Building & Running Models
Mediation Analysis
Testing indirect effects through a mediating variable.
Mediation analysis in ResearchPLS is bootstrap-based: it estimates the indirect effect of a predictor on an outcome through one or more mediators, alongside the direct and total effects.
Significance is judged from the bootstrap confidence interval on the indirect effect -- if the interval excludes zero, the indirect effect is considered statistically supported at that confidence level.
Mediation can be combined with moderation (moderated mediation / conditional process analysis) for models where the strength of an indirect effect depends on a third variable.