[Marsbar-users] large variation in contrast values

[Nicole D. <nic...@uk...>]

Hi,
I get large variations (high SD/SEM) between my subjects' contrast
values. Some have values in the 30ies to 40ies and some even have
negative values. Does that sound fishy? Is there an upper (conventional)
limit for a contrast values or is a value of 30 o.k.???
=A0
I have some heterogeneity within the design matrices of my subjects
'cause the number of trials per condition were dependent on=A0judgements
they made. In addition, I had 4 scanning sessions. Sometimes it happened
that they made only a few judgements per a certain condition in e.g.
session 1 and more in session 2. Could that somehow explain=A0the size and
variation regrading the contrast values?
=A0
It would be great if you could help ASAP...
Thanks so much,
Nicole
=A0
=A0
=A0
=A0


------------------------------------
Dipl.-Psych. Nicole David
Klinik und Poliklinik f=FCr Psychiatrie und Psychotherapie
Universit=E4t zu K=F6ln
Kerpenerstr. 62
50924 K=F6ln
Deutschland
Tel:+49-221-478-87146
Fax::+49-221-478-87702
http://www.medizin.uni-koeln.de/kliniken/psychiatrie/Bildgebung/
------------------------------------