Bayesian Mediation Analysis

By Ying Yuan, David P. MacKinnon. Psychological methods, 2009

613 citations recorded by the source corpus.

This article proposes Bayesian analysis of mediation effects. Compared to conventional frequentist mediation analysis, the Bayesian approach has several advantages. First, it allows researchers to incorporate prior information into the mediation analysis, thus potentially improving the efficiency of estimates. Second, under the Bayesian mediation analysis, inference is straightforward and exact, which makes it appealing for studies with small samples. Third, the Bayesian approach is conceptually simpler for multilevel mediation analysis. Simulation studies and analysis of two datasets are used to illustrate the proposed methods.

Subfields: Computer Networks and Communications, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/773d16f494c5edd967231702830f040344cf15eb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Statistics and Probability

Citations in the corpus

Advantages of Monte Carlo Confidence Intervals for Indirect Effects (2012, methodology): “Finally, in this article we did not consider Bayesian credible intervals for indirect effects, another promising method for obtaining asymmetric intervals for indirect effects (Biesanz et al., 2010; Yuan & MacKinnon, 2009).”
Meaningful Mediation Analysis: Plausible Causal Inference and Informative Communication (2017, background): “Bollen and Stine (1990) first proposed to use bootstrapping, and Yuan and MacKinnon (2009) and Zhang, Wedel, and Pieters (2009) later proposed to use Bayesian estimation to establish the significance of the indirect effect.”
Partial, conditional, and moderated moderated mediation: Quantification, inference, and interpretation (2018, methodology): “Various alternatives exist (such as Monte Carlo CIs or Bayesian credible intervals, see Hayes, 2018; Preacher & Selig, 2012; Yuan & MacKinnon, 2009), but here I recommend the bootstrap CI, as this method is already widely used in mediation analysis for inference about indirect effects (which are…”
Bayesian Estimation and Inference (2015, methodology): “To illustrate using informative priors and examine experimental effects with a Bayesian estimator, data from Zyphur et al. (2007) are re-analyzed with a path model in Mplus.”
Bayesian structural equation modeling: a more flexible representation of substantive theory. (2012, background): “Overview articles of an applied nature and with a latent variable focus include Scheines, Hoijtink, and Boomsma (1999); Rupp, Dey, and Zumbo (2004); Yuan and MacKinnon (2009); and Kaplan and Depaoli (in press).”
Natural soundscapes and tourist loyalty to nature-based tourism destinations: the mediating effect of tourist satisfaction (2018, background): “The purpose of a mediating effect analysis is to judge whether the relationship between the independent variable X and dependent variable Y is partly or wholly due to the mediating variable M (Baron & Kenny, 1986; Yuan & MacKinnon, 2009).”
Conditional Process Analysis: Concepts, Computation, and Advances in the Modeling of the Contingencies of Mechanisms (2020, methodology): “…interval (for a discussion of the mechanics, see Hayes, 2018a; Shrout & Bolger, 2002), the Monte Carlo confidence interval (MCCI; Preacher & Selig, 2012), the distribution of the product approach (MacKinnon, Fritz, Williams, & Lockwood, 2007), and Bayesian methods (Yuan & MacKinnon, 2009).”
Advances in mediation analysis: a survey and synthesis of new developments. (2015, methodology): “…The simple mediation model with M as a mediator of the effect of X on Y. (Bollen & Stine 1990, Shrout & Bolger 2002), Bayesian credible intervals (Yuan & MacKinnon 2009), Monte Carlo CIs (MacKinnon et al. 2004, Preacher & Selig 2012), and methods involving product distributions (MacKinnon et al.…”
Regression-based statistical mediation and moderation analysis in clinical research: Observations, recommendations, and implementation. (2017, methodology): “…& Selig, 2012), and the Bayesian credible interval (Yuan & MacKinnon, 2009).4 Research shows these all perform well while also outperforming the Sobel test (Fritz, Taylor, & MacKinnon, 2012; Hayes & Scharkow, 2013; Preacher & Selig, 2002; Williams & MacKinnon, 2008; Yuan & MacKinnon, 2009).”
Effects of Mindfulness-Acceptance-Commitment (MAC) on Sport-Specific Dispositional Mindfulness, Emotion Regulation, and Self-Rated Athletic Performance in a Multiple-Sport Population: an RCT Study (2019, background): “Due to the less restrictive distributional assumptions, related to the Bayesian framework, the normality assumption is not needed to be fulfilled to perform the analyses (Yuan and MacKinnon 2009).”
Moderated Mediation Analysis Using Bayesian Methods (2015, methodology): “This result is consistent with the findings from Bayesian methods with noninformative priors in Yuan and MacKinnon (2009) and the findings in Preacher et al. (2007) from frequentist resampling methods.”