Gifted Identification: How Children Are Actually Selected

What the threshold actually is

The conventional cutoff for a gifted programme is an IQ of 130. On a modern scale — mean 100, standard deviation 15 — that is two standard deviations above average, which places a child at roughly the 98th percentile. Some districts use 132, some use 125, and programmes for the profoundly gifted may use 145 or higher. There is no scientific law behind any of these numbers. They are administrative lines drawn on a continuous distribution. That last point matters more than it sounds. Nothing changes about a child between 129 and 130. The standard error of measurement on a good individually administered test is around three to five points, which means a child who scores 128 today might score 132 next month, and neither number is the 'real' one. Any cutoff applied to a measurement with error will misclassify children in both directions, near the line, every time. The scale matters as much as the number A score of 130 on a test with a standard deviation of 16 — the older Stanford–Binet scales — is not the same as 130 on a modern SD-15 scale. It sits at a lower percentile. Historic gifted scores and modern ones are not directly comparable, and programmes that quote a single number without naming the scale are being careless.

The finding that reframes the whole question

Here is the part almost no guide for parents mentions, and it is the most important thing on this page. In most school systems, a child is tested for giftedness because a parent or a teacher referred them. The test is a gate, but the referral is the gate to the gate. In 2016, David Card and Laura Giuliano published a study in the Proceedings of the National Academy of Sciences examining what happened when one large district stopped relying on referrals and simply screened every second-grader. The eligibility standards did not change. The tests did not change. The only thing that changed was who got looked at. The odds of being identified as gifted rose by 174% among disadvantaged students. Among Black students the increase was 74%; among Hispanic students, 118%. Girls, English language learners and children on free or reduced-price lunch were all systematically under-referred by the traditional parent-and-teacher system. The district also operated a two-tier threshold: 130 for most children, and a lower 116 for English language learners and children on free or reduced-price lunch. Even with that accommodation already in place, universal screening still found large numbers of eligible children the referral system had simply never sent for testing. And then the funding was cut, the screening stopped, and by 2011 the gifted share of third-graders had fallen back to where it had been before. The gains were entirely a property of the process, and they evaporated when the process did.

What this means for a parent

If you suspect your child is gifted and the school has not raised it, that is weak evidence about your child and strong evidence about referral systems. Ask directly whether the district screens universally or waits to be asked. It is the single most useful question you can put to them.

The General Ability Index, and why it exists

A Full Scale IQ on the WISC-V is built from seven subtests spanning five domains, including working memory and processing speed. Those two indexes are the most fragile things on the test. Processing-speed subtests in particular are simple, heavily timed, and clerical — which is precisely what makes them sensitive to attention, anxiety, medication, motor control and vision, none of which is reasoning ability. So a child with strong verbal and fluid reasoning but attentional difficulties can produce a Full Scale IQ that understates their reasoning quite badly, because two weak indexes drag the composite down. This is a well-recognised problem, and the profession's own answer to it is the General Ability Index: a composite built from the verbal comprehension and reasoning subtests, leaving working memory and processing speed out. The GAI is not a loophole for getting a child over a line. It is an acknowledgement, written into the test's own manuals, that the Full Scale IQ is not always the best estimate of a child's reasoning. Where a substantial gap exists between the reasoning indexes and the working-memory or processing-speed indexes, the GAI is often the more defensible number — and many gifted programmes accept it for exactly that reason. Related: The WISC-V test — how the children's Wechsler scale is structured, and why independent analysis does not support its five factors

Why very high scores get unreliable

A standardisation sample for a major IQ test contains a couple of thousand people, selected to represent the population. Two standard deviations above the mean, you are talking about the top 2% — in a sample of 2,200, that is a few dozen people. Three standard deviations up, it is a handful. Four standard deviations up, the sample contains essentially nobody. This is why scores above about 160 should be treated with real caution. There is no large body of people at that level against whom the norms were built, so the numbers are produced by extending the model beyond the data — extended norms — rather than by direct comparison. The further out you go, the more the score is an extrapolation and the less it is a measurement. It is also why the very high IQ figures that circulate for public figures are not meaningful in the way people assume. A score cannot be more precise than the data behind it, and out at the extremes the data run out.

What a gifted identification does and does not tell you

• It is a placement decision, not a diagnosis. Giftedness is not a medical condition, and an IQ cutoff is an administrative tool for allocating a limited number of places.
• It is domain-general by construction. A child can have exceptional mathematical or spatial ability and not clear a general threshold, because the composite averages across domains.
• It does not measure motivation, persistence, or opportunity — and those do a great deal of the work in what happens next.
• It is not a prediction about an individual life. IQ predicts outcomes on average, across groups; what it licenses you to say about one child is considerably less than most people assume.
• A single score is not a verdict. Every score is a range. A retest is a legitimate request, not an admission of doubt.

The most defensible way to read a gifted identification is as a piece of evidence that a particular child, on a particular day, performed at a level where a particular programme is likely to suit them. That is genuinely useful. It is not a statement about their worth, their ceiling, or their future — and the children the system misses are not less able, they are less referred.

Sources

• Card, D., & Giuliano, L. (2016). "Universal screening increases the representation of low-income and minority students n gifted education." Proceedings of the National Academy of Sciences, 113(48), 13678–13683. — https://doi.org/10.1073/pnas.1605043113
• Canivez, G. L., Watkins, M. W., & Dombrowski, S. C. (2017). "Structural validity of the Wechsler Intelligence Scale for Children–Fifth Edition." Psychological Assessment, 29(4), 458–472. — https://doi.org/10.1037/pas0000358
• American Educational Research Association, American Psychological Association, & National Council on Measurement in Education (2014). Standards for Educational and Psychological Testing. AERA.
• Pearson — Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V): technical and interpretive materials.
• Roid, G. H. (2003). Stanford–Binet Intelligence Scales, Fifth Edition: Technical Manual. Riverside Publishing.
• Deary, I. J. (2020). Intelligence: A Very Short Introduction (2nd ed.). Oxford University Press.