LSAT to IQ: What LSAT Scores Really Say About Intelligence
LSAT to IQ conversion is impossible: no peer-reviewed study links LSAT scores to IQ. See what the LSAT really measures and what the evidence supports.
Dr. Russell T. WarneChief Scientist
Share
No peer-reviewed study has ever correlated LSAT scores with IQ, so there is no valid way to convert an LSAT score to an IQ score. Every "LSAT to IQ" table online is built on guesswork rather than data. The real evidence is still worth knowing: the LSAT is a highly reliable reasoning test, it correlates .85 with the GRE among people who took both, and high-IQ societies accept top LSAT scores as admission evidence. This article lays out what the LSAT measures, why the missing study matters, what the indirect evidence shows, and what the percentile math honestly supports.
What the LSAT measures, according to its maker
The Law School Admission Council scores the LSAT on a scale from 120 to 180 and describes it as assessing "the comprehension of complex texts with accuracy and insight," the ability to manage information and draw reasonable inferences from it, critical thinking, and the analysis and evaluation of others' reasoning. LSAC never calls it an intelligence test. Since August 2024 the scored sections are two Logical Reasoning sections and one Reading Comprehension section; the famous "logic games" section was retired after a settlement with blind test takers who could not draw the diagrams the puzzles favored.
Whatever it is called, the LSAT behaves like a serious cognitive instrument. LSAC reports that test forms typically have reliability coefficients above 0.90, a figure in the same territory as professional IQ tests, and that the LSAT is the single best predictor of first-year law school grades, ahead of undergraduate GPA. That predictive power is remarkably stable: in the very first validity study in 1950, the LSAT alone correlated .44 with first-year grades, results LSAC describes as remarkably consistent with what it observes today.
The study that would justify a conversion does not exist
Turning LSAT scores into IQ scores would require a linking study: give the same people the LSAT and a full IQ battery, then publish the equation. For the SAT, that study exists. Frey and Detterman (2004) found the SAT correlated .82 with the general intelligence factor from a military test battery in 917 people, which is why our SAT to IQ conversion article can discuss real published equations. For the LSAT, no such study has appeared in any peer-reviewed journal. What circulates instead is blog arithmetic and calculator sites mapping taker percentiles straight onto IQ percentiles, a move that fails for reasons the next sections make concrete.
Researchers do treat the LSAT as a member of the cognitive-ability family. Kuncel and Hezlett (2010) class it among "standardized tests of cognitive abilities" alongside the SAT, GRE, and MCAT, and show that the common objections to these tests, such as the claim that they merely proxy socioeconomic status, do not survive the evidence. Being in the family is not the same as having a published IQ conversion, and for the LSAT only the first of those is true.
The best indirect evidence: the GRE-LSAT bridge
The strongest data point comes from ETS. To help law schools accept the GRE, ETS studied 1,587 admitted law students at 21 law schools who had taken both tests. GRE verbal and quantitative scores predicted LSAT totals with a correlation of .85, about what is observed when people simply retake the same test. Two different admissions exams, built by different organizations, measure nearly the same thing in the same people. That is strong evidence of a shared general-reasoning core.
The same study is also a warning. Even with real data on real dual-takers, ETS refused to publish a one-to-one conversion and instead reports predictions with an error band of about 5 points on the 120-to-180 scale. If the best-resourced testing organization in the world will not claim single-point precision between two tests it has actual data for, a website converting LSAT scores to IQ with no data at all deserves no confidence.
What high-IQ societies and percentiles honestly show
American Mensa accepts the LSAT as admission evidence today: a score at the 95th percentile qualifies for tests taken after 1982 (a 662 on the old scale before then). There is a revealing wrinkle in that rule. Mensa's standard is the 98th percentile of the general population, yet it accepts the 95th percentile of LSAT takers, implicitly conceding that people who sit for the LSAT are a smarter-than-average pool. The Triple Nine Society, aiming at the 99.9th percentile, currently requires a 179 for tests taken from January 2021, and its cutoff has moved era by era (177 for 2010 through 2020) as the score distribution shifted. Even the societies re-anchor constantly; a static conversion table cannot be right. Our guide to high-IQ societies covers the full admission rules.
