Categorical exogenous variable

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
PLS Junior User
Posts: 1
Joined: Tue Nov 08, 2022 10:42 pm
Real name and title: Phil Be

Categorical exogenous variable

Post by beckily »

Dear all,

I am currently working on my master thesis.
Before I go ahead with my final survey, I would like to make myself familiar with SmartPLS as I have never used this software before.

My model has one main exogenous variable, two assumed mediators, one dependent variable after the mediators and a subsequent DV after the first.
Furthermore, the two mediation paths are assumed to be moderated (by binary categorical variables).
I also assume a direct effect of the IV on the first DV.

Now, my question is: Can I calculate this model in SmartPLS as a SEM?
My exogenous variable reflects the three experimantel conditions and will therefore by reflecting three independent group levels.
I understand, that they would need to be transformed into dummy variables accordingly, right?
But how can I measure the effects? And does it matter which group is assigned as the reference group or can I simply switch it around to analyze all effects?

My other option would by to employ a measurement to capture my DV in a non categorical way, however this would not be an ideal representation of my research topic.

I would be delighted to learn from the knowledgable people in this forum.

Kind regards

PS: Please find a draft of my proposed model attached.
Post Reply