본문 바로가기

블로그Online and clinical tests

What is the Stanford-Binet test?

The Stanford-Binet Intelligence Scales, Fifth Edition (Roid, 2003) is a one-to-one clinical IQ test with 10 subtests and five factors (PRO-ED). How it is given and scored, and why online quizzes differ.

목차
  1. What is the Stanford-Binet test?
  2. What does the Stanford-Binet measure?
  3. How the Stanford-Binet is given
  4. How long it takes
  5. How the Stanford-Binet is scored
  6. Score ranges and the manual's labels
  7. Every score carries error
  8. Who uses the Stanford-Binet, and for what
  9. Stanford-Binet vs Wechsler
  10. Online "Stanford-Binet" tests and the real SB5
  11. A short history of the Stanford-Binet
  12. Choosing a test for what you want to know
  13. Frequently asked questions
  14. What is the Stanford-Binet test used for?
  15. What is a good score on the Stanford-Binet test?
  16. How much does the Stanford-Binet test cost?
  17. Who is the father of the IQ test?
  18. Is an IQ of 130 gifted, and how high can an SB5 score go?
  19. Is a free online Stanford-Binet test the real one?

핵심 요약

  1. The Stanford-Binet test is a clinical IQ test given one to one; its current edition is the SB5 by Gale H. Roid, published in 2003 (PRO-ED; WPS).
  2. The SB5 measures fluid reasoning, knowledge, quantitative reasoning, visual-spatial processing and working memory, each verbally and nonverbally (PRO-ED).
  3. Scores on the SB5 have an average of 100 and a standard deviation of 15 (PRO-ED); each is an estimate with a range that can differ between tests and sittings.
  4. The SB5 manual's labels describe score ranges on one test, not a person; for questions about giftedness or disability, ask a qualified psychologist.
  5. An online quiz named Stanford-Binet is not the published SB5, which a trained examiner gives in person (PAR; PRO-ED).

Short answer: The Stanford-Binet test is a clinical IQ test given one to one by a trained examiner. Its current version is the Stanford-Binet Intelligence Scales, Fifth Edition (SB5), written by Gale H. Roid and published in 2003 (PRO-ED; WPS). It covers ages 2 to 85 and older and uses 10 subtests to measure five factors: fluid reasoning, knowledge, quantitative reasoning, visual-spatial processing and working memory (PRO-ED). It reports IQ scores with an average of 100 and a standard deviation of 15 (PRO-ED), and each score is an estimate with a range.

SB5 at a glance:

  • Full name: Stanford-Binet Intelligence Scales, Fifth Edition, or SB5, the current edition as of October 2026 (PRO-ED catalog, read 11 October 2026).
  • Published: 2003, first by Riverside Publishing; PRO-ED publishes it now (Sansone and colleagues, Journal of Neurodevelopmental Disorders, 2014; Buros Center for Testing), and PAR and WPS also sell it (PAR; WPS).
  • Ages: 2 to 85 and older, in one kit (PRO-ED).
  • Format: in person and one to one (PRO-ED; PAR).
  • Structure: 10 subtests, five verbal and five nonverbal, forming five factor indexes (PRO-ED).
  • Scores: a Full Scale IQ, Verbal and Nonverbal IQ, Brief IQ and five factor indexes, each from 40 to 160 (PRO-ED).
  • Time: about 5 minutes per subtest (PRO-ED; PAR).

What is the Stanford-Binet test?

The Stanford-Binet line dates to Lewis Terman's 1916 revision of Binet's test (Sage Encyclopedia of Career Development), and the SB5 is its current edition (PRO-ED catalog, read 11 October 2026). PRO-ED, the publisher, calls it a "collection of 10 subtests, providing Nonverbal, Verbal, and Full Scale IQ scores and other diagnostic indexes" (PRO-ED product page, read 11 October 2026).

Your results are compared with people your own age. The SB5's norm sample is 4,800 people aged 2 to 85 and older, which PRO-ED says "closely matches the 2000 U.S. Census". No sixth edition was listed by PRO-ED, PAR or WPS when we checked on 11 October 2026.

What does the Stanford-Binet measure?

