[Eeglablist] MANCOVAN for ANCOVA

Eric HG erichg2013 at gmail.com
Wed Aug 25 10:52:47 PDT 2021


Dear Eeglablist,

As there is no function built into MATLAB that can perform ANCOVA, I have
been trying to use the mancovan toolbox by William Gruner:
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.mathworks.com_matlabcentral_fileexchange_27014-2Dmancovan&d=DwIBaQ&c=-35OiAkTchMrZOngvJPOeA&r=kB5f6DjXkuOQpM1bq5OFA9kKiQyNm1p6x6e36h3EglE&m=sCeM1vOX3fM9WVH8rVH8BFB9azxa_PydrbJj6aL-CYg&s=I1IjgZx1RhZaoTntzRK00LchdhugWA1yL0QXlGUjmxQ&e= 

However, the documentation is a little difficult to decipher and maybe some
of you have used it before (hence the post in the forum).

When I run the following code using the "Iris dataset":

*load fisheriris*

*[ T, p , F] = mancovan(meas(:,3), species, meas(:,1))*


I get the following two F-values:

624.9854

194.9496


First question: Is it correct to understand that 1) the first one is the
group comparison, which is used to understand if there is a significant
effect between groups while controlling for the effect of the covariate
while 2) the second one is the results of the covariate?


Or more simply put, is the first value usually the one used when comparing
groups in the EEG literature?


Best regards,


Eric


More information about the eeglablist mailing list