[Eeglablist] LIMO between subjects ANOVA with repeated factors

Dr Cyril, Pernet wamcyril at gmail.com
Wed Sep 1 00:24:21 PDT 2021

Hi Chrissy

> Good afternoon,
> I have a between subjects (3 grp) design with repeated factors (4 of them). I'd like to calculate custom contrasts, but I'm not sure what order the contrasts would go in. From the design matrix (and common sense), I can see that I would need 12 values in any contrast (13 if you include the constant).
> I'm finding the documentation for the between subjects ANOVA with repeated factors a little sparse.

Actually, you would not - at the 1st level (per subject) you should have 
5 columns, 4 conditions + the constant; then at the group level, you 
would have the groups and the conditions (no constant - because LIMO 
MEEG relies on a Hotelling T^2)

the documentation is 
-- I would love feedback to improve it though!

> Would these contrasts be in the following order (1) cond 1, grp 1, (2) cond 2, grp 1, (3) cond 3, grp 1, (4) cond 4, grp 1, (5) cond 1, grp 2 .... etc
> OR (1) cond 1, grp 1, (2) cond 1, grp 2, (3) cond 1, grp 3, (4) cond 2, grp 1, (5) cond 2, grp 2

those are therefore not needed - the repeated measure at the group level 
will compute for you the repeated factor effect, group effect and 
interaction. Of course post-hoc contrasts will be possible too here 
based on the design - happy to help this that if needed.

best wishes



Dr Cyril Pernet, PhD, OHBM fellow, SSI fellow
Neurobiology Research Unit,
Building 8057, Blegdamsvej 9
Copenhagen University Hospital, Rigshospitalet
DK-2100 Copenhagen, Denmark

wamcyril at gmail.com

More information about the eeglablist mailing list