Posted Content•
Interpretation and Identification of Causal Mediation
TL;DR: It is shown that natural effects can be identified by methods that go beyond standard adjustment for confounders, applicable to observational studies in which treatment assignment remains confounded with the mediator or with the outcome.
Abstract: This paper reviews the foundations of causal mediation analysis and offers a general and transparent account of the conditions necessary for the identification of natural direct and indirect effects, thus facilitating a more informed judgment of the plausibility of these conditions in specific applications. We show that the conditions usually cited in the literature are overly restricted, and can be relaxed substantially, without compromising identification. In particular, we show that natural effects can be identified by methods that go beyond standard adjustment for confounders, applicable to observational studies in which treatment assignment remains confounded with the mediator or with the outcome. These identification conditions can be validated algorithmically from the diagramatic description of one's model, and are guaranteed to produce unbiased results whenever the description is correct. The identification conditions can be further relaxed in parametric models, possibly including interactions, and permit us to compare the relative importance of several pathways, mediated by interdependent variables.
Citations
More filters
TL;DR: The mediation package implements a comprehensive suite of statistical tools for conducting causal mediation analysis in applied empirical research and implements a statistical method for dealing with multiple (causally dependent) mediators, which are often encountered in practice.
Abstract: In this paper, we describe the R package mediation for conducting causal mediation analysis in applied empirical research. In many scientific disciplines, the goal of researchers is not only estimating causal effects of a treatment but also understanding the process in which the treatment causally affects the outcome. Causal mediation analysis is frequently used to assess potential causal mechanisms. The mediation package implements a comprehensive suite of statistical tools for conducting such an analysis. The package is organized into two distinct approaches. Using the model-based approach, researchers can estimate causal mediation effects and conduct sensitivity analysis under the standard research design. Furthermore, the design-based approach provides several analysis tools that are applicable under different experimental designs. This approach requires weaker assumptions than the model-based approach. We also implement a statistical method for dealing with multiple (causally dependent) mediators, which are often encountered in practice. Finally, the package also offers a methodology for assessing causal mediation in the presence of treatment noncompliance, a common problem in randomized trials.
2,417 citations
Cites background from "Interpretation and Identification o..."
...See Pearl (2014) and Imai, Keele, Tingley, and Yamamoto (2014)....
[...]
TL;DR: This paper argued that the ability of the longitudinal design to reflect causality has been overstated and that it offers limited advantages over the cross-sectional design in most cases in which it is used.
Abstract: The cross-sectional research design, especially when used with self-report surveys, is held in low esteem despite its widespread use. It is generally accepted that the longitudinal design offers considerable advantages and should be preferred due to its ability to shed light on causal connections. In this paper, I will argue that the ability of the longitudinal design to reflect causality has been overstated and that it offers limited advantages over the cross-sectional design in most cases in which it is used. The nature of causal inference from a philosophy of science perspective is used to illustrate how cross-sectional designs can provide evidence for relationships among variables and can be used to rule out many potential alternative explanations for those relationships. Strategies for optimizing the use of cross-sectional designs are noted, including the inclusion of control variables to rule out spurious relationships, the addition of alternative sources of data, and the incorporation of experimental methods. Best practice advice is offered for the use of both cross-sectional and longitudinal designs, as well as for authors writing and for reviewers evaluating papers that report results of cross-sectional studies.
543 citations
Cites background from "Interpretation and Identification o..."
...The notion that the observed connection between X and Y might be attributable to additional causal variables has been discussed in other contexts such as endogeneity (Antonakis, Bendahan, Jacquart, & Lalive, 2010) and causal mediation (Imai, Keele, & Tingley, 2010; Pearl, 2014)....
[...]
TL;DR: This review provides a survey and synthesis of four areas of active methodological research: (a) mediation analysis for longitudinal data, (b) causal inference for indirect effects, (c) mediationAnalysis for discrete and nonnormal variables, and (d) mediation assessment in multilevel designs.
