I would like to run an anova with more than 500 subjects and 3 copes per subject. The Feat GLM gui doesn't seem to be able to handle this much information. I tried to make a design.fsf file myself using perl, and then just running the feat without having to open the GLM gui. This works with normal sized data sets, but once again with large data sets I get this error
Problem with processing the model: Can't find key fmri(evg1.2)
That line does exist in the file I created, so I'm at a loss for why I'm getting that error. Any ideas on what could be happening? Or perhaps other solutions to running such a large analysis?