How to interpret moderation results

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
thenotorious
PLS Junior User
Posts: 5
Joined: Wed Jan 02, 2019 4:13 pm
Real name and title: Aidin Yavari, Student

How to interpret moderation results

Post by thenotorious »

Hi everyone,

I have a model with the following variables:
- Independent Variable: Transparency
- Mediators: Trust & Reputation
- Moderators: Awareness
- Dependent Variables: Purchase intention & Brand attitude

Following the literature of Hair et al. (2016) I have assessed my model with the advice proposed and everything fits accordingly. However, I am not sure how to interpret some aspects of moderation in my model. When looking for R squared and the path coefficients with its p-values, should I also look at the relationship between ''awareness-->reputation'' for example? Or solely at ''Moderating effect 1--> reputation'' for example?

In other words, should I ignore ''awareness-->corresponding variable'' in the table or not? In my mind it's not logical to look at that because I am only interested in the moderated effect I created in SMARTPLS.

I have attached a screenshot of my model.

Thank you in advance.

Image
thenotorious
PLS Junior User
Posts: 5
Joined: Wed Jan 02, 2019 4:13 pm
Real name and title: Aidin Yavari, Student

Re: How to interpret moderation results

Post by thenotorious »

Anyone by any chance has an idea? I understand that the above mentioned phenomenon is called the interaction term but in literature suggested here on the forum, I cannot find any information on how to interpret the final result with the significance.
agalvez
PLS Expert User
Posts: 39
Joined: Mon Jul 04, 2016 10:17 pm
Real name and title: Alex

Re: How to interpret moderation results

Post by agalvez »

As far as I know, the structural path from the Moderating effect and the dependent variable is not significant (t < 1,65). Therefore, there is no moderating effect. However, wait for some expert to reply. Bests.
thenotorious
PLS Junior User
Posts: 5
Joined: Wed Jan 02, 2019 4:13 pm
Real name and title: Aidin Yavari, Student

Re: How to interpret moderation results

Post by thenotorious »

agalvez wrote: Thu Jan 03, 2019 6:10 pm As far as I know, the structural path from the Moderating effect and the dependent variable is not significant (t < 1,65). Therefore, there is no moderating effect. However, wait for some expert to reply. Bests.
Thanks for your response. I am aware that the path is not significant. Maybe I should rephrase my question: When looking at the path coefficients I see the relationship between the ''moderating effect'' and let's say the ''mediator''. However, I also see the relationship between my moderator variable (which is awareness here) and the mediator. What should I do with that relationship since I am only interested in the moderating effect, not on the direct relationship between my moderator variable and my other variables..
agalvez
PLS Expert User
Posts: 39
Joined: Mon Jul 04, 2016 10:17 pm
Real name and title: Alex

Re: How to interpret moderation results

Post by agalvez »

This post may help you

viewtopic.php?t=3965
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: How to interpret moderation results

Post by jmbecker »

I would have all the effects in the table and then I would say something like: "...while the moderator Awareness has a direct significant effect on Trustand Reputation both interaction terms are not significant, hence there is no moderation of the effect of Transparency on Trust or Reputation by Awareness."
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Post Reply