Commentary: Using Potential Outcomes to Understand Causal Mediation Analysis

In this commentary, we demonstrate how the potential outcomes framework can help understand the key identification assumptions underlying causal mediation analysis. We show that this framework can lead to the development of alternative research design and statistical analysis strategies applicable to the longitudinal data settings considered by Maxwell, Cole, and Mitchell (2011).     

Cross-Sectional Analysis of Longitudinal Mediation Processes

Statistical mediation analysis can help to identify and explain the mechanisms behind psychological processes. Examining a set of variables for mediation effects is a ubiquitous process in the social sciences literature; however, despite evidence suggesting that cross-sectional data can misrepresent the mediation of longitudinal processes, cross-sectional analyses continue to be used in this manner. Alternative longitudinal mediation models, including those rooted in a structural equation modeling framework (cross-lagged panel, latent growth curve, and latent difference score models) are currently available and may provide a better representation of mediation processes for longitudinal data. The purpose of this paper is twofold: first, we provide a comparison of cross-sectional and longitudinal mediation models; second, we advocate using models to evaluate mediation effects that capture the temporal sequence of the process under study. Two separate empirical examples are presented to illustrate differences in the conclusions drawn from cross-sectional and longitudinal mediation analyses. Findings from these examples yielded substantial differences in interpretations between the cross-sectional and longitudinal mediation models considered here. Based on these observations, researchers should use caution when attempting to use cross-sectional data in place of longitudinal data for mediation analyses.     

纵向数据的中介效应分析

目前中介效应检验主要是基于截面数据,但许多时候截面数据的中介分析不适合进行因果推断,因而需要收集历时性的纵向数据,进行纵向数据的中介分析。评介了基于交叉滞后面板模型、多层线性模型和潜变量增长模型的纵向数据的中介分析方法及其四个发展。第一,中介效应随时间变化,如连续时间模型、多层时变系数模型。第二,中介效应随个体变化,如随机效应的交叉滞后面板模型和多层自回归模型。第三,中介模型的整合,如交叉滞后面板模型与多层线性模型整合为多层自回归模型。第四,中介检验方法的发展,建议使用Monte Carlo、Bootstrap和贝叶斯法进行纵向数据的中介分析。总结出一个纵向数据的中介分析流程并给出相应的Mplus程序。随后展望了纵向数据的中介分析的拓展方向。     

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

Maxwell and Cole (2007) Maxwell, S. E. and Cole, D. A. 2007. Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12: 23–44. [Crossref], [PubMed], [Web of Science ®] , [Google Scholar] showed that cross-sectional approaches to mediation typically generate substantially biased estimates of longitudinal parameters in the special case of complete mediation. However, their results did not apply to the more typical case of partial mediation. We extend their previous work by showing that substantial bias can also occur with partial mediation. In particular, cross-sectional analyses can imply the existence of a substantial indirect effect even when the true longitudinal indirect effect is zero. Thus, a variable that is found to be a strong mediator in a cross-sectional analysis may not be a mediator at all in a longitudinal analysis. In addition, we show that very different combinations of longitudinal parameter values can lead to essentially identical cross-sectional correlations, raising serious questions about the interpretability of cross-sectional mediation data. More generally, researchers are encouraged to consider a wide variety of possible mediation models beyond simple cross-sectional models, including but not restricted to autoregressive models of change.     

Correlational Data,Causal Hypotheses,and Validity

A shared problem across the sciences is to make sense of correlational data coming from observations and/or from experiments. Arguably, this means establishing when correlations are causal and when they are not. This is an old problem in philosophy. This paper, narrowing down the scope to quantitative causal analysis in social science, reformulates the problem in terms of the validity of statistical models. Two strategies to make sense of correlational data are presented: first, a ‘structural strategy’, the goal of which is to model and test causal structures that explain correlational data; second, a ‘manipulationist or interventionist strategy’, that hinges upon the notion of invariance under intervention. It is argued that while the former can offer a solution the latter cannot.     

