There are other types of population distributions in nature other than the bell curve. It may be that a racial group's IQ distribution is not symmetrical but instead skewed to the left or to the right.

That's true, PlainTalk, there are other distributions, but they often (though not always) mean a population is currently undergoing natural selection. Such "skews" can persist for thousands, even hundreds of thousands, of years if the selection pressure is slight.

Or so I understand. However, since I'm a molecular biologist, not an evolutionary biologist, my opinion should not be taken as an expert's. Just as Mr. Saletan's shouldn't be taken as anything close to such.

(Not really relevant here, but thought I'd explain. )

IQ tests are normed to force a fit to a Gaussian distribution. This converts the scores from ordinal to equal interval. The reasons for doing this were not capricious. If you really want a complete answer, the best option for you is to read all of Chapter 4 in this:
Jensen, A.R. (1980). Bias in mental testing. New York: Free Press.

There is a reality even prior to heaven and earth; indeed, it has no form, much less a name. Eyes fail to see it; it has no voice for ears to detect. To call it Mind or Buddha violates its nature, for it then becomes like a visionary flower in the air. It is not Mind, nor Buddha; absolutely quiet, and yet illuminating in a mysterious way, it allows itself to be perceived only by the clear-eyed. It is Dharma truly beyond form and sound; it is Tao having nothing to do with words.

- Dai-o Kokushi

The bell curve point is a good one. It would be instructive to see the distributions of IQ scores in different ethnic groups plotted to show how large the overlaps are.

And of course, any serious student of statistics understands that an average is meaningless without context - the distribution around the mean. If I have a group of 4 people, 3 of whom are 4 feet tall and one who is 12 feet tall I have a group with an average height of 6 feet. A tall "race"? No, what I have is 3 dwarves and a giant. The average height is meaningless.

I'd also be interested in seeing mode IQ scores by ethnic group. That is, what is the most common score among white Americans, among African Americans, among Asian Americans? I wouldn't be surprised to see that the most common scores in all three groups fall into the same range.

Dear PK,

The bell curve point is a good one. It would be instructive to see the distributions of IQ scores in different ethnic groups plotted to show how large the overlaps are.

You can find such distributions here:
Jensen, A.R. (1980). Bias in mental testing. New York: Free Press.

I'd also be interested in seeing mode IQ scores by ethnic group. That is, what is the most common score among white Americans, among African Americans, among Asian Americans? I wouldn't be surprised to see that the most common scores in all three groups fall into the same range.

You can find answers here:
Lynn, Richard (2006). Race Differences in Intelligence: An Evolutionary Analysis, Washington Summit Publishers, Georgia.

You can find answers here:
Lynn, Richard (2006). Race Differences in Intelligence: An Evolutionary Analysis, Washington Summit Publishers, Georgia.

The findings in this book are bogus. Just read thru some of it briefly. It claims South African university students have an IQ of 81. We know that one can generally can only function at the university level with an IQ of 100 and above. Thus the findings make no sense.