What does your IQ score actually mean?

The average IQ is 100, by definition. IQ tests are periodically re-normed so that the population mean stays at 100. 50% of people score above 100 and 50% score below.

An IQ of 120 falls at approximately the 91st percentile, meaning roughly 91 in 100 people score below this on a standardised test. This is the "Superior" range on the Wechsler scale.

An IQ of 130 falls at approximately the 98th percentile, placing you in the top 2% of the population. This is the threshold for Mensa membership. Roughly 1 in 50 people qualifies.

Mensa requires a score at or above the 98th percentile. On the Wechsler scale (SD=15), this is approximately IQ 130. On the Stanford-Binet (SD=16), approximately 132. On the Cattell scale (SD=24), approximately 148.

Most free online IQ tests are not clinically validated and tend to inflate scores by 5–15 points. They are normed on self-selected samples, not representative populations. For a reliable result, the gold standard is an in-person evaluation by a licensed psychologist using a validated instrument such as the WAIS-IV or Stanford-Binet 5.

An IQ of 140 falls at approximately the 99.6th percentile, placing you in the top 0.4% of the population, or roughly 1 in 261 people. This is well above the Mensa threshold and falls in the "Very Superior" to "Exceptionally Gifted" range depending on which classification system is used. At this level, scores become increasingly imprecise because standardised tests have fewer items that discriminate at the extreme upper end of the scale. The WAIS-IV has a ceiling of 160, and scores above 145-150 are estimates rather than precise measurements. (Source: Wechsler 2008)

An IQ of 110 falls at the 75th percentile, meaning you score higher than approximately 75% of the population. In Wechsler classification, this is "High Average" (110-119). It is a solid, above-average score. For context, the average university graduate scores approximately 115, so 110 is slightly below the typical graduate but well above the general population mean. At the 110 level, factors like motivation, conscientiousness, and opportunity tend to have a larger influence on life outcomes than the IQ difference between 110 and, say, 120. See also the EQ test for a complementary measure of emotional intelligence percentile.

IQ scores are age-normed, meaning the average is 100 at every age by design. Raw cognitive ability does change with age: fluid intelligence (novel problem-solving) peaks in the mid-20s and gradually declines, while crystallised intelligence (accumulated knowledge) continues to increase into the 60s and 70s (Cattell 1963; Horn & Cattell 1967). Because IQ tests are normed by age group, your IQ score can remain stable even as the underlying cognitive profile shifts. A 70-year-old with an IQ of 100 is at the 50th percentile for their age group, even though their raw scores on speed-based tasks may have declined from their 25-year-old peak.

The Flynn Effect, named after researcher James Flynn (1984, 1987), is the observation that average IQ scores have risen approximately 3 points per decade throughout the 20th century across every country where longitudinal data exists. This means someone scoring 100 today would have scored approximately 115 by 1950s norms. The causes are debated but likely include improved nutrition, greater access to education, increased familiarity with abstract thinking and test-taking, and reduced childhood disease. The effect appears to be slowing or reversing in some high-income countries since the 1990s (Bratsberg & Rogeberg 2018). IQ tests must be periodically re-normed to keep the average at 100, which is why a score of 100 always means "average for today's population."

The Flynn Effect, documented by researcher James Flynn in 1984, refers to the sustained increase in raw IQ test scores observed across many countries throughout the 20th century. Average scores rose by approximately 3 points per decade in many developed nations between 1930 and 1990. The effect is attributed to improvements in education, nutrition, healthcare, and increasing familiarity with abstract thinking tasks rather than genuine increases in underlying intelligence. To keep the mean at 100, IQ tests must be periodically re-normed, which is why an IQ of 100 today represents a different absolute performance than it did in 1950. Since approximately 1990-2000, evidence from Scandinavian countries and the UK suggests the Flynn Effect has plateaued or reversed slightly — a "negative Flynn Effect" — though this is still debated and may reflect demographic changes, screen time effects, or changes in education system priorities.

The vast majority of online IQ tests are not reliable measurements of intelligence. Validated IQ tests (Wechsler Adult Intelligence Scale, Stanford-Binet, Cattell Culture Fair) require trained administrators, strictly controlled conditions, and have been calibrated against large normative samples over decades. Online tests typically lack standardisation, adequate normative data, and the breadth of subtests needed to measure the multiple components of intelligence accurately. Most online tests are also designed with a systematic upward bias — flattering the test-taker produces engagement and shares — and consistently produce scores that are 10-20 points higher than equivalent performance on validated instruments. If you want a meaningful estimate of your IQ, the only reliable route is a supervised test administered by a qualified psychologist. Mensa's supervised admission tests and the tests offered by licensed neuropsychologists are the most accessible validated options for adults.

IQ is one of the most studied predictors of life outcomes in psychology, with a substantial and consistent body of research showing correlations with educational attainment, occupational performance, income, and health. The correlation between IQ and job performance is approximately 0.4-0.5 across most occupations, which is among the strongest predictors available to employers. IQ is particularly predictive in complex, information-intensive occupations and less predictive in routine jobs. However, IQ is far from deterministic. The correlation with income, for example, is positive but leaves the large majority of variance unexplained — personality traits (particularly conscientiousness), motivation, social skills, and circumstances account for far more of the individual variation in life outcomes than IQ alone. At population level, higher IQ is associated with better health outcomes, longer life expectancy, and lower probability of criminal behaviour — but these are probabilistic associations driven by complex mediating factors, not fixed outcomes for individuals.