What is a full scale IQ (FSIQ)?

Short answer: A Full Scale IQ (FSIQ) is the overall score on a Wechsler intelligence test such as the WAIS-5, one number that estimates your general intellectual ability from several subtests taken together. It uses the standard IQ scale, where the average is 100 and the standard deviation is about 15 points (Hall v. Florida, 2014). On the WAIS-5, which Pearson published in 2024, seven subtests feed the FSIQ; on the WAIS-IV of 2008, ten did (Pearson's product pages, read November 1, 2026). Score reports usually print it with a percentile and a range, because every score carries some measurement error.

How a Full Scale IQ is built

Both editions build the FSIQ in five steps, as a 2025 WAIS-IV study by Albayrak Rasborg Hartogsohn and colleagues describes and Pearson's 2024 WAIS-5 sample score report shows:

1. Raw score. You earn points on each subtest you take.
2. Scaled score. Each raw score is converted, using tables for people your age, into a scaled score. "Scaled scores have a mean of 10 and a standard deviation (SD) of 3."
3. Sum of the core subtests. The scaled scores of the subtests that count toward the FSIQ are added together: seven on the WAIS-5, ten on the WAIS-IV.
4. Norm table. That sum is looked up in the test's norm table, which turns it into the FSIQ on a scale with a mean of 100 and an SD of 15. The FSIQ is never an average of your index scores.
5. Percentile and confidence interval. The report gives the FSIQ with a percentile rank and a range, on Pearson's sample report a 95% confidence interval.

Pearson's WAIS-5 sample score report, which we read on November 1, 2026, shows the steps for one sample examinee. The seven FSIQ subtests have scaled scores of 14 (Similarities), 15 (Vocabulary), 11 (Block Design), 12 (Matrix Reasoning), 10 (Figure Weights), 9 (Digit Sequencing) and 11 (Coding). They add up to 82, and the norm table converts that sum of 82 into an FSIQ of 111, at percentile rank 77, with a 95% confidence interval of 106 to 116.

The same report gives five index scores: Verbal Comprehension 124, Visual Spatial 103, Fluid Reasoning 105, Working Memory 97 and Processing Speed 108. Their mean, 107.4, is printed on the report as the comparison score mean, yet the FSIQ is 111: it comes from the subtest sum through the norm table, not from the indexes. By our calculation, 111 sits at the 76.8th percentile of an ideal normal curve, close to the report's 77th.

What are the five WAIS-5 index scores?

An index score groups related subtests on the same scale as the FSIQ, with a mean of 100 and an SD of 15 (2025 WAIS-IV study). Winter and colleagues, writing in the Journal of Intelligence in 2025, name the WAIS-5's five primary indexes, and the short description of each follows their paper:

• Verbal Comprehension Index (VCI): problem solving with language and verbally presented information.
• Visual Spatial Index (VSI): problem solving with mainly visual information, such as spatial puzzles.
• Fluid Reasoning Index (FRI): problem solving with new information (see fluid intelligence).
• Working Memory Index (WMI): holding information in mind and using it to solve a new problem.
• Processing Speed Index (PSI): solving problems quickly with known or easy-to-access information.

The WAIS-IV had four: Verbal Comprehension, Perceptual Reasoning Index (PRI), Working Memory and Processing Speed, according to the 2025 WAIS-IV study. Winter and colleagues (2025) say the WAIS-5 "split and restructured the PRI into the VSI and FRI, aligning more closely with CHC broad abilities".

Your report names its edition. If it lists four indexes including Perceptual Reasoning, it is a WAIS-IV report. Pearson's WAIS-5 FAQ, read November 1, 2026, says most practitioners move to a new edition within 8 to 12 months of release. The WAIS-5 came out in August 2024, according to Winter, Trudel and Kaufman's 2024 WAIS-5 review. So many reports written before about mid-2025 are WAIS-IV reports, by our reading of those dates.

