about 20 points higher than the IQ scores of other African Americans living in
lower income circumstances. Clearly, the family environment had a huge effect
on the development of skills underlying the performance on IQ tests. Further-
more, these and other adoption studies indicate that when African-American
children are adopted into European-American families, their average IQ per-
formance typically comes to be nearly equal to that of the European Americans
(Flynn, 1980; Eyferth, 1961; Tizard, Cooperman, Joseph, & Tizard, 1972; Scarr &
Weinberg, 1976, 1983). Yet the IQ correlation between the African-American
adopted children and their biological parents is greater than the correlation
between the African-American children and their adopted parents. Again, that
seemingly paradoxical result is because correlation reflects rank order. The IQ
scores of the adopted African-American children may have been improved by
their environment, but the environment did not affect their rank order on the
IQ test. The rank order of the IQ scores of the adopted children continued to
reflect the rank order of the IQ scores of their biological parents.
Other research that examines children in similar environments but with dif-
ferent racial backgrounds has also contradicted the hereditarian claim of a
racial difference in intelligence. Loehlin, Lindzey, and Spuhler (1975) examined
theIQscoresofchildrenborntoGermanmothersandAmericanfatherssta-
tioned in Germany after World War II. One group of children was fathered by
African Americans, while the other group was fathered by European Ameri-
cans. Both groups were raised by German mothers in roughly similar economic
circumstances. The averages of the IQ scores of the two groups of children were
equal, even though one group received half of its genes from people of African
descent. Furthermore, there is no correlation between degree of African ances-
try of African Americans and their performance on IQ tests (Scarr, Pakstis,
Katz, & Barker, 1977).
It is true that some Asian-American people, such as Japanese Americans,
score higher on average on IQ tests than do European Americans. But cross-
cultural studies that take into account cultural factors, such as the proportions
of rural and urban dwellers, suggest no difference between Asians and Euro-
peans in IQ test performance (Stevenson et al., 1985). Furthermore, some Asian
groups that have immigrated to the West and subsequently endured poverty in
theWestscoreloweronIQteststhandoEuropeans(seeMackintosh,1986).
Sometimes people use the high correlations in IQ performance between
twins reared apart as evidence that ethnic differences in IQ performance must
be due to genetic differences. In fact, though, even if one overlooks the inter-
pretation problems associated with this paradigm, the twin findings are per-
fectly consistent with an environmental explanation for group differences in IQ
performance.
To see why, consider the following hypothetical situation. Suppose we have
three sets of twins (Jerry and Gerry, Robin and Robyn, and Sara and Seri) who
are reared apart. On IQ tests, Jerry and Gerry both obtain 100, Robin and
Robyn both obtain 110, and Sara and Seri both obtain 120. So the correlation
between the IQ scores of the twins is 1.0. Now suppose the second member of
each pair (Gerry, Robyn, and Seri) is each given extensive training so that each
improves his or her IQ performance by 20 points. So now the IQ scores will be
800 R. Kim Guenther