I have been working really hard to find the answer to this question, but the statistical references I have read all say they won't go into detail about these techniques. They give a citation, but I don't have access to it. I have asked the statistics professor in my department, but he doesn't really work with ANOVA too much and wasn't able to help.
So here is my question: Which post hoc one-way ANOVA test, Tamhane's T2, Dunnett's C and Dunnett's T3, is better under what circumstances?
The Games-Howell test is too liberal for my situation as the sample sizes are less than twice as different between my groups.
Any help in this matter would be greatly appreciated!
So here is my question: Which post hoc one-way ANOVA test, Tamhane's T2, Dunnett's C and Dunnett's T3, is better under what circumstances?
The Games-Howell test is too liberal for my situation as the sample sizes are less than twice as different between my groups.
Any help in this matter would be greatly appreciated!