www.cmm.bristol.ac.uk/forum

Hello,

I am trying to fit 3 level multilevel model, measurements taken at two time points (pre and post treatment) nested within participants further nested within clinicians. Is it possible to fit such a model when there are only two time points at level 1? I am interested to assess if there is a change between the first and second measurement occasions on the outcome variable.

If there are any readings you can recommend on mlm models with only two time points (most resources refer to three or more time points) I would greatly appreciate it.

Many thanks

Two time points at level 1 should be no problem however you may consider constructing the change response post-pre and thus reducing your 3 levels to 2 levels. You could still use the pre response column as a predictor if you wanted to look at the impact of pre-value on size of change. Hope that makes sense,
Bill.