We use Model 3 to examine our hypotheses, but we do not know how can we code our moderator variable in SPSS, because the moderator is categorical (group norm: equity vs. favoritism) and when we code them as 0 and 1 results differ from another coding (e.g. -1 & 1).
We searched that which one true is and which coding system must used be, but unfortunetaly we could not find the solution. Especially, we read Hayes' book ( Introduction to Mediation, Moderation, and Conditional Process Analysis) and Aiken & West's book (Multiple Regression: Testing and Interpreting Interactions). We analyze coding systems like dummy coding, effect coding etc. in regression models but still we could not be sure.
When we use PROCESS Model 3, if our moderator variable is categorical, how should we code moderator variable? Is there any suggestion on this topic?
Thank you for answers.
Sami.