Multiple Moderators and Control variables

Frequently asked questions about PLS path modeling.
Post Reply
Alexg
PLS Junior User
Posts: 3
Joined: Fri May 18, 2018 11:34 am
Real name and title: Mr. Alessandro Grillo

Multiple Moderators and Control variables

Post by Alexg » Fri May 18, 2018 12:10 pm

Hello everyone,

I am running a quite simple model with two exogenous variables (A and B) pointing both at two endogenous variables (C and D).

A--->C
A--->D
B--->C
B--->D

However, I have three moderators all reflective and each of them is measured through several items on 7 points likert scale. They should affect all the relationships of the model. The main aim is to reveal the significance of the moderating effect. Hence as suggested in the book Hair et al. (2017) I am using a two stage approach.

I also read in the 2013 Editorial "Partial Least Squares Structural Equation Modelling: Rigorous Applications, Better Results and Higher Acceptance" that if there is theoretical support it would be possible to test one moderator at time. However, something is not clear to me and I could not find any answer to the below questions.

1) Since my moderators affect all the relationships in the model, Should I test my moderators one at the time and per each single relationship or for all relationships of the model? Meaning, if my Moderator is M should I include the interaction terms (MxA; MxB) for all the endogenous variables (C and D) at once?

or

2) Would be better to test each single moderator separately for each endogenous variable and relationship? So test M on the relationship between A--->C and then test M on the relationship between A--->D and so forth?

3) I have 5 control variables for my endogenous variables. Should I Include also those in the analysis or better to run the moderation test for significance without control variables and then include them?

I would like to thank you in advance for taking the time to read this!!

Your help will mean a lot to me.

Thank you!
Ale

jmbecker
SmartPLS Developer
Posts: 820
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Multiple Moderators and Control variables

Post by jmbecker » Fri May 18, 2018 2:27 pm

You should have the moderator on both paths at the same time. As they influence different variables they should also not interfere much, but it is simply more correct to do so.

You should also always include your controls. Or you may compare the two models with and without controls, but not accounting for the controls in the moderator analysis would be wrong if you think you need controls.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

Alexg
PLS Junior User
Posts: 3
Joined: Fri May 18, 2018 11:34 am
Real name and title: Mr. Alessandro Grillo

Re: Multiple Moderators and Control variables

Post by Alexg » Fri May 18, 2018 6:19 pm

Hello,

thank you very much for your clarifications!!!

Really helpful!

Post Reply