Recently I read an article by Rungtusanatham, Miller, & Boyer (2014) entitled Theorizing, testing, and concluding for mediation in SCM research: Tutorial and procedural recommendations.
In this paper, I came across recommendations for theorizing and testing mediation models. As recommended, when theorizing for the mediation process, it is not necessary to hypothesize direct effect except when when "other mediators, besides M1, may theoretically or empirically explain the total effect of X on Y or when the direct effect of X on Y may have a sign that is opposite of the sign for a specific indirect effect". However, it is recommended that when testing for mediation processes " Include the direct effect of X on Y when testing for mediation effects".
I have also read 'PLS-SEM using SmartPLS 3.0: Chapter 12: Assessment of Mediation Analysis by Ramayah et al. (2017).
I wish to know if it is necessary to include a path showing the relationship between X and Y even when we do not intend to measure it? What if we want to investigate only the indirect effect?
Yes it is required, in order to check the existence of partial mediation or complete mediation or no mediation.
You must include the direct effect to ensure the sum of the indirect and direct effect equate to the model with only the total effect. Otherwise your estimate of indirect effect is incorrect:
c'=c + ab
Hi Sood,
Nitzl et al. (2016, p.1853) claimed that "[...] researcher should regularly include the direct effect c' in their PLS to control and determine the type of mediating effect."
From my understanding, since significant mediating effects are no longer simply revealed by only looking at the indirect effect, determining the types of the mediations is imperative. To do so, researchers need to add direct relationships (X-Y) for ad hoc mediation-type testing, rather than changing the model for the entire parts of our research. I hope my understanding of Nitzl et al's recommendations is correct and you may find it helpful.
Reference:
Nitzl, C., Roldan, J.L. and Cepeda, G. (2016), "Mediation analysis in partial least squares path modeling: Helping researchers discuss more sophisticated models", Industrial Management & Data Systems, Vol. 116 No. 9, pp. 1849-1864. https://doi.org/10.1108/IMDS-07-2015-0302
Zhao, X., Lynch Jr, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of consumer research, 37(2), 197-206.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size