Non-parametric test for interaction

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Non-parametric test for interaction
Message posted by Mitch (via 192.38.5.67) on May 20, 2001 at 12:25 PM (ET)

How do I test for an interaction effect, in ANOVA, by using a non-parametric test. Do I compute dummy-variables for all of the categories on the variables and test by using Mann-Whitney or is it possible to use the Kruskall-Wallis test? Is it possible at all?

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Non-parametric test for interaction
Message posted by Didriana Leal (via 155.201.35.19) on May 25, 2001 at 10:29 AM (ET)

I would like to know about the advantages of the non parametric test over the parametric test. Also, I would like to know the differences

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.