Re: Race, intelligence, and anti-racist prejudice (Was: Genetic Evolution)

Warren Sarle (
Sat, 25 Feb 1995 22:02:40 GMT

In article <>, (Kaa Byington) writes:
|> Stephen Lajoie ( wrote:
|> ...
|> : Third, accept the definition of intelligence being what an IQ test
|> : measures. Since significant corelations between economic status, crime,
|> : teenage pregnancy, academic ability And IQ score have been shown to
|> : exist, it is a useful measure.
|> It looks to me like the correlation is with poverty, which is the
|> underlying cause of all those things.
|> A lesson from the first day of Statistics class: There is a perfect
|> correlation between ice cream sales and drownings. When one goes up, the
|> other goes up. Year after year. Does that mean that buying and eating
|> ice cream causes drowning? No. That's spurious causality. It means they
|> both go up in summer.

If you had stayed past the first day of statistics class, you might
have learned about partial correlation. If poverty is the "underlying
cause of all those things", then the partial correlation of any two
of "those things" controlling for poverty should be near zero. But
_The Bell Curve_ is full of analyses similar to that (but more
sophisticated) demonstrating that controlling for poverty leaves
IQ significantly associated with "those things". While these analyses
cannot show that IQ _is_ the underlying cause, they do show that
poverty _is not_ the underlying cause.


Warren S. Sarle SAS Institute Inc. The opinions expressed here SAS Campus Drive are mine and not necessarily
(919) 677-8000 Cary, NC 27513, USA those of SAS Institute.