Which subtests make up the FSIQ?

On the WAIS-5, the FSIQ is a composite of seven subtests, the 2024 WAIS-5 review says: Vocabulary, Similarities, Block Design, Matrix Reasoning, Figure Weights, Digit Sequencing and Coding. Pearson's FAQ, read November 1, 2026, says the seven-subtest model was judged "more related to general intelligence (g) psychometrically", among other reasons.

On the WAIS-IV, the FSIQ used all ten core subtests, the 2025 WAIS-IV study shows: Similarities, Vocabulary and Information; Block Design, Matrix Reasoning and Visual Puzzles; Digit Span and Arithmetic; Symbol Search and Coding. That changes what drives the total: by our calculation, working memory and processing speed subtests are 4 of the WAIS-IV's 10 and 2 of the WAIS-5's 7. The 2024 review reports that "the Processing Speed Index and Working Memory Index comprised 40% of the WAIS-IV Full Scale, a value that was reduced to 28.6% on the WAIS-5."

Examiners cannot pick your best results. Pearson's WAIS-IV FAQ answers "No." to using the highest subtest scores and allows at most two substitutions, and its WAIS-5 FAQ says "Only one substitution is permitted when deriving the FSIQ or the NMI" (both read November 1, 2026). The children's version is the WISC-V, which this article does not cover. For the WAIS itself, from its age range to how long it takes, see our guide to the WAIS.

Is a full scale IQ the same as "IQ"?

"IQ" is the everyday word for a score on an intelligence test. The FSIQ is the Wechsler name for that test's overall composite. A 2025 paper by Kim and colleagues opens: "The Full Scale IQ is a global estimate of a person’s general intelligence factor". The 2024 WAIS-5 review calls it "the g score" of the WAIS-5, the general factor that the subtests share.

Other tests have their own full-scale scores. The Stanford-Binet Fifth Edition, by Roid (2003), yields "not only a Full Scale IQ but also a Brief IQ, Verbal IQ, Nonverbal IQ", says the Texas Statewide Leadership for Autism Training summary (read November 1, 2026). Pearson's short WASI-II (2011) has two full-scale forms, FSIQ-4 and FSIQ-2.

The 2024 WAIS-5 review lists the FSIQ as one of four WAIS-5 global scores, beside the Nonverbal Index, the Nonmotor Index and the General Ability Index. Winter and colleagues (2025) note research suggesting the FSIQ "may overgeneralize an individual’s cognitive performance", which "remains a debated topic among psychologists." For what any IQ number covers, see what an IQ test measures.

How precise is a full scale IQ score?

The 2024 WAIS-5 review, reporting the test's technical manual, says that "The mean reliability of the FSIQ averaged 0.97 across the age range". Winter and colleagues (2025) put index reliabilities at 0.90 to 0.97 on the WAIS-5.

The standard error of measurement (SEM) turns reliability into points. The court in Hall v. Florida (2014) quoted the American Psychological Association: "the average SEM for the WAIS-IV is 2.16 IQ test points and the average SEM for the Stanford-Binet 5 is 2.30 IQ test points". By our calculation, a reliability of .97 gives an SEM of about 2.6 points.

The court explained the ranges: "1 SEM equates to a confidence of 68% that the measured score falls within a given score range, while 2 SEM provides a 95% confidence level that the measured score is within a broader range".

On Pearson's 2024 sample report, the FSIQ of 111 carries an SEM of 3.00 and a 95% interval of 106 to 116. On that SEM, a 95% range would be about ±5.9 points and a 90% range about ±4.9, by our calculation. Pearson's 106 to 116 is narrower. The report says its intervals are calculated with the Standard Error of Estimation, and they need not sit evenly around the score. Its Verbal Comprehension score of 124 has a 95% interval of 115 to 130, which by our calculation is 9 points below the score and 6 above. The court's summary holds for any edition: scores "should be read not as a single fixed number but as a range." Our IQ test accuracy guide covers error in more depth.

