An alternative is a loglinear model,
that is, the log cell frequency is a function
of the covariates mapped to dummy variables.
That would be most easily done with -glm-
or -poisson- after a -contract-. That
might be a sledgehammer, but it's a systematic
approach.
Nick
[email protected]
David Airey
> > I am using the tabulate, chi2 command to calculate a chi-square
> > test of independence for 3(race) x 3(school type) table.
> However, I
> > would like to know if each of the racial group differs
> > significantly from one another, not if the overall chi-square is
> > significant - sort of like a bonferroni for chi-square. Is the
> > tabulate command the right command to do this?
> Maybe use a model that allows 3 categories as dependent variable
> predicted by a 3 level categorical variable, like mlogit?
>
> xi: mlogit school i.race
>
> followed by use of "test" as in the examples in the help file...
*
* 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/
