Skip to content
Prev 10778 / 20628 Next

Nested Random Effects Model using lme4/lmer

Hi,
A late reply.

m2 is working because you estimate two variances, one for crop species 
and one for wild. Dont know the experimental set up here but if within 
species variation is not interesting, you may as well average over 
species (mean of the 6(?) replicates) to keep it simple. As you may have 
noticed, the fixed effect estimate will not change. Regarding 
pseudo-replication. This is important for the calculations of DFs. So if 
you use nlme or intend to use t/F-test, you have to make sure that they 
are correct ( F1,30 I guess).

Gustaf