Hi. 

In following model, ABCD impacts E. E impacts F. F impacts G. H and I are moderators. In most studies impact of ABCD on E is measured by experiment and measured using ANOVA. Can I measure ANOVA and SEM together in single model?

Similar questions and discussions