Hello,
I am working with a multinomial outcome for the first time, and its relation
to an ordered predictor, year of occurrence. First, I would like to look at
the unadjusted model. I ran -opartchi- and -mlogit-. I like -opartchi-
because the table shows all categories explicitly and the output is very
concise, which the investigator for this study appreciates. The help file
explains that the location component describes differences in the location
of the responses on the ordered scale, while the dispersion component
describes dispersion across the ordered scale. I interpret this to mean
that for variation in the number of cases per year receiving each form of
treatment, the chi-sq stat=2203 on 4 d.f. (see output below), and for
variation in the pattern of treatments over time, that is, which treatments
are getting more or less common, the chi-sq statistic=95.76 on 4 d.f. Is
that correct?
Thank you, Laurel
. opartchi year, by(treatment) table
...
Chi-square tests
df Chi-square P-value
Independence 20 2339 0.0000
-------------------------------------------------------
Components of independence test
Location 4 2203 0.0000
Dispersion 4 95.76 0.0000
Laurel A Copeland
Data Analyst
VA Ann Arbor Healthcare System
*
* 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/
