Hi all,
At the moment I am setting up a model in SmartPLS and I have a couple of questions that I can’t seem to solve myself. I have to say I am new to the program.
The model consists of six latent variables (LV). The first two are independent LVs (called A and B) drawing paths to the next three LVs (called C, D and E) which are all connected to the last LV (called F).
1. My first question regards the R squared values related to each of the dependent LVs C, D and E. I am experiencing weak R squared values (0,261 – 0,208 – 0,167) which I think means that A and B explains only little of the variance in C, D and E. If this is correct, would it then in general be fair to validate the model, despite the low R square values of A and B, given that my hypotheses are testing the effects of A and B on C, D, E and F?
2. Moreover, how do I calculate/find in the report a measure of the fit of the model, say R squared of the model and not only the singles LVs?
3. My last question concerns the path coefficients, inner model coefficients. How do I interpret the path coefficients? Are these allowed to be negative?
I know these questions are very beginner-like, but I hope you will take 2 minutes to help out.
Thanks in advance!
Kind regards,
Sune