In article <[EMAIL PROTECTED]>, Richard Ulrich <[EMAIL PROTECTED]> wrote: > - discussing IQ >On 22 Apr 2004 15:12:15 -0500, [EMAIL PROTECTED] (Herman >Rubin) wrote:
>> In article <[EMAIL PROTECTED]>, >> Robert J. MacG. Dawson <[EMAIL PROTECTED]> wrote: >[ snip much interesting comment] >> > Similarly, at the lower end, most standard adult IQ tests presumably do >> >not distinguish below a certain level at which the subject is incapable >> >of following the instructions. >> > Extremes of ability need to be measured with special instruments. >HR > >> Possibly special instruments, but still use the idea of a scale, >> not of an arbitrary distribution, or contortion to fit an >> inappropriate cookbook probability course. We do not use the >> same instruments to measure temperatures of .001 K as we do for >> the surface of the Sun, or for the interior of a thermonuclear >> reactor, but we are still using the same scale. >"temperature" is reasonably defined as one dimension. >IQ, merely for the parts important for living and school, >incorporates at least 6 dimensions which have been rather >well demonstrated as independent 'talents' that use different >parts of the brain. Yes and no; it has even recently been shown in articles in strong science journals (NOT psychology or education) that some aspects require the use of both sides simultaneously. MRI was used to find out what is happening. But this does not change the point; I still question whether there is any remotely fair reason to adjust scores on anything to resemble a normal distribution. >Well-measured IQs below about 80, I once was told by specialists, >demonstrate trauma or genetic defect. So there is (approximately) >a normal zero for humans, if you want to think of it that way. "Genetic defect" is genetic variation, not the absence of a key gene, or the presence of a bad one. On the usual normal IQ scale, about 9% are below 80. It goes much lower. BTW, if there are individual genes which can make substantial differences, the entire basis for normality, the assumption that enough factors are added for the central limit theorem to be an excellent (not just good) approximation is destroyed. Good studies have demonstrated that the most important factor is purely genetic. Identical twins reared apart are more similar that fraternal twins reared together, and in some cases than identical twins reared together. >And then there is this complication: Our measured IQ strongly >does reflect something cultural and educational, particularly for >those tests of 'abstract reasoning' (Raven's progressive matrices) >that still are (I think) regarded as having the strongest genetic >loading. I doubt that the psychologists can understand abstract reasoning, especially that it is not incremental. The place where it is most clear is in mathematical concepts, and few outside of abstract mathematicians even see these, which can be understood without too much difficulty by children, but apparently not by those in "education". > - By the way, Herman -- >Do you have any reference for your description of the 'better >scaling' for IQ, with bigger 'real' differences at the high end, or >is that personal, subjective opinion? I do not remember that. It is partly subjective, but all of the "old" IQ tests, produced before the introduction of the normal distribution, have IQs ranging well over 160, including estimates of more than 200. It was admitted that it was hard to measure at this level, but this does not mean that scores above 130 should not be reported. -- This address is for information only. I do not claim that these views are those of the Statistics Department or of Purdue University. Herman Rubin, Department of Statistics, Purdue University [EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
