putes a verbal IQ score and a performance IQ score; the Raven’s Matrices, a
nonverbal test of intelligence; and college entrance tests like the SAT.
Research on IQ tests demonstrates that various IQ test scores are positively
correlated with one another; for example, the Wechsler IQ score correlates
about .8 with the Stanford-Binet. IQ tests are also moderately correlated with
grades in school (the correlation is usually about .5), number of years of formal
education, occupational status, and, to a lesser extent, with success in an occu-
pation (Kline, 1991; Neisser, Boodoo, et al., 1996). The correlation between suc-
cess in an occupation (measured, for example, by supervisor ratings) and IQ
scores is typically about .3. So people who get good grades, go to school for a
long time, have professional jobs such as doctors and lawyers, and get higher
ratings from supervisors evaluating their work tend to score higher on IQ tests
than do people who get poor grades, drop out early, have jobs such as factory
workers, and get lower evaluations from their supervisors.
Keep in mind that these correlations do not tell us much about the causes of
the relationship between IQ scores and other measures, such as grades in
school. It could be, for example, that the superior intellect some people possess
causes them to score higher on IQ tests, do better in school, and get better jobs.
But there are other possibilities. Perhaps motivation to succeed is the cause (or
at least one of the causes) of the correlations—a generally motivated person
will try harder to do well on IQ tests, stay in school longer, and work harder on
the job. Or maybe health is a cause of the correlations—a generally healthy
person is more likely than an unhealthy person to be alert in school, acquire the
knowledge necessary to do well on IQ tests, and perform well on the job. It
couldalsobethattheeconomicadvantagesomepeopleenjoyiswhatenables
them to do better on IQ tests, do better in school, and get better jobs (McClel-
land, 1973).
It should be pointed out, however, that the relationship between IQ perfor-
mance and educational and occupational success cannot be attributed entirely
to socioeconomic factors (Barrett & Depinet, 1991). Parental background vari-
ables like parental income and education do not predict occupational achieve-
ment as well as do IQ test scores (Gottfredson & Brown, 1981). Grades in school
are more strongly correlated with SAT scores than with parental income (Baird,
1984).
36.2 Is Intelligence Unitary?
As I suggested at the beginning of this chapter, much of the recent work on the
nature of intellectual differences has taken the form of a reaction to the histori-
cally entrenched hereditarian theory of intelligence and the IQ enterprise it
established. In this section I will discuss the evidence that intelligence is uni-
tary, that it reflects a generic intellectual system. In the next major section I will
develop the argument for a multifaceted model of intellectual differences.
Evidence for the Unitary View
Charles E. Spearman (1863–1945) was one of the first psychologists to demon-
strate that people who do well on any one subtest of the IQ inventory tend to
do well on any other subtest. That is, the various subtests that make up the IQ
Individual Differences in Cognition 783