Dear all,
I'm dealing with data that doesn't come from a normal distribution. I wanted
to make a ANCOVA with baseline and treatment as covariates. Due to the fact
that the residuals are not normal at all, I have to use an other method.
I was thinking to do an ANCOVA an the ranks (because I want to include
covariates in my model).
But,
1) do I have to look at the normality of the residuals when I use an ANVOCA
on the ranks?
2) what kind of rank would you used?
3) I heard that BLOM ranks transform the data into a normal distribution?
But should I check the normality of the residual in that case?
Thanks a lot for your advise because I really don't know what to do.
V. Baudoux
_________________________________________________________________
Receive your Hotmail & Messenger messages on your mobile phone with MSN
Mobile http://www.msn.be/gsm/smsservices
|