Report on the role of standardized test scores in undergraduate admissions [pdf]
A new Dartmouth report finds that SAT and ACT scores meaningfully predict first‑year college GPA and can help identify high‑potential students from disadvantaged schools, challenging the recent shift to test‑optional admissions. Commenters debate how strong this predictive power really is, whether standardized tests are closer to IQ measures or just proxies for socioeconomic status and test prep, and how they compare to grades, essays, and extracurriculars as admissions signals. The thread also explores equity trade‑offs, arguing that removing tests may unintentionally favor affluent applicants while leaving deeper structural problems in K‑12 education and access to support unaddressed.
Predictive Power of SAT/ACT
- Many commenters say it’s unsurprising SAT/ACT correlate with first‑year GPA; both largely test ability to study known material and perform on exams.
- Several view SAT/ACT as rough proxies for “g”/IQ plus conscientiousness; others caution they differ from formal IQ tests and are highly coachable.
- Some note prior work (cited in the Dartmouth report) linking scores to later GPA, grad school, and earnings, but this specific report only validates first‑year GPA.
Critiques of the Dartmouth Analysis
- Skeptics highlight that SAT explains ~22% of variance in first‑year GPA, calling that “modest” and insufficient to claim test scores are “highly predictive.”
- Figure 1 (binned scatter) is criticized as visually overstating correlation and lacking variance information.
- Others respond that 22% is substantial in social science, and that SAT outperforms high‑school GPA (~9% of variance) within Dartmouth’s already filtered pool.
Equity, Class, and Race
- One camp argues standardized tests are among the fairest scalable tools:
- Less manipulable than GPAs, essays, extracurriculars, recommendations, which all heavily favor affluent families and legacy/donor applicants.
- Critical for spotting high‑ability students from weak or unknown schools.
- Another camp stresses structural bias:
- Wealth buys prep, stability, and time; disadvantaged students may lack sleep, food, quiet, or support.
- Test‑optional moves have sometimes pushed admissions toward résumés and essays that are even more class‑skewed.
- Strong debate over using neighborhood or school‑level SES to “boost” scores:
- Supporters see it as correcting for unequal opportunity.
- Critics see it as de facto race‑ and class‑based engineering, vulnerable to gaming and arbitrary weighting.
What Should Admissions Optimize For?
- Some emphasize predicting academic success and graduation; others question the relevance of GPA to job performance or “real‑world” success.
- Traits like grit, persistence, creativity, and work ethic are widely seen as important but hard to measure at scale.
- Several argue elite universities primarily function as prestige filters and social sorters rather than pure educators, so tests are valuable as transparent, comparable signals.
Broader System Concerns
- Discussion touches on grade inflation, social promotion in K‑12, capacity limits at elite schools, legacy admissions, and whether access should be expanded versus rationed by increasingly fine‑grained metrics.