Dear users,
I am new to the Stata world, and I am not an expert statistician at all. In my master's thesis, I have a 5x3 comparison table with expected frequencies low enough to make Fisher's exact preferred over chi-square. Unfortunately, there is no way I can merge columns/rows. The table is the result of the intersection of two clustering analyses performed. Each column corresponds to a clustering category (call this clustering A). Similarly, each row corresponds to another clustering category (call this clustering B). Once performed the Fisher Exact Test and found out a very low p-value, I am interested in performing a post-hoc test. I want to be able to say that, given a specific clustering category within clustering A, there is a significant difference in actual frequencies with respect to expected ones. What would be the most suitable test to do so?
Thank you very much!
I am new to the Stata world, and I am not an expert statistician at all. In my master's thesis, I have a 5x3 comparison table with expected frequencies low enough to make Fisher's exact preferred over chi-square. Unfortunately, there is no way I can merge columns/rows. The table is the result of the intersection of two clustering analyses performed. Each column corresponds to a clustering category (call this clustering A). Similarly, each row corresponds to another clustering category (call this clustering B). Once performed the Fisher Exact Test and found out a very low p-value, I am interested in performing a post-hoc test. I want to be able to say that, given a specific clustering category within clustering A, there is a significant difference in actual frequencies with respect to expected ones. What would be the most suitable test to do so?
Thank you very much!
