Charles A Pierce (capierce) wrote:
>>Is there a way, perhaps via a mixed-factor chi-square or loglinear model or HLM, to analyze grouped count data such as the following (N = 400; 200 cases in each treatment condition; treatment condition is between Ss and pre-post is within Ss; I don't have individual-level data, just grouped count data like you see below; DV = # of cases observed engaging in a behavior at pre and post):<<
I think this is a McNemar's test situation, unless I misunderstood you:
help symmetry
help symmi
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
