On modern tests the average is set at 100, and about 68 percent of people score between 85 and 115, so that range is what most people mean by normal. Many test manuals label 90 to 109 as Average. A score a few points either side of 100 is not a meaningful difference, because every test result carries a margin of error of a few points.

Can an IQ test measure all of intelligence?

No. An IQ test samples reasoning, knowledge, memory, and speed under standard conditions, which together capture general mental ability well. It does not measure creativity, practical judgment, social skill, motivation, or expertise. Most researchers treat IQ as a good estimate of one important part of intelligence rather than the whole of it.

Do IQ scores change with age?

An IQ is always calculated relative to people your own age, so your score can stay similar even as your raw abilities rise in childhood or some decline in old age. Rank relative to peers is fairly stable in adulthood, but scores can shift by several points with education, health, and practice, and short-term variation between sittings is normal.

Is a high IQ the same as being smart in everyday life?

Not exactly. Higher IQ makes it easier to learn and solve problems, and it predicts school and job outcomes on average. But everyday good judgment also depends on knowledge, experience, temperament, and habits such as checking your assumptions. Research on rational thinking shows that people with high IQs still fall for many common reasoning biases.

What Is IQ? What It Means and What It Measures

IQ stands for intelligence quotient. It is a standardized score from a test of reasoning, problem solving, knowledge, memory, and mental speed, scaled so that the average score for people your age is 100 and about two out of three people score between 85 and 115. An IQ score tells you how your performance compares with a large reference group, not how much intelligence you "have" in absolute terms.

What does IQ stand for?

The term comes from the German psychologist William Stern, who in 1912 proposed dividing a child's "mental age" (the age group whose typical performance the child matched) by their actual age. Lewis Terman multiplied that ratio by 100 for the 1916 Stanford-Binet, so a 10-year-old performing like a typical 12-year-old scored 120.

Modern tests no longer divide anything. Since David Wechsler's tests in the late 1930s, IQ has been a deviation score: your result is placed on a scale where the mean of your age group is 100 and the standard deviation is 15. The word "quotient" survived, but the arithmetic did not. The full backstory is in our history of IQ testing.

What does an IQ test measure?

A professionally administered IQ test, such as the WAIS-5 for adults (published in 2024) or the WISC-V for children, is a battery of short subtests. Each targets a different ability, and the results are combined into a full-scale score. The main areas covered by current batteries are:

Verbal comprehension: vocabulary, word knowledge, explaining how two concepts are alike.
Fluid reasoning: spotting rules in unfamiliar patterns, such as matrix puzzles, and applying them.
Visual-spatial processing: mentally assembling or rotating shapes.
Working memory: holding and manipulating information, for example repeating digits in reverse order.
Processing speed: doing simple visual tasks quickly and accurately.

These abilities are distinct, yet people who do well on one tend to do well on the others. That consistent overlap is what psychologists call the general factor, or g, and the full-scale IQ is largely an estimate of it. Our explainer on the g factor and general intelligence covers how that works and what it does and does not imply.

A widely quoted working definition of the trait these tests aim at comes from a 1994 statement signed by 52 researchers: intelligence is "a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience" (Gottfredson, 1997). IQ tests sample that capability. They do not capture all of it, and they were never meant to.

How are IQ scores scaled?

Because the scale is fixed at a mean of 100 and a standard deviation of 15, every score maps to a percentile. A score of 115 is higher than about 84 percent of the reference group; 130 is higher than about 98 percent; 85 is higher than about 16 percent. Roughly 68 percent of people fall between 85 and 115 and about 95 percent between 70 and 130.

No score is exact. Even on the best clinical tests, a full-scale result carries a margin of a few points, which is why reports give a confidence interval such as 104 to 112 rather than a single number. For what each range is called and how common it is, see IQ score ranges and what counts as a good score, or look up any individual score in the IQ scale table.

What does IQ predict?

IQ is one of the most studied measures in psychology, mainly because it predicts real outcomes better than most single measures do. The associations are moderate to strong at the level of large groups and loose at the level of any one person.

School performance. In a study of about 70,000 English children, general ability measured at age 11 correlated about .81 with national exam results at 16, when both were measured as latent factors (Deary et al., 2007). Correlations with ordinary school grades are typically lower, around .5 (Neisser et al., 1996).
Education, occupation, and income. A meta-analysis of longitudinal studies found correlations of about .56 with years of education, .43 with occupational level, and .20 with income (Strenze, 2007). Income, in other words, depends heavily on other things.

