Re: st: Logistic regression for multiple binary dependent

["Bryan W. Griffin" <bwgriffin@georgiasouthern.edu>]

Are the three dependent variables separate measures from each subject such 
that they could be viewed as clusters of responses from subjects? If yes, 
one possible option is to model those outcomes from a generalized linear 
mixed model framework. This is akin to repeated measurements type models in 
regression/ANOVA. An example of such modeling can be found in Agresti's 
text Categorical Data Analysis (2002, 2nd edition, Wiley, p. 504-506) where 
individuals were asked whether they support abortion (yes/no) in each of 
three different situations. You can also find other literature (primarily 
in medical statistics) of such models which are often referenced as 
generalized linear mixed models, but sometimes as multilevel logistic 
models, logistic regression with random effects, and the like. At least 
three possible commands in Stata exist for these analyses, xtlogit (option 
re for random effect and pa for population average (GEE model), xtgee, and 
gllamm. Careful about interpretations from these model since the population 
average interpretation differ from the cluster-specific interpretation.




> Date: Tue, 22 Mar 2005 10:10:25 -0500
> From: cbautista@hivresearch.org
> Subject: st: Logistic regression for multiple binary dependent
>
> Hi list,
>
> In situations where we have a binary (0/1) dependent variable and it is
> well known to use logistic regression for the data analysis. However, we
> have a study with three binary dependent variables and we are looking for
> some statistical procedure to analyze them together.
>
> Any idea?
>
> Thanks in advance,
>
> /Christian
___________________________________________________________________
Bryan W. Griffin
Curriculum, Foundations, & Reading
P.O. Box 8144
Georgia Southern University
Statesboro, GA 30460-8144

Phone: 912-681-0488
E-Mail: bwgriffin@GeorgiaSouthern.edu
WWW: http://coe.georgiasouthern.edu/foundations/bwgriffin/

*
*   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/