Skip to main content
×
Home
    • Aa
    • Aa

Explaining Causal Findings Without Bias: Detecting and Assessing Direct Effects

  • AVIDIT ACHARYA (a1), MATTHEW BLACKWELL (a2) and MAYA SEN (a2)
Abstract

Researchers seeking to establish causal relationships frequently control for variables on the purported causal pathway, checking whether the original treatment effect then disappears. Unfortunately, this common approach may lead to biased estimates. In this article, we show that the bias can be avoided by focusing on a quantity of interest called the controlled direct effect. Under certain conditions, the controlled direct effect enables researchers to rule out competing explanations—an important objective for political scientists. To estimate the controlled direct effect without bias, we describe an easy-to-implement estimation strategy from the biostatistics literature. We extend this approach by deriving a consistent variance estimator and demonstrating how to conduct a sensitivity analysis. Two examples—one on ethnic fractionalization’s effect on civil war and one on the impact of historical plough use on contemporary female political participation—illustrate the framework and methodology.

Copyright
Corresponding author
Avidit Acharya is Assistant Professor of Political Science, Stanford University (avidit@stanford.edu).
Matthew Blackwell is Assistant Professor of Government, Harvard University (mblackwell@gov.harvard.edu).
Maya Sen is Assistant Professor of Public Policy, Harvard University (maya_sen@hks.harvard.edu).
Footnotes
Hide All

Thanks to Adam Cohon, Allan Dafoe, Justin Esarey, Adam Glynn, Robin Harding, Gary King, Macartan Humphreys, Kosuke Imai, Bethany Lacina, Jacob Montgomery, Judea Pearl, Dustin Tingley, Teppei Yamamoto, and conference or workshop participants at Dartmouth, Harvard, Princeton, WashU, the Midwest Political Science Association meeting, and the Society for Political Methodology summer meeting for helpful discussions and comments. Thanks to Anton Strezhnev for valuable research assistance. Any remaining errors are our own. The methods in this article are available as an open-source R package, DirectEffects, at http://www.mattblackwell.org/software/direct-effects/. Code and data to replicate results in this article can be found at http://dx.doi.org/10.7910/DVN/VNXEM6.

Footnotes
Linked references
Hide All

This list contains references from the content that can be linked to their source. For a full set of references and notes please see the PDF or HTML where available.

Alberto Alesina , Paola Giuliano , and Nathan Nunn . 2013. “On the Origins of Gender Roles: Women and the Plough.” Quarterly Journal of Economics 128 (2): 469530.

Joshua D. Angrist , and Jörn-Steffen Pischke . 2008. Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton: Princeton University Press.

Matthew Blackwell . 2013. “A Framwork for Dynamic Causal Inference in Political Science.” American Journal of Political Science 57 (2): 504–20. http://www.mattblackwell.org/files/papers/dynci.pdf

Matthew Blackwell . 2014. “A Selection Bias Approach to Sensitivity Analysis for Causal Effects.” Political Analysis 22 (2): 169–82.

James D. Fearon , and David D. Laitin . 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97 (01): 7590.

Paul W. Holland 1986. “Statistics and Causal Inference.” Journal of the American Statistical Association 81 (396): 945–60. http://www.jstor.org/stable/2289064

Kosuke Imai , Luke Keele , Dustin Tingley , and Teppei Yamamoto . 2011. “Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies.” American Political Science Review 105 (04): 765–89.

Kosuke Imai , Dustin Tingley , and Teppei Yamamoto . 2013. “Experimental Designs for Identifying Causal Mechanisms.” Journal of the Royal Statistical Society. Series A (Statistics in Society) 176 (1): 551.

Kosuke Imai , and Teppei Yamamoto . 2013. “Identification and Sensitivity Analysis for Multiple Causal Mechanisms: Revisiting Evidence from Framing Experiments.” Political Analysis 21 (2): 141–71. http://pan.oxfordjournals.org/content/21/2/141.abstract

Marshall M. Joffe , and Tom Greene . 2009. “Related Causal Frameworks for Surrogate Outcomes.” Biometrics 65 (2): 530–8.

