Hi,

I have used a linear mixed model (in SPSS) on a longitudinal dataset with two groups (patients and controls) as fixed factor, two tasks (did seperate tests) and time with three levels (repeated measure) and subject (wasn't modeled as random factor as using time as repeated measure excludes the possibility to use subject as random factor). Dependant variable was surface area of activation in different ROI's. I performed seperate tests (on the square root of the data to reach compound symmetry) for each ROI to look for group, time and time*group interaction effects.

Now I would like to do a similar analysis on a node by node basis (I use SUMA) to look for the same effects. There are missing values, which is why I didn't use a repeated measures ANOVA in the first analysis. What test should I use?

Casper