Does the SAT measure intelligence?

Key takeaways
- On older versions of the SAT, scores correlated strongly with intelligence, about .82 in one 2004 study.
- In College Board's data, three or more Bluebook practice tests went with scores about 61 points higher.
- Treat a low score as a list of skills to work on.
- SAT scores plus high school GPA predict first-year college GPA better than either alone.
The short answer is partly. Scores on older versions of the SAT tracked general intelligence closely, and one widely cited study concluded that the SAT was mainly a measure of it. But today's SAT is built to test reading, writing and math skills taught in school, and those skills respond to practice. Your child's score report shows which of those skills to work on next.
Why the SAT became associated with IQ
The test's ancestor was the Army Alpha, a group intelligence test used to sort military recruits during World War I. Carl Brigham, a psychologist who worked on the Army Alpha, adapted it into the Scholastic Aptitude Test in the 1920s, and the word "aptitude" implied the test measured something innate.
Over the decades, the SAT moved away from that framing. It dropped analogies, added more curriculum-aligned content, and eventually changed its name to "SAT" (no longer an acronym for anything). The current digital SAT tests reading comprehension, grammar and usage, and math through algebra, geometry, and trigonometry. College Board describes the digital SAT as a measure of "students' skills and knowledge" for college and career readiness.
But the association with intelligence stuck. Mensa, the high-IQ society, still accepts old SAT scores (from tests taken through January 31, 1994) as proof of qualifying intelligence. One widely cited study, Frey and Detterman (2004), found a correlation of about .82 between SAT scores and a measure of general intelligence drawn from the military's ASVAB test, and about .72 between SAT scores and Raven's reasoning test after adjusting for the narrow range of scores in its college sample. The authors concluded that the SAT is "mainly a test of g," psychologists' term for general intelligence. Their data came from older versions of the SAT, not from the current digital test.
What the correlation means and what it doesn't
A correlation that strong means students with higher IQ scores tend to score higher on the SAT, the way taller people tend to weigh more. Height still measures one thing and weight another.
The overlap comes from shared skills. Both IQ tests and the SAT reward reading comprehension and working through structured problems under time pressure, so a student strong in those tends to do well on both. The SAT also tests a published set of school subjects that students can study directly. A student who hasn't covered trigonometry will miss trigonometry questions regardless of their IQ.
The bigger difference is what practice does. Scores on both kinds of test rise when people retake them. A 2018 meta-analysis of 122 studies found significant gains when people retook cognitive ability tests, leveling off after the third try. On the SAT, study adds to that. In College Board's 2025 study of the digital SAT, students who completed three or more full-length Bluebook practice tests scored about 61 points higher than similar students who took none, and about 79 points higher among students whose PSAT score was 900. College Board notes those estimates aren't causal. A practice test also shows which question types a student misses, such as transitions or systems of equations, and each of those can be practiced on its own.
The table below compares the two tests. The IQ column uses one IQ test for school-age children, the Wechsler Intelligence Scale for Children (WISC-V), as the example.
| Feature | Digital SAT | IQ test (WISC-V) |
|---|---|---|
| What it covers | Reading and Writing; Math (algebra, advanced math, problem-solving and data analysis, geometry and trigonometry) | Verbal comprehension, visual-spatial skills, fluid reasoning, working memory and processing speed |
| Who takes it | High school students applying to college | Children aged 6 to 16 |
| Length | 98 questions in 2 hours 14 minutes, plus a 10-minute break | 10 primary subtests; the core subtests take about 60 minutes |
| Scores | 400 to 1600 total; 200 to 800 per section | Index scores on a scale where 15 points is one standard deviation |
| How strongly the two line up | One study (Frey and Detterman, 2004) found a .82 correlation between older SAT scores and a general-intelligence score from the ASVAB in 917 people, and .72 between SAT scores and Raven's matrices in 104 undergraduates after correcting for their narrow score range | |
Sources: College Board, SAT structure; College Board, Digital SAT Suite specifications overview; Pearson, WISC-V; Frey and Detterman, Psychological Science (2004). Checked October 29, 2026.
How to use a low score to plan practice
The most damaging version of the SAT-IQ myth is the one that tells students their score is fixed. If a student believes the SAT measures innate intelligence, they have less reason to prepare. Studying for a test that measures fixed ability looks pointless.
The SAT tests a specific set of skills, and those respond to practice. Learning the grammar rules the SAT tests lets a student answer those questions more reliably, and timed practice on algebraic word problems builds speed. Four Bluebook practice tests also make the format familiar.
Read a low score as a list of skills that need work. A student who sees a 1050 and thinks "I need to work on transitions and algebra" has a clear next step, and our Bluebook score report guide shows how to find those skills. We've also written about how SAT prep builds skills students use in college.
What the SAT is designed to predict
College Board designed the SAT to predict first-year college grades, and its own research backs that up. In College Board's 2019 national validity study of more than 223,000 students who entered college in fall 2017, SAT scores and high school GPA correlated with first-year college GPA at .51 and .53, respectively, after adjusting for range restriction. Used together, the correlation rose to .61, so the two predict college performance better than either one alone. We've written about how colleges weigh SAT scores alongside grades.
Today's SAT isn't designed to measure intelligence, creativity, leadership, motivation or the other qualities that shape long-term success. It measures the skills and knowledge students need for college-level coursework, and colleges read it alongside grades, essays and everything else in the application.
Once a practice test shows which skills cost your child points, Sharp builds their practice around those skills.
Kim's take
A first diagnostic score mostly tells me how unfamiliar a student is with the test. The idea that the SAT measures IQ hurts strong students who score lower than they expected, because they decide they aren't smart enough and stop trying. Usually, the second practice test after a few weeks of focused work is the first score that means much.

Kim Strauch, SAT Tutor & Founder
Kim scored a perfect 1600 on the SAT, graduated summa cum laude from Dartmouth, and founded Sharp after years of tutoring students one-on-one.
How we write and fact-check our postsFrequently asked questions
Can you convert an SAT score to an IQ?
Not meaningfully. Researchers, including Frey and Detterman, have published equations for estimating IQ from SAT scores, but they rely on data from versions of the SAT that no longer exist. The current SAT has different content and a different scoring scale, so any SAT-to-IQ conversion is rough at best.
If SAT scores can improve with prep, does that mean IQ can too?
Somewhat. Scores on cognitive ability tests also rise when people retake them, and a 2018 meta-analysis found those gains level off after the third try. The SAT also tests content a student can study directly, like grammar rules and algebra.
Is a low SAT score a sign of low intelligence?
Not on its own. A low score shows which skills your child hasn't mastered yet and how familiar they are with the format and timing, and practice can change both.
Should I tell my child their SAT score doesn't matter?
No. The score matters for admissions and scholarships. Treat it as where your child stands today on skills that practice can raise, and keep it separate from their worth or intelligence.
Sources
- College Board: History of the SAT (2009)
- College Board: SAT structure
- Pearson: Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V)
- Frey & Detterman (2004): Scholastic Assessment or g? The Relationship Between the SAT and General Cognitive Ability
- American Mensa: Qualifying test scores
- College Board: Understanding SAT Scores
- College Board: Digital SAT Practice Tests and SAT Score Gains (2025)
- Scharfen, Peters & Holling (2018): Retest effects in cognitive ability tests: A meta-analysis, Intelligence
- College Board: Digital SAT Suite specifications overview
- College Board: National SAT Validity Study (2019)



