October 26, 2007

Mainstream sense on IQ

The article below from "The Times" of London gets it pretty right. One point they glide over, however, is that "culture fair" tests (one not dependant on a particular educational or social background) show very similar racial and class gaps to other tests -- and have been doing so for decades. In Australia there was even a test specifically designed for blacks in a black environment (the Queensland Test) and whites still scored way higher on that. Pesky!

Rows about race and IQ pop up about once a decade, and invariably cause a fuss. But do they really signify anything? The evidence is that IQ, as conventionally measured, does differ between racial groups. Many studies have shown differences of about 15 points between the mean scores of white and black groups, and some have shown that Chinese and Japanese groups score higher still.

The counter-argument is that the conventional measurement of IQ is heavily culture-dependent. A test developed originally to measure the intelligence of Caucasians may not be fair to those whose cultural heritage is different. Some tests do contain biases that may disadvantage people, either on the basis of culture or, indeed, socio-economic status. When the scores are corrected for wealth and social position, the gaps narrow.

A second argument - that IQ doesn't matter anyway -is easier to dismiss. A large body of evidence shows that IQ is linked to success in life, both educational and economic. This is why the gap narrows when corrections are made for socio-economic status, because in a mobile society success is largely determined by intelligence.

To some, it is enough to say that IQ tests are unfair, and that their conclusions must be disregarded, but that flies in the face of the evidence that such tests are in fact good predictors of success. So if we accept that group differences are real, do they matter? The first thing to say is that such group differences tell us nothing about individuals. The distribution curves for all groups overlap, and the range of scores within each group is far wider than the differences between them. So even if the group differences were to be real, they tell us nothing about individuals within those groups. To judge individuals by the nation, race, or sex to which they belong is prejudice. Everyone has the right to be judged on his or her own merits.

The difficulty is that campaigners for racial equality have insisted on the rights of whole groups, not of individuals. They have sought to be argue that entitlements should be conferred on individuals not on their own merits but simply because they belong to that group.... Charles Murray, the author of The Bell Curve, a book published in the 1990s that reviewed the evidence dispassionately, said that the only answer was "an energetic and uncompromising recommitment to individualism".

All people deserve equal treatment. But that is not quite the same as saying they are all equal. The error comes in taking a group difference, which may or may not be real, and using it to judge the worth of individuals. That is racism.

Source







Go to John Ray's Main academic menu
Go to Menu of longer writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page
Go to Selected pictures from John Ray's blogs