forum.smartpls.com

The direct effects are those results from the Path Coefficients table. You also use them for the VAF and not the total effect results.

That depends on you. If both dependent variables are equally important then averaging their weighted averages might make sense, but I would also prefer to see the results on both and then do the calculations so that differences become more obvious.

Based on your example, the weighted R² are 0.08 0.46 0.21 = (798*0.037+202*0.909)/(798+202) And the segmention solution do not look so different. PLS-POs seems to find a negative coefficient in the one segment and a positive in the other. The order of segments (seg1 vs. seg2) does not matter and can...

First, as we explain in the paper you have to repeatedly start the PLS-POS (as you have to do with most segmentation algorithms) because they may end in local optima. I would suggest at least 10 repetitions. Assessing the average weighted R² (weighted with sample size) is probably the best criterion...

If IV2 (the mediator) has only a very small f-sq it is not nessesarily bad. It only means that it does not explain additional variance over the direct effect of IV1. It can still be a good mediator (in fact, if it is a full mediation, then it should explain to much additional variance, because it on...

Yes, that is the preffered option.

The connection will most likely have an impact, but that is not bad. Without the moderator the interaction effect would simply be wrong. So you should include the moderator in your analysis! If you get a significant interaction without the moderator but not with the moderator, then the interaction i...

Yes, that sometimes happens with the PLSc Bootstrapping. Until now, there is no established way to treat these cases. It happens when part of the correction value gets negative and the square root is taken of these negative values. It is a bit similar to Heywood cases in CB-SEM where you can get neg...

I don't know what exactly you want to do that is not possible with SmartPLS and requires the correlation matrix of the LOC. I was just pointing at the reports where you can find what you were asking for.

There is a small bug with the Basic Bootstrapping option. Please choose the Complete Bootstrapping Option in the Bootstrapping dialog. Then you should not get that error. It will be fixed in the next release, but until then you have to use Complete Bootstrapping Results. EDIT: This problem is solved...

You must include the moderator in your model. If you do not include the moderator your interaction effect will be biased and not meaningfull. There is no way that a model without the moderator makes sense.
There is a bunch of literature on this topic.

You can use the latent variable scores and plug them into any other software that you might want to use, but in terms of correlations we also provide the correlation matrix of the latent variables. You will find these things in Smart 3 when you click on "Latent Variables" in the PLS results.

Hi Sussmek, Yes, the interface is a little bit different from Version 2. I heard that they are actually updating the book to the new version. However, mostly differences are not large. With regard to your question: In SmartPLS 3 the data is standardized by default (i.e., the mean 0, var1 option in S...

SmartPLS does not know whether your latent variable is a 2nd-order construct or a normal construct. Hence, the AVE of a 2nd-order construct is based on all repeated indicator variables. This AVE should not be used. Instead, as you have mentioned, you should use the path coefficients from the reflect...

These nice Histograms are now part of SmartPLS Version 3.2.0. You should use at least use 1,000 bootstapps to get a good resolution. (Examples here use 10,000 resamples).
In addition, we also fixed the problem with sign change options and indirect and total effects.