average IQ scores — by age, country, and population

The average IQ score is 100. This is not an empirical discovery but a mathematical certainty — IQ tests are specifically designed and periodically re-normed so that the mean score in the general population is always 100, with a standard deviation of 15. Roughly 50% of the population scores between 90 and 109, the range classified as "average" on most IQ classification scales.

But "average" is more nuanced than a single number. IQ scores vary systematically with age, differ across norming populations, and have been rising over historical time. Understanding these patterns requires distinguishing between the statistical properties of the test and the underlying cognitive abilities being measured.

In the context of IQ testing, "average" refers to the central region of the normal distribution. A score of 100 is at the 50th percentile — exactly the median. The "average range" (90-109) encompasses roughly half the population, and these individuals represent the broad middle of cognitive ability as measured by standardized tests.

It is important to understand that scoring "average" on an IQ test is not a negative result. The norming sample includes adults of all educational levels, professions, and backgrounds. An IQ of 100 means performing at the median of the entire adult population, which includes college graduates, skilled professionals, and experts in various fields.

For more on how norming works and how raw scores are converted to IQ, see how IQ is calculated.

Cognitive abilities do not remain constant throughout life, and the trajectory differs depending on the type of intelligence being measured. The Cattell-Horn-Carroll theory distinguishes between two primary forms:

Fluid intelligence — the ability to reason through novel problems, identify patterns, and think abstractly — tends to peak in the late twenties and begins a gradual decline thereafter. This is the type of intelligence primarily measured by matrix reasoning tests like Raven's Progressive Matrices.

Crystallized intelligence — accumulated knowledge, vocabulary, and learned problem-solving strategies — continues to increase well into middle age and often remains stable or even grows into the sixties and seventies. For a detailed comparison, see fluid vs crystallized intelligence.

Because IQ tests are normed within age groups, a 60-year-old who scores 100 has performed at the median for 60-year-olds, not for the population as a whole. This age-norming ensures that the natural decline in fluid intelligence does not artificially deflate older adults' scores.

In the 1980s, political scientist James Flynn documented a consistent phenomenon: average IQ scores had been rising across generations in every country where data was available. The increase averaged approximately 3 points per decade, though it varied by test type and country.

The gains were largest on tests of fluid intelligence, particularly matrix reasoning tasks, and smallest on tests of crystallized intelligence like vocabulary. This pattern suggests that the increase is not primarily driven by improved education (which would boost crystallized scores more), but rather by environmental changes that affect abstract reasoning capacity.

Proposed explanations include improved nutrition and healthcare, reduced exposure to environmental toxins (especially lead), increased cognitive stimulation from modern environments (visual media, technology, complex systems), smaller family sizes allowing more parental investment, and greater familiarity with standardized testing formats.

The Flynn effect has important implications: it means raw scores from different eras are not directly comparable without re-norming, and it demonstrates that IQ is responsive to environmental conditions at the population level. Some researchers have reported a slowing or reversal of the Flynn effect in certain developed countries since the 1990s, though the reasons remain debated.

Researchers have conducted IQ testing across many countries, and these studies show meaningful variation in average scores between nations. However, interpreting these differences requires substantial caution and context.

The most critical factor is that cross-national IQ differences correlate strongly with environmental variables: access to quality education, childhood nutrition, healthcare infrastructure, economic development, and exposure to environmental toxins. Countries with higher GDP per capita, better educational systems, and lower rates of childhood malnutrition consistently produce higher average test scores.

These correlations point to environmental causation, not innate differences between populations. The Flynn effect itself demonstrates this: countries that have undergone rapid economic development have seen their average IQ scores rise dramatically within a single generation — far too quickly to reflect genetic change. For example, gains of 15-20 points over 30 years have been documented in countries experiencing rapid industrialization.

Additional methodological concerns include: sample representativeness (many studies used small, non-representative samples), test cultural bias (items designed in Western contexts may not translate equivalently), language effects, and differences in test-taking motivation and familiarity with testing formats.

The most widely cited cross-national IQ dataset, compiled by Richard Lynn and Tatu Vanhanen, has drawn significant criticism from the psychometric community. Methodological objections include the use of small and non-representative samples, the extrapolation of scores for countries where no testing data existed, and the conflation of scores from different test instruments without proper equating.

Developmental psychologist Richard Nisbett and others have argued that cross-national IQ differences are best explained by environmental factors, noting that within-country improvements following environmental changes (better nutrition, expanded education) often match or exceed the between-country differences observed in cross-sectional studies.

The responsible interpretation of cross-national IQ data is that it reflects current environmental conditions and opportunities, not fixed or inherent characteristics of populations. For more on what determines IQ, see is IQ genetic? nature vs nurture in intelligence.