For a two factor GLM analysis with two levels, I'm uncertain as to why the Freesurfer GLM measurement of the interaction effect of say, group and gender, is a t-test and not an F-test. In my case, my factors are group (healthies vs patients) and gender (male vs female). For a 2x2 two-way ANOVA analysis, the measurement of an interaction is always an F-test.
For example, on the following Freesurfer FSGD Wiki page (Four Groups (Two Factors/Two Levels), No Covariates) http://surfer.nmr.mgh.harvard.edu/fswiki/Fsgdf4G0V under Contrast 7, it claims the gender-x-handedness is a t-test.
Is this contrast a t-test or actually an F-test?
Thanks, Chris
Hi Chris,
It is a t-test. In the case of one variate (a single row to the contrast matrix), an F-test is the same as an unsigned t-test.
doug
Chris Hyatt wrote:
For a two factor GLM analysis with two levels, I'm uncertain as to why the Freesurfer GLM measurement of the interaction effect of say, group and gender, is a t-test and not an F-test. In my case, my factors are group (healthies vs patients) and gender (male vs female). For a 2x2 two-way ANOVA analysis, the measurement of an interaction is always an F-test.
For example, on the following Freesurfer FSGD Wiki page (*Four Groups (Two Factors/Two Levels), No Covariates)* http://surfer.nmr.mgh.harvard.edu/fswiki/Fsgdf4G0V under Contrast 7, it claims the gender-x-handedness is a t-test.
Is this contrast a t-test or actually an F-test?
Thanks, Chris
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Chris Hyatt -
Douglas N Greve