SAT to IQ Converter: What Your Score Says About Your IQ, and What It Does Not

How the conversion works

The conversion rests on one idea: a percentile is a percentile. If 81 percent of a population scores at or below your SAT total, the IQ that sits at the 81st percentile of a normal curve is your IQ equivalent. With a mean of 100 and a standard deviation of 15, that score is about 113.

The College Board publishes two percentile columns for the current SAT. The nationally representative percentiles, in the Board's words, are "derived from a research study of U.S. students in 11th and 12th grade" and weighted to represent all students in those grades, "regardless of whether they typically take the SAT." The user group percentiles are based on the actual scores of students who graduated in the past three school years and took the test. Each percentile is the share of students with scores "equal to or lower than" a given score.

The nationally representative column is the better match for IQ, which is normed on the whole population rather than on volunteers. The user column runs lower at the top because test takers are a stronger group than students as a whole. The arithmetic for any percentile is in our IQ percentile calculator.

SAT to IQ conversion table

Percentiles below are the College Board's published figures for SAT total scores. IQ equivalents are computed by converting each percentile to a z-score on the normal curve and applying IQ = 100 + 15z, rounded to the nearest point.

Two readings stand out. A 1200, often treated as a solid score, lands near IQ 113 on the national column: High Average. A 1400 lands near 128, just under the 130 line that Mensa uses on a test with a standard deviation of 15. See IQ 130 for what that threshold means.

Why the top of the table blurs

The College Board reports whole-number percentiles, and that rounding costs precision where the curve is thinnest. On the nationally representative column, every total from 1450 to 1510 is listed at the 99th percentile, and everything from 1520 to 1600 at "99+." A 1460 and a 1510 therefore convert to the same IQ equivalent, about 135, and a perfect 1600 can only be placed above it.

The SAT also has a ceiling: 1600 is the highest score it can record, however far above that a student's ability lies. Ceilings blur rank at the top of any test: Frey and Detterman noted one on the Raven's scores in their own second study. Anyone claiming an IQ of 150 or more from an SAT score is reading more into the table than it holds. For rarity at that level, see the IQ rarity chart.

The research: Frey and Detterman (2004)

The case for treating the SAT as an intelligence proxy comes mainly from Meredith Frey and Douglas Detterman, writing in Psychological Science in 2004 under the title "Scholastic Assessment or g?"

Their first study used the National Longitudinal Survey of Youth 1979. For 917 participants who had taken both the SAT and the Armed Services Vocational Aptitude Battery, the correlation between SAT scores and a general-ability factor extracted from the ASVAB was .82, rising to about .86 when a curvilinear term was added. Their second study tested university students on Raven's Advanced Progressive Matrices; with 103 students left after one outlier was removed, the correlation with recentered SAT scores was .483, which the authors estimated at .72 once corrected for the students' restricted range. Their verdict: "the SAT is mainly a test of g."

Later work added nuance. Thomas Coyle and David Pillow reported in 2008 that SAT scores predict college grades even after general intelligence is removed, meaning the test also captures school-specific skills. For the matrix test they used, see Raven's Progressive Matrices.

Three changes that break old conversions

An SAT score belongs to its era. The scale has been reset several times, and a conversion built for one version misreads another.

• The 1995 recentering. By 1994, real section averages had drifted to about 425 in verbal and 475 in math. Beginning in April 1995, the College Board recentered the scale so that typical performance again landed near 500 per section. A pre-1995 score therefore understates the same performance on the later scale.
• The 2005 expansion. From March 2005 to January 2016, the SAT ran on a 600 to 2400 scale, and analogy questions were dropped.
• The 2016 redesign. In March 2016 the test returned to a 400 to 1600 total, rebuilt around high school coursework. The College Board released concordance tables in May 2016 to translate older scores.

The test then went digital: international centers switched on March 11, 2023, and US administrations on March 9, 2024. The percentile table above applies to the current 400 to 1600 test. For an older score, translate it with the College Board's concordance first, or accept a wider margin of error.

Why an SAT score is not an IQ score

The SAT measures reading, writing and mathematics taught in school. Coursework, coaching, practice tests and test-day conditions all move scores in ways that do not move general ability. Frey and Detterman's correlation of .82 is strong, which also means a real share of SAT variation is something other than g.

The populations differ, too. Even the nationally representative percentiles describe American high school juniors and seniors, not adults of all ages, and the user group reflects people who chose to sit the test, more than 2 million from the class of 2025 alone. IQ norms, by contrast, cover the full population at each age.

An SAT-based IQ is best read as a band of several points either side of the table value. A psychologist-administered test gives a direct measure; our free IQ test is a practice screener, and are online IQ tests accurate explains the difference. For context on the numbers, see what is a good IQ score and what is the average IQ, or browse our IQ guides.

Editorial standard: IQ figures for historical figures are retrospective estimates and are labeled as such; scores for living people appear only when the person disclosed them or a reputable outlet reported them. Corrections: contact the editors.