Hello I've encountered a problem while performing Logit analysis in SPSS. A analyze a 4*2*8 table formed by variables: education, sex, region of birth plus a covariate - age. Education is considered as dependent variable. I built a main effect model with a syntax below: genlog edu by sex reg with age /model= multi /print= estim freq cov /plot=none /design= edu edu*sex edu*region edu*age. The procedure is working but I've got some problems with interpetation. I'm able to interpret all parameters formed by categorical variables as they represent particular odds and odds ratios. But there are four parameters edu*age for each category of edu, what do they represent? I couldn't find an answer in SPSS Advanced Stats manual (ver 7.5) nor in books I have (Demaris: "Logit modelling"; Hagenaars: "Loglinear modelling with latent variables"; Vermunt: "LEM: A general program for analysis of categorical data". Please help me, I'm very depserate, because this analysis is a part of text I'm currently working on, which must be finished very fast. I would be very VERY gratefull, for eternity Michal Bojanowski mailto:[EMAIL PROTECTED] ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================