From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Tue Nov 08 2005 - 12:37:25 MST
Michael Vassar wrote:
> http://www.vanderbilt.edu/Peabody/SMPY/Top1in10000.pdf
> gives a good overview of the average life outcomes of a group of high
> performing pre-teens.
SMPY, yep. Them's the folks that administered me the SAT.
> Note that they are not a random sample, but
> rather a sample with parents wealthy and academically encouraging enough
> to send them to CTY.
I don't recall being sent anywhere. They just administered the SAT at
Hillel Torah North Suburban Day School.
> That said, has anyone from SIAI tried outreach to prodigies and/or
> ex-prodigies? Has the strategy been considered?
I considered outreach to other SMPY medalists on the theory that, as I
was one of them, there might be other Eliezers in there. I think I went
so far as to call the John Hopkins center and ask if there was some way
to mail something to the other SMPYers, without necessarily acquiring
their mailing addresses, and the answer was no. But this is long enough
ago that I'm not sure if I'm remembering making this phone call, or
remembering imagining it.
> Finally, I'm uncomfortable with the diagnosticity example. Reversion to
> the mean is a consequence of a non-uniform distribution of priors. With
> a uniform distribution of priors, say, GPA 3.5-4.0, 3.0-3.5, 2.5-3,
> 2-2.5, <2 (I'm guessing that those are rough quintiles) the subjects
> predictions should not revert, though their confidence intervals should
> broaden.
Priors in this case should be approximately Gaussian, in which case
reversion to the mean is appropriate.
-- Eliezer S. Yudkowsky http://intelligence.org/ Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Tue Feb 21 2006 - 04:23:19 MST