Should Colleges Use Test Scores as Admissions Criteria?

How Admissions Got Here

Free write (5 min): should colleges require test scores? Ask for the reasoning, not just the answer — this is a topic where students have strong instincts and thin arguments.

The context first. College attendance has grown dramatically — steeply for men from the 1920s and for women from the 1940s — though enrolment dipped slightly in the past decade, and more women than men now attend, reversing the historical pattern. Elite admissions rates have fallen steadily as class sizes stayed flat. Worth noting though: while most institutions have grown more selective since 2002, about a third are actually admitting more students. Selectivity is not a universal trend.

Then the history of admissions criteria, which shows how recent and how contingent the current system is:

• 17th–18th centuries — applicants screened on character, family background, and knowledge of Latin and Greek.
• 19th century — requirements proliferate and vary wildly by school.
• 1926 — the first SAT, created to standardise requirements amid a broader movement for mass intelligence testing. That origin is worth naming.
• Mid-20th century — national standardisation around coursework, class rank, recommendations, and interviews.
• Late 1950s — colleges lean hard into tests, treating them as at least as good a predictor as a student's educational background.
• 1960s onward — personal characteristics enter the process; in 1978 the Supreme Court upholds race as a factor; in the early 2000s elite schools adopt "holistic" review.

Where things stood before COVID: the Common App is used by over 900 colleges; in 2019, 55% of them required test scores and 76% of applicants reported them.

Assign the readings, a deliberately balanced set, due before Tuesday:

• Q&A: Stuart Schmill on MIT's Decision to Reinstate the SAT/ACT Requirement — MIT News
• Critics Foresee Armageddon if Colleges Don't Use SAT/ACT Scores. Here's Why That's Wrong — Washington Post
• Didn't Ace the SAT? Just Design Microbe Transplant Research — New York Times
• In Praise of Aptitude Tests — The Economist. Note: published in 2005, and worth flagging as such.

The Gaps — and the Slide That Complicates Them

Mindful moment (5 min), then the heart of the topic, in this exact order. The sequence matters more here than in any other Education topic.

• SAT scores by family income and parental education. Students from higher-earning, more-educated families score higher on average. There are also gaps by race and ethnicity.
• A statistical detail worth catching: the top distributions show "fat tails" and appear cut off at 1600. That means those groups are skewed further toward high scores than a normal distribution would produce — practically, some of those students would score above 1600 if the scale allowed it.
• Then high school GPA, across the same demographics. The same patterns appear. Higher-income and more-educated families, higher GPAs; the same racial and ethnic gaps. And GPAs have drifted upward over time — grade inflation, visible across all groups.

The Natural Experiment COVID Ran

Mindful moment (5 min), then the material that makes this deck unusual: an actual large-scale test of the counterfactual.

• Before the pandemic, 46% of US undergraduate institutions did not require test scores. For the 2021–22 cycle that rose to just under 80%, with roughly 60% confirming the policy would continue.
• Score reporting collapsed — the share of Common App applicants reporting scores fell by more than 40% between 2019–20 and 2020–21. Two explanations, both plausible: students couldn't sit the exam, and students chose not to report scores they thought were too low.
• The decline was steepest among under-represented minority and first-generation applicants — precisely the groups test-optional policies were meant to help. Again, two readings: less time and fewer resources to prepare and test, or lower scores and a decision not to report them.
• Applications rose across the board, and most at the most selective institutions. Without a score to screen themselves out, more students took a shot.

The discussion: what should we conclude from the test-optional experiment? One reading is that it was a genuine opening for students with fewer resources. Another is that selective colleges are now admitting students without the information that predicts whether they'll cope with the coursework. Both are defensible from the same data.

Before students leave: themes, lingering questions, and one new resource for Thursday.

Design the Admissions Process

Mindful moment (5 min), resources shared, then the deck's own framing — and note that it explicitly tells students their answer need not be a hard yes or no:

• Should colleges use test scores? Does your answer depend on which college — MIT against a small liberal arts college? (MIT reinstated its requirement, and the assigned reading explains why.)
• What are the short- and long-term effects of using or dropping tests — on students, colleges, employers, and society?
• Design an ideal admissions process. How would you weigh applicants from very different backgrounds? Which factors are genuinely fair?
• The deck's sharpest question: to what extent are people's opinions about standardised testing shaped by how well they personally did on it? Does that matter?
• And underneath all of it: what is college admission for? Should institutions admit the most academically prepared, those with the greatest potential to benefit, or those with the greatest need for upward mobility? Those three answers point in different directions.

Score Your Participation

The four reflective questions, then the participation matrix. Students also use the participation rubric to score the week and explain the score.