A Regression Framework for Causal Mediation Analysis with Applications to Behavioral Science

We introduce and extend the classical regression framework for conducting mediation analysis from the fit of only one model. Using the essential mediation components (EMCs) allows us to estimate causal mediation effects and their analytical variance. This single-equation approach reduces computation time and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations. Additionally, we extend this framework to non-nested mediation systems, provide a joint measure of mediation for complex mediation hypotheses, propose new visualizations for mediation effects, and explain why estimates of the total effect may differ depending on the approach used. Using data from social science studies, we also provide extensive illustrations of the usefulness of this framework and its advantages over traditional approaches to mediation analysis. The example data are freely available for download online and we include the R code necessary to reproduce our results.     

Methods for Mediation Analysis with Missing Data

Despite wide applications of both mediation models and missing data techniques, formal discussion of mediation analysis with missing data is still rare. We introduce and compare four approaches to dealing with missing data in mediation analysis including listwise deletion, pairwise deletion, multiple imputation (MI), and a two-stage maximum likelihood (TS-ML) method. An R package bmem is developed to implement the four methods for mediation analysis with missing data in the structural equation modeling framework, and two real examples are used to illustrate the application of the four methods. The four methods are evaluated and compared under MCAR, MAR, and MNAR missing data mechanisms through simulation studies. Both MI and TS-ML perform well for MCAR and MAR data regardless of the inclusion of auxiliary variables and for AV-MNAR data with auxiliary variables. Although listwise deletion and pairwise deletion have low power and large parameter estimation bias in many studied conditions, they may provide useful information for exploring missing mechanisms.     

Bayesian Causal Mediation Analysis for Group Randomized Designs with Homogeneous and Heterogeneous Effects: Simulation and Case Study

A fully Bayesian approach to causal mediation analysis for group-randomized designs is presented. A unique contribution of this article is the combination of Bayesian inferential methods with G-computation to address the problem of heterogeneous treatment or mediator effects. A detailed simulation study shows that this approach has excellent frequentist properties, particularly in the case of small sample sizes with accurate informative priors. The simulation study also demonstrates that the proposed approach can take into account heterogeneous treatment or mediator effects without bias. A case study using data from a school-based randomized intervention designed to increase parent social capital leading to improved behavioral and academic outcomes in children is offered to illustrate the Bayesian approach to causal mediation in group-randomized designs.     

Sociosexual Orientation,Commitment, and Infidelity: A Mediation Analysis

ABSTRACT

Individuals with an unrestricted sociosexual orientation (SO) are less committed to their romantic relationships and more likely to engage in infidelity. Furthermore, commitment is negatively associated with tendencies to cheat. However, no previous research has examined the possible mediating role of commitment in the relationship between SO and infidelity. The current study examined whether commitment mediated the relationship between SO and willingness to engage in three types of cheating behaviors (Ambiguous, Deceptive, and Explicit). Results suggested that commitment partially mediated the relationship between SO and infidelity, such that unrestricted individuals had lower commitment, which in turn led to a greater likelihood of cheating. Implications of this association and directions for future research are discussed.     

A Comparison of the Cross-Sectional and Sequential Designs when Assessing Longitudinal Mediation

Mediational studies are often of interest in psychology because they explore the underlying relationship between 2 constructs. Previous research has shown that cross-sectional designs are prone to biased estimates of longitudinal mediation parameters. The sequential design has become a popular alternative to the cross-sectional design for assessing mediation. This design is a compromise between the cross-sectional and longitudinal designs because it incorporates time in the model but has only 1 measurement each of X, M, and Y. As such, this design follows the recommendation of the MacArthur group approach, which stresses the importance of multiple waves of data for studying mediation. These 2 designs were compared to see whether the sequential design assesses longitudinal mediation more accurately than the cross-sectional design. Specifically, analytic expressions are derived for the bias of estimated direct and indirect effects as calculated from the sequential design when the actual mediational process follows a longitudinal autoregressive model. It was found that, in general, the sequential design does not assess longitudinal mediation more accurately than the cross-sectional design. As a result, neither design can be depended on to assess longitudinal mediation accurately.     

