Intensive Longitudinal Mediation in Mplus (original) (raw)
Related papers
Mediation variables are those intervening variables (M) that affect the relationship between the independent variable (X) and the dependent variable (Y). The causal relation between X and Y through the mediation variable M must be established to analyze mediation. M is a mediator, if it involves indirect effects among the set of original variables X and Y. A requirement for a variable to have a causal effect on another is that the cause must precede the outcome in time. This paper argues that the right way to prove the mediation effect is by analyzing different moments in time to support the relations among the variables, which can be done using longitudinal models. The use of traditional analysis in mediation models without the time effect (assumed to occur instantaneously) biases the parameter estimation and could create the possible cancelation of effects, if the paths have opposite signs. Mediation models (without the time effect) are quite frequently used in social sciences. In contrast, the longitudinal mediation model is a very uncommon methodology within the social science community. The aim of this paper is to shed some light about how to analyze the time framework and to provide a methodology to solve the possible limitations that could hinder the robustness of the mediation analysis.
Psychological methods, 2017
The study of mediation of treatment effects, or how treatments work, is important to understanding and improving psychological and behavioral treatments, but applications often focus on mediators and outcomes measured at a single time point. Such cross-sectional analyses do not respect the implied temporal ordering that mediation suggests. Clinical trials of treatments often provide repeated measures of outcomes and, increasingly, of mediators as well. Repeated measurements allow the application of various types of longitudinal structural equation mediation models. These provide flexibility in modeling, including the ability to incorporate some types of measurement error and unmeasured confounding that can strengthen the robustness of findings. The usual approach is to identify the most theoretically plausible model and apply that model. In the absence of clear theory, we put forward the option of fitting a few theoretically plausible models, providing a type of sensitivity analysis...
Impact of violations of measurement invariance in cross-lagged panel mediation models
Behavior Research Methods
When fitting cross-lagged panel mediation models, the assumption of longitudinal measurement invariance should be made. This simulation study investigated the impact of violations of measurement invariance on cross-lagged panel mediation analysis using the latent cross-lagged panel mediation model (L-CLPM). Results showed that estimates of direct and mediated effects tended to be less accurate and the type I error rate of testing mediated effects can be inflated under non-invariance conditions. On the other hand, power for detecting model misspecifications due to longitudinal non-invariance was high across all investigated conditions. Fit indices were not sensitive to violations of measurement invariance in the L-CLPM under the manipulated conditions. We end the article with suggestions, limitations, and future directions. Keywords Measurement invariance. Cross-lagged panel mediation. Model fit. Parameter estimation Mediation modeling has been widely used in behavioral and social sciences. Most applications of mediation models are based on cross-sectional designs where data are collected on the same occasion. However, researchers have argued against the practice of cross-sectional meditation analysis (e.g., Cole
A Meditation on Mediation: Evidence That Structural Equations Models Perform Better Than Regressions
Journal of Consumer Psychology, 2007
In this paper, we suggest ways to improve mediation analysis practice among consumer behavior researchers. We review the current methodology and demonstrate the superiority of structural equations modeling, both for assessing the classic mediation questions and for enabling researchers to extend beyond these basic inquiries. A series of simulations are presented to support the claim that the approach is superior. In addition to statistical demonstrations, logical arguments are presented, particularly regarding the introduction of a fourth construct into the mediation system. We close the paper with new prescriptive instructions for mediation analyses.
A Tutorial in Longitudinal Measurement Invariance and Cross-lagged Panel Models Using Lavaan
2020
In longitudinal studies involving multiple latent variables, researchers often seek to predict how iterations of latent variables measured at early time points predict iterations measured at later time points. Cross-lagged panel modeling, a form of structural equation modeling, is a useful way to conceptualize and test these relationships. However, prior to making causal claims, researchers must first ensure that the measured constructs are equivalent between time points. To do this, they test for measurement invariance, constructing and comparing a series of increasingly strict and parsimonious models, each making more constraints across time than the last. This comparison process, though challenging, is an important prerequisite to interpretation of results. Fortunately, testing for measurement invariance in cross-lagged panel models has become easier, thanks to the wide availability of R and its packages. This paper serves as a tutorial in testing for measurement invariance and c...
