Average IQ: What 100 Really Means

The average IQ score is 100. That is not a discovery about human intelligence: it is a choice built into how IQ tests are scored. Understanding that choice explains a lot about what IQ scores mean, why your result depends on who you are compared with, and why test makers have to keep updating their numbers. This guide walks through it step by step.

Average is set, not measured

When a test is developed, it is given to a large sample of people chosen to represent the population it will be used with. This sample is called the norm group. The test makers then convert raw results into a scale where the middle of the norm group is placed at 100.

In other words, a score of 100 means "you performed exactly like the typical person in the comparison group". If the test were made harder or easier, the raw scores would change, but the average would still be set at 100. That is why an IQ score is always relative: it describes your position, not an absolute quantity.

The 100 and 15 scale

The second number that defines the scale is the standard deviation, which is set at 15 points. The standard deviation describes how spread out scores are around the average. On a bell-shaped curve:

| Range | Distance from average | Share of people | |---|---|---| | 85 to 115 | within 1 standard deviation | about 68% | | 70 to 130 | within 2 standard deviations | about 95% | | 55 to 145 | within 3 standard deviations | about 99.7% |

This gives every score a precise meaning. A score of 115 is one standard deviation above average, which places you higher than about 84% of the norm group. A score of 85 is one standard deviation below, at about the 16th percentile. The formula behind it is simple: IQ = 100 + 15 × z, where z is how many standard deviations you are from the average.

Some older tests used a standard deviation of 16 or even 24, which is one reason old scores are hard to compare with modern ones. Today, 15 is the near-universal convention.

What counts as an average score

Strictly, only 100 is the average. In practice, people describe a band around it as average, most often 90 to 109. About half of all people score in that band. A slightly wider band, 85 to 115, holds about two thirds of people.

The important point is that small differences in this band rarely mean anything. Every test has some measurement error, often several points, so scores of 96 and 104 can reflect the same underlying ability. If your score falls in the average range, you are in the largest group of people, and your result is as normal as a result can be. Our IQ score chart shows how the other ranges are labelled.

How norms work

Because the average is defined by the norm group, the quality of the norm group matters a great deal. A good norm group has three properties:

• It is large. Hundreds or thousands of people per age group give stable estimates.
• It is representative. It should reflect the population in age, sex, education and region. A sample of university students, for example, would produce averages that are too high for the general population.
• It is age-specific. Performance on reasoning tasks changes across childhood and again in later adulthood. Scores are therefore compared within age bands, so a 10-year-old is compared with other 10-year-olds and a 60-year-old with other 60-year-olds.

Online tests face a particular problem here. The people who choose to take an IQ test online are not a random slice of the population: they tend to be younger, more educated and more interested in puzzles. Without correction, an online test would place its average too high, and ordinary performance would look below average. Careful test makers adjust for this by weighting results so the comparison group better matches the general population.

The Flynn effect

In the second half of the 20th century, researchers noticed something surprising. When people took older tests, they scored higher than the original norm groups had. Across many countries, average performance on IQ tests rose by roughly 3 points per decade. This became known as the Flynn effect, after the researcher who documented it most widely.

Researchers have suggested several reasons, including better nutrition and health, more years of schooling, and daily environments that demand more abstract thinking. No single cause is agreed on. More recent studies suggest that the rise has slowed, stopped or even reversed in some countries, which is still being studied.

For test users, the practical lesson is that norms age. A test normed decades ago compares you with a population that scored lower on average, so it can overstate your result. This is why professionally administered tests are re-normed regularly, and why a score is only as meaningful as the norm group behind it.

What the average does not tell you

The average describes a group, not any individual within it. Knowing that the average is 100 tells you nothing about a specific person's knowledge, talents, creativity or character. Two people with the same score can have very different strengths: one may be stronger at spatial tasks, the other at working memory. This is why a breakdown by ability says more than a single overall number.

Averages are also not targets. Scoring above or below 100 is not a goal or a failure: it is a description of how you performed on one set of tasks, on one day, compared with one reference group. To turn your score into a precise position, see our guide to IQ percentiles.

Compare yourself with the average

The IQTestScale test takes about 20 minutes and measures four abilities: fluid reasoning, spatial thinking, working memory and quantitative reasoning. Your result places you on the 100 and 15 scale, with a likely range and a percentile, so you can see how your performance compares with the average. It is a self-assessment for personal insight, not a clinical evaluation.

Take the IQ test