Using PROCESS macro vs. logistic regression

#1
Hi all,
I am trying to investigate a moderation effect of a binary categorical moderator variable on the effect of another independent variable (categorical, binary) on a categorical binary outcome variable. With binary categorical I mean that the variables can have basically two values (0 and 1, an event happens/does not happen).

I understood from the website that the PROCESS tool for SPSS can be used? I plugged it in and it showed results of a binary logistic regression but I am unsure if this is correct? I have heard other opinions that PROCESS cannot be used for categorical outcomes like mine? However, Hayes (2012) state: "Perhaps the most obvious limitation of PROCESS is that it is restricted to the analysis of dependent variables (Y) that are properly modeled with OLS or logistic regression, meaning outcomes that are continuous (or at least approximately so or treated as such) or binary. It has no procedures built in for properly modeling categorical mediators or multicategorical outcomes."

Would it be correct to use PROCESS or should it be a logistic regression with interaction effect?

Thank you very much!
 

hlsmith

Not a robit
#2
If you are investigating "moderation" and not "mediation", I would not see why just examining the multiplicative interaction term and plotted results from logistic regression wouldn't suffice.
 
#3
Ok, thank you for your answer. I was not sure if it yields the same results as in the PROCESS tool that officially measures moderation effects. Do you have any info about that?
 

hlsmith

Not a robit
#4
Not really familiar with the PROCESS Macro, I thought it was just for mediation and mediated moderation, but I am sure it has advanced. I would recommend doing both, running logistic regression and using PROCESS to confirm you know what is being calculated. Two variable moderation is fairly simple to examine on your own.