I have a little exercice to do for a stats class... And I need some help. (SPSS program)
-1 dicho dependant variable
-4 predictors (all categorical variables of two or 3 groups-that can be use as continuus variable) and 1 control (categorical 3 groups not continuus)
-and I want to do a logistic regression with that, but also had a moderator (categorical 3 groups)
I tried this regression. first Recode new variables for my 3 groups moderator, 0-1 with each group.
I compute interaction variable between my 3 new recode predictors and 1 of my predictor (3 groups but us as continuus...) so three interection variables
First bloc with my controle
Second bloc with all the other V (and my 3 recode moderators)
Third bloc just my interaction variable.
Am I doing something wrong? Is it possible to do something like that?
Thank you for your help!!
Last edited by TaB; 05-28-2012 at 12:50 PM.
Advertise on Talk Stats