what is the practical explanation for partial and full mediation?
For example, if x→Z→Y. if Z fully mediate the effect of x on y what is the practical explanation, and if Z partially mediate, what is the practical interpretation.
Regards,
Fazli Haleem
Assuming a causal model. . .
Full mediation:
X causes Y because of Z
Partial mediation:
X causes Y because of Z and some other reason.
Indeed, full mediation would imply that the mediator fully explains the association between x and y. Of course, the model needs to be based on a solid hypothesis based on previous findings or expected pathways.
Dear Haleem,
Kindly find a comprehensive recent guide on mediating analysis, you can easily understand partial and full mediation concept
Article Making Sense of Mediating Analysis: A Marketing Perspective
Practical explanation of full and partial mediation is explained in my blog post: https://t.co/kKFU8hD4Oj
Also, Series of post regarding confusing statistical terms like mediation, interaction, moderation, etc. can be found on http://learnerworld.tumblr.com/
Dear Haleem,
Partial Mediation: If IV has both direct and indirect effect on DV
Full Mediation: If Iv has only indirect effect on DV
Extract from: A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) ISBN: 978145221744417
VAF = indirect effect / total effect * 100
Total effect = indirect effect + direct effect
If the indirect effect is significant but does not absorb any of the
exogenous latent variable's effect on the endogenous variable, the VAF is rather low. This occurs when the direct effect is high and declines only very slightly after a mediator variable with a significant but very small indirect effect is included. In this situation, the VAF would be less than 20%, and one can conclude that (almost) no mediation takes place. In contrast, when the VAF has very large outcomes of above 80%, one can assume a full mediation. A situation in which the VAF is larger than 20% and less than 80% can be characterized as partial mediation. A suppressor effect, which characterizes the sign change of the direct relationship after the mediator variables have been included, is an exception to the VAF-based assessment of mediating effects. For example, when a significant positive path relationship without the mediator variable becomes significantly negative after the inclusion of the mediator variable, a suppressor effect occurs. The VAF becomes larger than one or, in some instances, even negative with suppressor effects and can no longer be interpreted. This kind of situation always represents full mediation.
- Badges
- Science topic
- Similar topics
- Psychology