HealthScorer

Free IQ Test (10-item pattern recognition)

Free 10-question IQ screener in 3-5 minutes. Pattern recognition only, no signup, no email, results stay in your browser. Estimated band on the.

Last updated: Sources verified:

Free, no-signup IQ test — what you are about to take

This is a 10-item pattern-recognition screener you can finish in 3 to 5 minutes. The questions test fluid reasoning — spotting the rule in a sequence of numbers or letters and choosing what comes next. The format is inspired by Raven’s Progressive Matrices (Raven 1936) and Cattell’s Culture Fair Intelligence Test (Cattell 1949), the two best-validated psychometric instruments for measuring fluid intelligence without depending on vocabulary or schooling.

Ten items. About 3 to 5 minutes. No signup, no email, no ads. Your answers stay in your browser.

✓ Pattern recognition only (no language or vocabulary required) ✓ Result mapped to the Wechsler-scale IQ band (M = 100, SD = 15) ✓ Honest framing — a 10-item screener cannot replace a full clinical test ✓ Private — results never leave your device

Scroll down to start.

How this IQ test is calculated

Your raw score is simply the number of items you answered correctly (0 to 10). The raw score then maps to one of five bands on the Wechsler scale (WAIS-IV technical manual, Wechsler 2008):

Raw correctEstimated IQ bandWechsler classificationPopulation share
9–10≥130Very superior~2% (Mensa cutoff)
7–8115–129High average → superior~14%
5–6100–114Average (upper half)~34%
3–485–99Average (lower half)~34%
0–2≤84Below average, OR test-conditions interfered~14%

The band is more reliable than the exact number. A 10-item screener has a wide confidence interval — your “true” IQ on a 60-90-minute WAIS-IV could plausibly land anywhere within ±10-15 points of the band mid-point. That is normal for brief instruments. The Raven Advanced Progressive Matrices, even at 36 items, runs a confidence interval of about ±5 points.

The reason this format works at all is the general factor of intelligence (Spearman’s g, supported by meta-analytic data since Jensen & Weiss 2002): a person’s performance on one cognitive task correlates moderately with performance on most other cognitive tasks. Pattern-recognition on number and letter sequences is one of the cleanest entry points to measure g, which is why the WAIS-IV Matrix Reasoning subtest looks remarkably similar to the items here.

What your score actually tells you (and what it doesn’t)

What this screener measures: fluid reasoning — your speed and accuracy at spotting an unfamiliar rule and projecting it forward. This is one of the four index scores in WAIS-IV (perceptual reasoning).

What it doesn’t measure:

  • Verbal comprehension — vocabulary, definitions, language reasoning
  • Working memory — how many items you can manipulate at once
  • Processing speed — how fast you can perform simple cognitive routines
  • Social cognition, creativity, practical intelligence — these are real abilities the Wechsler scale doesn’t fully capture either

Two people with the same band here can score very differently on a full clinical test if one is strong in verbal and weak in fluid, or vice versa. The band is one slice of cognitive ability, not the whole cake.

When this test is useful — and when it isn’t

Useful for:

  • Quick self-curiosity (“roughly what band am I in?”)
  • Comparing your fluid reasoning before and after sleep / stress / illness changes
  • A starting point before deciding whether a full assessment is worth the cost

Not useful for:

  • Job applications, court documents, or anything where a specific IQ number matters legally
  • Diagnosing intellectual disability, giftedness, or a cognitive disorder (only a licensed psychometrician can do this, using a full clinical test)
  • Comparing yourself with a child — paediatric IQ uses age-specific norms and a different test (WISC-V for ages 6-16)

IQ test scores vs the population — common reference points

A few numbers people search for and what they mean on the Wechsler scale:

IQPercentileNotes
702ndWechsler cutoff for “extremely low” / borderline intellectual disability
8516thOne standard deviation below the mean
10050thPopulation median by definition
11584thOne standard deviation above mean — “high average”
12091stLower edge of “superior” band
13098thMensa membership cutoff
14599.9thTwo standard deviations above 115; rare in unselected samples

These percentiles are population-relative. By construction the average shifts over time — the Flynn effect (Flynn 1987) showed industrialised countries gained ~3 IQ points per decade across the 20th century, attributed to nutrition, schooling, and abstract-thinking environments.

