Hi, I'm using glimmix but my questin is applicable to mixed too.
I'm trying to analyse animal behavioural data. I have tha data entered as binomial by animal but the fact is animals affects others behavior so I have to analyse by separated groups of animals.
How I tell SAS to use a group of animals (REP variable) as a sample unit instead of an individual animal?
Rather than attempting to make the herd the unit of measure and treating each animal as a repeat, the more traditional way of approaching the problem is to use the animal as the unit of measurement and specify the herd as a random effect.
Thanks Doc, for your quick answer. So, you say with this approach you solve the no-independence issue that exist inside each herd? I think there's no a herd effect but an individual effect...Thanks again