Hi Michelle,

Are you familiar with the interface of the command Glm (or Glm_gui in the Mac)?

1) Open the Glm_gui

2) Go to the tab "Contrasts and F-tests"

3) Using design1 as an example, set the number of t-tests to 24 and F-tests to 4.

4) Enter all the t-tests (contrasts 1-24)

5) In the buttons on the right of each contrast, select those that will constitute the F1, F2, F3, and F4, as indicated in the spreadsheet.

That's it!

If you don't want the t-tests, but only the F-tests, it suffices to enter just the 8 relevant contrasts, and mark them as indicated as F1, F2, F3, and F4. The others can be omitted.

Hope this helps.

All the best,

Anderson

On 2 December 2016 at 23:05, Michelle VanTieghem <[log in to unmask]> wrote:

Hi Anderson,

Ok thanks. I kept with the continuous age model. I wanted to ask how to actually test the omnibus F-test of the 2(group) x 3 (emotion) interaction. I have all of the contrasts but I don't have the omnibus test. I haven't been able to find an example on the FSL info page, and I am confused by your design matrices that just indicate f1,f2,f3,f4 next to some contrasts. How do I actually make a contrast that represents the F-tests, aka, what does the F-test file look like? I have a feeling it is very simple I just don't know how it should be coded!

Thank you,
Michelle