What Is the Average IQ? Average Intelligence, the Normal IQ Range and the US Average

The average IQ score is 100. That isn't a coincidence or a survey result — it's how IQ tests are built. Every major test is standardized so that the mean score of the norming sample is exactly 100, with a standard deviation of 15 points. About 68% of people score between 85 and 115 — the range most people mean by a normal IQ — and roughly 95% score between 70 and 130.

Why the average IQ score is always 100

When a new IQ test is developed (or an old one revised), it is given to a large, representative sample of people — the norming sample. Raw results are then mapped onto a bell curve centered at 100. Your IQ score is therefore not a count of correct answers; it's a statement about where you stand relative to everyone else. A score of 115 means you scored one standard deviation above the mean — higher than about 84% of the population.

What counts as a normal IQ?

There is no official definition of a normal IQ, but two conventions are in common use, and they differ only in how wide a net they cast. The narrower one is the classification band 90 to 109, which psychologists label "average" and which contains 49.5% of people. The broader one is 85 to 115 — one standard deviation either side of the mean, containing 68.3%. Either way, a normal IQ is not a threshold you pass or fail; it is the thick middle of a curve, and the great majority of people are in it.

What that means for an individual score is worth stating plainly: average intelligence is the single most common result there is. Half of all people score below 100 and half score above it, by construction. A result of 97 or 103 is not a meaningfully different outcome from 100 — the difference is smaller than the measurement error of the test itself, which is why how accurate IQ tests are matters more than the last few points of any score.

How IQ scores are distributed across the population

For a deeper breakdown, see our guides to every IQ range and IQ level and the IQ score chart, or convert any score to a population rank with the IQ percentile table.

What is the average IQ in the USA?

On any IQ test normed in the United States, the average American IQ is 100 — by construction, not by measurement. That is the answer to the question most people are actually asking, and it is the one figure here you can rely on. The reason you will also see 98, 97 and 101 quoted for the same country is that those come from a completely different kind of exercise: cross-country estimates that place every nation on one shared scale.

Both things can be true at once, and the distinction is the whole answer:

• The average IQ in America is 100 on an American-normed test. A publisher takes a nationally representative US sample and maps the raw scores so that sample averages 100. Asking what the average is afterwards is like asking how long a metre is.
• National estimates put the United States somewhere between about 97 and 101. These are attempts to compare countries with each other, built from pooled studies or from online-test traffic. They are estimates with wide uncertainty, and — as the next section shows — they disagree with each other badly.

So if you have just taken a test and want to know where you stand among Americans, compare against 100 and read your percentile. A score of 100 puts you ahead of about half the US population; 115 puts you ahead of roughly 84%. You can convert any specific score with the IQ percentile table.

Average IQ for adults

The average IQ for adults is 100, and it is 100 for every adult age group separately. Modern tests are age-normed: a 32-year-old is scored against a sample of other 32-year-olds, and a 68-year-old against other 68-year-olds, so the mean cannot drift with age even though the underlying abilities do. This is the single most common misreading of adult IQ data — people expect the average to fall in later life, and it does not, because the yardstick moves with the person being measured.

Average IQ by sex: men vs women

On a full-scale IQ test the average score for men and the average score for women are both about 100. There is no meaningful difference in the overall average. What the research does find is a difference in profile — which kinds of question each group tends to do slightly better on — and a difference in spread. Neither of those is a difference in the average, and both are far smaller than the variation between individuals of the same sex.

Three findings, kept separate here because they are constantly merged into one:

• The overall averages are effectively equal. Janet Hyde's review of 46 meta-analyses of psychological sex differences concluded that men and women are similar on most variables studied, with the majority of measured differences small or very small.
• The profiles differ a little. The 2007 consensus review by Halpern and colleagues reports that women tend to score higher on verbal measures, particularly writing, while men tend to score higher on most visuospatial measures such as mental rotation. A full-scale IQ adds both kinds of task together, so these largely cancel out.
• Male scores are more spread out. The same review finds greater male variability on quantitative and visuospatial measures — more men at both the top and the bottom of the distribution — while noting that the reasons for it remain unresolved. A wider spread is not a higher average, and the two are routinely confused.

