I have no idea about how regression to the mean figures into the voting
issue - but I do know a little about regression the mean. It refers to
the fact that when predicting one thing (Y) from another (X), and the
correlation between X and Y drops, predictions of Y from X converge on
the mean of Y.
Many people mistakenly assume that "regression to the mean" means
successive generations get more homogeneous - but this is not true.
Regression to the mean is a statistical phenomena - it has nothing to do
with how the world changes.
If you are not convinced, consider this. When you predict children's IQ
from their parents - children's IQs are - on average - closer to the
mean from the parents, at both ends. The very bright parents have, on
average, children closer to the mean, and very dull parents have
children closer to the mean as well. BUT, if you regressed in the other
direction - predicting parents IQ from children, the PARENTs IQ will
appear to regress toward the mean too! It has to. As r drops, more of
the variance is predicted by factors _other than_ the X you have
measured.
Michael Sylvester wrote:
> how does regression towards the mean figure out in all this?
--
---------------------------------------------------------------
John W. Kulig [EMAIL PROTECTED]
Department of Psychology http://oz.plymouth.edu/~kulig
Plymouth State College tel: (603) 535-2468
Plymouth NH USA 03264 fax: (603) 535-2412
---------------------------------------------------------------
"What a man often sees he does not wonder at, although he knows
not why it happens; if something occurs which he has not seen before,
he thinks it is a marvel" - Cicero.
