forum.smartpls.com

There are certainly many paper that use PLS to estimate reflective-reflective higher-order models. You might also find some in your field. Whether to use or not to use PLS in a purely reflective model is an ongoing debate. You will find some arguments and literature here: https://www.smartpls.com/do...

It is not the observations that you might want to inspect (i.e., the case id values), but the items (i.e., the indicators in your measurement models). You need to see if there are some exceptional low loadings that negatively influence the AVE.

Dummy variables may cause problems with the bootstrapping procedure if they are unevenly distributed. That is because bootstrapping is a random sampling procedure. It samples with replacement from the original dataset. Hence, there might be subsamples where one of your dummies has only ones or zeros...

It is an indication that there seems to be some form of mediation via your mediators in addition to the direct effect. However, you cannot establish mediation via any of your particular mediators. Individual mediators seem to have only very weak (or no) effects. Currently you cannot tell likely beca...

If you want to show that there is an effect of A on B controlling for C then you should estimate that model including C.
If you want to show that controlling for C does not change the effect of A on B then you need to estimate both models.

Well, there are also CB-SEM estimators that can handle non-normal data, especially if you have very large sample sizes. Depending on the reviewer of your paper you might get problems if that is your only reason to use PLS. Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use...

The central question is: Do you believe that all dimensions are part of a higher-order constructs that fully mediates all effects or not? If yes, you need to include the higher-order construct SQ and assess the effect of each sub-dimension of SQ and then the effect of SQ on satisfaction. However, if...

You need to distinguish between the conceptual (theoretical) orientation of the construct (reflective, formative) and the statistical tool to estimate constructs (composites or common factors). In standard PLS all constructs are estimated using composites. Composites are always determinate and thus ...

When you run the bootstrapping choose "Percentile Bootstrap" in the Advanced Settings for Confidence Interval Method

Of course a descriptive analysis of the path coefficients provides a first indication, but it would be better to really test if they are different. Rodríguez-Entrena, M., Schuberth, F., & Gelhard, C. (2018). Assessing statistical differences between parameters estimates in Partial Least Squares path...

What do you mean with estimating them together? How are they interconnected? Why would you want to estimate them separately?

Generally, such a model should be estimable, but I doubt that you really get meaningful results from such a small sample size. More data is always better, but you should also think about your population and whether your sample really represents that population.

The output "Indicator Data (Standardized)" shows your data (variables) included in the model in standardized form. Of course there will be negative values, because standardization implies mean centering. Hence all value below the mean for each variable are negative.

Based on the framework by Zhao et al. it is competitive mediation. Suppression is more a statistical term than a theoretical phenomenon. The more important question is: Does it make sense? You obviously have a negative direct and total effect, but there are aspects that also work positive and thereb...

The choice between reflective and formative should ideally been made by theoretical considerations and not empirical. If you want to do an empirical test you can do a CTA-PLS analysis. That would work for second-order constructs with the two-stage approach. The nomological network is important for s...