On Mon, 28 Jul 2008 14:33:44 +0100, G Markopoulos
<[log in to unmask]> wrote:
>Dear all,
>
> if at the first level analysis I want to define a t contrast that compares one
>condition with the average of two other conditions... would I weigh the
>contrasts thus: [1 -0.5 -0.5 0 0 0] OR [1 -1 -1 0 0 0] ? In other words, how
>do I collapse across two conditions and compare this to another condition?
Use [1 -0.5 -0.5 0 0 0]
>
> Thank you all in advance,
>
> Gerry
>
|