Dear Mark, Yes, that is right with respect to the command line, i.e., supply the "-f design.fts" to run the F-contrasts and, if you'd like cluster-level inference, also include the -F and/or -S options. With respect to the repeated measurements, check the randomise webpage for examples. All the best! Anderson 2012/11/27 Mark <[log in to unmask]> > Hi, > > I've set up a repeated-measures anova in GLM (5 t contrasts and 3 f > contrasts) and I want to do a small volumn correction by randomise in a > mask. I enter the directory of cope.feat and use filtered_func_data as the > input. If I run > > randomise -i filtered_func_data.nii.gz -o output_name -d design.mat -t > design.con -m mask.nii.gz -T -c 2.3 > > There will be only output of the statistics of 5 t contrasts (tstst1~5). > But what I want is those of f contrasts. So should I use the one below or > other command? > > randomise -i filtered_func_data.nii.gz -o output_name -d design.mat -t > design.con -f design.fts -m mask.nii.gz -T -c 2.3 > > Thanks. > > Mark >