What do the percentile and the label mean?

The percentile rank compares you with people your age. Pearson's WAIS-5 FAQ, read November 1, 2026, gives the example that "an examinee with a percentile rank of 15 performs better than 15% of other examinees the same age". Our IQ percentile guide converts any score, and the IQ bell curve explains why most scores bunch near 100.

The word next to your score depends on the edition. The 2025 WAIS-IV study lists these FSIQ bands:

• 130 and above: Very superior
• 120 to 129: Superior
• 110 to 119: High average
• 90 to 109: Average
• 80 to 89: Low average
• 70 to 79: Borderline
• 69 and below: Extremely low

The WAIS-5 sample report (Pearson, 2024) uses newer words: it calls a VCI of 124 Very high, an FSIQ of 111 Above average, and scores from 97 to 109 Average. We found no public full WAIS-5 descriptor table, so we print none. Pearson's own FAQ says "Qualitative descriptors are only suggestions and are not evidence-based; alternate terms may be used as appropriate."

What are the GAI and CPI on a report?

Scorza and colleagues (2023) describe the General Ability Index on the WAIS-IV as combining Verbal Comprehension and Perceptual Reasoning. They call it an "alternative global estimate of intelligence for individuals whose working memory and processing speed indexes are significantly lower than VCI and PRI". The Cognitive Proficiency Index "includes the subtests that form the working memory index (WMI) and the processing speed index (PSI)".

Pearson's WAIS-5 FAQ, read November 1, 2026, explains how the GAI is built. "The GAI has always been conceptualized as the same subtests as the FSIQ but removing subtests explicitly contributing to the WMI or PSI, so that it can be directly compared to the FSIQ." On the WAIS-5, cognitive proficiency is one of fourteen new ancillary index scores. American Mensa, which otherwise requires an FSIQ, accepts the GAI for the WISC-IV, WISC-V and WAIS-IV only, when two index scores differ by one and a half standard deviations or more (its qualifying scores page, read November 1, 2026).

In Pearson's 2024 sample report, the GAI is 115 and the CPI 104, beside the FSIQ of 111, so the GAI sits 4 points higher. A psychologist decides whether a gap like that matters for the questions the assessment was meant to answer.

What if your index scores are far apart?

Gaps between indexes are common. Kim and colleagues (2025) write that "large intrapersonal variability in a cognitive profile is quite common". They cite a simulation by Daniel and colleagues (2007): about 10% of a simulated general population would show 34 points or more between their lowest and highest index scores, even with everyone sharing the same ability structure. In the WISC-V's standardization sample of children aged 6 to 16, about 57% had a gap of 23 points or more between their highest and lowest index scores, Kim's team adds. That is a base rate, a measure of how often a gap occurs. It says nothing about what a gap means for one person.

Whether the FSIQ still sums you up well is where researchers disagree:

• Use caution. Scorza and colleagues (2023) argue that in clinical settings the FSIQ "must be regarded with caution because it may be negatively affected by a decrease in the WMI and the PSI". In that case, they write, "the GAI may be considered as a more appropriate measure".
• The FSIQ holds up. Kim and colleagues (2025) say "the current consensus among researchers is that the FSIQ is the most reliable and valid predictor of one’s overall cognitive functioning", even with scatter.
• Open questions. Kim's team also quotes Sattler (2018): "There is little research to guide us in determining whether an FSIQ becomes invalid when the index scores show extreme variability". Their own study used a clinical sample of young people aged 6 to 16 with neurodevelopmental disorders. In that sample, the FSIQ's link to academic achievement may weaken at extreme scatter. The authors say the specific cutoffs they found need cross-validation in future studies.

The call on your profile belongs to the psychologist who tested you. An uneven profile on its own is not a diagnosis.

Can your full scale IQ change on a retest?

