Has anyone used semPLS package in R? I am using this package to run a structural equation model (SEM) as my data showed multivariate non-normality, and Partial Least Square SEM is suggested. I was able to run the model but it only reports regression coefficients. How do we extract p-values and effect sizes of the coefficients?
Dear Ramesh,
it is a while ago that I worked with semPLS, but there is a functioon bootsempls which you can use for inferences.
Together with some colleagues we develop a R package called "cSEM" where PLS-PM in its current form is implemented:
https://github.com/M-E-Rademaker/cSEM
If you have questions about this package, feel free to contact me.
Moreover, there is the matrixpls which can be used and is particularly helpful when you want to conduct a simulation study.
HTH
Best regards,
Florian
Hi Ramesh,
I din't work with semPLS, but I used the "plspm" package before, a package developed by Gaston Sanchez. It is very useful, and you have all the details regarding the analysis (including inference) in this manual:
https://www.gastonsanchez.com/PLS_Path_Modeling_with_R.pdf
An alternative for you could be to try another software, Warp PLS, you have three months of free trial. The software is easy to use (although the econometrics behind it is very strong) and has a very friendly interface. The only restriction is that it works under Windows.
Dear all,
the issue of the older PLS R packages is that PLSc is not implemented and therefore you cannot consistently estimate common factor models. Moreover none of the packages except cSEM can consistently estimate higher-order models, non-linear factor models and deal with ordinal indicators. I am not sure about testing the overall model fit or MGA which is also possible in cSEM.
Best regards,
Florian
Dear Jian,
lavaan cannot be used to estimate a model PLS.
Best regards,
Flo
Thank you all for your helpful comments. I am trying to run a couple of suggested options and will see which one bet fits my data. Thanks Florian Schuberth Theo K. Dijkstra Elena Druica
Dear Ramesh Paudyal ,
what option did you use in the end? I am facing the same problem: I need to extract p-Values from my semPLS model and am not sure how to proceed.
Best regards,
Lisa
Hi Lisa Birkenbach, I finally went with the SmartPLS software. I found it very user-friendly.
Yes, Ramesh. Smart PLS is very user-friendly. In fact there are also many high quality journals available for its applications and updates.
All the best!
Regards
mei peng
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression