3 way chi square? Group differences

Posted by lken on Apr 13, 2012; 5:54pm
URL: http://spssx-discussion.165.s1.nabble.com/3-way-chi-square-Group-differences-tp5639038.html

I have found a few links to post-hoc comparisons when using a chi-square test, but I have not been able to apply many of them to my data. Can someone point me to a post or send me in the right direction?

I have three groups (A,B,C).
I have one categorical response variable (fall type) with three levels (1,2,3)

I want to know:
a) Is the frequency of fall type 1 different bewteen groups A,B,C
b) Do the frequencies of fall types 1,2,3 differ across a single group A

I know I can get the first answer from a chi-square, but I don't know how to do post hocs to find where the differences are. I am unsure about the second question, whether or not I should just break the analysis down farther and test one group at a time.

Any ideas?

Thanks,
lken