The SB5 measures five broad abilities, each with one nonverbal and one verbal subtest. PRO-ED says: "Two subtests (one verbal and one nonverbal) combine to form each of the 5 factor indexes." The ten subtests are named by domain and factor (Grieder and colleagues, Assessment, 2022; PRO-ED):

  • Fluid reasoning
    • Nonverbal Fluid Reasoning (Matrices, one of the two starting subtests)
    • Verbal Fluid Reasoning
  • Knowledge, which Radtke and colleagues (Health Psychology Report, 2025) map onto crystallized intelligence
    • Nonverbal Knowledge
    • Verbal Knowledge (Vocabulary, the other starting subtest)
  • Quantitative reasoning
    • Nonverbal Quantitative Reasoning
    • Verbal Quantitative Reasoning
  • Visual-spatial processing
    • Nonverbal Visual-Spatial Processing
    • Verbal Visual-Spatial Processing
  • Working memory, which Radtke and colleagues (2025) map onto short-term memory
    • Nonverbal Working Memory
    • Verbal Working Memory

The five nonverbal subtests make the Nonverbal IQ and the five verbal ones make the Verbal IQ (PRO-ED). PRO-ED says the nonverbal half "requires no (or minimal) verbal responses from the examinee".

That mix is typical. A 1996 APA task force report (American Psychologist) said that measures such as "the Wechsler tests and the Stanford-Binet, include many different types of items, both verbal and nonverbal." For how parts become one number, see what an IQ test measures. For the first factor on its own, see what fluid intelligence is.

How the Stanford-Binet is given

The SB5 is an in-person, print test (PAR). PRO-ED describes fold-out item books that stand between you and the examiner, pictures on your side and directions on theirs, plus a box of blocks, toys, chips and other materials. A session runs in a set order:

  1. Two starting subtests. Matrices and Vocabulary come first. WPS says they "function as an Abbreviated Battery, producing a Brief IQ score that determines the starting point for the remaining subtests."
  2. The nonverbal section, at the level those results suggest.
  3. The verbal section. PRO-ED's record form follows "the initial subtests, the Nonverbal sections, and Verbal sections in that order."
  4. Scoring, by hand or with PRO-ED's online scoring system for examiners.
  5. Interpretation. "A trained psychologist evaluates and interprets an individual's performance on the scale's subtests to discover strengths and weaknesses and offer recommendations" (Gale Encyclopedia of Children's Health).

Starting at your level is deliberate. PRO-ED says the design, built on item response theory, "increases the precision of measurement by tailoring the difficulty of the items to examinee levels, and saves time".

How long it takes

PRO-ED gives "Approximately 5 minutes for each of 10 subtests". By that arithmetic, testing takes about 50 minutes before breaks, and no publisher page gives one figure for a whole session.

Real sessions vary. In a dataset of 419,135 assessment results for children and youths aged 3 to 18 tested with the Polish adaptation of the SB5, testing averaged about 30 minutes for the youngest and up to about 90 minutes for older children and youths (Scientific Data, 2024). For the Early SB5, whose full battery covers ages 2 to 5, PRO-ED lists 30 to 50 minutes.

How the Stanford-Binet is scored

The SB5 uses a deviation IQ: your score shows where you stand among people your own age, on a scale with an average of 100 and a standard deviation of 15 (PRO-ED).

  • Subtest scaled scores: an average of 10, a standard deviation of 3, a range of 1 to 19 (PRO-ED).
  • Composite scores: the five factor indexes, the Verbal IQ, Nonverbal IQ, Brief IQ and Full Scale IQ, each on the 100 and 15 scale with a range of 40 to 160 (PRO-ED).
  • Extras: percentile ranks, confidence intervals, age equivalents and change-sensitive scores, which PRO-ED says "provide a means to compare changes in an individual's scores over time" (Texas Education Agency; PRO-ED).

The 1916 Stanford-Binet used a ratio IQ, mental age divided by chronological age (Sage Encyclopedia of Career Development), and Terman "multiplied the quotient by 100 to yield a ‘normal’ IQ score of 100" (Ackerman, Journal of Intelligence, 2022). Form L-M, issued in 1960 (Psychology's Feminist Voices), switched to a deviation IQ, according to the Sage Encyclopedia of Career Development, and the 1986 fourth edition used a standard deviation of 16 (Gale Encyclopedia of Mental Disorders). For what 100 and 15 mean, see what the average IQ is.

Score ranges and the manual's labels

The SB5 manual labels each Full Scale IQ range. This list is "the test publisher classification" as a 2018 study in Neuropsychiatric Disease and Treatment reports it, citing Roid (2003):

  1. 145 to 160: very gifted or highly advanced
  2. 130 to 144: gifted or very advanced
  3. 120 to 129: superior
  4. 110 to 119: high average
  5. 90 to 109: average
  6. 80 to 89: low average
  7. 70 to 79: borderline impaired or delayed
  8. 55 to 69: mildly impaired or delayed
  9. 40 to 54: moderately impaired or delayed

The labels describe score ranges on one test. They are not a diagnosis and do not decide whether a child is gifted or has a disability. For that question, ask a school psychologist or licensed psychologist, who can look at more than one score.

