IQ Percentile Calculator

IQ is a relative score, not a raw measurement. Test publishers collect raw scores from a representative norming sample, then transform that distribution so the population mean is exactly 100 and one standard deviation equals a fixed gap (15 points on Wechsler and Stanford-Binet 5, 24 points on Cattell Culture Fair). Because of that rescaling, an IQ of 115 always means 'one standard deviation above the mean of the norming sample', regardless of the underlying test difficulty. This calculator uses the same convention: it converts your score to a z-score against the chosen scale's mean and SD, then reads off the percentile from the standard normal distribution.

The choice is historical. Wechsler (WAIS, WISC) and modern Stanford-Binet 5 both standardize to SD 15, which is the most common modern convention. The Cattell Culture Fair test, designed by Raymond Cattell in the 1940s to minimize verbal and cultural content, uses SD 24, which is a wider spread. The same z-score maps to the same percentile across scales, but the IQ numbers look different: a score of 130 on Wechsler (z = 2.0, 97.7th percentile) corresponds to about 148 on Cattell. This calculator handles the conversion automatically when you switch the scale dropdown, so the classification labels stay comparable.

Mensa accepts the top 2 percent of the general population, which is the 98th percentile cutoff. Under a normal distribution with mean 100 and SD 15, the 98th percentile lands at IQ 130 (z = 2.054). On Cattell's SD 24 scale the same 98th percentile lands at IQ 148. The threshold is not arbitrary: it is rare enough to define a small selective group while still being measurable on standard supervised tests without major ceiling effects. Mensa only accepts scores from approved, in-person, professionally administered tests, so any online IQ result, including the estimate from this calculator, will not satisfy their entry requirements.

Well-normed IQ tests have a measurement standard error of roughly 3 to 5 points, which means a single sitting can read 5 to 10 points away from your 'true' score in either direction. Accuracy is also pulled down by test anxiety, sleep, illness, and unfamiliarity with the format. At the high end (above IQ 145) most tests run into ceiling effects, where the highest possible items can no longer distinguish between very high scorers. Cultural and linguistic bias remains a real issue for traditional Wechsler-style tests in non-native English speakers, which is part of why the Cattell Culture Fair scale exists. Treat any single IQ number, especially one from an online or self-administered test, as a rough estimate, not a diagnosis.

A percentile rank is the share of the norming population that scored at or below a given value. If you land at the 84th percentile, 84 percent of people in the norming sample scored at or below you. It does not mean you got 84 percent of the questions right or that you are '84 percent smart'. Percentiles are not evenly spaced in IQ points either: the jump from the 50th to the 84th percentile is the same as the jump from the 84th to the 97.7th (one standard deviation in both cases), but the percentile gap is 34 versus 13.7. That is why small IQ differences near the average matter less than the same number of points near the tails of the distribution.