From CIP's blog I learned of R.E. Nisbett's Intelligence and How to Get It. The library had a copy. I'm still in the thick of it. Of course, I like it because it confirms all my prejudices. Nisbett makes out a strong case that almost all population differences in IQ can be explained by environmental and developmental differences, and has little to do with genes. "Can be explained" is different from "is explained" but then that is about as far as we can get with the "soft" sciences.
One of the strongest indications that IQ is mutable and not genetically determined is the IQ increase observed in different cohorts of the same population. This is known as the Flynn effect. IQ tests are normed for a cohort, i.e., by definition, the 21-25 age group of white Americans have a normally distributed IQ with a mean IQ of 100 and a standard deviation of 15, and by definition the 45-49 age group of white Americans had the same distribution when they were 21-25. But since the two groups might have done exactly the same test, it is possible to compare absolute scores on the test, and by that token today's 21-25 year olds are significantly IQ-smarter than the 45-49 age group was at the same age.
The good news is that IQ can be developed - by better parenting, better teaching methods at school, better environments for children to grow up in, and so on.
The bad news is that IQ is, IMO, somewhat meaningless. Note the Flynn effect mentioned above. Such has been the increase in IQ raw test scores that by today's standards, the great-grandparents of today's young adult were barely intelligent enough (in IQ terms) to deal with the world. Since that is absurd, it is more likely that people have become much more proficient at taking tests. IQ in my opinion, is like rank in a school classroom. It has some predictive properties; is difficult to compare class ranks across a generation or even sometimes across schools.
The bad news also is that for a country like the US, improving the IQ of the population will take some heavy lifting. Devising effective programs is difficult. For a country like India with a long way to go with respect to literacy, school enrollment, public health, nutrition, etc., big gains in the IQ of the general population are possible because effective programs to remedy these problems can be devised (e.g., simply increase the measures that contribute to the Human Development Index). If we assume that IQ measures some quality that is of value in coping with the world, then IQ development is essential for the welfare of the people.
Tuesday, June 22, 2010
Logging you in...Comments by IntenseDebate
Subscribe to:
Post Comments (Atom)
CIP · 768 weeks ago
I think that Nisbett would rather strongly disagree with you about IQ being meaningless. It represents a real and valuable set of intellectual skills (however acquired) and that's why it is so closely correlated with academic and other success in a vast range of fields. IQ scores have increased because people have gotten better at solving the types of problems IQ tests measure - this is not the complete set of problems people have to solve, but it's an increasing important subset of them. In some ways, the increase in IQs is similar to the even greater increase seen in average height in developed countries - it's a function of their intellectual diets in a way analogus to the way height is to their nutrition.
Arun · 768 weeks ago
Yes, Nisbett would strongly disagree with me. But I must call them as I see them. Unlike height, there is no objective measure of what problems people are actually solving, and whether they're solving them with greater ease or faster than the previous generation.
CIP · 768 weeks ago
Moreover, skill at those solutions is highly correlated with a whole bunch of other success measures. Whether or not you can trace those quite objective differences to some physical changes in the brain is a question for the future, but objective doesn't necessarily mean concrete.
Arun · 768 weeks ago
As I wrote, IQ is like class rank. The top MIT undergraduate in physics today is not easily comparable to one from twenty-five years ago. Yes, e.g.. (s)he probably know a lot more mathematics. Is (s)he a lot more intelligent? I don't know. However, being at the top of the class is predictive.