I need to perform hierarchical regression with dummy variables. I also need to check moderation by introducing in the model interactions of these dummy variables and the moderator. My questions are:

  1. Do I introduce in the model a number of dummy variables which is equal to the total number of predictors minus 1, making the absent variable the reference category?
  2. How do I interpret the results, given presence of the reference category?
  3. For interactions, do I multiply the dummy variables (categorical level) by the moderator (interval level, measured on a 1-5 Likert scale)?

Thanks for your time and any ideas are much appreciated.

Kind regards Muzi

  • $\begingroup$ What is the nature of your response variable (dependent variable)? Is it a Likert scale, or a continuous variable or what? $\endgroup$ – gung Jun 12 '12 at 3:10
  • $\begingroup$ The dependent variable was measured on a 1-5 Likert Scale, Thanks $\endgroup$ – Muzi Jun 12 '12 at 20:47
  • $\begingroup$ What software package are you using? $\endgroup$ – Michael Bishop Aug 31 '12 at 2:32

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.