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

[adrienbaudouin]

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

 

[lumhearts]

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.

 

[adrienbaudouin]

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

 

[lumhearts]

I'm not sure - sorry. Probably not.

 

[mp83]

This shows how to achieve it in SPSS

http://books.google.com/books?id=1eTyuMDND94C&pg=PA145&lpg=PA145&dq=nonparam#v=onepage&q&f=false

Visit : Site|Stats Blog|Personal mini-site

Contact : Contact

 

[Dragan]

...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.

 

[nguyenkinh]

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

If you can use R, this is your solution
http://www.ytecongcong.com/2013/04/scheirer-ray-hare-test-kiem-dinh-phi-tham-so-two-way-anova/