[JMP] Scheirer-Ray-Hare extension of Kruskal-Wallis test

#1
Hello,
i was looking for a generalization of the multi-way ANOVA and i found that the Scheirer-Ray-Hare extension of Kruskal-Wallis test could work. but i don't know how to carry out this test with JMP (if it's possible).
does someone have any idea how i could do that test with JMP?

Thanks in advance

Adrien
 
#2
I'm not sure about the SRH test. I will look into it. But - how about converting your data to ranks. Then run a regular 2-way ANOVA on the ranks. This is a common approach.
 
#3
thanks for your answer but does the rank transformation work only for the 2 way ANOVA and not beyond? because in my case, i have 4 crossed factors

adrien
 

Dragan

Super Moderator
#6
...But - how about converting your data to ranks. Then run a regular 2-way ANOVA on the ranks. This is a common approach.
This approach can be used but, unfortunately, it is really only legitimate for a two-way ANOVA where you have only 2 levels of each factor i.e. a (2X2) design.
 
#7