Authoritarianism,Threat, and Prejudice: An Analysis of Mediation and Moderation

Right-wing authoritarianism is a central construct in individual differences approaches to prejudice. Its power to predict prejudice is often attributed to perceived threat. However, the exact moderating and mediating processes involved are little understood. In two studies (Ns = 53, 84), exposure to threatening versus nonthreatening information about an ethnic out-group had reliable indirect effects on prejudice in authoritarians, but not in nonauthoritarians, largely because authoritarians were more likely to perceive actual threat when they interpreted the information received to represent a threatening argument. Additionally, in Study 2, authoritarians reacted more strongly with negative emotions when they perceived actual threat.     

Identification and Sensitivity Analysis for Average Causal Mediation Effects with Time-Varying Treatments and Mediators: Investigating the Underlying Mechanisms of Kindergarten Retention Policy

Considering that causal mechanisms unfold over time, it is important to investigate the mechanisms over time, taking into account the time-varying features of treatments and mediators. However, identification of the average causal mediation effect in the presence of time-varying treatments and mediators is often complicated by time-varying confounding. This article aims to provide a novel approach to uncovering causal mechanisms in time-varying treatments and mediators in the presence of time-varying confounding. We provide different strategies for identification and sensitivity analysis under homogeneous and heterogeneous effects. Homogeneous effects are those in which each individual experiences the same effect, and heterogeneous effects are those in which the effects vary over individuals. Most importantly, we provide an alternative definition of average causal mediation effects that evaluates a partial mediation effect; the effect that is mediated by paths other than through an intermediate confounding variable. We argue that this alternative definition allows us to better assess at least a part of the mediated effect and provides meaningful and unique interpretations. A case study using ECLS-K data that evaluates kindergarten retention policy is offered to illustrate our proposed approach.     

Family-of-Origin,Sexual Attitudes,and Perceptions of Infidelity: A Mediation Analysis

Abstract

The purpose of this study was to examine the influence family of origin and sexual attitudes have on perceptions of infidelity through a mediation analysis to test whether sexual attitudes mediates the relationship between family of origin attitudes and perceptions of infidelity. Results indicated that family of origin attitudes were linked with higher permissive sexual attitudes and lower perceptions of infidelity. Additionally, results indicated an indirect link between family of origin attitudes and perceptions of infidelity through sexual attitudes. These findings align with previous research examining how family of origin influences the occurrence of infidelity and provides insight to helping professionals as they address infidelity in romantic relationships.     

Causal Closure,Causal Exclusion,and Supervenience Physicalism

This article considers the recent defense of the supervenience approach to physicalism due to Jaegwon Kim. Kim argues that supervenience supports physical causal closure, and that causal closure supports physicalism – indeed, a kind of reductive physicalism – and thus that supervenience suffices for physicalism. After laying out Kim's argument, I ask whether its success would truly vindicate the role of supervenience in defining physicalist positions. I argue that it would not, and that insofar as Kim's defense of supervenience physicalism succeeds, it does so by showing that supervenience physicalism is not a unique, nonredundant way to be a physicalist.     