Where the "men score 4.6 points higher" figure comes from

This number circulates widely enough to be worth stating precisely rather than ignoring. It comes from Paul Irwing and Richard Lynn, and the version most often linked is their 2006 item in Nature, which is a reply in a Brief Communications Arising exchange — a response to a critique of their earlier analysis published in the same journal. Three things about it are usually left out:

• The figure is drawn from university students, not from a nationally representative sample of adults.
• It is measured on the Progressive Matrices, a single non-verbal reasoning test — not on a full-scale battery covering verbal, spatial, memory and processing-speed tasks.
• It was published as one side of a documented scientific dispute, not as a settled result.

A selected sample measured on one non-verbal test is not the same thing as the average IQ of men and women, and other analyses of the same question find no difference in the mean. This is a genuinely contested corner of the literature, and the honest summary is the one at the top of this section: the overall averages are close to equal, and any real difference is small enough to be swamped by which test was used and who was tested.

For reading your own result none of this changes anything. Your score is compared against a mixed-sex norming sample, so 100 means the same thing whoever you are. Convert a specific score to a population rank with the IQ percentile table, or see how the same question works across age groups.

What is the world average IQ? Why sources say 100, 90 or 85

Nobody has measured the world average IQ, because no IQ test has ever been normed on a sample of the world. Every published norm belongs to one country — an American sample for the American edition of a test, a German sample for the German one — so the honest answer to "what is the average IQ in the world?" is that each national average is 100 on its own test, and there is no shared yardstick that turns them into one global number.

Search for the question and you will find 100, 90 and 85 all stated confidently on the same results page. They are not contradictory; they answer different questions.

100 is a property of the scale, not a finding about people. A test publisher administers a new test to a norming sample — usually a nationally representative sample of one country in one period — and maps the raw results so that sample's mean lands on 100. Say "the average IQ is 100" and you have said something true by construction about that test, that country and that year.

A quoted average human IQ of 85 or 88 is a different claim: an estimate of how the world's population would score if measured against one country's norms. Those figures come from compilations that pool many separate studies using different tests, different years and samples of very uneven quality and size, and the best-known of them have been widely criticised on exactly those grounds. Treat any single global figure as an estimate with wide uncertainty rather than a measurement.

The reference country is Britain. In Richard Lynn's national IQ datasets — the source behind most world maps and "global average" figures — the mean of the United Kingdom was set at 100 and every other country was expressed relative to it, a convention its users call the "Greenwich IQ". A world figure of 85 to 90 on that scale therefore means "some 10 to 15 points below a British norming sample", not "the average human scores 85 on the test you take". Move the reference to another country's norms, or to a later year of British norms, and the world figure moves with it.

The clearest demonstration that "the average" is a moving reference point rather than a fixed fact is the Flynn effect below. Raw performance rose by roughly three points a decade through the 20th century, so a person scoring exactly 100 today would have scored well above the mean of a norming sample drawn in the 1950s — on the order of 15 to 20 points above it, if that rate held steady across the whole span, which in some countries it has not. The number 100 stayed put only because the tests were re-normed underneath it.

What average IQ means in practice

What is the definition of average IQ?

Average IQ means the midpoint of a standardized IQ scale. On modern SD-15 scales, that midpoint is fixed at 100 by design, not estimated from a casual poll.

Is average IQ the same as "normal IQ"?

Not exactly. "Average" is usually the 90 to 109 classification band, while many people use "normal" for the broader 85 to 115 range. Both describe common ranges around the same center of 100.

Can you compare two people by a 2-point IQ difference?

Usually no. Small gaps are often within normal measurement error. Interpret results as ranges plus percentile, not as precise rankings by single-point differences.

Where can I compare a specific score quickly?

Use the IQ score chart for broad ranges and the IQ percentile table for point-by-point conversions.

