RUSHTON REPORTS MORE DATA THAT AFRICAN IQ IS ABOUT 70
\doc\web\99\02\rushafr.txt
Date sent:      	Tue, 9 Feb 1999 18:26:40 +0100
To:             	"J. P. Rushton" <rushton@julian.uwo.ca>
From:           	salter@erl.ornithol.mpg.de (Frank Salter)


Phil,


Thanks for reporting this important research. What standard deviations are
you getting for black and white samples? If the high-heritability theory
is correct, blacks should have larger SDs because they are more
genetically diverse than other populations.


Frank Salter




>Phil Rushton here reporting on my recent trip to South African
>universities in search of a high IQ African population. I gave Raven's
>Standard Progressive Matrices to 300 African and White first year
>university students. The African students scored at the 14th percentile
>with an IQ equivalent of 84. Assuming these students are 1 standard
>deviation above their population mean, this study corroborates the many
>findings of IQ = 70 for the general African population (e.g., as reviewed
>in The Bell Curve, see especially the Afterword to the paperback
>edition). The Raven's test should have been very easy for university
>students and indeed Africans got 44 out of 60 items correct so one cannot
>say they did not understand what was required of them. But this is what
>White 14 year olds get correct. Other item analyses showed the items
>"behaved" the same way for Africans as for Whites, its just that they
>were more difficult.  The test was untimed and Africans and Whites had
>one and a half hours to complete a test that most did in under 30
>minutes. This paper is under editorial review.
>
> In another study (also under review), I re-analyzed a set of data from
>154 13- to 15-year- old secondary school students from Soweto,
>Johannesburg, on the Wechsler Intelligence Scale for Children-Revised
>(WISC-R). Here, too, the Africans scored 1.5 to 2 standard deviations
>below U.S. norms (both blacks and whites combined), suggesting a
>population IQ of about 70 to 75. Importantly, the African-White
>differences were more pronounced on tests of high-g than it was on tests
>of low-g (g is the general factor of intelligence) (r =0.77, p = .05).
>The effect remained even when g was extracted from the Black rather than
>from the White standardization sample (r = 0.60, p <.05), as it also did
>using Spearman's rho instead of Pearson's r (g from Whites = 0.74, g from
>Blacks = 0.74, ps< .005).




______________________________________________________________________ Dr.
Frank Salter MAX-PLANCK-INSTITUT fuer VERHALTENSPHYSIOLOGIE Humanethologie
und Humanwissenschaftliches Zentrum der LUDWIG-MAXIMILIANS-UNIVERSITAET,
MUENCHEN Von-der-Tann-Straße 3 D-82346 ANDECHS Deutschland Tel. +49
(0)8152 373 55 Fax. +49 (0)8152 373 70