Introduction to Statistical Mediation Analysis

By David P. MacKinnon. 2008

5,374 citations recorded by the source corpus.

Subfields: Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/c1dad5822b1e635af593b5aeb887a06358b299a9

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

Citations in the corpus

Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies (2011, methodology): “To highlight the flexibility and transparency of our approach, we first provide a brief review of the standard approach to estimating mediation effects (e.g., Baron and Kenny 1986; MacKinnon 2008).”
A general approach to causal mediation analysis. (2010, neutral): “7 This use of residuals differs from the structural equation modeling tradition in that residuals are not being used to access the fit of statistical models as discussed in Mulaik (2007). Instead, they are used to assess the violation of the key identification assumption.”
Classroom Emotional Climate, Teacher Affiliation, and Student Conduct (2011, methodology): “Following the steps in mediation analysis (Baron & Kenny, 1986; MacKinnon, 2008), we tested the following hypotheses: (a) classrooms with higher observed CEC ratings have students with higher year-end conduct grades (direct link between predictor-outcome); (b) classrooms with higher CEC ratings have”
The Mediation Formula: A Guide to the Assessment of Causal Pathways in Nonlinear Models (2011, background): “(7) The intuition here is somewhat different, and represents a nonlinear version of the “product-of-coefficients” strategy in linear models (MacKinnon, 2008); the E(Y |do(x, z), w) term encodes the effect of Z on Y for fixed X = x and W = w, while the [P (z|do(x′), w) − P (z|do(x), w)] encodes the…”
On the use of partial least squares path modeling in accounting research (2011, methodology): “One of the key advantages of SEM versus regression is the capability of SEM to test mediating variables as part of a comprehensive model (MacKinnon, 2008).”
Estimation of a Semiparametric Natural Direct Effect Model Incorporating Baseline Covariates (2014, positive): “Conditional models for direct and indirect effects are of interest in making inferences about so-called moderated mediation effects, a topic of interest particularly in psychology (Muller et al., 2005; Preacher et al., 2007; Mackinnon, 2008).”
How does mindfulness-based cognitive therapy work? (2010, methodology): “In line the approach of Segal and colleagues, we calculated standardized residualized change scores for the mediating variables using a simple linear regression model in which Time 1 scores predicted Time 2 scores (Mackinnon, 2008; Segal et al., 2006).”
Standardization and Control for Confounding in Observational Studies: A Historical Perspective (2015, background): “Instead, a general area of Mediation Analysis has grown up; see MacKinnon (2008), Section 1.8, for a useful historical survey.”
Bayesian Mediation Analysis (2009, methodology): “For ease of exposition, MacKinnon (2008) added some context to the data set, as follows: Assume a study of a special exercise program for depression, where the independent variable X was the continuous measure of how conducive the person’s lifestyle and environment was to exercise, the mediating…”
Alternative Methods for Assessing Mediation in Multilevel Data: The Advantages of Multilevel SEM (2011, methodology): “…design for purposes of the simulation, we adopt the random-intercept, fixed-slope model for 2–1–1 designs described by Krull and MacKinnon (1999, 2001), MacKinnon (2008), and Pituch and Stapleton (2008): Mij D “M0j C eMij “M0j D ”M00 C ”M01Xj C uM0j (1) Yij D “Y 0j C “Y 1j Mij C eY ij “Y 0j D ”Y…”
Structural Equation Models and Mixture Models With Continuous Nonnormal Skewed Distributions (2016, positive): “(67) The usual definitions of a direct and indirect effect are β2 and β1 × β3 (see, e.g., MacKinnon, 2008).”
The relation between achievement goal and expectancy-value theories in predicting achievement-related outcomes: A test of four theoretical conceptions (2013, methodology): “Therefore, the acceptance of a model does not necessarily imply that other untested models would not provide a better fit to the data (MacKinnon et al. 2000) or that one variable in the model causes another one (MacKinnon 2008; Pearl 2000).”