Dear all,

I have googled quite a bit and looked into the Stata manual, but couldn't really find an answer to the following:

We are planning a longitudinal (pre, post, post) intervention study in an educational context.
We are interested to find out how different treatments (dummy coded) effect the students performance in a competence test (continous variable).
Various background variables (e.g. gender, SES) need to be controlled for.

We will have 3 timepoints (level 1) nested in students (level 2) nested in classes (level 3),
i.e. we do have repeated measures AND the students will be in groups.

My question is:
Can I model these 3 levels with the -mixed- command in Stata or is it restricted to two levels?

Thank you very much in advance and kind regards,
Hanne