Role of demographics as moderators in PLS SEM

[Shilpi Sachdeva]

Respected Sir

 In my PLS SEM model, I want to use demographics (age, marital status, income, sources of income) as moderators. I have tried using MGA but while running MICOM Compositional Invariance could not be established. Hence, I cannot move forward with MGA. 

Second, I tried checking the moderator effect individually. My doubt is do i need to check its effect on all the relationships (which are logically supported) and report the results because my IDV'sare 14 in number including one mediator. Out of 14, I can test this relationship on 9 IDVs, as five are BIG 5 PERSONALITY TRAITS. But I am failing to understand how to report this and calculate in the model, as it will give me (9*3=27) combinations. Please suggest how to proceed in this situation, as it will make the model very complicated. One important thing about my model is that it is based on women only (I have collected data from women participants only, as I want to understand women's saving behaviour).

Looking forward to your kind guidance.

Thanks & Regards