I'm looking for a way to do a post-hoc analysis for a Fisher-Freeman-Halton analysis (Fisher's Exact for contingency tables larger than 2x2). For example I've found a significant difference on a 6 (groups) x 2 (outcome) table. The next step would be to determine which of the groups differs significantly. Usually you would perform some sort of post-hoc test, but for this specific analysis (Fisher-Freeman-Halton/Fisher's Exact for mxn tables) I can't find any. All suggestions are welcome!
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
I do not believe that any formal post hoc tests for a Fisher-Freeman-Halton analysis have been implemented. I know of one case, however, in which this was accomplished simply by analyzing the individual 2 x 1 contingency tables and adjusting for multiple comparisons.
