A couple additional thoghts I didn't get around to before leaving for
my 8:30 lecture:
(1) The clearest way of looking at the stats side of things is probably
that one would expect a high enough r^2 between schools' performances in
one year and in the next that regression to the mean would be a rather
minor phenomenon.
(2) _Why_ were even the best schools expected to improve, with targets
that seem to have been overambitious?? I would hazard a guess that it
might be due to an inappropriate overgeneralization of the philosophy -
appropriate, in an eduational context, for individual students - that
progress should constantly be being made.
Getting further off-topic, we see the same thing in economics, where
the standard model for the Western economies is one of constant growth,
and our institutions seem unable to adapt to slight shrinkage - or even
slower-than-usual growth - without pain all round.
Our culture seems to concentrate on the idea that the moment an
institution stops growing it starts to die, and does not seem to put
much effort into maintaining "mature" institutions for which the
constant-growth paradigm is no longer appropriate. Maybe this cultural
neoteny is still appropriate and advantageous at this stage in history -
I don't know.
-Robert
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
