Stanford-Binet 5 (SB5) Scores Explained: Ranges, Subtests and Age Range

The Stanford-Binet Intelligence Scales, Fifth Edition (SB5, also written Stanford-Binet 5 or SB-5) reports IQ scores from 40 to 160 on a scale with a mean of 100 and a standard deviation of 15. It tests people from age 2 to 85 and older with ten subtests: five factors, each measured in a nonverbal and a verbal form. The Full Scale IQ uses all ten, the short Abbreviated Battery IQ (ABIQ) uses two, and an Extended IQ in a separate manual reaches down to 10 and up to 225. We found no announcement of a sixth edition. This guide sets out the subtests, the score scales, the real ceiling, how SB5 compares with the Wechsler tests and what a school can ask of it.

What is the Stanford-Binet 5, and what is its age range?

The Stanford-Binet line is the oldest in intelligence testing: Alfred Binet and Theodore Simon published the first version in 1905, and Lewis Terman's Stanford revision of 1916 gave it its name. The fifth edition was published in 2003 and written by Gale Roid, after a seven-year revision. It is individually administered by a trained examiner, and the manual asks for graduate or professional-level training in psychological assessment. The publisher's description of it is the first intellectual battery to cover five cognitive factors in both the nonverbal and the verbal domain.

The age range is 2 to 85 and older. The norm sample was 4,800 people aged 2 to 96, with 85 and older as the top age group in the norm tables, and a further 1,365 people from documented special groups were tested during standardisation. That is a wider span than any single Wechsler scale: the WAIS-5 covers 16 to 90, the WISC-V 6 to 16 and the WPPSI-IV 2:6 to 7:7 (see our guides to WAIS-IV and WAIS-5 and WPPSI-IV scores). Buros and Riverside's own reference lists now name PRO-ED as the publisher, and the kit is sold by PRO-ED and listed in Riverside Insights' store. For the history of the earlier editions, our blog's Stanford-Binet IQ test overview covers the ground.

What are the ten SB5 subtests and the five factors?

SB5 crosses five factors with two domains. Each factor has one nonverbal and one verbal subtest, which is why there are ten. The titles below are from the Essentials of SB5 Assessment excerpt, which reproduces the technical manual's list; some subtests change content at the lowest levels, so the names show the sequence.

• Fluid Reasoning: nonverbal Object Series/Matrices; verbal Early Reasoning, followed by Verbal Absurdities and Verbal Analogies
• Knowledge: nonverbal Procedural Knowledge (gestures), followed by Picture Absurdities; verbal Vocabulary
• Quantitative Reasoning: Nonverbal Quantitative Reasoning and Verbal Quantitative Reasoning
• Visual-Spatial Processing: nonverbal Form Board, followed by Form Patterns; verbal Position and Direction
• Working Memory: nonverbal Delayed Response, followed by Block Span; verbal Memory for Sentences, followed by Last Word

The test starts with two routing subtests, Object Series/Matrices and Vocabulary. Their results decide the starting level, from six functional levels, for the other eight, so the test adapts and has no fixed question order. An older version of this page said the subtest names were not public; they are, in the Wiley excerpt, though the items and the manual stay with the publisher.

What are the SB5 score scales?

Every composite is a standard score with a mean of 100 and a standard deviation of 15, and a range of 40 to 160. That covers the Full Scale IQ (FSIQ, all ten subtests), the Nonverbal IQ (NVIQ, the five nonverbal subtests), the Verbal IQ (VIQ, the five verbal ones), the five factor indexes and the Abbreviated Battery IQ. The ten subtests are reported as scaled scores with a mean of 10, a standard deviation of 3 and a range of 1 to 19. The 15-point scale is new in SB5: the 1960 third revision, and the fourth edition according to an encyclopedia entry, used a standard deviation of 16, so an older Stanford-Binet score is not on the same ruler. On the SD-16 ruler a 132 is two standard deviations above the mean, which is a 130 on the SD-15 scale; our guide to SD 15 against SD 16 shows the conversion.

• Percentile of the composites, if scores are normal: 85 is the 16th, 100 the 50th, 115 the 84th, 130 the 98th (about 1 in 44), 145 the 99.87th, 160 about 1 in 31,600
• Reliability: FSIQ .98 (standard error of measurement 2.30), VIQ .96, NVIQ .95, ABIQ .91 (4.55), factor indexes .90 to .92
• Subtest reliabilities: .84 to .89 on average across age groups

The percentiles are arithmetic on a normal curve, from our percentile calculator logic, not a table from the manual. They show a limit the publisher's own numbers imply: a normal curve would put about 0.15 people at or above 160 in a sample of 4,800, so the top of the standard scale rests on extrapolation, not on a measured percentile.

