> Carl Huberty wrote:
>
> Why do articles appear in print when study methods, analyses,
> results, and conclusions are somewhat faulty?... I can think of two
> reasons: 1) journal editors can not or do not send manuscripts to
> reviewers with statistical analysis expertise; and 2) manuscript
> originators do not regularly seek methodologists as co-authors. Which
> is more prevalent?
I would say that both are more or less necessary conditions
for the appearance of a statistically illiterate paper. Neither is
sufficient. Any difference in prevalence would be hard to spot,
as the affected papers would be (in case 1\2) well-written but as
pearls cast before swine; and (in case 2\1) sometimes well-written
despite the lack of consultation, and sometimes rejected. 1 1/2 of
these cases are hard to spot!
-Robert Dawson
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================