Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: GEE and ANOVA

From   David Airey <david.airey@Vanderbilt.Edu>
Subject   st: GEE and ANOVA
Date   Wed, 23 Nov 2005 21:37:44 -0600


I have a simple data set with a small number of experimental groups with a balanced number of animals per group. Group differences are the interest. The measures on each animal consist of 20 to 40 binary repeated measures. I can analyze these data two ways. First, I can use GEE with binomial family, logit link function, and exchangeable correlation structure. Second, I can average the 0s and 1s for the dependent variable to a summary statistic for each animal (proportion or percentage) and use a simple oneway ANOVA. Each animal in this data set has not been equally sampled. Averaging to a summary statistic ignores this. Does GEE treat clusters with unequal sizes differently?


* For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index