Sources verified 2026-05-13

The instruments behind this screener and the IQ-scale framework it maps to are all peer-reviewed:

  • Raven JC. Mental tests used in genetic studies: The performances of related individuals in tests mainly educative and mainly reproductive. MSc Thesis, University of London (1936). Original Raven Progressive Matrices.
  • Cattell RB, Cattell AKS. Handbook for the Culture Fair Intelligence Test, Scale 2 & 3. Institute for Personality and Ability Testing (1949 with later editions).
  • Wechsler D. Wechsler Adult Intelligence Scale — Fourth Edition (WAIS-IV) Technical and Interpretive Manual. Pearson (2008). Source of all band classifications used here.
  • Flynn JR. Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin 1987 (PMID 3322503).
  • Jensen AR, Weiss V. The general factor of intelligence — Meta-analytic evidence. Intelligence 30 (2002), pp. 273-289 (PMID 12126457).
  • Roberts BW, Kuncel NR, Shiner R, Caspi A, Goldberg LR. The power of personality: The comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life outcomes. Perspectives on Psychological Science 2 (2007).

Privacy

The test runs entirely in your browser. Inputs, answers, and the result band are computed locally; nothing is sent to our servers. We send one anonymous event to a privacy-respecting analytics service — your locale code and the band string (e.g. average_upper), nothing else. No raw answers, no per-item data, no identifiers.