What is the highest SB5 score possible?

On the standard scales it is 160. The publisher offers more through the Extended IQ (EXIQ), described in the SB5 Interpretive Manual as supporting Full Scale IQ scores substantially lower than 40, for example to an IQ of 10, or higher than 160, up to an IQ of 225. It exists for examinees who run off either end of the standard tables. It matters when you read a claim such as a Stanford-Binet 180 or 200: a standard SB5 composite cannot produce it. Such a number is either an Extended IQ, a score from an older edition, or a figure with no source. We could not read the Extended IQ tables themselves, so how well the norms support the tail is not something we can vouch for. Our guide to the highest IQ score a test can report covers the general problem.

How is the Abbreviated Battery IQ (ABIQ) scored, and how reliable is it?

The ABIQ is built from the two routing subtests only, Object Series/Matrices and Vocabulary, and the publisher gives 15 to 20 minutes for it against 45 to 75 for the full battery; each half, nonverbal or verbal, takes about 30 minutes alone. The reliability gap is the cost. With a standard error of measurement of 4.55, an ABIQ of 130 carries a 95% band of roughly 121 to 139 by our arithmetic (the standard error times 1.96), while the FSIQ's 2.30 gives about 125 to 135. The ABIQ is a screening estimate, the same caution our guide to full-scale IQ against index scores gives for any composite built from fewer subtests than the total, and our margin of error guide explains the band.

What do SB5 scores mean, and what are the score ranges?

Read as any score on a mean-100, SD-15 scale: 100 is average and about two-thirds of people fall between 85 and 115. The descriptive labels below are the ones commonly published with the test; the examiner's manual with its own table is sold, so we could confirm them only at the level of a reference work and not from the manual. A shorter test normed again on 2023 census data is covered in RIAS-2 scores explained.

• 145 to 160: Very gifted or highly advanced
• 130 to 144: Gifted or very advanced
• 120 to 129: Superior
• 110 to 119: High average
• 90 to 109: Average
• 80 to 89: Low average
• 70 to 79: Borderline impaired or delayed
• 55 to 69: Mildly impaired or delayed
• 40 to 54: Moderately impaired or delayed

The table stops at 160 and 40, so a score from the Extended IQ range has no label here. For other scores see the guide to an IQ of 110 and our Mensa requirement guide, which lists the SB5 and Wechsler lines side by side. For a test given without spoken instructions, see Leiter-3 scores explained.

What SB5 score counts as gifted?

The publisher sets no single gifted cutoff; schools and programmes do. One checkable example is the Davidson Academy of Nevada, which lists the Stanford-Binet Intelligence Scales, Fifth Edition among its accepted tests and asks for a standard score of 145 or above on at least one of the verbal, nonverbal or full scale. That is 15 points above the 130 where the SB5 label turns to Gifted. A threshold should always name the composite and the edition, since the two tests do not always agree, as the next section shows. See also our guide to universal screening for gifted programs.

How does the SB5 compare with the Wechsler scales (SB5 vs WAIS)?

The families share a scale but not a battery. In the manual, the SB5 Full Scale IQ correlated .84 with the WISC-III (N = 66), .82 with the WAIS-III (N = 87) and .83 with the WPPSI-R (N = 71), reproduced from the Technical Manual in the Wiley excerpt; we found no manual correlations with the WISC-V or the WAIS-5. A peer-reviewed comparison of 30 typically developing 12- to 14-year-olds (Wilson and Gilmore, 2012) found Full Scale IQs significantly higher on the WISC-IV than on the SB5 and concluded the two cannot be presumed interchangeable. A high correlation does not mean the same score for the same child, which is why a single result is an estimate. Our guide to WISC-V index scores covers the current Wechsler children's scale, and DAS-II scores explained a third individual test.

Is there a Stanford-Binet 6, and is the SB5 still used?

We found no announcement and no release date for a Stanford-Binet 6. PRO-ED, Riverside Insights' store and Stoelting each list the fifth edition only, and none mentions a successor. An earlier version of this page cited a norming project for a sixth edition; PRO-ED's norming portal needs a login and its public pages list no projects, so we have removed that claim. The SB5 is still sold, and its 2003 norm sample is more than twenty years old, which is the issue behind the Flynn effect: our guide to why IQ norms expire explains why the age of a norm sample matters. The WISC-V has norms from 2013 and 2014 and a successor is in preparation; see the WISC-6 release date and what is unknown.

An SB5 score is a band on the mean-100, SD-15 scale, best read as five factor scores plus a total, and only as current as its norm sample. The IQ Metrics test reports a score on the same scale with its percentile shown beside it. For a test built on four processing scales, see CAS2 scores explained.

Common questions