Genes, Brains, and Human Potential The Science and Ideology of Intelligence

332 THE PROB LEMS OF EDUCATION ARE NOT GE NE TIC

Ripley largely extolls the “hard- work” Chinese and South Korean educa-
tional methods: “In an automated, global economy, kids needed to be
driven; they need to know how to adapt, since they would be doing it all
their lives.”^21
In my view there is a sad irony here. Is this the kind of potential we
want our children to develop? Should we not be promoting in them the
education and intelligence through which they can cultivate a more con-
genial world for all, rather than become the toilers in someone’s “ham-
ster wheel” (an image Ripley also uses)? Susan Engel puts the contrasting
view that American schools have allowed “the pursuit of money to
guide our educational practices,” and in so doing “we have miseducated
every one.”^22

LITTLE CONNECTION WITH JOB PER FOR MANCE

As just explained, we would expect correlations between school and
higher education per for mance because they are partly testing the same
learning rather than potential. Moreover, since educational level is a con-
dition of level of entry to the job market, there is bound to be some as-
sociation between educational attainment and occupational level. School
or college per for mance, that is, automatically predicts level of occupation.
A more impor tant test of whether they refl ect genuine potential, as op-
posed to a host of other background variables, is whether they predict
job per for mance.
As explained in chapter 3, it has always been the case that data have to
be stretched a great deal to demonstrate the existence of any relationship
between test or exam per for mances and job per for mance. I described
how investigators have resorted to almost any “test” as surrogates of
potential (or “g”) to fi nd such a relationship. Among these have been
educational attainment test scores, including simple reading scores. What-
ever they use, the correlations have always been tiny, around 0.2. And the
causes of that cannot be distinguished from noncognitive factors known
to aff ect such correlations.
Likewise, surveys going back to the 1960s have routinely shown that
neither school nor university grades are good predictors of per for mance

This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:58:56 UTC

http://www.ebook3000.com