open-discussion > ANOVA:1-factor 3-levels (Repeated Measures)
Showing 1-2 of 2 posts
Display:
Results per page:
Sep 21, 2017  06:09 AM | zhao rui
ANOVA:1-factor 3-levels (Repeated Measures)
Dear all,

I have 20 subejcts each of whom have 3 sessions.

I want to analyze the difference between these 3 session.

I setup the design matrix and contrasts just like the example of 'ANOVA 1-factor 4-levels (Repeated Measures)' in the fsl mannual (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/F...). For the design matrix, see the attach file.

However, I do not find the outputs about the F-test. All the output files are t-stat per contrast and the corrected p-value images. I do not know what was the problem.

How should I do?

Thanks a lot

yours

Rui
Attachment: design-FTEST.zip
Sep 29, 2017  05:09 AM | Jorge Manuel S├ínchez - Hannover Medical School
RE: ANOVA:1-factor 3-levels (Repeated Measure
Dear Rui,

unfortunately the current version of the Toolbox does not support F-tests.
Dual-Regression runs the following randomise command:
  • randomise -i [input] -o [output] -m [mask] -d [design] -t [t-contrast] -n 500 -x -T -V
If you want to perform F-tests, you will have to edit some commands manually. Open dR/scripts+logs/drD. At the end of each line, you will find the randomise command. Append -f [design.fts] and then rerun it.
You will also have to edit drE to merge the output.


We are currently working to include these features in the next release. It will be possible to change the number of permutations and the -f and -e tags will be supported.


Best regards,

Jorge Manuel