团队领导行为与团队心理安全：以信任为中介变量的研究

已有的团队领导研究着重于领导行为对团队绩效的影响,而缺乏其作用机制和路径的研究。本文选取信任与团队心理安全两个变量,以142个团队为样本,对团队领导行为是否以及如何通过信任影响团队心理安全进行实证研究。研究结果表明：(1)转换型领导行为与团队心理安全之间存在正相关关系,显著地影响团队心理安全感;(2)信任是形成团队心理安全感的重要因素之一;(3)信任在团队领导行为、团队心理安全感之间的中介效应显著。     

Well-Being and Physical Health: A Mediation Analysis

This paper’s aim is to propose a mediation framework and test whether lifestyle choices and social capital are pathways through which baseline levels of well-being affect subsequent physical health among older adults. Using large-scale panel data for Australia, we find that past levels of well-being have strong direct effects on present physical health. We also show that more frequent socialization and more frequent participation in physical activity are two pathways through which higher levels of well-being lead to better physical health. These mediating effects vary across gender. Our findings highlight a protective role of subjective well-being in physical health. Interventions taking into account not only the direct but also the indirect effects of well-being are promising avenues for physical health maintenance in the older population.

     

Understanding How Creative Thinking Skills,Attitudes and Behaviors Work Together: A Causal Process Model

Managers (N = 112) from a large international consumer goods manufacturer participated in a field experiment in which they learned and applied the Simplex process of creative thinking to solve real management problems. The interrelationships among six attitudinal and behavioral skill variables learned during the training were measured to improve understanding of how these variables contribute to the process. Predicted relationships were tested and a best‐fit causal model was developed. Behavioral skill in generating quantity of options was the most important variable overall: it was directly associated with behavioral skill in both generating quality options and evaluating options. The key attitudinal skill and the second most important variable overall was the preference for avoiding premature evaluation of options (deferral of judgment). The other attitude measured, the preference for active divergence, played only an indirect role in the process.     

The Gaussian Graphical Model in Cross-Sectional and Time-Series Data

We discuss the Gaussian graphical model (GGM; an undirected network of partial correlation coefficients) and detail its utility as an exploratory data analysis tool. The GGM shows which variables predict one-another, allows for sparse modeling of covariance structures, and may highlight potential causal relationships between observed variables. We describe the utility in three kinds of psychological data sets: data sets in which consecutive cases are assumed independent (e.g., cross-sectional data), temporally ordered data sets (e.g., n = 1 time series), and a mixture of the 2 (e.g., n > 1 time series). In time-series analysis, the GGM can be used to model the residual structure of a vector-autoregression analysis (VAR), also termed graphical VAR. Two network models can then be obtained: a temporal network and a contemporaneous network. When analyzing data from multiple subjects, a GGM can also be formed on the covariance structure of stationary means—the between-subjects network. We discuss the interpretation of these models and propose estimation methods to obtain these networks, which we implement in the R packages graphicalVAR and mlVAR. The methods are showcased in two empirical examples, and simulation studies on these methods are included in the supplementary materials.     

Personality,Positivity and Happiness: A Mediation Analysis Using a Bifactor Model

Recent studies suggested an important role of neuroticism and extraversion facets as incremental predictors of subjective well-being outcomes. Research has shown that positive cognitions mediated the relation between personality traits and well-being. The present study examined the relationship between neuroticism and extraversion, measured as general and group factors, and subjective happiness through a general positivity factor. 770 community participants (69.4 % females; M = 55.34; SD = 16.01) completed personality, satisfaction with life, optimism, self-esteem, and subjective happiness scales. A bifactor model was used to parse general and specific variance components for multifaceted constructs. The general positivity factor completely mediated neuroticism-subjective happiness relationships and overlapped with general neuroticism, whilst it partially mediated extraversion-subjective happiness ones. Other paths to happiness involved cheerfulness and enthusiasm. Assertiveness, activity level and excitement-seeking had a weak relationship with subjective happiness and only through positivity. Gregariousness and friendliness had neither direct nor indirect effects on subjective happiness. Life satisfaction had a twofold role as a component of positivity as well as providing an independent contribution to variance in subjective happiness. In keeping with previous research, neuroticism acted as a sort of general negativity factor. Cheerfulness and extraversion made an incremental contribution to variance in subjective happiness. Our findings support the utility of a multifaceted approach to study pathways from personality to well-being. Theoretical and practical implications for promoting well-being were discussed.