March 13, 2006

I'm sure this study will get a lot of favorable publicity,

because, judging from the abstract, it sounds intentionally stupid, and thus is likely to become part of the conventional wisdom:

Testing for Racial Differences in the Mental Ability of Young Children

Roland G. Fryer, Steven D. Levitt
NBER Working Paper No. 12066 Issued in March 2006

---- Abstract -----

On tests of intelligence, Blacks systematically score worse than Whites, whereas Asians frequently outperform Whites. Some have argued that genetic differences across races account for the gap. Using a newly available nationally representative data set that includes a test of mental function for children aged eight to twelve months [emphasis mine - I would refer to them as "infants," not as "young children," but Levitt knows how to spin a study], we find only minor racial differences in test outcomes (0.06 standard deviation units in the raw data) between Blacks and Whites that disappear with the inclusion of a limited set of controls. The only statistically significant racial difference is that Asian children score slightly worse than those of other races. To the extent that there are any genetically-driven racial differences in intelligence, these gaps must either emerge after the age of one, or operate along dimensions not captured by this early test of mental cognition.

I haven't seen Levitt and Fryer's paper, but their gimmick looks pretty obvious -- you make the IQ gap disappear, hesto presto, by pointing at something that's not an IQ test. There are no IQ tests that purport to measure IQ much below age 3, and even age 3 or 4 is pretty shaky. At "eight to twelve months," you are mostly measuring physical coordination and/or personality development, not advanced rational problem solving ability, which is the main point of IQ testing. It's been known for decades that black infants tend to be ahead of white infants on those dimensions, with Asians in third place on average.

I've got a great idea for Levitt and Fryer's next study. They could prove that the racial difference in long jumping ability seen in the Olympics can't be genetic by showing that for young children of 8 to 12 months old, there is zero difference by race in long jumping ability. The mean long jump distance for all three races among 8 to 12 month olds is zero, so that means any later differences that emerge can't be genetic. Right?

Then they could prove that the racial slam dunking gap can't be genetic because there are no racial differences in the ability to slam dunk among 8-12 month olds.

A reader writes:


Maybe I could pitch a study that seeks to prove group differences in IQ tend to disappear during periods of unconsciousness. Grant money, here I come!


Perhaps Levitt and Fryer got the idea for studying the IQ of 8-12 month olds from this article:


Study: Dolphins Not So Intelligent On Land

GAINESVILLE, FL—Although dolphins have long been celebrated for their high intelligence and for appearing to have a complex language, a team of researchers at the University of Florida reported Monday that these traits are markedly less evident on dry land.

According to study researchers, a group of 25 bottlenose dolphins removed from their holding tanks failed 11 exercises designed to test their basic cognitive abilities and reasoning skills.

"The dolphins were incapable of recognizing and repeating simple gestures," said study co-author Dr. Scott Lindell. "Their non-verbal communications were limited to a rapid constriction and expansion of the blowhole, various incomprehensible fin motions, and heavy tremors while they lay prone on the lab table."

After capturing the dolphins from the ocean, Lindell and his colleagues tagged them and placed them under the intense, high-wattage lights of a moisture-proof lab. The researchers then administered an extensive battery of tests designed to measure everything from the dolphins' self-awareness to their aptitude for writing and reading comprehension.

"Dolphins have a popular reputation for being excellent communicators," Lindell said. "But our study group offered only three types of response to every question we posed: a nonsensical, labored wheezing, an earsplitting barrage of unintelligible high-pitched shrieks, and in extreme cases, a shrill, distressed scream."

Even the dolphins' proven ability to navigate through a form of sonar called echolocation was ineffective on land.

"The military has claimed great success in training these mammals, utilizing their echolocation skills to detect mines that have been placed underwater," said Lindell, who conducted a similar experiment in a concrete parking lot. "We were unable to replicate this finding ourselves." -- The Onion


My published articles are archived at iSteve.com -- Steve Sailer

No comments: