I know David Lubinski, one of the investigators in that correlational study, reasonably well. (He and I are alumni of the same school district and same undergraduate university, and have many mutual friends besides meeting each other in conferences from time to time.) It's important to distinguish the claim about the Study of Mathematically Precocious Youth (of whom my son is one participant) and my claim about mathematical "genius" in general. The claim you cite is that frequency of eminent accomplishment in that study population increases as SAT scores at age twelve increase. That may be. My claim is that the threshold IQ score for eminent accomplishment in mathematics and other domains that recognize "genius" is no higher than 120, and I have Arthur Jensen and Hans Eysenck and the Terman longitudinal study on my side as I make that claim.
My understanding was that many of the claims of threshold effects at least partially stemmed from the imprecision of many tests, even at the 1.5-2 sigma levels.
Would you be able to send me some more info? Email's in my profile.
Comments
I know David Lubinski, one of the investigators in that correlational study, reasonably well. (He and I are alumni of the same school district and same undergraduate university, and have many mutual friends besides meeting each other in conferences from time to time.) It's important to distinguish the claim about the Study of Mathematically Precocious Youth (of whom my son is one participant) and my claim about mathematical "genius" in general. The claim you cite is that frequency of eminent accomplishment in that study population increases as SAT scores at age twelve increase. That may be. My claim is that the threshold IQ score for eminent accomplishment in mathematics and other domains that recognize "genius" is no higher than 120, and I have Arthur Jensen and Hans Eysenck and the Terman longitudinal study on my side as I make that claim.
My understanding was that many of the claims of threshold effects at least partially stemmed from the imprecision of many tests, even at the 1.5-2 sigma levels.
Would you be able to send me some more info? Email's in my profile.