"IQ test" is not one test. The label stretches across individually administered clinical batteries, supervised group papers, nonverbal matrix tests and the free screeners you meet online — instruments with different authors, publishers, lengths and jobs.
The two families every IQ test belongs to
Everything below sits in one of two families.
Standardised batteries are published instruments, sold only to qualified purchasers and administered one person at a time by a psychologist or another trained professional. Each comes with a manual, a norming sample and a fixed administration procedure, and the resulting score is only meaningful because that procedure was followed.
Screeners and online tests are shorter, self-administered, and free or cheap. They use the same item formats — that is what makes them recognisable — without the administration or the norming that makes a battery's score official. Most questions in this area ("is the WAIS online?", "is a verbal IQ test a separate test?") resolve as soon as you know which family you are looking at.
Stanford-Binet Intelligence Scales
The oldest name in the field. It descends from the Binet-Simon scale that Alfred Binet and Théodore Simon built in France in 1905 to identify schoolchildren who needed extra support, adapted for English speakers at Stanford University by Lewis Terman in 1916 — which is where the "Stanford" comes from.
The current edition is the fifth, SB5, authored by Gale Roid and published by Riverside Insights. It is administered one-to-one across a very wide age range, from early childhood into late adulthood, and it is organised around five factors — fluid reasoning, knowledge, quantitative reasoning, visual-spatial processing and working memory — each sampled with both verbal and nonverbal items. That structure is why it reports more than a single number: a full-scale composite, verbal and nonverbal composites, and a profile across the five factors.
Who administers it: a qualified professional, in person, usually over about an hour. It is not sold to the public, and there is no legitimate online version.
The Wechsler scales: WAIS, WISC and WPPSI
The other name you meet constantly. David Wechsler built a family of age-banded scales rather than one test: the WAIS for adults, the WISC for school-age children and the WPPSI for preschoolers. The adult scale is now in its fifth edition, WAIS-5, published in 2024; the fourth edition, WAIS-IV, is still in wide use. All of them are published by Pearson.
A Wechsler administration runs a series of short subtests — vocabulary, similarities, block design, matrix reasoning, digit span, coding and others — which roll up into index scores covering verbal comprehension, visual-spatial reasoning, fluid reasoning, working memory and processing speed, and then into a full-scale composite. WAIS-IV grouped the visual-spatial and fluid items into a single perceptual reasoning index; the newer edition separates them. The index profile is the part professionals actually work with, because a single composite hides the variation that makes an assessment useful.
Who administers it: a qualified professional, one-to-one, typically over an hour or more.
Can you take the WAIS IQ test online?
Not in any real sense, and this is worth being blunt about. The WAIS is a restricted instrument — Pearson sells it only to purchasers who can document the required qualifications, and its scores depend on an examiner following a scripted administration, handling physical materials and observing the person being tested. Digital administration does exist: Pearson's Q-interactive delivers Wechsler subtests on tablets. There is still an examiner present, or on a supervised video link.
Anything advertised as "the WAIS, online, free" is not the WAIS — usually it is a matrix-reasoning screener wearing the name. Which makes it a different thing, worth judging as one.
Kaufman Brief Intelligence Test (KBIT-2)
Written by Alan and Nadeen Kaufman and published by Pearson, the KBIT-2 is a screener rather than a full battery, and it is open about it. It covers a wide age range, from young children to older adults, and runs in a fraction of the time a full battery takes. It has two scales: a verbal one, built from word knowledge and riddles, and a nonverbal one built from matrices, which combine into a composite.
Its job is triage — deciding whether a full assessment is warranted, re-checking an earlier result, or screening a research sample — rather than producing the detailed profile a full battery gives you. It is still administered by a trained examiner, not taken at home.
Culture-fair IQ tests: Cattell and Raven's
A culture-fair test — more carefully, a culture-reduced one — tries to strip out everything that depends on language, reading skill and culturally specific knowledge, leaving abstract figures that in principle mean the same thing anywhere.
