RE: Is IQ usefully predictive? (was: Re: the term "eugenics")

From: Harvey Newstrom (mail@HarveyNewstrom.com)
Date: Fri Aug 24 2001 - 07:47:38 MDT


Damien Broderick wrote,
> I think Anders stated that it's not correlated with other desiderata such
> as success at college or in life. This is not supported by, say:
>
> http://www.sciam.com/1998/1198intelligence/1198gottfredbox2.html

This is an interesting graph at the top. I think I could agree with it
mostly, but I do have a couple of concerns about this graph.

First, it only shows negative things, such as unemployment, divorce, etc. I
guess the graphs prove that bad things happen to dumb people. It doesn't
show that intelligent people actually produce good things. I wish they had
shown the graph for positive measures such as grades, salary, etc.

Second, the negative items are all binary events. Unemployed or not? I
wish they had shown graphs of unemployment time, salary rate, grades or some
other curves. These would be easier to compare to the bell curve. A single
binary point, such as "poverty: yes/no" are not as useful for analysis as a
complete curve would be. Restricting the data to a single cut-off point,
and then graphing all those cut-off points on a curve is a very common way
to blur, distort or misinterpret data. If the original curve of
unemployment rates, salary, number of divorces, etc. were graphed, we could
directly see whether the curves follow the same IQ curve. The graph as
presented is very imprecise and may or may not support the conclusion as
given. I wish they had shown all the data on a curve instead of just a
single cut-off point such as "poverty: yes/no".

I realize that this is more of a populist piece, but I wish the statistical
analysis and science were more precise in the article and more completely
disclosed. I can't say it is wrong, but I can't say that it proves anything
either.

--
Harvey Newstrom <http://HarveyNewstrom.com> <http://Newstaff.com>


This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:10:02 MST