The official LSAC percentile table for 2021 through 2024 makes the selection problem explicit. These figures are rarity among LSAT takers, not IQ equivalents: a 170 beats 95.69 percent of takers, a 173 beats 98.15 percent, and a 180 beats 99.90 percent, while the median taker scores about 152. If takers were a random sample of the population, a 173 would sit near the 98th percentile, roughly IQ 131 territory on a standard-deviation-15 scale. But takers are college graduates who chose law, a positively selected group, so the true population standing of a 173 is higher than that arithmetic suggests, by an amount no published study quantifies. The defensible conclusion is a direction, never a point estimate, which is also why the average IQ of lawyers runs well above 100.
A coachable score is not an IQ reading
LSAT scores respond to preparation; that is the premise of an entire industry. The effect is real enough to see in the brain: Mackey, Whitaker, and Bunge (2012) found that about three months of intensive LSAT preparation changed white-matter connectivity in frontal and parietal regions, with larger score gains tracking larger changes. Training sharpens the specific reasoning skills the LSAT rewards. It does not raise general intelligence, and a number that moves with a prep course cannot double as a stable IQ.
If the goal is an actual IQ score, the tool is a test built and normed for that purpose. The RIOT IQ test (the Reasoning and Intelligence Online Test), developed by intelligence researcher Dr. Russell T. Warne, states its uncertainty the way LSAC and ETS do and conversion tables never do: the free version gives an estimate in about 8 minutes with a margin of error of 14.9 IQ points, the Basic version narrows that to 5.6 points in about 13 minutes, and the Full version reaches 3.7 points across 15 subtests in about 52 minutes. An honest instrument publishes its error band.
Frequently asked questions
Is the LSAT an IQ test?
No. LSAC describes it as measuring reading, reasoning, and argument skills for law school. It is highly reliable and cognitively loaded, and researchers group it with cognitive-ability admissions tests, but it is not normed on the general population.
What IQ is a 170 or 175 LSAT?
No published conversion exists. What can be said: a 170 beats about 95.7 percent of LSAT takers and a 175 about 99.0 percent, and the population equivalent is somewhat higher than those figures suggest, imprecisely.
Does Mensa accept the LSAT?
Yes. A 95th-percentile score qualifies for tests taken after 1982, and a 662 qualifies for earlier tests. Triple Nine accepts era-specific cutoffs, currently 179 for tests taken from January 2021 onward.
Did removing logic games change what the LSAT measures?
Barely, by LSAC's numbers. Across more than 200,000 test sessions, dropping the section changed the mean score and the correlation with first-year grades by about a hundredth of a point each, though it did remove the test's most abstract task type.
If I raise my LSAT score with prep, did my IQ go up?
No. Prep builds test-specific skill, and research shows it can even alter brain connectivity, but trained performance on one exam format is not a change in general intelligence.
References
1. Educational Testing Service. (2018). Frequently asked questions about the GRE Comparison Tool for Law Schools. ets.org
2. 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. doi.org/10.1111/j.0956-7976.2004.00687.x
3. Kuncel, N. R., & Hezlett, S. A. (2007). Standardized tests predict graduate students' success. Science, 315(5815), 1080-1081. doi.org/10.1126/science.1136618
4. Kuncel, N. R., & Hezlett, S. A. (2010). Fact and fiction in cognitive ability testing for admissions and hiring decisions. Current Directions in Psychological Science, 19(6), 339-345. doi.org/10.1177/0963721410389459
5. Law School Admission Council. (2024). LSAT percentile table, 2021-2024 testing years. lsac.org
6. Law School Admission Council. (n.d.). LSAT reliability and validity in brief. lsac.org
7. Mackey, A. P., Whitaker, K. J., & Bunge, S. A. (2012). Experience-dependent plasticity in white matter microstructure: Reasoning training alters structural connectivity. Frontiers in Neuroanatomy, 6, 32. doi.org/10.3389/fnana.2012.00032
8. American Mensa. (n.d.). Qualifying test scores. us.mensa.org
9. Triple Nine Society. (n.d.). Qualifying test scores. triplenine.org
