[Phys-L] Re: Math SAT

[jsd at AV8N.COM (John Denker)]

jbellina wrote:
> I am seeing all this wonderful news of an increase in the average SAT
> from 518 to 520.  Am I wrong or is this nothing hype.  Isn't a change
> of 2 points in the noise?

Well, it depends on what sort of "noise" we're talking about.

In purely empirical terms, the SDoM (standard deviation of the mean)
is smaller than the halfwidth of the distribution ... smaller by a factor
of sqrt(N).  The halfwidth of their distribution is a few hundred points,
and their N is huge ... something like 400,000.
   http://www.collegeboard.com/press/article/0,,46851,00.html

But statistics, schmatistics, we are still left with the question of
whether the increase in score means anything.  Maybe the students got
slightly smarter (whatever that means) ... or maybe the test got
slightly easier ... or .........

And of course we are left with the even bigger question of what
the SAT score itself (nevermind delta score) is supposed to mean.

I reckon the score means something, but I'm not entirely sure what,
and it probably is not nearly as meaningful as most people think
it is.  In particular, it seems unlikely that a dumb person would
get a super-good score ... but I know lots of smart people (reeeally
smart, reeeally successful people) who consistently got mediocre
scores.