--- On Sun, 2/8/09, Alison McCarthy wrote:
> I was wondering whether there was a problem if I reported
> standardised coefficients? I have many variables that are
> scaled differently and do not appear to be comparable in the
> using the usual b scores. I am using GEE, pa f(nb)
> corr(exch)
In principle there is nothing wrong with standardizing
variables to make their coefficients comparable. I'd be a bit
hesitant to compare the effects of variables that are very
different on substantive reasons. I'd rather report the
effect of getting a year older and getting a euro/dollar more
income on a dependent variable. Standardizing means that you
loose these substantively interesting metrics, which is often
more important to me than finding the "strongest" effect.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
