Saturday, December 29, 2018

Cosmo Shalizi takes on IQ

IQ


Because You Really Wished I'd Write More About Books
Books to Read While the Algae Grow in Your Fur, August 2008
Books to Read While the Algae Grow in Your Fur, June 2008
Last Words on Saletan
Reading Skills
In Which I Demand That Slate Refund My Subscription
Uncle Fritz Explains How It Feels to Argue about Intelligence Tests
g, a Statistical Myth
Yet More on the Heritability and Malleability of IQ
Those Voices Again
...In Different Voices
On the Superiority of Sociology to String Theory


PS:

Q: So the analogy suggests that IQ scores are...?
A: A proxy for the skills and habits encouraged by a bureaucratic society; skills and habits which can be at once highly heritable (because of strong transmission through family and neighbors) and highly learned (within the scope of what it is biologically possible for humans to learn and internalize). Innate ability needn't enter into it at all. The implications for democracy would be nearly nil.
Q: And the famous g?
A: Is a statistical artifact, or better yet a myth; but that is another story for another time.
Posted by Arun at 9:46 PM
Labels:

Comment (1)

Loading... Logging you in...
Dashboard | Edit profile | Logout
  • Logged in as
0 Vote up Vote down
Guest's avatar

Guest · 326 weeks ago

Shalizi is always interesting, but I found the "g is a statistical artifact" argument too facile. Yes, a random collection of data is likely to have some linear combination of values that explains a bunch of the variance, but the real question is the residuals. He argues that he can find combinations of variables that explain more life success than IQ, but if you add, say, (too use one of his choices), sexiness rating, you might get a slightly higher "a" (his new variable), but what about the residuals? In particular, if this addition found another vector that explained a lot of the remaining variance, you would be justified in regarding the variables as independent. Of course the opposite result would also be possible, but this needs to be checked by experiment, not speculation.

Post a new comment

Comments by

Subscribe to: Post Comments (Atom)