To Test or Not to Test? What Research Has to Say About the SAT and ACT

Sep 14, 2026

Uploaded by Jiven

Newsletter examining whether SAT and ACT scores should be used in college admissions, especially in light of pandemic-era changes and concerns about fairness. Includes views from FairTest and analyses of survey findings, noting correlations with socioeconomic status and unequal learning opportunities. Summarizes a research review in Education Policy Analysis Archives based on 72 peer-reviewed articles, weighing both predictive value and risks of differential predictions and coaching effects.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages3

Reading Time8 min

AccessFree to download

To Test or Not to Test?

Pandemic and access concerns

Evidence from surveys and advocacy groups

Research review: balanced findings

What decision-makers should consider

Thursday, August 7, 2025 Newsletter To Test or Not to Test? What Research Has to Say About the SAT and ACT During the pandemic, many colleges and universities stopped requiring students to submit SAT or ACT scores, as lockdowns made it challenging to access testing centers. The decisions were applauded by those who view the exams as unfair barriers to college access, especially for students from low-income families, English learners, and other groups of students who earn, on average, lower scores. “What do the tests tell us?” FairTest director Harry Feder said last year.“They tell us how wealthy your parents are. To near perfect correlation.” A 2024 survey by FairTest, which aims to promote “equitable and reasonable assessment,”found that 80 percent of four-year institutions do not require either exam for admissions. However, there is evidence that the pendulum may swing back in the opposite direction, as officials at some highly selective colleges have expressed concerns that, rather than making their processes fairer and more equitable, the elimination of exam requirements has actually disadvantaged applicants from low-income families. “We found that applicants without scores from lower socioeconomic backgrounds were less likely to be admitted than others, because they were less likely to have that evidence in other parts of their applications,” Yale’s dean of undergraduate admissions told the Christian Science Monitor. The inequitable opportunities to learn that result in lower test scores also [http://nepc.colorado.edu/publication/newsletter-sat-act](http://nepc.colorado.edu/publication/newsletter-sat-act) 1 of 3 result in other unequal outcomes relevant to applications. But on the question of whether use of these tests in the admissions process is useful and equitable, there may not be a clear answer. A recent research review led by NEPC Fellow Audrey Amrein-Beardsley of Arizona State University offers evidence that supports both maintaining and eliminating use of the exams. Published in April in the peer-reviewed journal, Education Policy Analysis Archives, the piece incorporates 72 peer-reviewed articles published between 1969 and 2019. Amrein-Beardsley initiated the study as a result of being asked to serve as a witness for the plaintiffs in Smith et al. (Plaintiffs) v. Regents of the University of California, Janet Napolitano, a 2020 case that was settled in 2021 after the system agreed to halt the use of the ACT and SAT in admissions decisions at least through 2025. Plaintiffs in that suit alleged that the use of the exams in admissions decisions: • Could be eliminated because high school grade point averages—which the plaintiffs described as having “less adverse impact on disadvantaged groups”—predict college outcomes, including first-year grades, just as well as if not better; • Gave a leg up to applicants who could afford test coaching; • Used a normative score distribution that “artificially compare[d] students to one another” rather than “measuring individual merit” and “tended to iteratively discard test items on which ‘underrepresented minority students perform[ed] well,’ creating bias;”and • Led to a “starkly disparate [set] of student outcomes.” Amrein-Beardsley and her coauthors did find evidence to support these contentions. For instance, they note that the scores’ ability to predict individual students’ college outcomes“is smaller than often assumed, and often smaller than that of high school GPAs.” The scores are best at predicting college outcomes when considered alongside high school GPA and class rank, they suggest. They also found that the tests “can yield unfair measures of academic abilities and merit, especially when considering the potential impacts of coaching or training by different groups of students.” And they described as “troubling” their finding that: [B]oth the SAT and ACT yield differential predictions for sets of dissimilar students. These differen

FAQs

Why did many colleges stop requiring SAT/ACT scores during the pandemic?

Lockdowns made it difficult to access testing centers. Colleges paused requirements amid concerns about unfair barriers to college access.

What does the FairTest analysis suggest about how SAT/ACT relate to socioeconomic status?

FairTest’s director argues the tests reflect how wealthy parents are, and a 2024 survey indicates many institutions do not require either exam for admissions.

What does the NEPC research review conclude about keeping or eliminating SAT/ACT use?

The review presents evidence supporting both approaches: the tests are psychometrically strong but can produce differential predictions that may disadvantage racial minority and socioeconomically disadvantaged students when used for admissions.

Sep 14, 2026

Uploaded by Jiven