Frequently asked questions

How accurate is a free online IQ test?
A 10-item online screener gives a wide band, not a precise IQ. Brief pattern-recognition tests correlate roughly 0.5–0.7 with full-scale WAIS-IV scores (Raven 1962 manual; Hayslip 2003 brief-form review). The screener can reliably place you in one of five bands (≤84, 85–99, 100–114, 115–129, ≥130), but the exact IQ number within a band has a confidence interval of ±10–15 points. For a precise number, only a 60–90-minute clinical test administered by a licensed psychometrician (WAIS-IV, Stanford-Binet 5) is reliable.
What IQ is considered high?
On the Wechsler scale (mean 100, SD 15), IQ scores are classified: ≥130 very superior (top 2%), 120–129 superior (next 7%), 110–119 high average (next 16%), 90–109 average (50% of adults), 80–89 low average (16%), 70–79 borderline (7%), ≤69 extremely low (2%). Mensa's standard cutoff for membership is IQ ≥130 on a Mensa-supervised standardised test. WAIS-IV 2008 Technical Manual is the source.
What does an IQ of 130 mean?
IQ 130 puts you at the 98th percentile — about 2 in 100 adults reach this score on standardised tests. It is Mensa's membership cutoff and is classified as 'very superior' on the WAIS-IV scale. Practically, it suggests strong fluid reasoning, but it does not predict career, relationships, or happiness on its own — those depend more on conscientiousness, social skills, and circumstance than on a single IQ number (Roberts 2007, Meta-analysis of personality traits and life outcomes).
What does an IQ of 120 mean?
IQ 120 sits in the 'superior' band (120–129), at roughly the 91st percentile. About 1 in 11 adults score in this range on the Wechsler scale. People with IQ 120 typically do well in abstract problem-solving and many professional fields. It is comfortably above average but below Mensa's 130 cutoff. The confidence interval on any single IQ measurement is around ±5 points, so a 120 today and 124 next year is the same person.
What does an IQ of 100 mean?
IQ 100 is the population mean by definition (Wechsler 2008). On a normal-distribution scale with SD = 15, an IQ of 100 means you scored exactly at the median — half of adults score lower, half higher. This is the most common single IQ value, around the 50th percentile. There is no 'should' about IQ 100; it is the calibration point of the entire scale.
Is IQ 115 above average?
Yes. IQ 115 sits at the 84th percentile — one full standard deviation above the population mean (Wechsler scale, M = 100, SD = 15). About 1 in 6 adults score at or above 115. It is the lower boundary of the 'high average' band (110–119) and represents comfortable above-average performance on standardised IQ tests.
What is the IQ test used by Mensa?
Mensa accepts a range of standardised supervised tests. The most-used worldwide are the Cattell Culture Fair III (CFIT), the Cattell B, Raven's Advanced Progressive Matrices, and full Wechsler tests (WAIS-IV / WISC-V). The Mensa membership threshold is the 98th percentile, equivalent to IQ ≥130 on the Wechsler scale (or ≥148 on the Cattell B scale, which uses SD = 24 instead of 15). Sources: Mensa International policy, Cattell technical manual.
How is IQ calculated from a 10-question test?
On this screener, your raw score (items correct, 0-10) maps to a Wechsler-scale band: 0-2 → IQ ≤84 (or test conditions interfered); 3-4 → IQ 85-99; 5-6 → IQ 100-114; 7-8 → IQ 115-129; 9-10 → IQ ≥130. The mapping uses the assumption that a 10-item pattern-recognition test has limited resolution (~5 bands rather than a precise number) and that the population mean for adults on this kind of task is roughly 5/10 correct in unsupervised conditions. Full clinical tests use item-response theory and norming on tens of thousands of adults to give a more precise score.
Can I take an IQ test for free online?
Yes — this one is free with no signup, no email, no ads. The trade-off is precision: a 10-item screener cannot match a full clinical test (WAIS-IV takes 60-90 minutes and costs €100-€300 in private practice). For most people, the band is the useful information. The exact number matters mainly for specific applications: gifted-programme admission, neuropsychological assessment, Mensa membership — where a supervised standardised test is required anyway.
What is fluid intelligence?
Fluid intelligence (Gf) is the ability to reason about new problems without relying on previously learned knowledge. Cattell (1971) split intelligence into fluid (Gf — abstract pattern recognition, working memory) and crystallised (Gc — accumulated knowledge, vocabulary). Raven's Matrices and Cattell's Culture Fair Test were designed specifically to measure Gf, minimising language and cultural background. This screener focuses on Gf via number and letter patterns.
Can IQ change over time?
Individual IQ is fairly stable from adolescence through middle adulthood — test-retest correlation runs around 0.8 over 10 years (Deary 2000, IQ and ageing). But population averages have shifted: the Flynn effect (Flynn 1987) documented a gain of about 3 IQ points per decade across the 20th century in industrialised countries, attributed to nutrition, education and abstract-thinking environments. Since the early 2000s, several countries have seen this trend slow or reverse slightly.
Why do I get different IQ scores on different tests?
Three reasons. First, tests sample different domains — Raven measures fluid reasoning, WAIS-IV adds verbal and working-memory subtests, and a 10-item screener has a wide confidence interval (±10–15 points). Second, test conditions vary — fatigue, language fluency, time of day, anxiety all shift scores by 5–10 points. Third, practice effects: doing the same test twice raises the second score by ~5 points even without true ability change. A 60–90-minute supervised test gives the most consistent number.
Is my data private?
Yes. The test runs entirely in your browser. Your individual answers are never sent to our servers. We log only a single anonymous event with the band (e.g. 'average upper') — never the raw score, never per-item answers, never anything that identifies you.

More health calculators

Calculators →

Sources

  1. Raven's Progressive Matrices and Vocabulary Scales — Section 1: General Overview — Raven JC (original 1936; updated editions by Pearson) (peer reviewed, retrieved 2026-05-13)
  2. Cattell Culture Fair Intelligence Test (CFIT) — Scale 2 & 3 technical manual — Cattell RB, Cattell AKS (original 1949) (peer reviewed, retrieved 2026-05-13)
  3. Wechsler Adult Intelligence Scale — Fourth Edition (WAIS-IV) Technical and Interpretive Manual — Wechsler D (2008), Pearson (guideline, retrieved 2026-05-13)
  4. Massive IQ gains in 14 nations: What IQ tests really measure (the Flynn effect) — Flynn JR, Psychological Bulletin (1987) (peer reviewed, retrieved 2026-05-13)
  5. The general factor of intelligence — Meta-analytic evidence — Jensen AR, Weiss V — Intelligence (2002) (peer reviewed, retrieved 2026-05-13)