Thank you for your suggestion, Maarten.
Olga wrote:
We are trying to use xtmixed as a mixed effect anova taking advantage of
maximum likelihood estimation.
In xtmixed Stata seems to interprete categorical (cardinal) fixed factors
(such as colour) as ordinals. How do we get Stata to treat these as a
simple categorical?
e.g.
xtmixed y colour || individual
where y is dependent continuous variable, colour is {red,green,blue}(coded
with an integer) and individual is a random factor (coded with an integer)
produces output in terms of regression coefficients etc. implying that
STATA interpreted the categoricals as ordinal variables. As though the
order was important, but of course it is not for categorical.
*
* 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/
