Cattell Culture Fair Intelligence Test Explained

The Cattell Culture Fair Intelligence Test (CFIT) is a nonverbal IQ test built to measure reasoning ability without leaning on the vocabulary, general knowledge or schooling that verbal tests reward. Its four subtests are pictures and shapes, not words, and its standard score uses an unusual spread — a standard deviation of 24 rather than the 15 used by the Wechsler scales and the Stanford-Binet 5, which this piece explains alongside how the test is scored, how well it actually avoids cultural bias, and where it is still used three-quarters of a century after it was built.

What is the Cattell Culture Fair Intelligence Test?

Psychologist Raymond Cattell created the CFIT in 1949 as an attempt to measure cognitive ability without the sociocultural and environmental influences that shape performance on more traditional, language-heavy tests. It grew out of Cattell's own theory that general intelligence splits into two parts: fluid intelligence (gf), the biologically based capacity to reason with new problems, and crystallized intelligence (gc), the knowledge and skill built up through culture and schooling. The CFIT targets gf specifically — our guide to CHC theory covers how that fluid/crystallized split became the backbone of most modern test batteries, and our blog's piece on fluid versus crystallized intelligence explains the distinction on its own.

The CFIT was originally published by Cattell's own Institute for Personality and Ability Testing (IPAT), which he founded the same year the test was released, and its catalogue later passed through several rounds of assessment-industry consolidation, so no single easily citable "current publisher" can be given for the original English-language test here. A related instrument is easier to pin down: the publisher Hogrefe currently sells the CFT 20-R, a German-language revision built on the same series-classification-matrices-conditions structure, for ages 8 to 17 — evidence the underlying design is still considered useful enough to keep revising rather than retire, more than three-quarters of a century after Cattell first published it.

What do the four CFIT subtests measure?

Scales II and III of the CFIT share the same four nonverbal, timed subtests. Scale I, built for young children and adults with intellectual impairment, uses a different eight-task structure built around mazes, copying and matching drawings instead.

• Series: pick the figure that continues a sequence
• Classification: pick the figure that does not belong with the others
• Matrices: complete a grid of geometric patterns
• Conditions: pick the design that satisfies the same spatial rule shown in a sample

Scale 1 is commonly described as suited to ages four to eight and to adults with intellectual impairment; Scale 2 to children around eight to thirteen and to adults of average ability; and Scale 3 to more able teenagers, college students and higher-ability adults. Scales 2 and 3 each come in two parallel forms, A and B, which is what lets a single combined score be built from both forms together, the way the high-range society mentioned further down requires.

Every item is a picture, never a word, which is the whole design premise: a test-taker with limited English or limited formal schooling is not supposed to be at a disadvantage answering it. Our guide to whether IQ is just pattern recognition covers the broader question of what nonverbal, figural items like these actually measure.

How is the CFIT scored?

Like most IQ tests, CFIT standard scores are centered on a mean of 100. Unlike most, Cattell built the test with a standard deviation of 24 rather than the 15 used by the Wechsler scales, the Stanford-Binet 5 and most other modern instruments. Our guide to standard deviation 15 versus 16 covers why the spread a test uses changes what a given number means; a CFIT spread of 24 goes further still. A 15-point difference from 100 is under a full standard deviation on the CFIT's own scale, where it would be a full standard deviation on a Wechsler scale — so a score cannot be compared across tests without converting it first, which our score converter does directly.

Does the CFIT actually eliminate cultural bias?

Not entirely, and Cattell's own test proved it. A 1986 study administered the CFIT to a large sample of American, Nigerian and Indian adolescents and examined the items for cultural bias; the test's claim to measure a single fluid-intelligence factor held up well across all three groups, but 59% of its individual items were flagged as showing measurable bias, leading the researcher to conclude the test could not claim universal validity as a culture-fair measure. The problem sat at the item level rather than in the test's basic structure — the CFIT was still measuring one coherent factor across all three groups, which means the finding is about which specific pictures favored which culture, not about whether the test measured anything real.

That finding sits alongside a broader, honest caveat about the whole category of "culture-reduced" tests, CFIT and Raven's Progressive Matrices included: neither fully escapes the influence of culture and learning, even though both were built to minimize it. Our guide to Raven's Progressive Matrices scores covers the best-known test in the same category.

Where is a culture-reduced test like this most useful?

A test built to minimize the influence of language and schooling matters most for exactly the populations a verbal test disadvantages: people being assessed in a language they are still learning, people with limited formal education, and any comparison across countries or cultures where a shared vocabulary test would tilt toward whichever group speaks it natively. That was Cattell's stated purpose in 1949, and it remains the CFIT's main niche today — cross-cultural comparison research, like the 2025 study below, rather than the general clinical role the Wechsler scales and the Stanford-Binet 5 fill in most English-language clinics. A test with a genuinely different item type also gives psychologists a second, independent estimate of reasoning ability when a verbal score looks out of step with everyday functioning.

Is the CFIT still used today?

Yes. A 2025 study in PLoS ONE used CFIT Scale 3, Form A, to measure fluid intelligence in 106 undergraduates — 53 from HSE University in Moscow and 53 from the American University of Sharjah in the UAE — as part of a cross-cultural comparison of intelligence and creativity. That is precisely the use case Cattell built the test for: comparing reasoning ability across two groups who do not share a first language or an education system, where a vocabulary-heavy test would tell you more about which group had studied more English than about which group reasoned better.

The test also has a foothold in high-range testing: the Triple Nine Society, membership open only to people scoring at or above the top 1 in 1,000, accepts a combined raw score of 85 across the CFIT's two parallel forms, A and B, on Scale III. Our guide to the Mensa IQ score requirement covers the far more common 1-in-50 threshold for comparison — a reminder that "high-IQ society" covers a wide range of actual rarity depending on which one is doing the admitting.

The CFIT set out to measure reasoning ability on a level field, and mostly succeeds at being nonverbal even where it falls short of erasing cultural influence entirely. The IQ Metrics test reports a score on the familiar mean-100, standard-deviation-15 scale with its percentile shown alongside it.

Common questions