In policy evaluations, interest may focus on why a particular treatment works. One tool for understanding why treatments work is causal mediation analysis. In this essay, I focus on the assumptions needed to estimate mediation effects. I show that there is no “gold standard” method for the identification of causal mediation effects. In particular, mediation effects will always have the character of estimates from observational data since they are generally subject to a specific form of confounding. Additionally, I demonstrate how randomization of the mediator and instrumental variables methods lead to fundamentally different quantities than causal mediation analyses. I also review recent work that discusses how the assumptions of mediation analyses differ when there is treatment noncompliance or when there are multiple mediators. Throughout, I motivate concepts using path diagrams and an example of a classroom intervention designed to increase test scores.
AngristJ. D.ImbensG. W.RubinD. B. (1996). Identification of causal effects using instrumental variables. Journal of the American Statistical Association, 91, 444–455.
2.
AvinC.ShpitserI.PearlJ. (2005). Identifiability of path-specific effects. In Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence (pp. 357–363). Edinburgh, Scotland. Morgan Kaufman.
3.
BaronR. M.KennyD. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.
4.
HeckmanJ.PintoR.SavelyevP. (2013). Understand the mechanisms through which an influential early childhood program boosted adult outcomes. American Economic Review, 103, 2052–2086.
5.
HollandP. W. (1986). Statistics and causal inference. Journal of the American Statistical Association, 81, 945–960.
6.
HuberM.LechnerM.MellaceG. (2015). The finite sample performance of estimators for mediation analysis under sequential conditional independence. Journal of Business & Economic Statistics. Advance online publication. doi:10.1080/07350015.2015.1017644
7.
ImaiK.KeeleL.TingleyD. (2010). A general approach to causal mediation analysis. Psychological Methods, 15, 309–334.
8.
ImaiK.KeeleL.TingleyD.YamamotoT. (2014). Comment on pearl: Practical implications of theoretical results for causal mediation. Psychological Methods, 19, 482–487.
9.
ImaiK.KeeleL.YamamotoT. (2010). Identification, inference, and sensitivity analysis for causal mediation effects. Statistical Science, 25, 51–71.
10.
ImaiK.TingleyD.YamamotoT. (2013a). Experimental designs for identifying causal mechanisms. Journal of the Royal Statistical Society: Series A (Statistics in Society), 176, 5–51.
11.
ImaiK.TingleyD.YamamotoT. (2013b). Experimental designs for identifying causal mechanisms (with discussions). Journal of the Royal Statistical Society, Series A (Statistics in Society), 176, 5–51.
12.
ImaiK.YamamotoT. (2013). Identification and sensitivity analysis for multiple causal mechanisms: Revisiting evidence from framing experiments. Political Analysis, 21, 141–171.
13.
KeeleL. J.TingleyD.YamamotoT. (2015). Identifying mechanisms behind policy interventions via causal mediation analysis. Journal of Policy Analysis and Management. Advance online publication. doi:10.1002/pam.21853
14.
MatzkinR. L. (2007). Nonparametric identification. In HeckmanJ. J.LeamerE. E. (Eds.), Chapter 73 in handbook of econometrics (Vol. 6b, pp. 5307–5368). Amsterdam: Elsevier.
15.
PearlJ. (1995). Causal diagrams for empirical research. Biometrika, 82, 669–710.
16.
PearlJ. (2001). Direct and indirect effects. In Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (pp. 411–420). San Francisco, CA. Morgan Kaufmann.
17.
PearlJ. (2014). Interpretation and identification of causal mediation. Psychological Methods, 19, 459–481.
18.
ReardonS. F.RaudenbushS. W. (2013). Under what assumptions do site-by-treatment instruments identify average causal effects?Sociological Methods & Research, 42, 143–163.
19.
ReardonS. F.UnluF.ZhuP.BloomH. S. (2014). Bias and bias correction in multisite instrumental variables analysis of heterogeneous mediator effects. Journal of Educational and Behavioral Statistics, 39, 53–86.
20.
RobinsJ. M. (2003). Semantics of causal DAG models and the identification of direct and indirect effects. In GreenP. J.HjortN. L.RichardsonS. (Eds.), Highly structured stochastic systems (pp. 70–81). Oxford, England: Oxford University Press.
21.
RubinD. B. (1986). Which ifs have causal answers. Journal of the American Statistical Association, 81, 961–962.
22.
RubinD. B. (1990). Comments on “On the application of probability theory to agricultural experiments. Essay on principles. Section 9” by J. Splawa-Neyman translated from the Polish and edited by D. M. Dabrowska and T. P. Speed. Statistical Science, 5, 472–480.
23.
RubinD. B. (2013). Discussion of: Experimental designs for identifying causal mechanisms. Journal of the Royal Statistical Society, Series A, 176, 5–51.
24.
ShapkaJ. D.KeatingD. P. (2003). Effects of a girls-only curriculum during adolescence: Performance, persistence, and engagement in mathematics and science. American Educational Research Journal, 40, 929–960.
25.
ShpitserI.VanderWeeleT. J. (2011). A complete graphical criterion for the adjustment formula in mediation analysis. The International Journal of Biostatistics, 7, 1–24.
