JASP Tutorial: PROCESS Module -- Complex Mediation & Moderation (Episode 61)

In this JASP video, I preview the new BETA Module of PROCESS, a full-feature mediation and moderation package that got its early roots in software like SPSS, as add-on syntax. Now you can run any kind of general linear regression model with predictors, mediators, and moderators.
In the video, I describe the process (heh) of completing a simple mediation and a simple moderation. More complex models are up to you to explore!
**The module is in beta, so expect issues and errors!**
Be sure to ask your questions, and drop your comments, suggestions, and feedback in the comment section!
JASP: jasp-stats.org
NOTE: This tutorial uses the new preview/beta build of 0.18. This build contains slightly more functions/features than the previous builds used for tutorials on this channel, but it is functionally the same for the purposes of this tutorial.
Socials & Sites: linktr.ee/alexbswan
Discuss this video and others on my Discord channel: / discord

[@riya_r28]

Precisely what I was searching for. Thank Prof. Swan!
Is it possible to test if a variable is a mediator or moderator? For instance, you have shown that anxiety is a mediator. Can we test if anxiety acts as a moderator?

[@AlexanderSwan]

I would caution against this - mediators are third variables and moderators are interactions. These complex regressions are designed to be backed by logic and theory, so you need to have that in place BEFORE you run a model. One should not go back and forth just to see. Further, it really depends on how you’re measuring the variables. In your example, anxiety would not really work as a moderator because of how it is usually measured .

[@jonrrobinson]

Thanks for another great video! Really informative. It would be good to know how this analysis should be reported (although I am sure you likely have a video planned on this). Thanks again 👍👍

[@AlexanderSwan]

Nothing planned on that, but it’s a good idea - I can work off a paper I have where we used PROCESS. It’s not as bad as you might think, tbh, and not much different from regular APA results section for regression models.

[@italiancannonli2490]

If you selected "model number" when "input type for model 1", you should input the correct Hayes model number (which you can find from the model templates in Hayes' handbook). For example, model 1 is simple mediation (here the number "1" means the first model template), and model 4 is simple moderation.

[@AlexanderSwan]

Yeah that becomes clearer if you look up the original Hayes handbook for PROCESS. I think it could be clearer in the JASP module! Thanks for the clarification!

[@italiancannonli2490]

@@AlexanderSwan I agree! The current process (beta) module in JASP should be clearer, cause it makes people confuse model number and (model) template number. Hopefully it will be fixed in the next release.

[@ericjanwagenmakers]

Thanks for the video and suggestions for improvement/clarification. We are taking this into account!