Page 1 of 1

### How can I interpret my PLS-MIMIC model?

Posted: Sun Feb 10, 2019 9:40 pm
I am a PhD student and I am studying PLS-MIMIC model. I have just one latent and two construct, one of them is formative and the other one is reflective. I have 10 indicators for formative construct, and 3 causes for reclective one . I am using SmartPLS 3. I understand I have just direct effect. Both indicators and causes have their loadings so how can I specify the relationship between causes and indicators?

### Re: How can I interpret my PLS-MIMIC model?

Posted: Mon Feb 11, 2019 2:28 pm
I don't quite understand the problem, but you might look into the PLS SEM book for redundancy analysis and also the following paper might help you:
Cheah, J. H., Sarstedt, M., Ringle, C. M., Ramayah, T., & Ting, H. (2018). Convergent validity assessment of formatively measured constructs in PLS-SEM: on using single-item versus multi-item measures in redundancy analyses. International Journal of Contemporary Hospitality Management, 30(11), 3192-3210.

### Re: How can I interpret my PLS-MIMIC model?

Posted: Tue Feb 12, 2019 6:34 am
I attached my model. I would lile to know that how I can know the relationships between eee and fff or ddd and hhh? Is it possible multiply their loadings of them to understand their(indicators) effects on causes)?

### Re: How can I interpret my PLS-MIMIC model?

Posted: Wed Feb 13, 2019 12:18 pm
Yes, you could multiply outer_weight*path_coefficient*outer_loading, e.g., 0.866*-0.371*0.833 = -0.268 for the effect of eee on fff in a sence of a total effect that matches partial regression coefficient logic.
Another option would be the use of outer_loading*path_coefficient*outer_loading, which would match a canonical correlation. You could find that also under Indicator Data (Correlations) > Model Implied Estimated Correlation Matrix