Michael Mitchell replied to Daniel Waxman
> More information about "vibl" can be found at
>
> http://www.ats.ucla.edu/stat/stata/seminars/stata_vibl/default.htm
And John Moran replied to Daniel Waxman
>> The two Stata modules "vibl" and "inteff" with the
>> discussions in Stata Journal are good ways to begin.
Although it was good of Michael and John to reply with helpful solutions
to Daniel's problem, I fear I have to chuck a spanner in the works.
Essentially, it has been shown that the use of interactions in logistic
regression may lead to invalid conclusions if residual variation differs
across groups. Consequently, such variation can produce apparent
differences in coefficients that are not indicative of true differences in
causal effects. Not my words, but those lifted straight out of the first
two pages of the much quoted paper by Allison (1999).
What to do? Fortunately, Glenn Hoetker's -complogit- allows you to test
for the presence of such residual variation. Go to:
net from http://www.business.uiuc.edu/ghoetker/
and click on -complogit-
or -findit complogit- if that doesn't work. Hope that helps!
CLIVE NICHOLAS |t: 0(044)7903 397793
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
Reference:
Allison PD (1999) "Comparing Logit and Probit Coefficients Across Groups",
SOCIOLOGICAL METHODS AND RESEARCH 28(2): 186-208
*
* 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/
