Re: I still think the glass is too big



Les Cargill wrote

Ech. Revenge of The Bell Curve.

http://krugman.blogs.nytimes.com/2009/08/28/heredity-environment-justice/

Krugman says "neener, neener, you Social Darwinist." in response to
http://gregmankiw.blogspot.com/2009/08/least-surprising-correlation-of-all.html

and deLong sez:
http://delong.typepad.com/sdj/2009/08/if-you-are-so-rich-why-arent-you-smart.html

which eventually refers to:
http://www.epi.org/economic_snapshots/entry/webfeatures_snapshots_20051012/

Okay. So if you don't wanna click all the links, it's the usual stuff about predicting socially desirable outcomes in
education.

So we end up with (I think) three tangled metrics which make no real single claim on the truth. These are:

1) income
2) IQ
3) educational success.

A partially complete list of (not-orthogonal) factors which are not mentioned includes:

- Personality. In order to be a good student, one must be able to willingly suspend disbelief.

And must give a damn about what is being taught.

- Rough perception of the value of the exercise. A good
student will have more faith that the process will be worth it.
Obviously, parents who did better will better send that message.

Not necessarily with that last.

- Adherence to the social-signaling expectations of the school
system itself. People can trade extracurricular participation
for actual performance. Given that people strenuously game the
system ( to a level you would have to see to believe ), this can
result in bizarro behavior.

But is largely irrelevant to whether they get much out of the education.

I still think the glass is too big. By this I mean that the mapping
from high level goals to detail level goals in education is incredibly fuzzy, full of self-reference and fraught with
irrational goof.

And the MUCH more fundamental point is that those who are
quite capable of educating themselves when adults will do fine
in any education system as long as they have access to information.

If I read the epi.org link properly... then what is broken is not the effect of income on educational performance, but
the ability of performance in 8th grade math to predict what is desired.

Its terminally stupid to believe that it ever could.

Which is fascinating - as measures of raw talent in academia go, 8th grade math scores are pretty close, SFAIK, to the
gold standard.

Like hell it is, most obviously with those who have no need for maths.

Look at the rough sensitivities - scores are still *as* much determined by 8th grade math scores as anything else.

And scores are a useless measure of anything much that matters.

I think outcomes are most likely about social signaling than anything else, and that higher income is more correlated
with that than raw ability of the student.

And high income is completely irrelevant to what matters for most.

I think, ultimately, Krugman is right - it's not quite a meritocracy out there.

Nothing like it in fact.

But the three of these fellas probably could not agree on what "merit" means to begin with.

They're completely irrelevant.


.