Abstract: Mediation processes are fundamental to many classic and emerging theoretical paradigms within psychology. Innovative methods continue to be developed to address the diverse needs of researchers studying such indirect effects. This review provides a survey and synthesis of four areas of active methodological research: (a) mediation analysis for longitudinal data, (b) causal inference for indirect effects, (c) mediation analysis for discrete and nonnormal variables, and (d) mediation assessment in multilevel designs. The aim of this review is to aid in the dissemination of developments in these four areas and suggest directions for future research.
517 citations
TL;DR: This article reviewed applications of covariance-based structural equation modeling (SEM) in the Journal of Advertising starting with the first issue in 1972 and identified 111 articles from the earliest application of SEM in 1983 through 2015, and discuss important methodological issues related to the following aspects: confirmatory factor analysis (CFA), causal modeling, multiple group analysis, reporting, and guidelines for interpretation of results.
Abstract: In this article, we review applications of covariance-based structural equation modeling (SEM) in the Journal of Advertising (JA) starting with the first issue in 1972. We identify 111 articles from the earliest application of SEM in 1983 through 2015, and discuss important methodological issues related to the following aspects: confirmatory factor analysis (CFA), causal modeling, multiple group analysis, reporting, and guidelines for interpretation of results. Moreover, we summarize some issues related to varying terminology associated with different SEM methods. Findings indicate that the use of SEM in the JA contributes greatly to conceptual, empirical, and methodological advances in advertising research. The assessment contributes to the literature by offering advertising researchers a summary guide to best practices and a reminder of the basics that distinguish the powerful and unique approach involving structural analysis of covariances.
402 citations
TL;DR: The possibility of extracting causal interpretations from black-box machine-trained models, and three requirements to make causal interpretations: a model with good predictive performance, some domain knowledge in the form of a causal diagram and suitable visualization tools.
Abstract: The fields of machine learning and causal inference have developed many concepts, tools, and theory that are potentially useful for each other. Through exploring the possibility of extracting causa...
266 citations
References
More filters
TL;DR: This article seeks to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating the many ways in which moderators and mediators differ, and delineates the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena.
Abstract: In this article, we attempt to distinguish between the properties of moderator and mediator variables at a number of levels. First, we seek to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating, both conceptually and strategically, the many ways in which moderators and mediators differ. We then go beyond this largely pedagogical function and delineate the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena, including control and stress, attitudes, and personality traits. We also provide a specific compendium of analytic procedures appropriate for making the most effective use of the moderator and mediator distinction, both separately and in terms of a broader causal system that includes both moderators and mediators.
80,095 citations
TL;DR: The authors discusses the central role of propensity scores and balancing scores in the analysis of observational studies and shows that adjustment for the scalar propensity score is sufficient to remove bias due to all observed covariates.
Abstract: : The results of observational studies are often disputed because of nonrandom treatment assignment. For example, patients at greater risk may be overrepresented in some treatment group. This paper discusses the central role of propensity scores and balancing scores in the analysis of observational studies. The propensity score is the (estimated) conditional probability of assignment to a particular treatment given a vector of observed covariates. Both large and small sample theory show that adjustment for the scalar propensity score is sufficient to remove bias due to all observed covariates. Applications include: matched sampling on the univariate propensity score which is equal percent bias reducing under more general conditions than required for discriminant matching, multivariate adjustment by subclassification on balancing scores where the same subclasses are used to estimate treatment effects for all outcome variables and in all subpopulations, and visual representation of multivariate adjustment by a two-dimensional plot. (Author)
23,744 citations
Book•
28 Apr 1989
TL;DR: The General Model, Part I: Latent Variable and Measurement Models Combined, Part II: Extensions, Part III: Extensions and Part IV: Confirmatory Factor Analysis as discussed by the authors.
Abstract: Model Notation, Covariances, and Path Analysis. Causality and Causal Models. Structural Equation Models with Observed Variables. The Consequences of Measurement Error. Measurement Models: The Relation Between Latent and Observed Variables. Confirmatory Factor Analysis. The General Model, Part I: Latent Variable and Measurement Models Combined. The General Model, Part II: Extensions. Appendices. Distribution Theory. References. Index.