The Cattell Culture Fair Intelligence Test, built by Raymond Cattell from 1949 and published across three scales for different ages and ability ranges, uses series, classification, matrices and topological "conditions" items, with no words at all.
Raven's Progressive Matrices, first published by John C. Raven in 1938 and now published by Pearson, does the same job with a single item type: a grid of abstract figures with one piece missing and a set of candidate pieces below it. The Standard, Coloured and Advanced versions differ in difficulty and age range rather than in kind.
The caveat matters: "culture-fair" is an aim rather than an achievement. Performance on abstract matrices still moves with schooling and with format familiarity, which is why researchers prefer the weaker word "culture-reduced".
Verbal IQ tests
There is rarely a standalone verbal IQ test. Verbal IQ is a score that comes out of a battery: the Wechsler scales reported a Verbal IQ alongside a Performance IQ up to the third edition, before index scores replaced that split.
The items look the same wherever you meet them — vocabulary, synonyms and antonyms, analogies of the A-is-to-B-as-C-is-to form, sentence completion, odd-one-out and short syllogisms. They sample reasoning expressed in language, which also makes them the item type most affected by which language you sat the test in.
Two things commonly sold as verbal reasoning tests are not IQ tests at all: graduate admissions sections and employer aptitude batteries. Same formats, different purpose, different norms.
Numerical and math IQ tests
A search for a math IQ test lands somewhere slightly different from what it sounds like. Quantitative items on a real battery — the Stanford-Binet's quantitative reasoning factor, the Wechsler's arithmetic subtest — test reasoning with numbers, not mathematical knowledge. Nobody is checking whether you still remember the quadratic formula.
The items are number series, ratios and proportions, percentages, and short word problems in which the arithmetic is deliberately easy and the structure is the hard part. That is the format our numerical reasoning screener uses, and it is why no calculator is needed for it.
Pattern recognition and spatial IQ tests
The item type most people picture when they think "IQ test": a sequence or grid of abstract shapes with a rule to find. Matrix reasoning, figure series, block design, visual puzzles and mental rotation all live here, and Raven's is the purest example of the type.
Pattern items need no reading, so they travel across languages, and they load heavily on fluid reasoning — working out a novel rule rather than retrieving a learned one. That is why almost every free online IQ test is built mostly out of them.
Proctored vs online IQ tests
A proctored test is supervised. Someone verifies who you are, controls the clock, keeps aids off the table and makes sure the conditions match the ones the test was normed under. That supervision takes three usual shapes: a psychologist administering a battery one-to-one, an invigilator running a scheduled group session, or a remote proctor watching over video.
An unsupervised online test has none of it: no identity check, no control over how long you took, what you looked up, or how often you have sat it before. That is a measurement limit, and it is why a result from one — ours included — is indicative rather than official.
Side by side
| Test | Who administers it | Item types | What it is for |
|---|---|---|---|
| Stanford-Binet (SB5) | Qualified professional, one-to-one | Verbal and nonverbal across five factors | Full individual assessment |
| Wechsler (WAIS, WISC, WPPSI) | Qualified professional, one-to-one | Many short subtests, index profile | Full individual assessment |
| Kaufman Brief (KBIT-2) | Trained examiner | Word knowledge, riddles, matrices | Screening and re-checks |
| Cattell Culture Fair | Trained examiner, often in groups | Abstract figures only | Language-reduced measurement |
| Raven's Progressive Matrices | Trained examiner or supervised session | Matrix completion only | Fluid reasoning, research, selection |
| Online IQ-style screeners | Yourself | Pattern, verbal, numerical, logical | Indicative self-assessment |
Which type you actually need
If you want a formal, documented result — a school placement, a diagnostic question, a high-IQ society application — you need a supervised administration by someone qualified to give it, and which instrument gets used is their decision rather than a menu you choose from.
If you only want to know which formats suit you, a screener does that in minutes. Ours runs forty items across pattern, verbal, numerical and logical reasoning, and the verbal and numerical screeners go deeper into one format each. Treat the output as a band and a profile of strengths, and read the score-ranges guide before attaching a number to it.
