Should Colleges Use Test Scores as Admissions Criteria?
This is the rare week where you're not studying a system from the outside — you're in it. SAT scores line up closely with family income and parents' education, which looks like a knockout argument against the test. Then you'll see the next chart: high school GPA does exactly the same thing. And COVID accidentally ran the experiment, when most colleges dropped the requirement and something surprising happened to who reported scores.
Reading List
Two of these defend testing and two attack it. The MIT one is worth reading closely whatever you think — it's a university explaining why it brought the requirement back after dropping it.
Required
- 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 the date: 2005. Ask yourself what's changed.
If You Want More
- The Problem with America's College Entrance Exam — Vox (video)
- What's Less Fair Than the SAT? You Might Be Surprised — LA Times
- The College-Admissions Process Is Completely Broken — The Atlantic
One-Week Plan
Here's the shape of the week. Your guide will set the exact timing for your section. One thing worth saying up front: nobody will be asked to share their own test scores. The topic is the policy, not you.
How Admissions Got Like This
Write first: should colleges require test scores? Give a reason, not just an answer.
Then some context. Far more people go to college than a century ago — and for the first time in history, more women than men. Admissions rates at elite schools have fallen steadily because class sizes stayed the same while applications multiplied. But here's a fact that gets lost: while most colleges have become more selective since 2002, about a third are actually admitting more students. The frenzy is concentrated at the top.
Then the history, which shows how arbitrary the whole thing is. Early American colleges screened applicants on character, family background, and knowledge of Latin and Greek. The first SAT was given in 1926, partly as an attempt to standardise a chaotic process and partly as part of a broader enthusiasm for mass intelligence testing. Colleges leaned hard on tests from the late 1950s. Personal qualities entered in the 1960s. "Holistic" review is barely twenty years old.
None of it is natural. All of it was decided by people, fairly recently.
The Gaps, and the Chart That Complicates Them
First, SAT scores broken down by family income and by parents' education. The pattern is clear and steep: more money and more parental education, higher average scores. There are gaps by race and ethnicity too.
Take a moment there. Does that prove the test is biased? Write down your answer before the next slide.
Then the next slide: high school GPA, broken down exactly the same way. Same patterns. Same gaps.
That doesn't mean the test is fine. It means the explanation has to be bigger than the test. Maybe both measures are biased. Maybe both are picking up real differences in preparation caused by very unequal schools and circumstances. Maybe some of each. Whichever you land on, "get rid of the SAT and it's fixed" doesn't survive this pair of charts — and figuring out what would help is a much more interesting problem.
The Experiment COVID Ran by Accident
Before the pandemic, 46% of US colleges didn't require test scores. By the 2021–22 cycle, that was just under 80%. For once, we got to see what happens.
- Score reporting fell by more than 40%. Some students couldn't take the test; others could have reported a score and chose not to.
- The drop was sharpest among first-generation and under-represented minority applicants — the students test-optional policies were supposed to help most. Worth thinking hard about why.
- Applications went up everywhere, and most at the most selective colleges. Without a score telling them not to bother, more students applied.
So what did we learn? One reading: it was a real opening for students without expensive test prep. Another: selective colleges are now choosing without the information that helps predict who'll handle the workload. Both come from the same data, which is why this is a discussion and not a lookup.
Before you leave, find a resource to bring on Thursday.
Design It Yourself
Your answer here doesn't have to be a hard yes or no — the deck says so explicitly.
- Does it depend on the college? MIT reinstated its requirement and explained why. Would you give the same answer for a small liberal arts college?
- What happens long-term, to students, colleges, employers, and everyone else, if tests go away — or stay?
- Design your own admissions process. How would you compare two applicants with completely different backgrounds? Which factors are actually fair — and does your alternative also favour students with money?
- An uncomfortable one worth answering honestly: how much does your opinion depend on how well you personally test? Does that matter?
And the question underneath everything: what is college admission for? Admitting the best-prepared students, the students who'd gain the most, or the students who most need a way up are three different goals — and they'd produce three different classes.
Score Your Own Week
What went well, what to improve, what you learned, and whether anything you believed changed. Then the participation matrix. Be honest with yourself.