Hi This is far from being a simple question given the Within-S aspect of the continuous variable. If there are relatively few levels of the continuous variable (or the values can be reduced without losing info, producing empty cells, etc.) then GLM or MANOVA with partitioning of the interaction might still be a worthwhile approach.
Such questions are best put to the SPSS newsgroup. For this and other sources of help, see Raynald's SPSS Tools page at http://www.spsstools.net/FAQ.htm#FindHelp Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax [EMAIL PROTECTED] Department of Psychology University of Winnipeg Winnipeg, Manitoba R3B 2E9 CANADA >>> "Helweg-Larsen, Marie" <[EMAIL PROTECTED]> 18-Aug-08 1:36 PM >>> I am having difficulty finding information about how to set up a regression in SPSS. I am testing an interaction between a continuous and a dichotomized variable (which is why I'm not using ANOVA). However, the continuous variable is measured within subjects. I cannot figure out how to setup the regression with the repeated measures variable. I've consulted Aiken and West's book "Multiple Regression: testing and interpreting interactions" (which is excellent in terms of dummy coding variables, centering them and interpretation of interactions), looked at the SPSS manuals, and searched the internet (many useful SPSS resources but none address using a regression with a repeated measure variable). Can any of you folks direct me to a source with the correct syntax. Thanks Marie **************************************************** Marie Helweg-Larsen, Ph.D. Department Chair and Associate Professor of Psychology Kaufman 168, Dickinson College Carlisle, PA 17013 Office: (717) 245-1562, Fax: (717) 245-1971 http://alpha.dickinson.edu/departments/psych/helwegm **************************************************** --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED]) --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED])
