What Is the Average IQ? (By Age, Country & Job)

What is the average IQ?

The average IQ is 100. That is not a finding about humanity; it is how the scale is built. Every major modern IQ test takes a large sample of people that represents the population, scores them, and then converts raw scores so that the middle of that group lands on 100.

The second number that defines the scale is the standard deviation, which is 15 points on the Wechsler tests (WAIS for adults, WISC for children) and on the current Stanford-Binet (fifth edition). That gives you a familiar shape:

• About 68% of people score between 85 and 115 (within one standard deviation of the mean).
• About 95% score between 70 and 130 (within two).
• About 99.7% score between 55 and 145 (within three).

So when someone asks “is 105 good?” the honest answer is “it’s average, which is where most people are.” For the full set of bands and labels, see the IQ scale and ranges chart.

Why is the average exactly 100?

Because modern IQ is a deviation score: a measure of how far you sit from the average of your reference group, translated into points. The test doesn’t measure some absolute quantity of intelligence the way a scale measures kilograms. It ranks you.

The process looks like this:

1. Build the test. Publishers write and trial dozens of subtests (vocabulary, pattern completion, working memory, processing speed and so on).
2. Norm it. They give the final version to a standardization sample, typically a couple of thousand people chosen to match the census on age, sex, education, region and ethnicity.
3. Convert raw scores. Within each age group, the raw score at the middle of the sample becomes 100, and each standard deviation becomes 15 points.

The name “IQ” (intelligence quotient) is left over from an older method. In the early 20th century, IQ really was a quotient: mental age divided by chronological age, times 100. That ratio worked poorly for adults, so David Wechsler replaced it with the deviation method in 1939. The name stayed. If you want the full history, read what an IQ test is and where it came from.

What is the average IQ by age?

The average IQ is 100 at every age. Scores are age-normed, which means a 10-year-old is compared with other 10-year-olds and a 70-year-old with other 70-year-olds. Each age band gets its own conversion table, and each table is built so the middle of that band scores 100.

This trips people up because raw ability does change with age. A typical 8-year-old solves far fewer puzzles than a typical 25-year-old, and on average, speed-based tasks decline across later adulthood while vocabulary and general knowledge hold up much longer. Age norming hides all of that on purpose. An IQ score answers “how do you compare with your peers?”, not “how much can you do?”

Two practical consequences:

• A child’s IQ score is not a forecast of adult raw ability. It is a ranking among children their age, and childhood scores are less stable than adult ones, especially in the early years.
• An older adult with an IQ of 100 is doing exactly as well as their age group. If you want to compare raw performance across ages, IQ is the wrong number.

What is the Flynn effect?

The Flynn effect is the long rise in raw IQ test performance across the 20th century, roughly 3 points per decade in many developed countries. It is named after James Flynn, who documented it in US data in 1984 and later across many countries.

Because of it, a person who scored exactly 100 against norms from the 1950s would score noticeably lower against norms from the 1990s. Test publishers deal with this by renorming every 10 to 20 years, which resets the average to 100 against the current population. That is one reason the average stays at 100: whenever performance drifts, the yardstick is moved.

What caused the rise is still debated. The candidates include better nutrition and health, more and longer schooling, smaller families, and a culture that drills people in abstract, hypothetical thinking. Flynn himself leaned toward the last one: modern life trains the kind of reasoning the tests reward.

In some countries the rise has stopped or reversed. The best-known evidence is from Norway, where Bratsberg and Rogeberg (2018) used military conscription test data and found declining scores among men born since the mid-1970s, including within families (younger brothers scoring lower than older brothers). That pattern points to environmental causes rather than to changes in who was having children. Findings elsewhere are mixed.

What is the average IQ by country?

There is no reliable, agreed-upon average IQ for each country. The figures you see in viral maps mostly trace back to datasets compiled by Richard Lynn and Tatu Vanhanen (starting with IQ and the Wealth of Nations, 2002) and later revisions of that work. These datasets are heavily contested, and you should treat them as such.

