Mediation or moderation analyses may be executed separately. However many models require the combined integrated effect to be assessed. Which software environment is preferable (ie yield the best results). PLS-SEM ? Covariance-Based SEM ? standard and hierarchical Multiple Regression applications (eg SPSS) ? any other ? The domain of interest is the work organisation: including issues of individual innovativeness, self-efficacy, leadership, engagement, commitment, empowerment, pro-activity, trust, support etc
PLS-SEM and CB-SEM is a different thing. You have to ensure what is the main purpose in your study. PLS-SEM is used for exploratory purpose (theory development) and CB-SEM is used for confirmatory study (Theory-driven).
Hi Silburn,
I think you may find the attached article helpful. I compare and contrast several methods, using both simulation and empirical data, across many different data types.
This should be a helpful guide for you, especially since I use some of these constructs as examples of mediators in practice
Ariel
Article Using mediation analysis to identify causal mechanisms in di...
I prefer Amos due to its simplicity, user friendly, and robust algorithm.
I prefer Smartpls to assess mediation and moderation by PLS SEM (it is even more symbol, user friendly and robust than Amos and more flexible in modeling). However, I agree with colleague that the major concern may be your research purpose and types of data.
Thanks Wan and Mark
The remarks on PLS-SEM and CB-SEM are appreciated
To clarify, the objective of the study is for theory testing; particularly Self Determination Theory
Silburn
Dear Zainudin,
Thanks for the feedback on AMOS and also on the appropriate context for the use of SEM.
Another great paper on the topic of appropriate use of SEM , particularly PLS, is from Lowry & Gaskin's IEEE paper "Lowry Gaskin 2014 Partial Least Squares (PLS) Structural Equation Modeling (SEM) for Building and Testing Behavioral Causal Theory"
http://bit.ly/1LZKGNe
Silburn
Dear Ariel,
your fascinating article pushed the envelope on mediation analysis. truly enjoyed reading and gained several insights
Silburn
The recent literature on integrated mediation and moderation models (mediated moderation and/or moderated mediation) also has strong influences from the work of Prof Andrew Hayes in Conditional Process Analysis which now appears in several journal articles. The PROCESS macro is an add-in for SPSS and SAS for executing mediated moderation and/or moderated mediation. The references below:
http://www.processmacro.org/index.html
https://scholar.google.com/scholar?cites=2148876485119216290&as_sdt=5,36&sciodt=0,36&hl=en
http://www.afhayes.com/
http://www.ats.ucla.edu/stat/spss/seminars/process_macro/process_training.htm
Silburn
Hi Silburn,
Both PLS-SEM and Covariance-based SEM can do moderation / mediation. But based on your remark "the objective of the study is for theory testing; particularly Self Determination Theory", sounds like Covariance-based SEM like AMOS can be suggested as you try to test / confirm the existed / existing theory. As what Wan had shared, there are differences between the two SEMs as per RG link below:
https://www.researchgate.net/post/Which_would_be_the_best_tools_to_measure_the_relationship_of_five_constructs_in_a_corporate_entrepreneurship_conceptual_framework_SEM-CVB_or_SEM-PLS
Regards,
Fung
Thanks for sharing Han,
In the context of the CB-SEM vs PLS-SEM trade-off, the MISQ Editor's "An Update and Extension to SEM Guidelines for Administrative and Social Science Research" also provides further guidance on how to ensure rigour when using SEM.
http://www.misq.org/skin/frontend/default/misq/pdf/appendices/2011/EdCommentsV35N2Appendices.pdf
Silburn
Hi Silburn, Prof Andrew Hayes's macros are his marvelous contribution in the field of testing moderation and mediation using SPSS, and you can perform your analysis by following Prof Andrew Hayes approach, I agree with you.
Silburn,
Thank you for your kind words. I am glad that you enjoyed my paper.
Ariel
I also suggest Prof. Andrew Hayes's macros, for measuring moderator and mediator variables. You can find on youtube. .
https://www.youtube.com/watch?v=RqkGMqDU20Q
Find the attached file for your satisfaction
Article Making Sense of Mediating Analysis: A Marketing Perspective
- Badges
- Science topic