Hi Michael,

What you describe sounds like a varying covariate which is usable using MANOVA syntax. The basic commands :

MANOVA time1 to time12 with timex_1 to timex_12 /wsfactors treat(2) block(2) time(3) /wsdesign.

GLM appears to treat the covariates as unvarying within subject.

This was accessible via the menu in version 6 of SPSS windows but then mysteriously disappeared from it.

HTH

Peter

On Wed, 14 Aug 2002, Michael Bersick wrote:

> Hi all, > I have a data set which is a 2x2x3 within subject design, with the 3 level > factor being time. Basically, subjects came in every quarter over the course > of 3 quarters to be tested. In addition, subjects were tested within each > quarter at varying times, and I would like to use the # of days into each > quarter as a covariate to control for differing amounts of experience. The > only thing I have been able to come up with is to regress subjects' data > individually at each quarter and then run a 2x2x3 repeated measures GLM on > the residuals for the 3 quarters, but I have a lingering sense that this is > not quite sound. Is there a better way to do this in GLM? > Thanks for any advice. > MB > > > _________________________________________________________________ > Chat with friends online, try MSN Messenger: http://messenger.msn.com >

--------------------------------------------------------------------------

" " A pinch of probability is worth a pound of perhaps

James Thurber

Peter Watson MRC Cognition and Brain Sciences Unit 15 Chaucer Road Cambridge CB2 2EF

Tel: +44 01223 355294 Ext. 801 Fax: +44 01223 359062 Email: peter.watson@mrc-cbu.cam.ac.uk