I have a question regarding the interpretation of an interaction using categorical variables where one is dummy coded (0, 1) and the other is contrast coded.

The variables are:

  • Var1: 3 levels, coded using planned orthogonal contrasts (level 1 vs. level 2+3, and level 2 vs. level 3)
  • Var2: 2 levels, dummy coded

The results from a logit mixed model regression using R and Lme4 gives the following general results:

Var11       n.s.    
Var12       n.s.
Var2        ***
Var11:Var2  ***
Var12:Var2  n.s

Now, as I understand it, the main effects are in fact simple effects and gives the result for the contrasts in Var1 on the first (0) level of Var2 only (link). But how do I interpret the interaction? Is it a "simple interaction" as well, and how should that be interpreted? How would this differ from an analysis where Var2 is coded in another way (say -1, 1)?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.