PANEL DATA Cain, M.K.a,b Time and Other Considerations in Mediation DesignArticle
This article serves as a practical guide to mediation design and analysis by evaluating the ability of mediation models to detect a significant mediation effect using limited data. The cross-sectional mediation model, which has been shown to be biased when the mediation is happening over time, is compared with longitudinal mediation models: sequential, dynamic, and cross-lagged panel. These longitudinal mediation models take time into account but bring many problems of their own, such as choosing measurement intervals and number of measurement occasions. Furthermore, researchers with limited resources often cannot collect enough data to fit an appropriate longitudinal mediation model. These issues were addressed using simulations comparing four mediation models each using the same amount of data but with differing numbers of people and time points. The data were generated using multilevel mediation models, with varying data characteristics that may be incorrectly specified in the analysis models. Models were evaluated using power and Type I error rates in detecting a significant indirect path. Multilevel longitudinal mediation analysis performed well in every condition, even in the misspecified conditions. Of the analyses that used limited data, sequential mediation had the best performance; therefore, it offers a viable second choice when resources are limited. Finally, each of these models were demonstrated in an empirical analysis.
Mediation from Multilevel to Structural Equation Modeling
Annals of Nutrition and Metabolism, 2014
Background/Aims: The purpose of this article is to outline multilevel structural equation modeling (MSEM) for mediation analysis of longitudinal data. The introduction of mediating variables can improve experimental and nonexperimental studies of child growth in several ways as discussed throughout this article. Single-mediator individual-level and multilevel mediation models illustrate several current issues in the estimation of mediation with longitudinal data. The strengths of incorporating structural equation modeling (SEM) with multilevel mediation modeling are described. Summary and Key Messages: Longitudinal mediation models are pervasive in many areas of research including child growth. Longitudinal mediation models are ideally modeled as repeated measurements clustered within individuals. Further, the combination of MSEM and SEM provides an ideal approach for several reasons, including the ability to assess effects at different levels of analysis, incorporation of measureme...
Mediation Analysis: A Retrospective Snapshot of Practice and More Recent Directions
Journal of General Psychology, 2009
R. Baron and D. A. Kenny's (1986) paper introducing mediation analysis has been cited over 9,000 times, but concerns have been expressed about how this method is used. The authors review past and recent methodological literature and make recommendations for how to address 3 main issues: association, temporal order, and the no omitted variables assumption. The authors briefly visit the topics of reliability and the confirmatory-exploratory distinction. In addition, to provide a sense of the extent to which the earlier literature had been absorbed into practice, the authors examined a sample of 50 articles from 2002 citing R. Baron and D. A. Kenny and containing at least 1 mediation analysis via ordinary least squares regression. A substantial proportion of these articles included problematic reporting; as of 2002, there appeared to be room for improvement in conducting such mediation analyses. Future literature reviews will demonstrate the extent to which the situation has improved.
Journal of Personality and Social Psychology, 2018
In light of current concerns with replicability and reporting false-positive effects in psychology, we examine Type I errors and power associated with 2 distinct approaches for the assessment of mediation, namely the component approach (testing individual parameter estimates in the model) and the index approach (testing a single mediational index). We conduct simulations that examine both approaches and show that the most commonly used tests under the index approach risk inflated Type I errors compared with the joint-significance test inspired by the component approach. We argue that the tendency to report only a single mediational index is worrisome for this reason and also because it is often accompanied by a failure to critically examine the individual causal paths underlying the mediational model. We recommend testing individual components of the indirect effect to argue for the presence of an indirect effect and then using other recommended procedures to calculate the size of that effect. Beyond simple mediation, we show that our conclusions also apply in cases of within-participant mediation and moderated mediation. We also provide a new R-package that allows for an easy implementation of our recommendations.
AStA Advances in Statistical Analysis, 2010
We propose a three step procedure to investigate measurement bias and response shift, a special case of measurement bias in longitudinal data. Structural equation modelling is used in each of the three steps, which can be described as (1) establishing a measurement model using confirmatory factor analysis, (2) detecting measurement bias by testing the equivalence of model parameters across measurement occasions, (3) detecting measurement bias with respect to additional exogenous variables by testing their direct effects on the indicator variables. The resulting model can be used to investigate true change in the attributes of interest, by testing changes in common factor means. Solutions for the issue of constraint interaction and for chance capitalisation in model specification searches are discussed as part of the procedure. The procedure is illustrated by applying it to longitudinal health-related quality-of-life data of HIV/AIDS patients, collected at four semi-annual measurement occasions.