By the normal curve with an average of 100 and a standard deviation of 15, 130 sits about at the 98th percentile and 145 about at the 99.9th (our calculation). Beyond the standard range, PRO-ED says the Interpretive Manual adds an Extended IQ that reaches "to an IQ of 10" at the low end and "to an IQ of 225" at the high end.

Every score carries error

A Stanford-Binet score is an estimate with a range, and it can differ between tests and between sittings. Schmukle (Assessment, 2025) writes that test administrators should "not take a participant’s observed score at face value" because of measurement error "that occurs due to imperfect reliability." The SB5 report includes confidence intervals (Texas Education Agency).

PAR and PRO-ED report that the SB5's reliabilities for the Full Scale, Nonverbal and Verbal IQ "range from .95 to .98". Our calculation from those figures and a standard deviation of 15: a 95% band runs roughly ±4 to ±7 points around the score.

A short form and the full test can disagree too. In the norm sample of the German adaptation, 1,829 people, the Full Scale IQ and the two-subtest Abbreviated Battery IQ differed by 8.12 points on average (Grieder and colleagues, 2022), and individual agreement was "often unsatisfactory, especially toward the tails of the IQ distribution." More in are IQ tests accurate?.

Who uses the Stanford-Binet, and for what

Schools, psychologists and clinics use the SB5 as one part of an evaluation:

  • Disability and special education evaluations. PAR calls it "Helpful in diagnosing a wide variety of developmental disabilities and exceptionalities". PRO-ED says that under IDEA (2004) it documents "the cognitive strengths and weaknesses of children, adolescents, and adults with learning difficulties, delays, and disabilities."
  • Giftedness assessment. PRO-ED calls it "one of the major instruments in assessment of intellectual giftedness".
  • Tracking development. The Sage Encyclopedia of Career Development lists school placement and "tracking intellectual development".

There is no single gifted cutoff. A 2024 systematic review in Frontiers in Psychology found that studies using the Stanford-Binet set criteria that "mostly fell within the scores 120–135". These are research criteria; for a real child, ask a school psychologist or licensed psychologist.

Stanford-Binet vs Wechsler

  • Age range. One SB5 kit covers ages 2 to 85 and older (PRO-ED). Pearson splits its Wechsler tests: the WISC-V (2014) for ages 6 to 16, the WAIS-5 (2024) for ages 16 to 90 (Pearson product pages, read 11 October 2026).
  • What they measure. Only the SB5 has quantitative reasoning among its five main factors. Pearson's WISC-V page (read 11 October 2026) lists it only as an extra, ancillary score, the Quantitative Reasoning Index. The Wechsler tests, the WISC-V and the WAIS-5, both have a processing speed index among their five primary indexes, and the SB5 has no processing speed factor (Pearson's WISC-V and WAIS-5 pages, read 11 October 2026; Radtke and colleagues, Health Psychology Report, 2025). The SB5 also measures each factor twice, once verbally and once nonverbally (PRO-ED).
  • Who can buy them. Pearson lists the WISC-V and WAIS-5 at Qualification Level C, and PAR and WPS list the SB5 at the same level (Pearson; PAR; WPS).

Whichever test is given, the score is an estimate with a range (Schmukle, Assessment, 2025).

Online "Stanford-Binet" tests and the real SB5

The SB5 is sold in print and given in person (PAR; PRO-ED). Its online products, such as a scoring and report system, are for examiners (PRO-ED catalog, read 11 October 2026). No publisher page we read offers an online or self-administered SB5 to the public, so a web quiz called "Stanford-Binet" is not the published SB5 and does not use its norms.

You also cannot simply buy the kit. PRO-ED sells Level C tests to people with a doctorate in psychology, education or a closely related field with formal assessment training, or a state license or certification in a related field, or full active membership in a professional organization such as the APA or NASP "that requires training and experience in the relevant area of assessment" (PRO-ED qualifications policy). WPS adds that it "only qualifies individuals for test purchase, not for test use." For more on the difference, see online and clinical IQ tests.

