Dear Boris,
you may want to test for all main effects and interaction effects at
the first level, i.e. run these t-contrasts
1 1 -1 -1; 1 -1 1 -1; 1 -1 -1 1;
plus any simple effects you are interested in (say A1 > B1, etc.).
Then take the corresponding contrast images to the 2nd level and just
conduct one-sample t-tests on each of them. The weightings of the
t-test at the 2nd level (i.e. '1' or '-1') will then allow you to
assess the different directions of first level effects, e.g. positive
versus negative interactions and so forth.
Good luck,
Tobias.
On 7/18/06, boris suchan <[log in to unmask]> wrote:
>
>
>
> Hi all,
>
> I am analysing a 2 x 2 factorial design at the first level and want to take
> the con images to a second level analysis.
>
> There are 2 factors A and B with each 2 levels A1 A2 B1 B2.
>
> Normally an f-contrast is used for the main effects and interactions, but f
> contrasts don't create con images…
>
> To create con images using a t-contrast at the first level I have two
> possibilities
>
>
>
> (A1 + A2) – (B1 + B2)
>
>
>
> and (B1 + B2) – (A1 + A2)
>
>
>
> In an f contrast these two t-contrasts would be combined, but which
> t-contrast should be used for the 2nd level analysis?
>
> Both?
>
>
>
> Many thanks
>
>
>
> boris
>
>
>
>
>
> Priv.-Doz. Dr. Boris Suchan
>
> Institute of Cognitive Neuroscience
>
> Ruhr University Bochum
>
> 44780 Bochum
>
> Tel.: + 49 234 3227575
>
> Fax: + 49 234 3214622
>
> mailto: [log in to unmask]
>
> http://www.ruhr-uni-bochum.de/neuropsy/mitarbeiter/boris_suchan.html
>
>
--
Tobias Egner, Ph.D.
Cognitive Neurology and Alzheimer's Disease Center
Feinberg School of Medicine
Northwestern University
320 East Superior, Searle 11
Chicago, IL 60611
Tel: (+1) 312 503 1749
Fax: (+1) 312 908 8789
|