Hi all,
In my model with three reflective exogenous variables and three endogenous formative variables, for one variable I am getting an R-square of 1.00. I don't know how to interpret that. Any suggestions? Can it be due to collinearity?
-Sumit
R-square = 1.00
-
Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi Sumit,
I already saw this happen when we have 2nd order LV formative as endogenous LV.
See viewtopic.php?t=719&highlight=2nd+order+formative
Best regards.
Bido
I already saw this happen when we have 2nd order LV formative as endogenous LV.
See viewtopic.php?t=719&highlight=2nd+order+formative
Best regards.
Bido
Thanks, but in that case, r-square for all endogenous formative variables should be 1, IF the scale is complete. An r-square less than one means that some aspect of the formative construct is missing from the scale. is there anyway we can get the residuals directly in SmartPLS to analyze them?
I also wanted to ask what is the optimum number of iterations that PLS algorithm should be allowed. I can see stark differences in loadings when the model was allowed 300 iterations vis-a-vis 3000 iterations.
I also wanted to ask what is the optimum number of iterations that PLS algorithm should be allowed. I can see stark differences in loadings when the model was allowed 300 iterations vis-a-vis 3000 iterations.