What IQ Is in the Top 1 Percent? About 135

Why 135 marks the top 1 percent

IQ tests are built so that scores follow a normal curve with a mean of 100 and a standard deviation of 15. That is the scale of the Wechsler tests and the Stanford-Binet 5. On that curve, the top 1 percent begins 2.326 standard deviations above the mean. Multiply 2.326 by 15, add 100, and the cutoff is 134.9.

The neighbors show how sharp the line is. An IQ of 134 sits at the 98.8th percentile, about 1 in 85. An IQ of 135 reaches 99.0, about 1 in 102. By 136 the figure is 99.2, and at 137 it is 99.3, about 1 in 147. Two points either side of 135 shift rarity by nearly half.

The cutoff is a property of the scale, not a publisher's judgment. Change the standard deviation and the line moves: 137 on the old Stanford-Binet scale (SD 16), 156 on Cattell-style scales (SD 24). The person has not changed. The ruler has.

Top 1, 2, 5 and 10 percent: the cutoffs

The table converts each common "top X percent" into its minimum score on three scales. The figures are theoretical values from the normal curve; the last column links to the nearest score page.

Two practical notes. Mensa rounds its top-2-percent line to 130 on Wechsler tests, 132 on the old Stanford-Binet and 148 on the Cattell. And searches for the "top 10th percentile" usually land on 120, which is slightly stricter than the exact 119.2 cutoff: an IQ of 120 beats about 90.9 percent of people.

What the 99th percentile actually tells you

A percentile rank is the share of a reference group that scored below you. A 99th-percentile result means you outscored about 99 of every 100 people in the test's norm sample, a census-matched group of a few thousand, not the entire population.

The label is coarse at the top. In whole-number terms, a 139, a 145 and a 160 are all simply above the 99th percentile. The gap between 1 in 215 and 1 in 31,574 disappears into one label. Pearson's published WAIS-5 sample reports print percentile ranks as whole numbers, which is why the standard score, not the percentile, carries the information at the extreme.

Psychologists therefore report the standard score together with a confidence interval. The percentile is a translation for lay readers. The score is the measurement.

The 100th percentile does not exist

People search for "100th percentile IQ" because some websites print it. No such rank exists on a properly scored test. The standard definition counts the percentage of scores below yours, plus half the frequency of your own score. Even the single highest scorer in any group lands short of 100.

Scale it to the planet. With about 8 billion people, the top scorer on Earth would sit near the 99.99999999th percentile. A site that reports "100" has rounded a figure such as 99.6 up to a number that means higher than everyone, including yourself. Treat it as a sign of careless scoring, not of a record result.

How many people are in the top 1 percent

One percent sounds exclusive. In absolute numbers it is not. Of roughly 8 billion people, about 80 million would score 135 or higher if the whole world sat the same well-normed test. A city of 1 million would hold about 10,000.

The group thins fast above the line. Only about 1 in 741 people reach 145, and about 1 in 31,574 reach 160, the ceiling of the WAIS-IV. For the full ladder from 100 to 200, see the IQ rarity chart.

The longest-running study of people in this range is the Study of Mathematically Precocious Youth, founded by Julian Stanley at Johns Hopkins on 1 September 1971. It has followed more than 5,000 people identified between 1972 and 1997, ranging from the top 3 percent to the top 0.01 percent in quantitative or verbal reasoning, and has produced more than 400 articles. One consistent finding: people at the very top of the range tend to accomplish more in school and work than the moderately gifted. Ability differences still matter inside the top few percent.

Proving a top 1 percent score

A single score is an estimate. On modern professional batteries the confidence interval spans about 10 points, so a measured 135 means the true score most likely lies between roughly 130 and 140. That is anywhere from the top 2.3 percent to the top 0.4 percent.

The top of the curve is also the thinnest part of any norm sample. The WAIS-IV was standardized on 2,200 US adults aged 16 to 90. Statistically, only about 22 of them would fall in the top 1 percent and about 3 above 145. Scores up there rest on few reference cases, one reason high IQ scores are less reliable than scores near the median.

A qualifying result needs a supervised, professionally administered test: a psychologist-given WAIS-5 or Stanford-Binet 5, or a national Mensa's proctored session. Our free practice IQ test is a fast screener. It helps decide whether a formal assessment is worth the fee; it does not replace one. See how accurate online IQ tests are and how to join Mensa.

Top 1 percent, Mensa and the genius label

Three lines get confused. Mensa's line is the top 2 percent, which it sets at 130 on Wechsler tests. The top 1 percent begins at 135. Lewis Terman's 1916 classification for the Stanford-Binet called scores above 140 "near genius or genius," a line that corresponds today to about 1 in 261 people.

Terman dropped the word by the 1937 revision of his test, and no major test has used it since. David Wechsler wrote in 1939 that he was "rather hesitant about calling a person a genius on the basis of a single intelligence test score." Modern labels are descriptive: Very Superior on the WAIS-IV, and Gifted or very advanced on the Stanford-Binet 5 for 130 to 144. For the full chart, see IQ scale classifications. For the threshold debate, see what IQ counts as genius.

Editorial standard: IQ figures for historical figures are retrospective estimates and are labeled as such; scores for living people appear only when the person disclosed them or a reputable outlet reported them. Corrections: contact the editors.