What does an IQ of 100 mean?

An IQ of 100 falls in the average band of the standard scale, at approximately the 50th percentile — as high as or higher than about 50% of the population. About 1 in 2 people score 100 or higher. Here is what that actually means, and just as importantly, what it does not.

The score in context

IQ scores sit on a scale constructed with a mean of 100 and a standard deviation of 15. A score of 100 therefore sits 0.00 standard deviations from the population average. About 68% of people score between 85 and 115, and 95% between 70 and 130 — the further from 100 a score sits, the faster the surrounding population thins out.

Scored close to 100 rather than exactly on it? Anything from 98 to 102 reads essentially identically: the gap is inside a single test's margin of error, so everything on this page — the classification, the percentile region, and the interpretation — applies across an IQ of 98, 99, 100, 101 or 102 alike.

What this band looks like in practice

Scores from 90 to 109 form the average band — the broad middle of the bell curve where half of all people sit. If your score lands here, your measured reasoning ability is typical of the population that runs the world: most teachers, nurses, tradespeople, managers, soldiers, and parents score in this band. Average, on a scale anchored to the human norm, means normal — not mediocre.

Practically, this band is compatible with almost everything. The everyday cognitive demands of most jobs, most study, and most life admin fall comfortably within it. Research linking cognitive scores to outcomes finds that within this band, differences of a few points predict essentially nothing about any individual life — effort, reliability, opportunity, and interest dominate completely.

It is also the band where measurement noise matters most, in a statistical sense: because the population is densest here, a swing of five points — well within a single test's error margin — moves you past tens of millions of people on paper while reflecting no real difference in ability. Treat scores in this band as "solidly typical" and resist reading meaning into single-digit differences.

Playing to the strengths of this band

The practical meaning of a score in the broad average band is freedom: essentially every ordinary path — trades, business, most degrees, management, creative work — is cognitively open, and the deciding factors are interest and persistence rather than measured ability. If there is a strategic lesson in a score here, it is to stop treating raw cleverness as the constraint and invest in the compounding assets instead: deep skill in one domain, a reputation for finishing things, and the communication ability that multiplies everything else. In study terms, people in this band get the highest return from technique — spaced repetition, active recall, deliberate practice — because technique is precisely what converts typical ability into top-quartile results, and it is entirely learnable.

How 100 compares with the anchor scores

The standard deviation anchors — 70, 85, 100, 115, 130, 145 — are the reference points every chart is built on. Here is 100 placed among them.

How much to trust a score of 100

Every IQ measurement carries error. Professionally administered tests report a confidence band of several points around any score, and single online sittings are noisier still: fatigue, distraction, rushing, or unfamiliarity with matrix puzzles can move a result meaningfully in either direction. The sensible reading of a 100 is "somewhere in the average region, give or take" — a good estimate, not a serial number.

The percentile is the part worth remembering. Scores can be rescaled; the fact of performing at roughly the 50th percentile of the comparison population is the stable, plain-English content of the result. When comparing results across different tests, always compare percentiles, never raw numbers, because different tests norm against different populations.

100 on other scales

Different tests have historically used different standard deviations, so the same underlying performance carries different numbers on different scales. A modern SD 15 score of 100 corresponds to approximately 100 on the older SD 16 Stanford-Binet convention and 100 on the SD 24 Cattell scale — identical percentile, different arithmetic. This is why comparing raw numbers across tests misleads, and why every comparison should run through percentiles.

Getting the most from a matrix reasoning test

If you plan to test or retest, a few things measurably help performance without gaming anything: take it rested and undistracted, since working memory is the first casualty of fatigue; read each grid row by row and column by column, naming the rule to yourself before looking at the options; and manage the clock — early questions should take seconds, banking time for the multi-rule items at the end where the discrimination between scores actually happens. Guessing carries no penalty on this test, so never leave a question unanswered as time runs out. And space any retest by weeks rather than hours: immediate retakes inflate scores through memory of specific items, which tells you nothing new.

What a single number leaves out

One score compresses many abilities. Verbal reasoning, spatial visualisation, working memory, and processing speed correlate, but imperfectly — two people at 100 can have quite different cognitive profiles, and most people have a spread of 10 to 20 points between their strongest and weakest domains. A matrix reasoning score specifically targets the fluid, pattern-based core of ability that correlates most strongly with the general factor, which is why single-format tests of it are meaningful — but no single number is a full portrait of a mind, at 100 or anywhere else.

Frequently asked questions