In this tutorial I demonstrate how to test causal models in PLS-graph. This is done using bootstrapping to produce t-statistics, which can then be used to calculate p-values.
As much as I think your PLS Graph videos are of great value for researchers new to PLS, you must take into account that you can't actually infer causality from PLS analysis, just model predictability. Bad as it is, it's what we've got :(
And I wouldn't take an R-square value of 0.23 as actually good, (except in very eary and exploratory research stages)
I also wouldn't just delete the non-significant path in my analysis. If it's non-significant, it just is ;)
As much as I think your PLS Graph videos are of great value for researchers new to PLS, you must take into account that you can't actually infer causality from PLS analysis, just model predictability. Bad as it is, it's what we've got :(
And I wouldn't take an R-square value of 0.23 as actually good, (except in very eary and exploratory research stages)
I also wouldn't just delete the non-significant path in my analysis. If it's non-significant, it just is ;)
Good job with the vids, anyway!
mastegas 7 months ago