Categorical Variable(With more than two categories) as Moderator in MGA PLS-SEM

[Neeraj Kaushik]

Dear Friends

In the last week, at least 3 scholars asked me how to take a categorical variable (with more than two categories) in Multi-Group Analysis (MGA) in Structural Equation Modeling (SEM).

Incidentally, in the last week, I read the paper Multigroup analysis of more than two groups in PLS-SEM: A review, illustration, and recommendations by Cheah Jun Hwa, Suzanne Amaro, and José L. Roldán in Journal of Business Research 156(4):1-19 DOI: 10.1016/j.jbusres.2022.113539 December 2022 

Available on https://www.sciencedirect.com/science/article/pii/S0148296322010049 

So far we can run Multi-group analysis (MGA) with 2 categories in AMOS, SmartPLS, and Adanco. But for a variable with more than two categories, we can use Non-parametric Distance Test (NDT) which is available only in the R-package cSEM.

I've explained this in these 2 videos. The first video on the mandatory requirement of MICOM for the MGA is explained in this video:

R-Studio Part-25.5 Why Invariance test: https://youtu.be/yYuBFf6Jm94

Happy Learning

Neeraj

[Neeraj Kaushik]

Dear Friends

In this second video, I've explained the working of MICOM and MGA using cSEM R-package.

CCA-9.2 Multi Group Analysis More than 2 categories in cSEM: https://youtu.be/2dkqfIg2mso

Happy Learning

Neeraj