The SAT's average student is not the population's average person
An SAT percentile ranks you against the students who chose, or were required, to take it. Many teenagers never do. If takers average a little above the population on the abilities both tests share, the median SAT taker is not an IQ-100 person, and a straight rank match runs low.
Nobody publishes the exact gap, and it changes as states adopt school-day testing (which pulls in nearly everyone) or colleges go test-optional (which thins the pool). The slider lets you set the assumption and see how much it matters.
For an SAT of 1350: shifting the pool by 0.00 SD adds about 0 IQ points to every equivalent. Bars run from IQ 55 to 160.
Five scales have carried the same three letters
A new test with new content and a new scale. The College Board published concordance tables to translate older scores.
Two hundred simulated students with your IQ in the crowd
Each dot is one simulated person, drawn so SAT and IQ correlate at r. Dots within 5 points of IQ 124 light up; their SAT spread is the honest answer to "what would I score?"
Frey and Detterman (2004): "Scholastic Assessment or g?"
917 participants in the 1979 National Longitudinal Survey of Youth had both SAT scores and the ASVAB. SAT correlated about .82 with a g factor extracted from the ASVAB.
Undergraduates took Raven's Advanced Progressive Matrices. After correcting for the narrow ability range of a college sample, SAT and Raven correlated about .72.
The authors concluded the SAT works largely as a measure of general ability and offered equations to estimate IQ from pre-1995 SAT scores. Those equations do not carry over to later scales.
Figures checked against the published abstract (Psychological Science 15(6), 2004). Study 1 used the original pre-1995 SAT; Study 2 used the revised, recentred SAT. The paper's IQ-conversion equations were calibrated on these older scales, not on today's digital SAT.
SAT totals and their rank-matched IQ
Naive match (no pool adjustment), current 1600 scale. Click a tile to load it above.
- Frey, M. C., & Detterman, D. K. (2004). Scholastic Assessment or g? The relationship between the Scholastic Assessment Test and general cognitive ability. Psychological Science, 15(6), 373–378.
- College Board (2023). SAT Suite of Assessments Annual Report: Total Group (class of 2023). Mean total about 1028.
- Dorans, N. J. (2002). Recentering and realigning the SAT score distributions: How and why. Journal of Educational Measurement, 39(1), 59–84.
- College Board (2016). Concordance tables: redesigned SAT to old SAT.
- College Board (2024). Digital SAT score scale and reporting.