forum.smartpls.com

Can you merge the problematic constructs?

At the end of this article, you find some recommendations (flow chart): http://link.springer.com/article/10.100 ... 014-0403-8

You find a chapter on moderation in PLS-SEM and the results interpretation here: https://www.smartpls.com/documentation/ ... s-sem-book

The book usually is available in university libraries.

In that case you may need an additional formative indicator that helps to better cover the domain of formative construct.

Avoid the circle path and/or use the latent variables scores from a model without a circular relationship and estimate the structural model with the circular path by using 2SLS (e.g. in the R program).

I would use the actual age as a single continuous moderator. Thereby, you do not lose any information. Alternatively, compare the third of the youngest against the third of the oldest (leave the middle third out) in a multigroup analysis.

Please take a look at this article which explains how to specify, estimate, and validate higher-order constructs in PLS-SEM: https://www.sciencedirect.com/science/a ... 8219301223

Insignficant total effects and negative ones (when you expect positive ones) are not reasonable for running an IPMA.

That's usually the case when there are some string values in your dataset, where SmartPLS expects numerical values. Please inspect your dataset, again, and take a look at these instructions: https://www.smartpls.com/documentation/ ... ata-import

There is no established way to conduct the marker variable approach in PLS-SEM. Sometimes you see that all items become single item constructs and their constructs second-order constructs. Thereby, you can also create a relationship from the marker variable to every single indicator (besides the rel...

You may not use PLSc (PLSc, GoF and higher-order models may not fit well; in that combination, results are not available). If you really want to mimmic factor-models by using PLSc, as a work-araound, you may use two-stage approach for higher-order constructs.

We did not follow-up on this direction. As as work around, you may extract the unstandardized latent variable scores from the IPMA results report and run a "standard multilevel analysis" for the partial regression models.

Thanks. Use the one or the other approach but with 5,000 bootstrap samples. However, your upper bound is above 1 which means that discriminant validity has not been established. Your confidence intervals are very wide. I suppose that you use a small sample (which would be a data problem) and/or a fu...

You would like to do a one-sided test (or use a 10% probability of error with the 2-sided test) to obtain the upper bound at the 95% point (and not at 97.5%.

You may exclude the potential explanatory variables form the model before running FIMIX-PLS. @FIMIX-PLS and segmentation: The book on advanced PLS-SEM issues by Hair et al. (2018) has a chapter on this topic: https://www.smartpls.com/documentation/learn-pls-sem-and-smartpls/book-on-advanced-pls-sem-...

Thanks. To get started on second-order models, you may want to take a look at the respective chapter in the book advanced PLS-SEM issues by Hair et al. (2018): https://www.smartpls.com/documentation/learn-pls-sem-and-smartpls/book-on-advanced-pls-sem-issues "To begin with, in my Ph.D. thesis model, ...