Hi S.PLS experts,
I need to run a model that has sequential mediators. I am wondering whether it is possible to run this model on smartpls.
If so, can anyone suggest me some reading materials (i.e., how to run a model on smartpls, how to interpret..)?
Thank you very much for your help!
Sequential mediation
- cringle
- SmartPLS Developer
- Posts: 818
- Joined: Tue Sep 20, 2005 9:13 am
- Real name and title: Prof. Dr. Christian M. Ringle
- Location: Hamburg (Germany)
- Contact:
Re: Sequential mediation
Yes, it is. The SmartPLS provides the indirect effects that you need to conduct the mediations analysis as explained in the bool by Hair et al. (2017): https://www.smartpls.com/documentation/ ... s-sem-book
Best
CR
Best
CR
Prof. Dr. Christian M. Ringle, Hamburg University of Technology (TUHH), SmartPLS
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de
-
- PLS Junior User
- Posts: 3
- Joined: Sun Nov 05, 2017 12:38 am
- Real name and title: Anuragini Shirish, Assistant Professor
Re: Sequential mediation
I had a question regarding sequential mediation. In my case I have a 2 models A-B-C-D AND A1-B1-C1-D1. B and C mediate A and D and B1 and C1 mediate A1 and D1. As per the PLS3 results the find that path c-d is the opposite sign to the path AD, therefore I understand based on latest readings that it can be a case of competitive partial mediation. In the 2nd model it is a complementary partial mediation with no significant change noted. However, I need to know if evaluating CI is also important in reporting the results or not. Because for some relationships the CI for the competitive mediation included zero but have significant p and t values.
I am not sure what needs to be reported the direct and indirect effect alone as reported in PLS3 along with R2 and Q2 or do we need to also verify the CI of these effects before reporting?
Thanks for any clarifications.
I am not sure what needs to be reported the direct and indirect effect alone as reported in PLS3 along with R2 and Q2 or do we need to also verify the CI of these effects before reporting?
Thanks for any clarifications.
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: Sequential mediation
You should also evaluate the significance of the effect using Confidence intervals or P-Values.
Confidence intervals are usually the better approach, but usually they also agree with the p-values if the parameter distribution is not vary non-normal and the effect on the edge of being significant.
Confidence intervals are usually the better approach, but usually they also agree with the p-values if the parameter distribution is not vary non-normal and the effect on the edge of being significant.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
-
- PLS Junior User
- Posts: 3
- Joined: Sun Nov 05, 2017 12:38 am
- Real name and title: Anuragini Shirish, Assistant Professor
Re: Sequential mediation
Thank you for your response. P values are significant but CI is not. Moreover, I am not sure how to change the parameter in the PLS3 software which displays only CI for 97.5 percentage. I would like to see if 95 percent confidence interval will vary the results. How do I get the nonvariation on the edge? You advise is appreciated
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: Sequential mediation
If you have a CI with 97.5% and 2.5% bounds you have a two-tailed 95% confidence interval. This is the default setting in the bootstrapping options (i.e., 5% error level and two-tailed).
However, you can also specifiy one-tailed or other error levels in the settings and thereby get other confidence bounds.
However, you can also specifiy one-tailed or other error levels in the settings and thereby get other confidence bounds.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de