Culture-fair IQ tests: what matrix reasoning measures

A culture-fair IQ test measures reasoning with shapes and patterns instead of words, so that your vocabulary, your first language and what you learned in school matter less. It reduces cultural and language bias. It does not remove it, and a short online version gives a soft estimate, not a precise score.

How a matrix item works

Most culture-fair tests use matrix puzzles. You see a grid, usually three by three, with one cell empty. The figures in the grid change according to rules that run across the rows, down the columns, or both. Your job is to work out the rules and pick the figure that completes the pattern.

The rules come from a small, standard set:

• A shape changes from one row position to the next.
• A count goes up or down (more dots, more lines).
• A figure rotates by a fixed amount.
• A fill or shading pattern alternates.
• Two figures combine, or one is subtracted from another.

Easy items use one rule. Hard items stack two or three rules at once, and the wrong answer options are built to match some rules but not all.

Where culture-fair tests came from

Two tests set the format. John C. Raven published the first version of Raven's Progressive Matrices in the UK in 1938. Raymond Cattell described a "culture-free" intelligence test in a 1940 paper, and his institute released it commercially in 1949. It became known as the Culture Fair Intelligence Test, built with the explicit goal of measuring intelligence with as little cultural loading as possible.

Cattell also proposed the split that explains what these tests target, first in 1943 and in a formal test in 1963: fluid intelligence, the ability to reason through new problems, and crystallized intelligence, the knowledge and skills you have built up over time. In his 1963 study, culture-fair tests loaded on the fluid factor and traditional tests on the crystallized one. Vocabulary tests lean crystallized. Matrix tests lean fluid.

What pattern completion measures

A matrix item asks you to do several things at once:

• Spot which features of the figures are changing.
• Form a hypothesis about the rule.
• Hold that rule in mind while you look for a second one.
• Check each answer option against every rule.

Carpenter, Just and Shell (1990) analyzed how people solve Raven's items, using think-aloud reports, eye movements and errors. What separated stronger from weaker solvers was the ability to induce abstract rules and to manage many problem-solving goals in working memory at once.

So a matrix score says something about how you handle new, abstract problems: seeing structure, juggling several constraints, and checking your work. It says little about what you know.

Why "culture-fair" is not "culture-free"

Removing words removes a big source of bias. It does not remove all of it.

Familiarity with the format. Someone who grew up with puzzle books and multiple-choice tests starts with an advantage over someone who never saw an abstract grid before.

Schooling. Formal education trains the habit of looking for abstract rules. Brouwers, Van de Vijver and Van Hemert (2009) analyzed Raven's scores from 798 samples in 45 countries and found that a country's level of education and the samples' own years of schooling were each linked to higher scores.

Rising scores over time. Flynn (1987) documented gains of 5 to 25 points in a single generation across 14 nations, with some of the largest on Raven's-type tests. Gains that fast cannot be genetic. They point to changes in environment, such as schooling, nutrition and daily exposure to abstract thinking.

Speed and pressure. Timed tests reward people who stay calm under the clock. Test anxiety lowers scores for reasons that have nothing to do with reasoning.

"Culture-fair" is best read as "less culture-loaded". That is still a real improvement over a vocabulary test for anyone working in a second language.

How to take a matrix test well

A few habits give you a fairer estimate of your own reasoning:

• Work out the rule before you look at the options. The wrong answers are designed to look almost right. If you pick by elimination first, they will pull you in.
• Check rows and columns. Many items have one rule across and a different rule down. Confirm your answer fits both.
• Do not stall. If an item stumps you, make your best choice and move on. Time spent stuck on one item drains the focus you need for the rest.
• Take it rested and uninterrupted. A matrix item needs you to hold several rules in mind at once. Tiredness and distraction cost more here than on a vocabulary question.

What a culture-fair score can and can't tell you

It can tell you:

• Roughly how you reason with unfamiliar, abstract material.
• Whether your non-verbal reasoning is ahead of or behind your verbal reasoning, if you take both kinds of test.

It can't tell you:

• How much you know, or how well you apply it in your field.
• How you reason about people, money or messy real-world trade-offs.
• Anything clinical. For a formal assessment, see a licensed psychologist.

Why we pair it with a classic reasoning test

Our assessment includes two reasoning tests: a classic test with verbal and visual items, and a culture-fair matrix test. Taking both is more informative than either alone.

If your matrix range is clearly higher than your classic range, language or schooling may be holding the classic score down, which is common for people working in a second language. If the classic range is higher, your strength may lie more in verbal reasoning and learned knowledge. A gap is information, not a verdict.

Where it sits in our assessment

The culture-fair test is one of eight in our assessment. Cognitive reasoning sits in our validated evidence tier, with the Big Five. Our version is short and unproctored, so we report a range, not a single score, and it never headlines your report. A soft estimate like this should never decide who gets hired. For why short tests need that caution, read how accurate is an online IQ test?. For the personality side of the assessment, see the Big Five personality test, explained and are personality tests scientific?

See how you reason and how you operate

A matrix score tells you how you handle abstract puzzles. Our assessment runs all eight tests in one sitting, takes about an hour, and compiles them into one report that puts your reasoning next to your personality. Founder OS is live today at $9.