Dear Group:
This question is about whether one has to use multivariate statistics
or univariate statistics given the description below.
I have 2 populations- patients and controls.
They each perform 2 tasks that test a similar function but in two
different ways. In this case it was 2 language tasks each testing a
different aspect of the language system. Each task is a box car with
active and control conditions.
What I want to do is to say in the patient population across the two
tasks what is different about the patients vs. controls. I'm not
looking for areas that are necessary similar in the two tasks for the
patients vs. controls.
My 2 thoughts are this.
1) set up a random effects set of con* images. Use a two sample
t-test, and enter all patient images (both tasks) as one condition,
and all controls as the other.
However this would seem to ignore the repeated measures aspects.
2) use multivariate analysis and designate each of the tasks as a variate.
Any thoughts.
thanks,
Darren
--
-------------------------------------------------------------------------
Darren R. Gitelman, M.D.
Cognitive Neurology and Alzheimer's Disease Center
E-mail: [log in to unmask] WWW:
http://www.brain.northwestern.edu
Voice: (312) 908-9023 Fax: (312) 908-8789
Northwestern Univ., 320 E. Superior St., Searle 11-470, Chicago, IL 60611
-------------------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|