A little. In the WAIS-5 retest study, 201 people aged 16 to 90 took the test twice about 29 days apart, and the stability coefficient was 0.93 (the 2024 WAIS-5 review). On average, the FSIQ rose by 3.7 points for ages 16 to 34 and 35 to 69, and 2.3 points for ages 70 to 90.

For young and middle-aged adults, the same paper reports gains of about 4 to 6 points on the WAIS-III and 4.5 to 5 on the WAIS-IV, so the effect has shrunk. It also gives a wider general figure: on a second administration, Full Scale IQs rise by 5 to 8 points. Treat 3.7 as the current WAIS-5 estimate and 5 to 8 as the older, general range. Hall v. Florida (2014) lists "practice from earlier tests" among the reasons a score can move.

How an online test's single score differs

An online test that reports one overall score shows you a single number, without the subtest and index scores behind a Wechsler FSIQ. A test built from one puzzle type, such as matrix reasoning, samples a single kind of task. The WAIS-5 FSIQ combines seven subtests, from Vocabulary to Coding (the 2024 WAIS-5 review). The American Psychological Association says a psychological assessment can include tests, interviews, records and observation. It adds: "A psychologist determines what information to use based on the specific questions being asked." American Mensa does not accept unsupervised internet-based tests as proof of eligibility, its qualifying scores page said on November 1, 2026. For a quick estimate before deciding on a full assessment, WhatIQ's test has "24 picture puzzles, about 15 minutes" (its homepage, read November 1, 2026) and gives one score on the standard scale with a 90% range. Its FAQ says the test "gives an estimate from 24 puzzles, not a clinical assessment".

Frequently asked questions

What is the scale of IQ scores?

Hall v. Florida (2014) says "The mean IQ test score is 100" and puts the standard deviation at "approximately 15 points", so two standard deviations is about 30 points. Index scores on the Wechsler tests use the same scale, and subtest scaled scores use a mean of 10 and an SD of 3, according to the 2025 WAIS-IV study.

What does a 120 mean on a full scale IQ?

On the WAIS-IV labels in the 2025 WAIS-IV study, an FSIQ of 120 falls in the 120 to 129 band called Superior. The WAIS-5 uses different words, and Pearson calls its descriptors "only suggestions". Your report's percentile column says where 120 sits among people your age. Read it with its confidence interval.

Is a low index score bad if my FSIQ is fine?

Gaps are common: Kim and colleagues (2025) say large differences between a person's index scores are "quite common", and researchers disagree about how much they change what the FSIQ means. Scorza and colleagues (2023) urge caution when working memory and processing speed are lower; Kim's team says the FSIQ stays the best overall predictor. Ask the psychologist who tested you what the profile means for your questions.

What is the difference between FSIQ and GAI?

The GAI uses the FSIQ's subtests minus the ones that count toward working memory and processing speed, "so that it can be directly compared to the FSIQ," Pearson's WAIS-5 FAQ said when we read it on November 1, 2026. Scorza and colleagues (2023) describe it as an alternative overall estimate when those two indexes are well below the others. In Pearson's 2024 sample report, the GAI is 115 and the FSIQ is 111.

Can your full scale IQ change?

Yes, within limits. The WAIS-5 stability coefficient was 0.93 over about a month, and retaking it raised the FSIQ by an average of 3.7 points for ages 16 to 69 (Winter, Trudel and Kaufman, 2024). The same review also gives a wider general figure: on a second administration, Full Scale IQs rise by 5 to 8 points. A few points between sittings is expected.

Does Mensa accept a full scale IQ?

Yes, from a Wechsler test: American Mensa says "Mensa requires a Full Scale IQ score (FSIQ)", with every subtest needed for it given, and lists a qualifying FSIQ of 130. Its Wechsler list named the WAIS through WAIS-IV, and not the WAIS-5, when we read its page on November 1, 2026. A WAIS-5 score is a question to put to American Mensa. Our Mensa IQ test guide covers the other routes.