Short answer:
You don't.
Slightly longer answer:
LR tests are not appropriate for clustered data. Any survey analysis text
(i.e. Korn and Graubard, Smith et al) will explain why, and there is a FAQ
on the Stata web site. Use linear combinations and contrasts. Depending on
the version of Stata (8 or pre-8) the syntax is something like -lincom and
-test.
Bryan Sayer
Statistician, SSS Inc.
[email protected]
-----Original Message-----
From: [email protected] [mailto:[email protected]]
Sent: Thursday, February 20, 2003 11:17 AM
To: [email protected]
Subject: st: svylogit procedure and covariate interactions
We are analyzing survey data using the svylogit procedure in Stata. In the
model of an outcome, we want to examine whether an interaction term is
significant. We think a likelihood ratio test comparing the model with the
interaction and without the interaction would be appropriate. However, I
have been unable to determine how to get the likelihood values in Stata. We
would like to know how you perform a likelihood ratio test using svylogit.
Please provide references if you have.
Thank you for your help.
*
* 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/
*
* 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/
