Bias in Cross-Sectional Analyses of Longitudinal Mediation: Partial and Complete Mediation Under an Autoregressive Model

By S. Maxwell, David A. Cole, M. Mitchell. Multivariate Behavioral Research, 2011

1,248 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/069828b02503b09be971e0f297ea90a3e0b384c4

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

The Merits of Using Longitudinal Mediation (2016, positive): “Based on logic and statistical constraints (Cole & Maxwell, 2003; Maxwell & Cole, 2007; Maxwell et al., 2011), it is even possible that none of the six would have been supported.”
Advances in mediation analysis: a survey and synthesis of new developments. (2015, background): “Moreover, the size of effects (autoregressive or otherwise) usually will vary as a function of lag (Cole & Maxwell 2003; Gollob & Reichardt 1987, 1991; Maxwell & Cole 2007; Maxwell et al. 2011).”
The Relationships Among Gratitude, Self-esteem, Social Support and Life Satisfaction Among Undergraduate Students (2014, background): “The first limitation is that the study was a cross-sectional design such that the temporal sequence of the independent variables, mediator, and dependent variables cannot be verified (Maxwell and Cole 2007; Maxwell et al. 2011).”
Exploring pathways linking greenspace to health: Theoretical and methodological guidance (2017, background): “Maxwell et al. (2011) illustrated how cross-sectional estimates of the indirect path between an independent and dependent variable through a single mediator could be substantial even when the coefficient for the indirect path was known to be zero in a longitudinal design.”
Mendelian randomisation for mediation analysis: current methods and challenges for implementation (2019, result): “This approach is preferred even when time varying effects are not a focus of the analysis, as cross-sectional data does not reflect the implied temporality suggested by mediation analysis [57–59].”
An Index and Test of Linear Moderated Mediation (2015, methodology): “211–212) is that at least one of the paths in a causal sequence (either the X → M or M → Y path) should be moderated by the standards of some kind of inferential test in order to claim that an indirect effect is moderated, though they do not state this as a formal requirement as Muller et al. (2005)”
Occupational Future Time Perspective: A Meta-Analysis of Antecedents and Outcomes (2018, methodology): “The general reliance on cross ‐ sectional methodologies in this literature is also a limitation to the interpretation of the path model we present herein (see Maxwell & Cole, 2007; Maxwell, Cole, & Mitchell, 2011).”