The Highest IQ Ever Recorded: Separating Fact From Claim

What People Mean by "Highest IQ Ever Recorded"

Search for the "highest IQ ever recorded" and you'll find lists of historical figures supposedly scoring 200, 250, even close to 300. These numbers get repeated across quiz sites and trivia articles, but almost none of them come from a documented, standardized test administered under controlled conditions by a qualified examiner. Most trace back to biographical estimates, childhood test results calculated with an outdated formula, or simple internet folklore that has been copied from one site to the next without verification.

The Guinness World Records Case

The most frequently cited real example is Marilyn vos Savant, who was listed in the Guinness Book of World Records during the late 1980s under a "Highest IQ" category, based on a childhood Stanford-Binet test. Guinness later retired that category altogether, stating that IQ tests are not precise enough at the extreme upper end to support a meaningful "world record." That single decision from the record-keeping organization most associated with the claim is itself strong evidence of how the psychometric community views scores far above the norm: not as a reliable ranking, but as a number that stops meaning very much once you get far enough from the center of the curve.

Why Extremely High Scores Are Unreliable

The Ceiling Problem

IQ tests are built and normed around a mean of 100 with a standard deviation of 15. That design works well for describing where most people fall — about 68% of the population scores between 85 and 115, and around 95% falls between 70 and 130. But very few people in any norming sample score above 145 or 160, so test publishers simply don't have enough data at that extreme to calibrate scores precisely. Above a certain point, a test can tell you someone answered almost everything correctly, but it cannot reliably tell you how far above the ceiling their true ability sits.

Ratio IQ vs. Deviation IQ

Many of the highest historical claims come from an older method called "ratio IQ," once used in early Stanford-Binet testing on children: mental age divided by chronological age, multiplied by 100. A gifted seven-year-old answering questions written for a fourteen-year-old could generate a startlingly high ratio score — but that arithmetic breaks down completely once applied to adults, and it has been abandoned in favor of the deviation IQ scoring used by modern tests, which compares a person only to same-age peers.

Standardization at the Extremes

A trustworthy IQ score requires a standardized instrument, a controlled testing environment, and a large, representative norming sample. Self-reported scores, scores from unsupervised online tests, or scores calculated from an unusual or discontinued instrument don't meet that bar — no matter how impressive the resulting number looks.

What "Above Average" Actually Looks Like

Rather than chasing an unverifiable "record," it's more useful to understand the practical scale most tests share. On the standard 100-mean, 15-SD scale, a score of roughly 130 sits near the 98th percentile — the threshold commonly used by high-IQ societies such as Mensa for membership eligibility. That's already a genuinely rare result. Claims of scores 60, 80, or 100 points beyond that should be treated with real skepticism, since no widely accepted, properly normed instrument can support that level of precision.

How IQTesta Approaches This

IQTesta's free test is designed to give you a quick, indicative estimate of your reasoning performance — a fun and informative benchmark, not a clinical or diagnostic assessment, and not a Mensa admission test. If you want a rigorously documented score, especially at the higher end of the scale, that requires in-person testing with a licensed psychologist using a professionally normed instrument. Our goal is simpler: to give you an engaging, evidence-informed snapshot of where your results fall on a well-understood, normally distributed scale — without pretending to chase a record that psychometrics itself has largely stopped taking seriously.