A few comments on this:
For non-pairwise comparisons I was tend to use the Scheffe adjustment
although Bonferoni or Sidak are also common.
There seem to be more issues concerning pairwise comparisons for which
the adjustments listed above are considered too restrictive. Many of the
the pairwise adjustments are based on the Studentized Range distribution.
I have been using John Gleason's -qsturng- program from 1997 to obtain
quantiles of the Studentized Range. It would be a step forward if Stata
were to "officially" include this distribution among its collection of
probability distributions.
The SPSS approach of including checklists of every possible pairwise
test seems to encourage users to shop around for the one that
produces the most significant comparisons.
On Mar 11, 2008, Nick Cox wrote:
Disappointment and embarrassment being noted, what would be your top
candidates for inclusion?
(And will others then agree, showing a consensus among people who use
this
kind of thing on what is justifiable and useful?)
StataCorp has a certain historical timidity about implementing stuff
when the practitioners of a field seemingly
can not even decide on procedures and every expert contributes a
different approach. Rightly or wrongly, multiple
comparisons has that reputation.
--
Phil Ender
Statistical Consulting Group
UCLA Academic Technology Services
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
