IQ score chart: every range and percentile

One table answers most questions about IQ scores: where a score sits on the bell curve, what percentage of people it beats, and how rare it is. This page gives you that table for every score from 55 to 160, then explains the machinery behind it so the numbers actually mean something.

The classification bands at a glance

The pattern to notice: the bands are symmetric around 100, and each 15 point step matters more than the last. That is the bell curve at work — scores bunch in the middle and thin out fast toward both tails, which is why half the population shares the 90 to 109 band while whole bands at the extremes hold only a few percent.

Full percentile chart, IQ 55 to 160

The table below maps each score to its z-score (how many standard deviations from the mean), its percentile (the share of people scoring at or below it), and its rarity in plain English. Values assume the standard scale: mean 100, standard deviation 15.

How to read the chart correctly

Z-scores: the distance from average

The z-score is simply how many standard deviations a score sits from the mean, and on this scale the arithmetic is trivial: subtract 100, divide by 15. An IQ of 115 is z = +1.0; an IQ of 85 is z = -1.0; an IQ of 130 is z = +2.0. Z-scores are why the anchor numbers on every IQ chart are 85, 100, 115, 130 and 145 — they are whole standard deviation steps, and each step has a known share of the population attached to it.

Percentiles: the plain-English translation

A percentile tells you the percentage of the population scoring at or below a given score. It is the single most useful number on the chart because it survives every rescaling: whatever units a test uses, the 84th percentile always means the same thing — ahead of about five people in six. When someone asks whether a score is good, the percentile is the honest answer.

Rarity: where intuition breaks

Rarity accelerates faster than most people expect. An IQ of 120 is roughly 1 person in 11. An IQ of 130, ten points higher, is 1 in 44. Another ten points, 140, is about 1 in 260. And 150 is 1 person in roughly 2,300. Equal steps in score are wildly unequal steps in rarity, which is why claims about very high IQs deserve scepticism — the population simply does not contain many people out there, and no test used at scale is well calibrated at those extremes.

The famous 68-95-99.7 rule

The normal distribution gives IQ its most quoted statistics. About 68% of people score within one standard deviation of the mean — between 85 and 115. About 95% score within two — between 70 and 130. And 99.7% score within three — between 55 and 145. Anything beyond those bounds is genuinely exceptional in the statistical sense, occurring in fewer than 3 people per 1,000 at both tails combined.

Why every chart says mean 100, SD 15

Because the scale is manufactured that way. Test designers give their question set to a large norming sample, observe the raw score distribution, and map it onto a normal curve fixed at mean 100 and standard deviation 15. Periodic re-norming keeps 100 glued to the current population average, which also produces the well-documented Flynn effect: raw performance drifted upward across the twentieth century, so the same raw score earned fewer IQ points each time the tests were re-anchored. Your IQ is always a comparison with your contemporaries, never an absolute measurement.

Converting between scales: SD 15, SD 16 and SD 24

Not every test in history has used a standard deviation of 15, and this single fact explains most of the inflated IQ claims you will ever encounter. Older Stanford-Binet versions used SD 16, and the Cattell scale used SD 24 — same mean of 100, wider spread. A performance at the 98th percentile is IQ 130 on the modern SD 15 scale, 132 on SD 16, and 148 on Cattell's SD 24. Identical rarity, wildly different-sounding numbers. When someone reports an IQ of 148, the first question is always: on which scale?

Every figure on this page, and every score our test reports, uses the modern SD 15 convention — always convert to a common scale, or better, to percentiles, before comparing numbers from different sources.

Percentile, percentage and rank: three words people mix up

A percentile is the share of the population at or below your score: 84th percentile means ahead of about 84 in 100. A percentage correct is a property of one test paper — getting 84% of questions right could be brilliant or unremarkable depending entirely on the paper's difficulty, which is why raw percentages are never comparable across tests and why serious tests do not report them. And a rank is a position in a specific group: third in your class says nothing without knowing the class. The percentile is the only one of the three anchored to the population, which is what makes it the universal translation layer for every scale in the table above.

A brief history of the scale

The chart above is the end point of a century of revision. Alfred Binet built the first practical intelligence test in 1905 to identify French schoolchildren needing extra help, expressing results as a mental age. William Stern proposed dividing mental age by chronological age in 1912 — the original intelligence quotient — and Lewis Terman's 1916 Stanford revision of Binet's test multiplied it by 100 and gave the world the familiar three-digit number. The modern deviation IQ arrived with David Wechsler in 1939: rather than mental-age ratios, which break down for adults, scores were placed on a normal distribution against age peers, with mean 100 and a fixed standard deviation. Every serious test since — and every row of the chart on this page — uses Wechsler's statistical construction, periodically re-normed so that 100 keeps tracking the current population average.

Frequently asked questions