Ricardo Ovaldia wrote:
I have data from an experiment where 50 pregnant rats
were scarified and their embryos randomly assigned to
two culture mediums. However, since the number of
embryos from each rat varies, the number of embryos
allocated to each media from each rat also varies.
Thus I have a M-N match within rat. (All rats have 2
or more embryos)
If the outcome was dichotomous, I could use -clogit-
to analyze these data. However, the outcome, "embryo
grade", is an ordered variable ranging from 1 to 5.
What I would need is then a conditional ologit, but
that does not seem to exist. Any suggestions would be
appreciated.
--------------------------------------------------------------------------------
A couple of possibilities come to mind. One would be -gllamm- or -reoprob-,
e.g.,
-xi: gllamm embryo_grade i.culture_medium, i(dam) family(binomial) ///
link(oprobit) adapt . . .-
Another, if you're interested only in hypothesis-testing, would be -npt_s-,
another user-written command, e.g.,
-npt_s embryo_grade, by(culture_medium) strata(dam)-.
Joseph Coveney
*
* 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/