forum.smartpls.com

Hi,

I am working on hypothesized structural model with same measurements but different models (Single order vs. higher order).

My fundamental question is that would it be possible to have two models with same measurements so that I evaluate both? First, overall relationship between higher order construct and the endogenous variables. Second, evaluate the detailed relationship between each dimension (lower order construct) and the endogenous variables.

However, in this case, which path coefficients should be more valuable such as path of lower order construct -> endogenous variable or path of lower order construct -> higher order construct -> endogenous variables?

For your reference, attached please see the hypothesized models.

Thanks

Jeong

I think you should choose one of them. Conceptually you either argue for a higher-order construct or not. You need to decide which model better represents your theory and the underlying data generating (causal) process. Both models will yield different results.
You are also able to assess the impact of each dimension on the final endogenous variable using the total effects of the LOC in a higher-order model.