Dear Nádia,
if you find a considerable correlation between TIV and a parameter of interest (which is used in your contrast), the suggested alternative approach is to use TIV with global scaling.
Best,
Christian
On Wed, 11 Jul 2018 12:24:14 +0100, Nádia Canario <[log in to unmask]> wrote:
> Hello
>
>I´m Nádia and i have a question regarding the use of TIV as a covariate in
>CAT12. So, if i understood correctly i have to check whether this variable
>is independent from the other variables (like my groups) and if it is not i
>should run the Design again using global scalling with TIV, right?
>
>
>Thanks for your help
>Best regards
>
|