Title page -- Table of Contents -- Preface -- Part I: OVERVIEW -- 1 Introduction -- 1.1 Concepts of moderation, mediation, and spill-over -- 1.2 Weighting methods for causal inference -- 1.3 Objectives and organization of the book -- 1.4 How is this book situated among other publications on related topics? -- References -- 2 Review of causal inference concepts and methods -- 2.1 Causal inference theory -- 2.2 Applications to Lord's paradox and Simpson's paradox -- 2.3 Identification and estimation -- Appendix 2.1: Potential bias in a prima facie effect -- Appendix 2.2: Application of the causal inference theory to Lord's paradox -- References -- 3 Review of causal inference designs and analytic methods -- 3.1 Experimental designs -- 3.2 Quasiexperimental designs -- 3.3 Statistical adjustment methods -- 3.4 Propensity score -- Appendix 3.A: Potential bias due to the omission of treatment-by-covariate interaction -- Appendix 3.B: Variable selection for the propensity score model -- References -- 4 Adjustment for selection bias through weighting -- 4.1 Weighted estimation of population parameters in survey sampling -- 4.2 Weighting adjustment for selection bias in causal inference -- 4.3 MMWS -- Appendix 4.A: Proof of MMWS-adjusted mean observed outcome being unbiased for the population average potential outcome -- Appendix 4.B: Derivation of MMWS for estimating the treatment effect on the treated -- Appendix 4.C: Theoretical equivalence of MMWS and IPTW -- Appendix 4.D: Simulations comparing MMWS and IPTW under misspecifications of the functional form of a propensity score model -- References -- 5 Evaluations of multivalued treatments -- 5.1 Defining the causal effects of multivalued treatments -- 5.2 Existing designs and analytic methods for evaluating multivalued treatments -- 5.3 MMWS for evaluating multivalued treatments -- 5.4 Summary.

Appendix 5.A: Multiple IV for evaluating multivalued treatments -- References -- Part II: MODERATION -- 6 Moderated treatment effects: concepts and existing analytic methods -- 6.1 What is moderation? -- 6.2 Experimental designs and analytic methods for investigating explicit moderators -- 6.3 Existing research designs and analytic methods for investigating implicit moderators -- Appendix 6.A: Derivation of bias in the fixed-effects estimator when the treatment effect is heterogeneous in multisite randomized trials -- Appendix 6.B: Derivation of bias in the mixed-effects estimator when the probability of treatment assignment varies across sites -- Appendix 6.C: Derivation and proof of the population weight applied to mixed-effects models for eliminating bias in multisite randomized trials -- References -- 7 Marginal mean weighting through stratification for investigating moderated treatment effects -- 7.1 Existing methods for moderation analyses with quasiexperimental data -- 7.2 MMWS estimation of treatment effects moderated by individual or contextual characteristics -- 7.3 MMWS estimation of the joint effects of concurrent treatments -- References -- 8 Cumulative effects of time-varying treatments -- 8.1 Causal effects of treatment sequences -- 8.2 Existing strategies for evaluating time-varying treatments -- 8.3 MMWS for evaluating 2-year treatment sequences -- 8.4 MMWS for evaluating multiyear sequences of multivalued treatments -- 8.5 Conclusion -- Appendix 8.A: A saturated model for evaluating multivalued treatments over multiple time periods -- References -- Part III: MEDIATION -- 9 Concepts of mediated treatment effects and experimental designs for investigating causal mechanisms -- 9.1 Introduction -- 9.2 Path coefficients -- 9.3 Potential outcomes and potential mediators -- 9.4 Causal effects with counterfactual mediators.

9.5 Population causal parameters -- 9.6 Experimental designs for studying causal mediation -- References -- 10 Existing analytic methods for investigating causal mediation mechanisms -- 10.1 Path analysis and SEM -- 10.2 Modified regression approaches -- 10.3 Marginal structural models -- 10.4 Conditional structural models -- 10.5 Alternative weighting methods -- 10.6 Resampling approach -- 10.7 IV method -- 10.8 Principal stratification -- 10.9 Sensitivity analysis -- 10.10 Conclusion -- Appendix 10.A: Bias in path analysis estimation due to the omission of treatment-by-mediator interaction -- References -- 11 Investigations of a simple mediation mechanism -- 11.1 Application example: national evaluation of welfare-to-work strategies -- 11.2 RMPW rationale -- 11.3 Parametric RMPW procedure -- 11.4 Nonparametric RMPW procedure -- 11.5 Simulation results -- 11.6 Discussion -- Appendix 11.A: Causal effect estimation through the RMPW procedure -- Appendix 11.B: Proof of the consistency of RMPW estimation -- References -- 12 RMPW extensions to alternative designs and measurement -- 12.1 RMPW extensions to mediators and outcomes of alternative distributions -- 12.2 RMPW extensions to alternative research designs -- 12.3 Alternative decomposition of the treatment effect -- References -- 13 RMPW extensions to studies of complex mediation mechanisms -- 13.1 RMPW extensions to moderated mediation -- 13.2 RMPW extensions to concurrent mediators -- 13.3 RMPW extensions to consecutive mediators -- 13.4 Discussion -- Appendix 13.A: Derivation of RMPW for estimating population average counterfactual outcomes of two concurrent mediators -- Appendix 13.B: Derivation of RMPW for estimating population average counterfactual outcomes of consecutive mediators -- References -- Part IV: SPILL-OVER -- 14 Spill-over of treatment effects: concepts and methods.

14.1 Spill-over: A nuisance, a trifle, or a focus? -- 14.2 Stable versus unstable potential outcome values: An example from agriculture -- 14.3 Consequences for causal inference when spill-over is overlooked -- 14.4 Modified framework of causal inference -- 14.5 Identification: Challenges and solutions -- 14.6 Analytic strategies for experimental and quasiexperimental data -- 14.7 Summary -- References -- 15 Mediation through spill-over -- 15.1 Definition of mediated effects through spill-over in a cluster randomized trial -- 15.2 Identification and estimation of the spill-over effect in a cluster randomized design -- 15.3 Definition of mediated effects through spill-over in a multisite trial -- 15.4 Identification and estimation of spill-over effects in a multisite trial -- 15.5 Consequences of omitting spill-over effects in causal mediation analyses -- 15.6 Quasiexperimental application -- 15.7 Summary -- Appendix 15.1: Derivation of the weight for estimating the population average counterfactual outcome E[Y(1,p,0,M−(p))] -- Appendix 15.2: Derivation of bias in the ITT effect due to the omission of spill-over effects -- References -- Index -- End User License Agreement.