IQ, health, and longevity

A field called cognitive epidemiology, built largely on Scottish cohorts tested as children in the 1930s and 1940s and followed for decades, studies how intelligence relates to health. Its most consistent finding is that higher childhood test scores go with longer lives. A meta-analysis of studies totaling over a million people found that each 15-point (one standard deviation) advantage in youth was associated with roughly a 24 percent lower risk of death during follow-up (Calvin et al., 2011). Follow-ups of the Scottish cohorts have reported similar gradients for several specific causes of death, including cardiovascular disease.

Researchers are careful about what this means. Part of the link runs through education, income, and occupation. Part may come from health literacy: understanding medical instructions, weighing risks, and sticking to treatment. A third proposal, which Ian Deary and colleagues call "system integrity," is that a well-functioning brain and a well-functioning body partly reflect the same underlying good health. These explanations are not mutually exclusive, and none means a lower score dooms anyone to worse health. The gradients describe populations, and many of the pathways, like education and access to care, are modifiable.

IQ and everyday decisions

Higher IQ is linked to somewhat better performance on tests of practical decision making, such as applying decision rules, resisting framing effects, and recognizing social norms. In one study of adults, a "decision-making competence" battery correlated with cognitive ability, but it also predicted fewer self-reported negative decision outcomes in everyday life after cognitive ability was accounted for (Bruine de Bruin et al., 2007).

The psychologist Keith Stanovich has spent decades documenting the gap between intelligence and rational thinking. Many reasoning errors, including myside bias (evaluating evidence in favor of what you already believe), are only weakly related to IQ. People with high scores make them too, and sometimes defend them more skillfully. Stanovich and colleagues argue that rationality needs its own measures (Stanovich et al., 2016). The practical lesson: a high IQ makes good reasoning easier, but habits like seeking disconfirming evidence and slowing down on important choices do work that IQ does not.

What IQ does not measure

An IQ test is a narrow, well-calibrated instrument. It leaves out a lot that matters:

Personality and drive. Conscientiousness, curiosity, and persistence predict achievement independently of IQ.
Creativity. Generating original, useful ideas correlates only modestly with IQ.
Social and emotional skills. Reading people and managing emotions are separate abilities; emotional intelligence is studied as its own construct.
Expertise and wisdom. Deep knowledge in a field and sound judgment built from experience are not on the test.

An IQ score also is not a measure of worth, and it is not a diagnosis on its own. Clinicians who assess intellectual disability or giftedness combine test scores with developmental history and everyday functioning. This article is general information, not medical or educational advice.

Is IQ fixed?

IQ is fairly stable in adulthood in the sense that people tend to keep their rank relative to peers, but it is not fixed. Schooling raises it: a meta-analysis of quasi-experimental studies estimated a gain of roughly 1 to 5 IQ points per additional year of education (Ritchie & Tucker-Drob, 2018). Average scores in many countries rose by roughly 3 points per decade during much of the 20th century, the Flynn effect, which is far too fast to be genetic; in several countries the gains have stalled or reversed since the 1990s.

Genes matter too. Twin and adoption studies consistently find that the heritability of intelligence rises from childhood to adulthood. But heritability describes how much of the variation within a particular population tracks genetic variation in that population. It says nothing about why groups differ, and it does not mean a trait cannot change. For the evidence on change, see whether IQ can change over time.

IQ in the age of AI

Large language models can now answer many questions of the kind found on IQ tests, and headlines sometimes assign them an IQ. Treat those numbers with caution. IQ norms describe how humans of a given age perform under standard conditions; a system that may have seen similar items during training, does not tire, and has a very different profile of strengths is outside the population the scale was built for.

For people, the open question is different: whether the abilities IQ tests measure will matter more, less, or differently as AI tools change the work people do. The studies on school, work, and health described above were done before these tools were common, so there is not yet good evidence either way. What seems unlikely to change is the value of reasoning well enough to judge an answer, including one a machine gives you.

Online tests and what they can tell you

Clinical IQ tests are given one-on-one by a psychologist and normed on large, representative samples. Free online tests can give a useful rough picture, but most have not been normed that way. Ours, for example, compares you only with its own test-takers; its result page shows an estimated IQ: your position among people who took the same form, converted to the mean-100 scale, labelled as an estimate and shown with a range. It is not a normed IQ, and the methodology page explains how it works. If you want to see where you stand on six reasoning domains, you can take the free test.

What Is IQ? What the Score Means and What It Measures