Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Dirk Enzmann <dirk.enzmann@uni-hamburg.de> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | RE: st: increasing variance when adding covariates (xtmelogit) |
| Date | Thu, 19 May 2011 15:35:42 +0200 |
Robert de Vries asked: http://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.1105/Date/article-891.html and Garry Anderson answerd: http://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.1105/Date/article-965.htmlIn this context (and not only for multilevel logistic regression models) the article by Mood referenced to below is also relevant. It shows that logistic regression estimates are affected by omitted variables even when these variables are unrelated to the independent variables in the model. See:
Mood, C. (2010). Logistic regression: Why we cannot do what we think we can do, and what we can do about it. European Sociological Review, 26, 67–82.
Dirk * * 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/