Marshall M. Joffe , Dylan Small , and Chi-Yuan Hsu . 2007. “Defining and Estimating Intervention Effects for Groups that will Develop an Auxiliary Outcome.” Statistical Science 22 (1): 7497.

Jerzy Neyman . 1923. “On the Application of Probability Theory to Agricultural Experiments. Essay on Principles. Section 9.” Statistical Science 5: 465–80. Translated in 1990, with discussion.

Judea Pearl . 2001. “Direct and Indirect Effects.” In Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence. UAI’01 San Francisco: Morgan Kaufmann Publishers Inc., 411–20. http://dl.acm.org/citation.cfm?id=2074022.2074073

James M. Robins 1994. “Correcting for Non-Compliance in Randomized Trials Using Structural Nested Mean Models.” Communications in Statistics 23 (8): 2379–412. http://www.hsph.harvard.edu/james-robins/files/2013/03/correcting-1994.pdf

James M. Robins 1997. “Causal Inference from Complex Longitudinal Data.” In Latent Variable Modeling and Applications to Causality, ed. M. Berkane . Vol. 120 of Lecture Notes in Statistics. New York: Springer-Verlag, 69117. http://biosun1.harvard.edu/~robins/cicld-ucla.pdf

James M. Robins 2003. “Semantics of Causal DAG Models and the Identification of Direct and Indirect Effects.” In Highly Structured Stochastic Systems, eds. P. J. Green , N. L. Hjort and S. Richardson . Oxford: Oxford University Press, 7081.

James M. Robins , Miguel A. Hernán , and Babette A. Brumback . 2000. “Marginal Structural Models and Causal Inference in Epidemiology.” Epidemiology 11 (5): 550–60. http://www.jstor.org/stable/3703997

Paul R. Rosenbaum 1984. “The Consequences of Adjustment for a Concomitant Variable That Has Been Affected by the Treatment.” Journal of the Royal Statistical Society. Series A (General) 147 (5): 656–66.

Donald B. Rubin 1974. “Estimating Causal Effects of Treatments in Randomized and Nonrandomized Studies.” Journal of Educational Psychology 6: 688701.

Tyler J. VanderWeele 2009. “Mediation and Mechanism.” European Journal of Epidemiology 24 (5): 217–24.

Tyler J. VanderWeele 2011. “Controlled Direct and Mediated Effects: Definition, Identification and Bounds.” Scandinavian Journal of Statistics 38 (3): 551–63.

Tyler J. VanderWeele and Eric J. Tchetgen Tchetgen . 2014. “Attributing Effects to Interactions.” Epidemiology 25 (5): 711–22.

Sijn Vansteelandt . 2009. “Estimating Direct Effects in Cohort and Case–Control Studies.” Epidemiology 20 (6): 851–60.

Sijn Vansteelandt . 2010. “Estimation of Controlled Direct Effects on a Dichotomous Outcome Using Logistic Structural Direct Effect Models.” Biometrika 97 (4): 921–34.

Sijn Vansteelandt , and Marshall Joffe . 2014. “Structural Nested Models and G-estimation: The Partially Realized Promise.” Statistical Science 29 (4): 707–31.

Sijn Vansteelandt , and Tyler J. VanderWeele . 2009. “Conceptual Issues Concerning Mediation, Interventions and Composition.” Statistics and Its Interface 2: 457–68.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

American Political Science Review
  • ISSN: 0003-0554
  • EISSN: 1537-5943
  • URL: /core/journals/american-political-science-review
Please enter your name
Please enter a valid email address
Who would you like to send this to? *
×
Type Description Title
PDF
Supplementary Materials

Acharya supplementary material
Acharya supplementary material 1

 PDF (95 KB)
95 KB

Metrics

Altmetric attention score

Full text views

Total number of HTML views: 53
Total number of PDF views: 713 *
Loading metrics...

Abstract views

Total abstract views: 1608 *
Loading metrics...

* Views captured on Cambridge Core between 13th September 2016 - 21st July 2017. This data will be updated every 24 hours.