Intelligence is measured primarily through standardized tests that assess multiple cognitive abilities, then combine those scores into a single number: the IQ score. The average IQ is set at 100, with a standard deviation of 15 points, meaning about 95% of people score between 70 and 130. But “intelligence” isn’t one thing, and the way we measure it has grown more sophisticated over the past century. Different tests capture different mental abilities, and no single score tells the whole story.
What IQ Scores Actually Mean
IQ scores follow a bell curve centered on 100. A score of 100 means you performed exactly at the population average. Each 15-point jump represents one standard deviation from the mean, so a score of 115 places you roughly in the top 16% of the population, while 85 puts you in the bottom 16%.
Scores between 70 and 130 are considered within the normal range and account for about 95% of people. Only about 5% of the population scores above 130 or below 70. Scores beyond 145 or below 55 are extremely rare, occurring in less than 1% of people. Organizations like Mensa require a score at or above the 98th percentile for membership, which translates to a full-scale IQ of 130 on the Wechsler scales or 132 on several other common tests.
The Wechsler Scales: The Most Widely Used Tests
The Wechsler tests are the backbone of modern intelligence testing. There are separate versions for adults and children, and both break intelligence into distinct cognitive areas rather than treating it as a single ability.
Adults: The WAIS-IV
The Wechsler Adult Intelligence Scale (WAIS-IV) uses 10 core subtests to produce four index scores and one overall full-scale IQ. The four areas it measures are:
- Verbal Comprehension: tested through vocabulary definitions, identifying how concepts are similar, and general knowledge questions. This index reflects your ability to understand and use language, and the vocabulary subtest is considered especially informative because it serves as a reliable indicator of overall cognitive baseline.
- Perceptual Reasoning: tested through arranging physical blocks to match a design, identifying patterns in visual matrices, and solving visual puzzles. This captures your ability to analyze and organize visual information without relying on words.
- Working Memory: tested through repeating number sequences and solving arithmetic problems in your head. This reflects how much information you can hold and manipulate mentally at one time.
- Processing Speed: tested through coding (matching symbols to numbers under time pressure) and searching for target symbols in rows. This measures how quickly you can scan, sequence, and discriminate simple visual information. The coding subtest is thought to tap a wider range of cognitive skills than symbol search alone, since it requires both visual-spatial scanning and short-term memory.
Children: The WISC-V
The Wechsler Intelligence Scale for Children (WISC-V) is structured similarly but splits the adult version’s perceptual reasoning area into two separate categories: visual-spatial ability and fluid reasoning. This gives five primary index scores instead of four, which helps clinicians identify more specific learning strengths and weaknesses. The five areas are Verbal Comprehension, Visual Spatial, Fluid Reasoning, Working Memory, and Processing Speed. This distinction matters for children because a kid who struggles with abstract pattern recognition but excels at spatial tasks (or vice versa) will show a meaningful difference across those two scores, pointing toward more targeted support.
The Stanford-Binet: The Oldest Modern IQ Test
The Stanford-Binet Intelligence Scales, now in their fifth edition (SB-5), take a slightly different approach. The test uses 10 subtests to measure five cognitive factors: fluid reasoning, knowledge, quantitative reasoning, visual-spatial processing, and working memory. One notable feature is its enormous age range. It’s validated for people from age 2 through 85 and older, making it one of the few tests that can track cognitive ability across nearly an entire lifespan. It’s commonly used in schools for gifted identification and in clinical settings to assess intellectual disability.
Non-Verbal Tests and Fluid Intelligence
Some tests deliberately strip away language to measure reasoning ability in its purest form. Raven’s Progressive Matrices, originally published in 1938 and still widely used, is a non-verbal test where each item shows a geometric pattern with a missing piece. You select which option completes the pattern from several choices. The patterns grow progressively more complex, requiring you to identify rules governing shape, orientation, shading, and quantity simultaneously.
Raven’s is considered a strong measure of what psychologists call fluid intelligence: the ability to solve novel problems without relying on previously learned information. This contrasts with crystallized intelligence, which reflects accumulated knowledge and skills (the kind captured by vocabulary and general knowledge subtests). Fluid intelligence tends to peak in early adulthood and gradually decline with age, while crystallized intelligence often continues growing well into middle age.
Because Raven’s doesn’t require reading, speaking, or cultural knowledge, it’s frequently used in cross-cultural research and with people who speak different languages. The Cattell Culture Fair Intelligence Test was designed with similar goals, using abstract figural tasks to minimize the influence of verbal ability, education level, and cultural background on scores. In practice, though, no test is completely culture-free, and performance on these “culture-fair” instruments still correlates with educational experience.
Beyond Traditional IQ: Broader Theories
Standard IQ tests focus heavily on analytical thinking, the kind of reasoning rewarded in school. Robert Sternberg’s triarchic theory argues this captures only one dimension of intelligence. His model identifies three relatively independent abilities: analytical (the logic and problem-solving that conventional tests measure), creative (generating novel ideas and solutions), and practical (the street smarts needed to navigate real-world situations effectively).
The Sternberg Triarchic Abilities Test (STAT) was developed to measure all three across verbal, quantitative, and figural domains using 36 multiple-choice items. The core argument behind this approach is that conventional intelligence tests are primarily measuring analytical ability and missing creative and practical strengths entirely. Schools, Sternberg notes, also tend to emphasize analytical achievement at the expense of creative and practical skills, which may explain why IQ correlates so strongly with academic performance but predicts real-world success less reliably.
Howard Gardner’s theory of multiple intelligences takes an even broader view, proposing that abilities like musical talent, bodily-kinesthetic coordination, and interpersonal skill represent distinct forms of intelligence rather than separate talents. These broader models haven’t replaced traditional IQ testing in clinical or educational practice, but they’ve influenced how educators think about ability and potential.
Where Intelligence Testing Happens
Full IQ testing is typically administered one-on-one by a trained psychologist and takes one to two hours. The most common reasons people encounter these tests are school placement (gifted programs or learning disability evaluations), clinical assessment (diagnosing intellectual disability, ADHD, or the cognitive effects of brain injury), and occasionally personal curiosity.
Schools often use group-administered tests like the Cognitive Abilities Test (CogAT) or the Naglieri Nonverbal Ability Test (NNAT) for screening. These are faster and cheaper than individual testing but provide less detailed information. If you’ve taken one of these school-based tests, a military aptitude test, or certain college entrance exams (like the GRE before May 1994 or the ACT before September 1989), you may already have a qualifying score on file. Mensa accepts scores from roughly 150 different standardized tests, and nearly two-thirds of their members join by submitting prior testing evidence rather than taking a new exam.
What IQ Tests Don’t Capture
IQ tests reliably measure a specific set of cognitive skills: reasoning, pattern recognition, verbal ability, memory, and processing speed. They’re well-validated predictors of academic performance and correlate moderately with job performance in complex occupations. But they don’t measure emotional regulation, motivation, social skill, creativity in open-ended domains, or the kind of adaptive problem-solving that keeps someone functioning well in unpredictable environments.
The gap between your highest and lowest index scores on a Wechsler test can be just as informative as your full-scale IQ. A person with a verbal comprehension score of 125 and a processing speed score of 85 has a very different cognitive profile than someone who scores 105 across all four areas, even if both produce a similar overall number. That pattern of strengths and weaknesses, rather than the single composite score, is often what clinicians find most useful when making educational or diagnostic decisions.