The highest score a modern IQ test can give
Every IQ test has a ceiling: the highest score its norm tables contain. For the major clinical tests that ceiling is around 160, four standard deviations above the mean. Statistically, about 1 person in 31,000 would score that high, and about 1 in 2,300 would reach 150.
Even these figures stretch the method. A norm sample of roughly 2,000 to 2,500 people contains at most a handful of people above 145, so scores at the very top are extrapolated from the shape of the curve rather than measured against real peers. A difference between 150 and 158 on a clinical test is within the uncertainty of the method. A short online test, with 25 to 40 items, cannot separate scores above about 145 at all; on our site the result scale stops at 145 for that reason.
The ceiling also depends on age. On children's tests a very able child can answer nearly every item in a subtest correctly, so the score is capped by the test rather than by the child's ability. For that reason the publisher of the WISC has released supplementary extended norms that go above 160. They are used mainly in gifted assessment, rest on very few children at that level and are far less precise than scores in the normal range.
Where claims of 200 and above come from
Most record-breaking numbers trace back to the old ratio IQ: mental age divided by real age, times 100. Under that formula a 6-year-old who performed like a 12-year-old scored 200. Exceptionally advanced children could reach very high ratio scores, but the numbers were not comparable to adult scores and were not rare in the way a deviation IQ of 200 would be. A deviation IQ of 200 would be more than six standard deviations above the mean, rarer than one in a billion, and no test has norms that reach it.
Other claims come from tests designed by enthusiasts for the very high range. These are untimed, often taken at home over weeks, and normed on small groups of volunteers. Their scores cannot be converted to a standard scale with any confidence.
Why there is no official record
Guinness World Records listed a "highest IQ" record for some years and retired the category in 1990, on the grounds that tests at that level are too unreliable to rank people. No scientific body keeps a register of top scores, and clinical results are confidential. Anyone described as having "the highest IQ in the world" is therefore holding a title nobody can verify.
Historical figures: estimates, not test results
Many famous IQ numbers belong to people who never took an IQ test:
- Albert Einstein and Isaac Newton were never tested. The figures of 160 or 190 attached to them are guesses that spread online.
- Johann Wolfgang von Goethe received one of the highest estimates in Catharine Cox's 1926 study: about 185 for his childhood and 200 for his early adulthood. Cox estimated IQs for 301 eminent people from biographical records of their childhood and youth; she said herself that the method was rough and that the figures depended on how much was recorded about each childhood.
- William James Sidis, a child prodigy who entered Harvard at 11 in 1909, is often given a score of 250 to 300. No test record supports it; the number appeared in later popular writing.
- Stephen Hawking, asked his IQ in a 2004 interview, said he had no idea and that people who boast about their IQ are "losers".
These estimates can be interesting history, but they are not measurements and should not be compared with a test score.
How to read a very high result
If an online test gives you a score above 145, read it as "well above average, exact level unknown". The number may be real, but no short test can tell 150 from 165. A practice effect, an easy item set or a test with flattering norms can each add 10 points or more. If an accurate figure at the top matters, for example for a gifted assessment, a psychologist administering a full battery is the only route, and even then the report will give a range rather than a single number.
For a harder check of the top of your range, the Mensa-format matrix test and the WAIS-format test on this site use more difficult items than the standard IQ test; both are modelled on the format of the published tests, not copies of them.
Sources
Cox, C. M. Genetic Studies of Genius, Volume II: The Early Mental Traits of Three Hundred Geniuses. Stanford University Press, 1926.
Roid, G. H. Stanford-Binet Intelligence Scales, Fifth Edition: Technical Manual. Riverside Publishing, 2003.
Wechsler, D. WAIS-IV Technical and Interpretive Manual. Pearson, 2008.
Solomon, D. "The science of second-guessing" (interview with Stephen Hawking). The New York Times Magazine, 12 December 2004.
Frequently asked questions
Who has the highest IQ in the world?
Nobody can say. Clinical results are confidential, no organisation keeps a verified register, and claims above about 160 come from tests that cannot measure that range reliably. Lists online repeat unverified numbers.
What is the highest IQ?
On today's major clinical tests the highest reportable score is about 160, roughly 1 in 31,000 people. Higher numbers exist only on old ratio scales or unstandardised tests.
Who has the highest IQ?
There is no verified answer. Guinness World Records retired its "highest IQ" category in 1990 because results at that level were too unreliable to rank.
What is the highest IQ ever recorded?
No reliable record exists. The 200-plus figures that circulate come from the old mental-age formula used with children, from historical estimates based on biographies, or from tests built by enthusiasts without proper norms.
What is the highest IQ score?
The ceiling on the WAIS-5, WISC-V and Stanford-Binet 5 is about 160. Short online tests stop lower, typically around 145, because they have too few hard items to separate scores beyond that.
What was Einstein's IQ?
Unknown. There is no record that Einstein ever took an IQ test, and the figure of 160 often quoted is a guess, not a result.