Interpretation of MGA and MICOM 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
PLS Junior User
Posts: 2
Joined: Fri Sep 08, 2023 12:44 pm
Real name and title: Kini.S PhD student

Interpretation of MGA and MICOM results

Post by kini »

Dear All,
I have observed heterogeneity in my data. So I separated my data into two data groups based on the categorical moderator. I ran MICOM and I have achieved configural and compositional invariance. However, I do not have equality of composite means and variances. Thus my measurement model shows partial invariance and hence I can proceed to MGA. After this, I checked for MGA and none of the paths show any significant difference across the two groups, establishing structural invariance of my model.
I am confused how to interpret this case. What does it mean to have partial invariance in the measurement model but complete invariance in the structural model? How to proceed further? Any guidance would be much appreciated. Thanks.

Henseler, Jörg, Christian M. Ringle, and Marko Sarstedt. "Testing measurement invariance of composites using partial least squares." International marketing review 33.3 (2016): 405-431.
PLS Junior User
Posts: 1
Joined: Tue Dec 05, 2023 8:46 am
Real name and title: Gabriel Manson

Re: Interpretation of MGA and MICOM results

Post by Nick32 »


When reporting your results, it's crucial to detail fit indices and parameter estimates. Just as in a cover letter where you highlight your qualifications and specific skills, every detail of your analysis reinforces the credibility and clarity of your conclusions. Chi-square difference tests and effect sizes offer valuable insights into the impact and significance of your results.

As far as partial invariance in the measurement model is concerned, identifying non-invariant indicators and testing possible sources of bias is an essential step. This reminds me of the way you adjust a cover letter to suit the target position, by identifying and highlighting the most relevant skills.
How might incorporating a comparative analysis between models improve our understanding of the validity and reliability of the measures in your study? Could this approach provide additional relevant information, similar to the way a well-targeted cover letter highlights a candidate's specific skills?

For more information on cover letters or business letters, click here.
Post Reply