I am trying to compute a PlS-SEM model using seminr, but it doesn't seem to work well: I can't even show the summary of the model. Is it because the structural model isn't well defined? These are the path coefficients in my model:
| Depression | Anxiety | WB | |
|---|---|---|---|
| Depression | 0.00 | 0.59715665 | -0.72310198 |
| Anxiety | 0.2889459 | 0.00 | -0.03600999 |
| WB | -0.43997596 | -0.04427055 | 0.00 |
I have another two variables in my model.
Also, it seems impossible to bootstrap this model.
When I simplify the model and remove some of the paths, it works. Is it because the SEM theory? I would like to create a explorative model so my aim is to relate all the variables between them. Is this possible using SEM? Does the matrix need to be triangular?