Skip to content
🔍 Paths

Talent identification

How systems notice — SMPY and CTY, 예고, AIS, grammar gates — and who they miss.

Talent identification — conceptual illustration

Talent identification is the set of tests, nominations, camps and schools that decide who is treated as having a file worth extra time.

Julian Stanley’s SMPY and Johns Hopkins CTY used above-level SAT. Gagné and Renzulli offered models that refuse IQ-only gates. Korea’s 영재교육 and 예고, Japan’s 英才 talk and 部活, China’s conservatories and 少年班, Finland’s late tracking, UK grammar and specialist schools, Brazil’s football academies and African running camps are working systems — not a single science.

Identification measures now, plus who was invited. This page is the methods chapter for the whole atlas. It will not rank groups by IQ or race.

ClassicSMPY / CTY
RiskReferral bias
SportAIS / academies
Arts예고 / conservatoire
EthicNo group rank
ToneEducational

Profile

Innate weight
40
Trainability
70
Early window
72
Evidence
62
Transfer
55
Visibility
78

Seven chapters

Frequently asked questions

What did SMPY do differently?
It tested young adolescents with a college SAT, lifting the ceiling. Stanley, then Benbow and Lubinski, followed maths-precocious youth. Access still depended on who heard about the test.
Is CTY the same as giftedness?
CTY is a search-and-course organisation. Giftedness is a construct (see that page). Paying for a summer course is not a soul.
Why not IQ only?
Renzulli’s rings and Gagné’s catalysts: commitment, creativity, practice, and opportunity. IQ-only gates miss arts, 2e, and the under-referred.
Do sport academies identify talent?
They identify current performance and size, with relative-age bias (see athletic-talent). AIS is a famous national machine with known false cuts.
Is Finland “anti-identification”?
It delays hard tracking and trusts teachers. That is a different ID theory: later, broader, less test-gated — not a claim that talent is unreal.
What about Cyril Burt?
A cautionary tale of fraud and class in British testing history. Do not build a modern system on a tainted ancestor. Binet’s original brief (help, not caste) is the better ancestor.
Can ID be fair?
Fairer: multiple measures, trial teaching, translated invitations, review dates, no group ranks. Perfect: no. Systems always leak access.
Is this page a ranking tool?
No. It is a map of how ranking tools fail and sometimes help.

Open practice curve →