I've also seen this comment copied in various discussions of WISC: "Unfortunately, many psychologists also use the WISC to identify gifted children. This is unfortunate because the test was not designed for such use."
Both of these comments, however, are from the mid to late 1990s (at the latest), before the WISC-IV and certainly before the Extended Norms for the WISC. (And, hoo-boy, the Extended Norms are a whole different can of worms!)
This is not back to the original conversation
but a thought that I've had periodically over the years. I do wonder how the test designers design questions to tease out that tail end. Sure, I can see knowledge tests or speed tests where you can look up the info or create something to be done faster than you, yourself, could do it, but I'm not sure how the designer creates an abstract reasoning problem, for instance, that tests a level of abstract reasoning beyond that which he, himself, possesses. Maybe we need the DYS kids as adults to be the ones to design IQ tests so they aren't designing a test to test greater intelligence than that which they possess.
I'm not suggesting that IQ test designers aren't very smart, just that I doubt that they are 99.9+ people. Maybe it's just a matter of testing stuff on enough 99.9+ people to ascertain what they can and can't answer. Maybe not.
