Dear All,
recently, while setting me on the straight and narrow about linear
contrasts for a linear mixed effect model, Prof Ripley pointed out that my
interpertation of the call
anova(mymodel.lme)
was not correct, because I was meant to add
type = "marginal", as in anova(mymodel.lme, type =
"marginal")
I tried to look deeper in the issue, asking people, checking on the R-list
archives and googling around, but the best I could find was another old
post reiterating that the anova call for fixed terms in a mixed model
requires *type = "marginal"*. I also checked MASS IV and Pinheiro
& Bates,
but I am not sure how much I got from my read.
As I am not a statistician, I'd simply like to ask whether is this the
*standard* way of generation an anova table for a mixed effect model, or,
if it is not, what are the general rules underlying the application of a
marginal approach for a mixed model effect.
Regards,
Federico Calboli
========================
Federico C.F. Calboli
Department of Biology
University College London
Room 327
Darwin Building
Gower Street
London
WClE 6BT
Tel: (+44) 020 7679 4395
Fax (+44) 020 7679 7096
f.calboli at ucl.ac.uk
