What is the average IQ?

100. That is not a discovery; it is how the scale is built. Test makers give a test to a large representative sample and set that sample’s average to 100.

What is the average IQ in the USA?

On any test normed on a representative US sample it is 100, because the norm sample defines the scale. Other figures quoted for the US come from studies that compare countries against a foreign reference sample, and their methods are disputed.

What is the average IQ by age?

100 at every age. IQ compares you with people your own age, so the average 20-year-old and the average 70-year-old both score 100, even though raw performance on reasoning puzzles is higher at 20.

It is above average: higher than about 75 in 100 people. It falls in the range usually labelled “above average” or “high average”.

Is average IQ going up or down?

It rose through most of the twentieth century, by roughly two to three points a decade. Recent studies in the United States and parts of Europe find the rise has stalled or reversed on some kinds of test.

Average IQ: What It Is, by Age, and What Changes It

Why the average is always 100

An IQ score is a comparison. To build a test, its makers give it to a large sample chosen to represent a population, find the average raw score, and call that 100. Scores are then spread so that 15 points equals one standard deviation.

So “the average IQ is 100” is a definition, in the same way that the average height percentile is 50. About two thirds of people score between 85 and 115. The full breakdown is on the IQ scale page.

Average IQ by age

The average is 100 at every age, because scores are compared within age groups. What changes with age is the raw performance behind the score:

Reasoning with new problems (the kind tested by pattern puzzles) tends to peak in early adulthood and decline gradually afterwards.
Knowledge and vocabulary keep growing for decades and hold up well into later life.

A 60-year-old and a 25-year-old with the same IQ are not solving the same number of puzzles. Each is at the same position among people their own age.

Has the average changed over time?

Yes, and by a lot. Raw scores rose throughout the twentieth century, an observation known as the Flynn effect. A meta-analysis put the gain at about 2.3 points per decade overall and 2.9 on modern tests (Trahan et al., 2014). A person who scored 100 against 1950 norms would score well below 100 against today’s.

The rise is not guaranteed to continue. A study of 394,378 American adults tested online between 2006 and 2018 found scores falling on several kinds of reasoning item (Dworak et al., 2023), and German samples show a similar reversal on figural reasoning between 2012 and 2022 (Oberleiter et al., 2024).

This is why tests are re-normed, and why a score always belongs to a particular set of norms and a particular year.

Average IQ by country: handle with care

Lists ranking countries by average IQ are widely shared and poorly founded. Many of the national figures come from small or unrepresentative samples, from tests given in a second language, or from estimates borrowed from neighbouring countries. Differences in schooling, nutrition, health and familiarity with testing all move scores, and none of them is a fixed property of a population.

The honest summary is that such tables measure the conditions under which people were tested at least as much as anything else. We do not publish one.

What your own score means against the average

90 to 109: average. Half of all adults are here.
110 to 119: above average, ahead of roughly 75 to 90 in 100.
120 to 129: high, ahead of roughly 91 to 97 in 100.
130 and above: very high, ahead of about 98 in 100.

For any exact score, use the IQ percentile calculator.

Sources

Trahan et al. (2014). The Flynn effect: a meta-analysis. Psychological Bulletin.
Pietschnig et al. (2015). One century of global IQ gains. Perspectives on Psychological Science.
Dworak et al. (2023). Looking for Flynn effects in a recent online U.S. adult sample. Intelligence.
Oberleiter et al. (2024). Measurement-invariant fluid anti-Flynn effects in population-representative German student samples (2012–2022). Journal of Intelligence.