 |
 |
Hi Andres, If I understand you correctly, your 2nd level is just combining across runs of the same subject. In that case, handedness doesn't change across run (same as gender), so you would not include handedness in the 2nd level. You could include both in a 3rd level analysis (across subjects) however. Best, -MH On Tue, 2011-02-15 at 17:20 +0000, Andres Roman wrote: > Hi All,a > I am doing the group analysis for a task that has 36 trials divided in two > runs. I was wondering about the use of covariates (sex and handedness) in > feat when you have such a model. I think that since I am doing a > multisession multisubject analysis I should add them as independent EV's in > the third level analysis after I merged my two runs in the second level > analysis using a fixed effect model. I figured that's the right way to > covariate for gender but im not so sure about handedness since that might > affect my second level analysis. Do you have any tips or hints for this? > Thank you very much >