#^#^#^#^#^#^#^#^# FROM CHUCK TUCKER 941201 #^#^#^#^#^#^#^#^#
I have been using the PCT argument on use of statistics
in the behavioral sciences and today read Stephen Jay
Gould's review of the book in THE NEW YORKER 941128
which states: "Herrnstein and Murray's correlation
coefficients are generally low enough by themselves
to inspire lack of confidence. .... Although low figures
are not atypical for large social-science surveys involving
many variables, most of H and M's correlation are very
weak - often in the 0.2 to 0.4 range. Now, 0.4 may
sound respectably strong, but - and this is the key point -
R2 is the squarre of the correlation coefficient, and the
square of a number between zero and one is less that the
number itself, so a 0.4 correlation yields an r-squared
of only .16. In Appendix 4, then, one discovers that the
vast majority of the conventional measures of R2, excluded
from the main body of the text, are less that 0.1. These
very low values of R2 expose the true weakness, in any
meaningful vernacular sense, of nearly all the relationships
that form the meat of 'The Bell Curve.'" (p. 147) Sound
familiar to anyone!
BTW, Newsweek in their report on this book interpreted R
as R2 and said something like ".4 correlation accounts
for 40% of the variation" - I was amazed.
On Modelling
I would really appreciate it if someday one of you modellers
could provide some instructions for using these model
program for teaching and perhaps research. I think it would
a nice topic for an essay - "How to use ..... "
Regards, Chuck