The main criticisms:

• Small and unrepresentative samples. Some national figures rest on a few hundred people, sometimes schoolchildren from one city, sometimes groups that don’t resemble the national population.
• Imputed values. For countries with no data, early versions estimated a figure from neighboring countries.
• Mixed tests and decades. Scores from different tests, given decades apart, are lined up as if they were comparable, even though the Flynn effect alone can move scores by many points over that time.
• Cultural and schooling effects. Familiarity with test formats, years of schooling and language all affect results, and they vary hugely between countries.
• Contested use. The datasets have been used to argue for claims about national and ethnic differences that many researchers reject. Critics, including Jelte Wicherts and colleagues in their reanalyses of the African data, argued that more careful sample selection produced substantially higher estimates for some regions.

Rankings built on these tables commonly place East Asian countries near the top. Even if you take that broad pattern seriously, the precise numbers and the gaps between neighbors in a ranking carry far more uncertainty than the decimals suggest. International school assessments such as PISA and TIMSS are better documented, but they measure school achievement, not IQ, and they shouldn’t be converted into IQ points.

For the US specifically, see the average IQ in the USA.

What is the average IQ by job?

Groups of people in different occupations do show different average scores on cognitive ability tests, but the overlap between occupations is enormous, and the data is older and patchier than the confident charts online imply.

What exists:

• Military testing. The largest classic datasets come from US Army testing in the two World Wars. In 1945, Thomas and Margaret Harrell published scores on the Army General Classification Test for enlisted men by civilian occupation. Professions requiring long training (accountants, lawyers, engineers) had higher medians than manual trades, and the range within each occupation was wide. Many people in “lower-scoring” jobs scored higher than many in “higher-scoring” ones.
• Employer test norms. Publishers of workplace ability tests report typical scores by job family. These describe applicants or hires, not everyone in a field, and they use their own scales rather than IQ points.
• Job performance research. A well-known 1998 meta-analysis by Frank Schmidt and John Hunter concluded that general mental ability was one of the best single predictors of job performance. A 2022 re-analysis by Paul Sackett and colleagues argued those estimates had been inflated by statistical corrections and put the relationship noticeably lower. Ability matters for work; how much is still argued.

The limits are worth spelling out:

• Selection, not destiny. Occupations that require years of education filter on credentials, which partly filter on test-taking ability. The averages describe that sorting process.
• Ranges beat averages. Knowing a job’s average score tells you almost nothing about any one person in it.
• Old data, changed jobs. Many job categories in the classic tables barely exist in the same form today.

If you see a table claiming that a specific job has an average IQ of exactly some number, ask where the sample came from, when it was tested, and on which test. Usually the answer is missing.

What affects where you land relative to 100?

An IQ score is one measurement of a few specific abilities, on one day, under test conditions. Several things move it that have little to do with “how smart you are” in everyday terms:

• Measurement error. Even the best clinical tests carry a standard error of a few points, which is why psychologists report a range. More on that in what your IQ test score actually means.
• Familiarity and practice. Taking similar tests recently tends to raise scores.
• Sleep, stress, illness, language. All of them affect performance on timed puzzles.
• The test itself. Online quizzes that aren’t normed on a representative sample can report almost any number.

If you want to see what these question types feel like, try the free Mentarium IQ practice test. It is a practice tool, not a clinical assessment, and it won’t give you a diagnosis or a certified number.

The Mentarium take

“Average IQ” is the most searched question in this whole topic, and the answer is almost a trick: 100, always, by construction. The more useful lesson is about how to read numbers in general. An IQ score is a ranking against a reference group, so the first question is always “compared with whom, measured how, and when?” That is the same habit behind good base-rate thinking and the map vs territory distinction: the score is a map of one slice of your abilities, drawn with a known margin of error.

Country rankings and job averages fail that test more often than they pass it. They compress shaky, decades-old samples into crisp decimals, and the crispness is what makes them spread. If you want to get sharper, IQ is mostly fixed terrain. The reasoning habits in critical thinking and the mental models library are where effort actually pays off.