forum.smartpls.com

SmartPLS implements the bootstrapping procedure to test the significance of path coefficients. It is a very established method that is well published and widely used, not only in the PLS context. As far as I know, WarpPLS uses several methods to test the significance of coefficients. Some of them ar...

You need a specific numeric value as missing value. I would suggest something like -999 Actually, your data file should not validate, but at the moment there is a problem with missing values that are really missing (as in your case). We will try to fix that in one of the next releases. Nevertheless,...

First, please open a support request. Second, please provide more information: 1) Which algorithm, 2) What kind of model, 3) When exactly does the error ocurr? If posible you may also send an example dataset and model with your support request or at least a screen shot, so that we can find the probl...

Hi BG, 1) There is no build in test for unequal variances, as there is no established test for a PLS path model. You have to find out for yourself, which test is the right one in your situation, given the advice in the literature. Generally, I would also look if there is consistency between the thre...

SmartPLS 2 does not run on MAC. However, SmartPLS 3 does run on MAC if you download the MAC Version.

The correlations between the indicators of a construct and hence the collinearity within a block will not change if analyzed separately or together. However, the weights do depend to some degree on the nomological network of a formative construct (often discussed as interpretational confounding). So...

There is a small bug in 3.1.5 which makes the signs of indirect and total effect incorrect. We will release 3.1.6. soon where this bug is fixed.

SmartPLS 3 provides information on the VIF for the inner and outer model.

Generally, you should include all IVs and interactions simultaneously. Otherwise you have an omitted variable bias. Sometimes this causes multicollinearity problems so that some researchers test each interaction one by one (but still having all IVs in the model). But this procedure is also suboptima...

Generally, you should include all interactions simultaneously. Sometimes this causes multicollinearity problems so that some researchers test each moderator one by one. But this procedure is suboptimal and should be clearly explained in a paper so that others know what you have done.

The bootstrapp intervals for HTMT are not corrected by the Bonferroni-method. At the moment you have to adjust the p-level to match your model.

You should use the Latent Variable Scores from the results report.

There can be several reasons that such a singular matrix problem occurs during the calculations. Possible reasons could be 1. a variable that has zero variance, i.e. it contains only the same values for each case/observation. This can happen, for example, if one uses a grouping variable in the path ...

Click right on an endogenous latent variable that should be the dependent variable in a moderated relationship. You will have there the option to add a quadratic or interation term.

You can see the number of cases, for example, if you copy your "Indicator Data" into Excel. It will show you the row numbers of each case.