Let's open with a quote in the book from David Flynn on page 180 "Psychologists should stop saying that IQ tests measure intelligence. They should say that IQ tests measure abstract problem-solving ability."
Beyond that, Stephen Murdoch shows that IQ tests can honestly be seen as measuring KQ (knowledge quotient), SQ (socialization quotient), and SESQ (socio-economic status quotient), but not intelligence quotient.
First, Murdoch points out that, contra people like Arthur Jensen and Charles Murray, there's not a lot of consensus that some abstract measure of intelligence called "g" exists, let alone exactly what it is, doubly let alone how to test for it. Once one gets out of the field of psychology and into the more rigorous fields of cognitive science and neuroscience, this "no-g" stance becomes even stronger.
Beyond that, Murdoch goes back to the origins of testing for intelligence, then the history of the original Binet, the Stanford-Binet, the Wechsler, and the SAT. One slight quibble is that Murdoch could have done even more to put this enamoration with intelligence testing of early 20th-century Americans into even broader context, i.e. things such as Taylorism in analyzing worker performance.
As Murdoch reports, according to IQ tests, blacks today are smarter than 1930s whites. We know evolution doesn't work that fast; ergo, whatever IQ tests measure, it ISN'T innate/hereditary intelligence. And, that holds true whether or not there's justification, or even strong justification, for the idea of "race."
Murdoch also reports on how IQ tests can be, and have been, tweaked to eliminate the black-white 15-point gap, or even have blacks scoring higher. Again, one minor quibble: Murdoch could have internationalized these findings more, such as higher-caste vs. lower-caste IQ test differences in India.
Related to this test-tweaking, Murdoch shows that specific-skill tests have more predictive value of future life success than the SAT. One other minor quibble here: having taken both tests, I'm disappointed that Murdoch had no discussion of the ACT test, which IS more subject-specific than the SAT. I have feeling that if the ACT were the standardized test, and were tweaked to address different cultural backgrounds, we'd be on a lot better footing than the recently revised SAT II.
Therefore, don't believe any reviewers who claim this isn't a balanced book. People like Arthur Jensen and Charles Murray have opened themselves up for proper, non-"politically correct" critique by making statements that in some cases are not scientifically verifiable and in other cases, can be and have been refuted.