19,019 citations
TL;DR: The art and science of cause and effect have been studied in the social sciences for a long time as mentioned in this paper, see, e.g., the theory of inferred causation, causal diagrams and the identification of causal effects.
Abstract: 1. Introduction to probabilities, graphs, and causal models 2. A theory of inferred causation 3. Causal diagrams and the identification of causal effects 4. Actions, plans, and direct effects 5. Causality and structural models in the social sciences 6. Simpson's paradox, confounding, and collapsibility 7. Structural and counterfactual models 8. Imperfect experiments: bounds and counterfactuals 9. Probability of causation: interpretation and identification Epilogue: the art and science of cause and effect.
12,606 citations
"Interpretation and Identification o..." refers background or methods in this paper
...It can be shown that the availability of such intermediate measurements can make up for the unobservability of all factors that confound T and M (Morgan and Winship, 2007, Ch. 3; Pearl, 2000a, Ch. 3)....
[...]
...10, we can condition on W = w, and, using the truncated product formula (Pearl, 2000a, p. 72), we can write CDE(m) = E[Y | do(T = 1,M = m)]− E[Y | do(T = 0,M = m)] = ∑ w E[Y | T = 1,M = m,W = w]P (T = 1,W = w) − E[Y | T = 0,M = m,W = w]P (T = 0,W = w) The reason such conditioning does not work for…...
[...]
...The separation condition in (40) is somewhat stronger than (36), since the former implies the latter for every pair of values, t and t∗, of T (see (Pearl 2000, p. 214))....
[...]
...…regression analysis of mediation, most notably the one advanced by Baron and Kenny (1986), can be stated as follows: To test the contribution of a given mediator M to the effect of T on Y , first regress Y on T and estimate the regression coefficient RY T , to be equated with the total effect τ ....
[...]
...Substituting (39) into (35) and using the law of composition E(Yt∗) = E(Yt∗Mt∗ ) (Pearl 2000, p. 229) gives (37), and completes the proof of Theorem 2....
[...]
TL;DR: A discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation is presented in this paper, where the objective is to specify the benefits of randomization in estimating causal effects of treatments.
Abstract: A discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation is presented. The objective is to specify the benefits of randomization in estimating causal effects of treatments. The basic conclusion is that randomization should be employed whenever possible but that the use of carefully controlled nonrandomized data to estimate causal effects is a reasonable and necessary procedure in many cases. Recent psychological and educational literature has included extensive criticism of the use of nonrandomized studies to estimate causal effects of treatments (e.g., Campbell & Erlebacher, 1970). The implication in much of this literature is that only properly randomized experiments can lead to useful estimates of causal effects. If taken as applying to all fields of study, this position is untenable. Since the extensive use of randomized experiments is limited to the last half century,8 and in fact is not used in much scientific investigation today,4 one is led to the conclusion that most scientific "truths" have been established without using randomized experiments. In addition, most of us successfully determine the causal effects of many of our everyday actions, even interpersonal behaviors, without the benefit of randomization. Even if the position that causal effects of treatments can only be well established from randomized experiments is taken as applying only to the social sciences in which
8,377 citations
"Interpretation and Identification o..." refers background or methods in this paper
...…from the perspective of non-parametric structural equation models.5 This approach integrates the potential outcome framework of Neyman (1923) and Rubin (1974) with that of SEM, thus combining mathematical rigor with the merits of staying intimately informed by the data-generating process or its…...
[...]
...Counterfactual thinking in statistics (Holland, 1988; Rubin, 1974) and epidemiology (Robins and Greenland, 1992), together with a formal semantics based on non-parametric structural equations (Balke and Pearl, 1995; Halpern, 1998; Pearl, 2001) have given causal mediation analysis a sound theoretical basis and have extended its scope from linear to nonlinear models....
[...]
...Counterfactual thinking in statistics (Holland, 1988; Rubin, 1974) and epidemiology (Robins and Greenland, 1992), together with a formal semantics based on non-parametric structural equations (Balke and Pearl, 1995; Halpern, 1998; Pearl, 2001) have given causal mediation analysis a sound…...
[...]