also an element in the black-box “enrollment management” algorithms that most private colleges, and increasingly many public ones, use to maximize how much tuition students pay. The first wave of test optionality was exclusively a private school phenomenon because it was all about marketing and recruitment, giving students with low scores and generous bank accounts another reason to apply.

Any prediction of where things will go after the mass move to test optionality has to take these complex motivations into account.

What standardized testing is increasingly being used for

Ever since the federal No Child Left Behind Act was enacted in 2002, public schools have been required to administer standardized tests to high school students.

At first, every state developed its own academic standards and tests, but that was pretty quickly revealed to be a bad idea — geometry is geometry, no matter where you live. So states began adopting common standards and exams, an idea that was integrated into an updated law, the Every Student Succeeds Act (ESSA), in 2015.

The College Board and ACT are technically nonprofits. But they make millions of dollars selling tests and saw the new law as a business opportunity. They had already divided up the testing market along regional lines (93 percent of Wisconsin high schoolers take the ACT, for example, compared to 4 percent in California). In the case of ESSA, they had actually lobbied for a provision that allows states to use the SAT or ACT as their required high school test.

Then they lobbied states to adopt their tests, with significant success. There are currently 14 states where more than 90 percent of high schoolers take the ACT, and 10 more that administer the SAT to comply with ESSA, according to the College Board.

And how that test is administered is starting to look different. David Coleman, CEO of the College Board, says the SAT is rapidly changing from a paper-based exam that college-bound students elect to take in high school gyms on weekends to a shorter digital assessment that’s given to everyone as a part of regular schooling. While only 36 percent of SATs were administered during the regular school day in 2018, he says, it will be 68 percent this year.

In other words, the College Board and ACT spent the years prior to a pandemic disruption no one could have foreseen insulating themselves against just such an eventuality, by taking the demand for their tests out of the hands of individual students and colleges and embedding it into public policy. Test-optional may be gaining ground in the college admissions process — but standardized testing is firmly established in American high schools anyway.

Who the SATs help and hurt

As long as states continue administering the SAT and ACT as a matter of course, the tests aren’t going away. And as long as colleges find them useful, they will continue to play a significant role in admissions.

That said, their importance still seems on a downward trajectory.

For testing critics, this is all good. Akil Bello, senior director of advocacy at FairTest, believes that the rise of test optionality shows the exams were never as important as they seemed.

“There isn’t a high school transcript-optional movement,” he observes. “The SAT clearly advantages certain groups of students,” referring to studies that consistently find lower scores for Black and Hispanic students.

But others aren’t so sure that getting rid of the SATs would actually do much — and might even be harmful. Jay Caspian Kang, a New Yorker writer, has argued that eliminating tests is mostly an empty gesture compared to reforms that would actually move the needle on improving equity for underprivileged students.

Meanwhile, Harvard economist Susan Dynarski has written persuasively about the benefits of universal test administration, arguing it can help surface high-performing students. Michigan began giving the ACT to all high school juniors in 2007. “The results were surprising,” she writes. “For every 1,000 low-income students who had taken the test before 2007 and scored well, another 480 college-ready, low-income students were uncovered by the universal test.”

Test supporters also say that eliminating the exams will put too much pressure on the other major determinant of admission: high school transcripts. ACT CEO Janet Godwin cites ACT research that high school grade inflation is a sizable and growing problem.

In 2016, 42 percent of ACT test takers had an “A” average. By 2021, the ranks of A students had grown to 55 percent. If the range of high school grades continues to collapse, they will be less useful for distinguishing students from one another. Essays and teacher recommendations are also subject to pressure from parents determined to help their children, particularly if those factors fill the gap that test optionality creates.

There are two small but significant “divergent” groups for whom test optionality might make a marked difference: students with high GPAs and low test scores, and students with low GPAs and high test scores. We’ll call them “High Grade” and “High Test,” respectively.

High Test students are much more likely to be white, male, and suburban, with parents who are wealthy and college-educated. High Grade students are more likely to be female and Black or Hispanic, from rural areas and households with more poverty and less education. In a 2016 ACT study of divergent students, High Test students were more than twice as likely as High Grade students to come from families earning more than $100,000 per year. High Grade ACT students were three times more likely than High Test students to come from a high-poverty school.

High Grade students are most likely to benefit from test optionality. They’ll be able to put their best feet forward without being penalized for low scores. And according to the findings from a 2021 paper by Yale economist Zachary Bleemer, not only is that good for High Grade students — for whom admission to elite universities is more consequential than it would be for High Test students — but it’s also good for society, because the High Test students displaced aren’t hurt as much as High Grade students are helped.

Where standardized testing goes from here

The diminishment of admissions testing is happening just as the school privatization movement is flourishing in a number of conservative states.

Schools don’t control SAT and ACT scores. They do control grades, to the point that college admissions officers already routinely adjust raw high school grade point averages up and down to make them more comparable among high schools with different academic standards. The more public schools are subject to market pressures, the more they will contort themselves to deliver the grades that families demand.

Coleman points to a College Board research finding that high school grade inflation has grown the most in private schools, with no corresponding increase in SAT scores. “We have to be thoughtful as a society about checks and balances,” he says. “What does it mean to rely on grades when there is no other widely available source of academic information?”

At the same time, selective colleges may have another reason to move away from testing: the imminent destruction of affirmative action.

If the Supreme Court’s conservative majority makes race-based admissions preferences illegal later this year, some colleges will use other means to maintain the desired racial composition of their freshman classes — which could expose them to legal scrutiny. When the Trump administration sued Yale over affirmative action in 2020, it included a table showing the combined test scores and grades for major racial/ethnic groups in the admitted class as evidence that Black and Hispanic students were less qualified.

But under a test-optional regime, such analyses will be less accurate, which would make it hard to point to test scores in a legal challenge against a university’s diversification efforts.

With everything in sudden flux, it can be hard to arrive at a clean takeaway on the messy state of standardized testing. But, complicated though it may be, there’s a case to be made that this new normal strikes a good, if uneasy, balance.

The rise of universal high school testing means that, per Dynarski’s point, more students whose talents were obscured by nonconformity or class bias or something else will have a chance to shine.

The rise of test optionality means that more students with years of solid academic success won’t be hamstrung by a small, standardized snapshot of their whole self.

Neither development will fundamentally change the complex calculi that determine college admissions. But more young people will have the chance to present the best of who they really are.

  Read More 

Advertisements