Average IQ by age

Because scores are age-normed, the average is 100 at every age: a 25-year-old and a 70-year-old are each compared to their own age group. Underneath the norming, raw abilities do change — fluid reasoning (solving novel problems) typically peaks in the 20s and gradually declines, while crystallized knowledge (vocabulary, facts, skills) keeps growing into the 60s — the distinction explained in fluid and crystallized intelligence. We cover the age pattern in detail in average IQ by age.

Average IQ by country: why the rankings disagree

There is no reliable ranking of average IQ by country, and the simplest way to show it is to put the popular lists side by side. They do not merely differ at the margins — they disagree about which country is first by more than ten IQ points, and about the United States by four. All figures below were read from the live pages on August 23, 2026.

The second row is the one that settles the question. On brght.org's list, the country ranked second in the world is Togo (116.06) and third is the French Southern Territories (115.71) — an Antarctic overseas territory with no permanent civilian population. Tokelau, home to roughly 1,500 people, sits tenth. No plausible theory of human cognition produces that list. What produces it is a small, self-selected handful of visitors from each of those places, which is what the site is actually measuring.

That is the flaw running through every online-traffic ranking, including the largest: a table built from people who searched out an IQ test and chose to finish it is a ranking of that site's visitors, not of nations. World Population Review states the basis of its own column plainly — data from people "who took the same IQ test on the website".

The academic lineage has a different problem rather than no problem. In IQ and the Wealth of Nations, the most-cited source of national IQ figures, 104 nations had no IQ study at all and were assigned a value estimated from surrounding countries. Where studies did exist, several were tiny and nowhere near representative: a test of 48 children aged 10–14 in Equatorial Guinea and of 50 aged 13–16 in Colombia were each taken as that country's national IQ. Those numbers have been re-published for two decades without the sample sizes travelling with them.

Where real differences in measured averages do exist, they track education access, childhood nutrition, health care and familiarity with timed abstract tests — the same factors behind the score gains that schooling and practice produce within a country, and behind the Flynn effect below, which moved whole populations by roughly three points a decade without the populations changing. Treat a country IQ ranking as a claim about data collection, not about people.

The same caution applies to average IQ by state tables in the US, which circulate widely as maps and rest on the same mix of small samples and self-selected online testing.

One thing this section is not about: the highest score ever recorded by an individual. That is a separate question, with its own problems of verification, covered in the highest IQ ever recorded.

IQ Revealed is not affiliated with, endorsed by or connected to World Population Review, brght.org or International IQ Test. They are named only to identify the published rankings discussed, every figure above comes from their own live pages on the date stated, and the sources are linked at the end of this article so you can check them.

The Flynn effect: averages have been rising

Raw performance on IQ tests rose steadily through the 20th century — about 3 points per decade — a phenomenon named the Flynn effect after researcher James Flynn. Better schooling, nutrition, smaller families and a more abstraction-heavy world are the leading explanations. Because tests are re-normed, the published average stays at 100 even as raw performance shifts; some wealthy countries now show a plateau or slight reversal.

Source and interpretation notes

This page summarizes mainstream psychometric conventions. For careful interpretation, prioritize test manuals and peer-reviewed sources over social-media score claims or unsourced ranking lists.

• Definitions and score bands here align with common SD-15 frameworks used by major IQ tests.
• Country-level comparisons are noisy and often sample-limited, so they should not be treated as fixed truths about populations.
• Online scores are estimates; for formal decisions, use professionally administered assessments.

What "average" means for you

A single number can't capture a mind. Two people with identical IQ 100 scores can have very different cognitive profiles — one strong in working memory, the other in visual-spatial reasoning. A single overall score cannot tell those two people apart, and that is the honest limit of any one number, ours included: it places you on the scale, it does not map your strengths ability by ability. What it gives you is a baseline — and a baseline is what focused training moves, by a few points rather than by a transformation.

If you want a practical baseline to place your own result on the same scale, you can start with the IQ test and score report and then interpret the result using the IQ testing methodology guide. At the other end of the distribution from the average sits the Mensa qualifying score, which roughly 2% of people reach.