A short history of the Stanford-Binet

  • 1905: Alfred Binet and Théodore Simon build a test in France to find school children who needed special education (Sage Encyclopedia of Career Development).
  • 1916: Lewis Terman at Stanford University completes the first US revision (Sage).
  • 1937: Terman and Maud Merrill publish two parallel forms, L and M, according to Psychology's Feminist Voices.
  • 1960: Merrill issues Form L-M, combining the best items of both (Sage; Psychology's Feminist Voices).
  • 1973: Robert Thorndike releases revised norms (Sage).
  • 1986: Thorndike, Elizabeth Hagen and Jerome Sattler publish the fourth edition, a point scale (Sage) with 15 subtests (Gale).
  • 2003: Gale H. Roid's fifth edition, the SB5 (PRO-ED; WPS).

Holden and Tanenbaum (Journal of Intelligence, 2023) write that by 1939 Binet's scale had 77 versions, and that the Stanford-Binet was "the most widely used intelligence test in the U.S. in the 1920s and 1930s."

Choosing a test for what you want to know

For a school, health or legal question, ask a school psychologist or licensed psychologist which clinical test fits, and let them interpret the result.

For a quick, non-clinical reading of one kind of reasoning, WhatIQ's test uses 24 timed picture puzzles, takes about 15 minutes and gives adults an estimate on the standard IQ scale with the range it most likely sits in. The site says it is "not a clinical assessment" and its terms say it is not given by a psychologist (whatiq.com and whatiq.com/terms, read 11 October 2026).

Frequently asked questions

What is the Stanford-Binet test used for?

Psychologists in schools and clinics use the current Stanford-Binet, the SB5, in evaluations for learning difficulties, delays, disabilities and giftedness, according to its sellers PAR and PRO-ED (read 11 October 2026). PRO-ED says the SB5 documents "the cognitive strengths and weaknesses of children, adolescents, and adults". The Sage Encyclopedia of Career Development adds school placement and tracking intellectual development. A test score is one input; for what it means for a child or adult, ask a school psychologist or licensed psychologist.

What is a good score on the Stanford-Binet test?

There is no pass mark. SB5 scores have an average of 100 and a standard deviation of 15 (PRO-ED). In the manual's labels, as a 2018 study in Neuropsychiatric Disease and Treatment reports them, 90 to 109 is "average", 110 to 119 "high average" and 120 to 129 "superior". Each score is an estimate with a range, and it can differ between tests and sittings (Schmukle, Assessment, 2025). What a score means for a person is for a psychologist to interpret.

How much does the Stanford-Binet test cost?

It depends on who gives it. PRO-ED, PAR and WPS sell the SB5 kit only to qualified professionals at Qualification Level C, such as people with a doctorate in psychology or education or a state license in a related field (PRO-ED qualifications policy; PAR; WPS, read 11 October 2026). So what you pay for is an evaluation, and the provider sets its price. Ask the psychologist or clinic for their fee and ask your insurer what it covers. The University of Washington's LEARN Clinic notes that "Many insurance companies do not fully cover this type of evaluation."

Who is the father of the IQ test?

The Stanford-Binet began with Alfred Binet: with Théodore Simon he built a test in France in 1905 to find school children who needed special education (Sage Encyclopedia of Career Development). William Stern first presented the intelligence quotient in 1912 (Heinemann, Journal of Intelligence, 2023). Lewis Terman at Stanford University turned Binet's test into the Stanford-Binet in 1916 (Sage) and multiplied Stern's quotient by 100, so an average score became 100 (Ackerman, Journal of Intelligence, 2022).

Is an IQ of 130 gifted, and how high can an SB5 score go?

Ask a school psychologist or licensed psychologist whether a person is gifted; they can look at more than one score. The SB5 manual labels the 130 to 144 range gifted or very advanced, as a 2018 study in Neuropsychiatric Disease and Treatment reports it, but that describes a score range on one test, not a person. A 2024 review in Frontiers in Psychology found gifted criteria in studies using the Stanford-Binet mostly between 120 and 135. "Genius" is not one of the manual's labels. The standard SB5 scale stops at 160, and PRO-ED says the optional Extended IQ reaches 225. Each score is an estimate with a range.

Is a free online Stanford-Binet test the real one?

No. The current Stanford-Binet, the SB5, is sold in print and given in person by a trained examiner (PAR; PRO-ED, read 11 October 2026). Its online products, such as a scoring system and digital manuals, are for examiners (PRO-ED). None of the publisher pages we read offers an online or self-administered SB5 to the public, so a web quiz with the name is not the published SB5 and does not use its norms. For a clinical score, see a psychologist.

글쓴이 소개

WhatIQ Editorial

The team behind the WhatIQ test

We build and score the WhatIQ test, and write here about how it works and what a score can and cannot say.

내 위치 확인하기

24문제, 약 15분, 짧은 워밍업으로 시작해요.

테스트 시작하기