[R] Multi-level Models with NLME

[Magill, Brett]

Hello,

I have a question about setting up a multi-level model in NLME.  I have the
P&B book, but I am still having trouble specifying the model.  I am using
the most recent version of NLME on R 1.3.0.  I was referred to NLME after a
previous query on multi-level models.  Yet, I think part of my confusion is
that I do not have a mixed model as far as I can see.  I have random effects
nested within levels of a random effect?

The problem is this:  I have a sample of classes (class being the nesting
factor).  Within each class I have a sample of students.  I have separate
random effects models at each level.  That is, at the class level, I have:

	random=~class_size+number_teachers+class_length

and at the student level I have:

	test_score~treatment+experience+age.

What I want o assess is the contribution of class level variables
simultaneously with individual level variables on test scores.
 

Now, my problem is specifying them both as a random effect.  I tried to use:

	random=list(class=~class_size+number_teachers+class_length,
=~treatment+experience+age)

But NLME did not like the second argument, no surprise, I really didn't
expect it too as random=list is used for multiple grouping factors.  Yet, I
can't figure out how to specify this.  Can someone offer me some advice on
how to fit this model.

Thank You.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._