Raven's Progressive Matrices — the gold standard IQ test

Raven's Progressive Matrices (RPM) is one of the most widely used and extensively researched nonverbal intelligence tests in psychology. Since its creation in 1936, it has been administered to millions of people across dozens of countries and has become a foundational tool in both clinical assessment and cognitive research.

Its enduring popularity stems from a combination of psychometric strength and practical simplicity. The test requires no language, no cultural knowledge, and no formal education to attempt. It measures pure reasoning ability — the capacity to perceive patterns, infer rules, and apply logic to novel visual problems. This makes it one of the strongest available measures of fluid intelligence and Spearman's g factor.

The test was created by John C. Raven, a British psychologist, in 1936. Raven developed the test as part of his research into the genetic and environmental contributions to intelligence, conducted under the supervision of Charles Spearman at University College London.

Spearman had proposed the existence of a general factor of intelligence (g) based on the observation that people who perform well on one type of cognitive test tend to perform well on others. Raven set out to create a test that would measure this general factor as directly as possible, with minimal contamination from language ability, educational background, or cultural familiarity.

The result was a test consisting entirely of visual pattern completion problems. The original version, known as the Standard Progressive Matrices (SPM), was first published in 1938. It rapidly gained acceptance in both clinical and military settings, and was used extensively during World War II for personnel selection in the British armed forces.

Each item in Raven's Progressive Matrices presents a visual pattern arranged in a matrix format, typically a 3x3 grid with the bottom-right element missing. The patterns follow specific rules governing attributes such as shape, size, color, orientation, number, and position. The test-taker must analyze the rows and columns of the matrix, identify the governing rules, and select the correct missing element from a set of six to eight options.

The test is "progressive" in that items increase in difficulty as the test proceeds. Early items may involve a single, straightforward rule (such as a shape that rotates 90 degrees in each cell). Later items involve multiple simultaneous rules that must be identified and applied together — for example, a shape that both rotates and changes color while a separate element follows an independent numerical progression.

This progressive structure means the test spans a wide difficulty range and can discriminate effectively across most of the IQ distribution, from below average to highly gifted. For details on how raw scores from such tests are converted to IQ scores, see how IQ is calculated.

standard progressive matrices (SPM)

The original version, containing 60 items organized in five sets of 12 (Sets A through E). It is suitable for the general population aged 6 to 80 and covers the full range of intellectual ability from the 5th to the 95th percentile. The SPM is the most commonly administered version in educational and occupational settings.

coloured progressive matrices (CPM)

A simplified version designed for young children (ages 5 to 11), elderly adults, and individuals with moderate or severe learning difficulties. It contains 36 items in three sets (A, Ab, B), with colored backgrounds to make the patterns more visually engaging and accessible. Items from the CPM correspond to the easier sections of the SPM.

advanced progressive matrices (APM)

A more difficult version designed for individuals of above-average ability, typically used with university students, professionals, and in high-stakes selection contexts. It contains 48 items in two sets and provides better discrimination at the upper end of the ability distribution. The APM is particularly useful for identifying giftedness and for research into high-level cognitive processes.

A test is considered culture-fair when its content does not advantage or disadvantage test-takers based on their cultural background, language, or educational experience. Raven's Progressive Matrices achieves this through several design choices.

The test uses only abstract geometric patterns. There are no words, numbers, or culturally specific symbols. The instructions can be communicated through demonstration rather than verbal explanation. The reasoning required is inductive and spatial, drawing on cognitive processes that are not contingent on specific educational curricula.

No intelligence test is perfectly culture-free. Familiarity with multiple-choice formats, experience with timed testing, and comfort with abstract visual stimuli can all vary across populations. However, RPM minimizes these biases more effectively than most alternatives, which is why it has been the instrument of choice for cross-cultural intelligence research for nearly a century.

Factor-analytic studies consistently show that Raven's Progressive Matrices has one of the highest g-loadings of any single test. In other words, performance on RPM is more strongly correlated with the general factor of intelligence than performance on most other cognitive measures.

This makes RPM an efficient tool: a single test, taking 20 to 45 minutes, can provide a reasonably accurate estimate of general cognitive ability. Full IQ batteries like the WAIS-IV take several hours and assess multiple domains (verbal, perceptual, working memory, processing speed), but the composite score they produce correlates very highly (typically r = 0.7 to 0.8) with RPM scores.

For a broader discussion of what IQ measures and how different tests approach the same underlying construct, see our introductory article.

In research, RPM is used to study group differences in intelligence, the heritability of cognitive ability, the effects of environmental interventions on IQ, and the neurological underpinnings of reasoning. The Flynn effect — the observed rise in IQ scores over generations — was first documented using RPM data, among other tests.

In clinical settings, RPM is used for cognitive screening, educational placement, giftedness identification, and as part of neuropsychological assessment batteries. It is particularly useful when testing individuals with language disorders, hearing impairments, or limited English proficiency, because it requires no verbal response.

The test has been normed and validated in dozens of countries, with population-specific percentile tables available for most major demographics. For an explanation of how these norms translate raw scores into standard IQ scores, see the IQ scale and score chart.