Violation of Normality

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

Violation of Normality
Message posted by Nicole (via 168.191.249.109) on January 4, 2002 at 1:42 PM (ET)

I'm trying to run a MANOVA with 6 dependent variables, where 4 of the 6 are significantly non-normal according the K-S test. How conservative is this test?...and any suggestions for transforming the data?

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

Re: Violation of Normality
Message posted by JG (via 172.163.237.19) on January 5, 2002 at 3:09 AM (ET)

The appropriate transformation depends on the actual distribution that you have. Can you transform one variable at a time or do we have a multidimensional situation?

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