Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies

American Political Science Review - Tập 105 Số 4 - Trang 765-789 - 2011
Kosuke Imai1, Luke Keele2, Dustin Tingley3, Teppei Yamamoto4
1Princeton University#TAB#
2Pennsylvania State Univ#TAB#
3Harvard University
4Massachusetts Institute of Technology,

Tóm tắt

Identifying causal mechanisms is a fundamental goal of social science. Researchers seek to study not only whether one variable affects another but also how such a causal relationship arises. Yet commonly used statistical methods for identifying causal mechanisms rely upon untestable assumptions and are often inappropriate even under those assumptions. Randomizing treatment and intermediate variables is also insufficient. Despite these difficulties, the study of causal mechanisms is too important to abandon. We make three contributions to improve research on causal mechanisms. First, we present a minimum set of assumptions required under standard designs of experimental and observational studies and develop a general algorithm for estimating causal mediation effects. Second, we provide a method for assessing the sensitivity of conclusions to potential violations of a key assumption. Third, we offer alternative research designs for identifying causal mechanisms under weaker assumptions. The proposed approach is illustrated using media framing experiments and incumbency advantage studies.

Từ khóa


Tài liệu tham khảo

10.2307/2669290

10.1007/978-1-4419-1764-5_8

10.2307/440290

10.1080/02699939508408966

10.1097/00001648-199203000-00013

10.2307/271055

10.2307/2111475

Collier, 2004, Rethinking Social Inquiry:Diverse Tools, Shared Standards

10.1017/S0003055409090212

Brady, 2004, Rethinking Social Inquiry: Diverse Tools, Shared Standards

10.2307/2111633

10.1111/j.1540-5907.2008.00353.x

Kinder, 1996, Divided by Color: Racial Politics and Democratic Ideals

10.3162/036298006X201896

10.1111/j.1468-2508.2006.00452.x

10.1257/jep.24.3.3

10.2307/270723

10.1017/S0003055401003045

10.1177/0002716209351526

10.1177/1740774507083434

Robins, 2010, Causality and Psychopathology: Finding the Determinants of Disorders and Their Cures, 103

10.1214/10-STS321

10.2307/2647913

10.1037/a0018933

Robins, 2003, Highly Structured Stochastic Systems, 70, 10.1093/oso/9780198510550.003.0007

10.1111/j.1540-5907.2010.00446.x

Imai K. , and Yamamoto T. . 2011. “Sensitivity Analysis for Causal Mediation Effects under Alternative Exogeneity Assumptions.” http://imai.princeton.edu/research/medsens.html. (accessed September 1, 2011).

10.1146/annurev.polisci.10.072805.103054

10.1257/jel.48.2.424

10.1002/sim.3016

Imai K. , Tingley D. , and Yamamoto T. . N.d. “Experimental Designs for Identifying Causal Mechanisms.” (With discussions). Journal of the Royal Statistical Society, Series A (Statistics in Society). Forthcoming.

10.1126/science.1113551

Jacobson, 1987, The Politics of Congressional Elections

10.2307/2669316

Glynn A. N. 2010. “The Product and Difference Fallacies for Indirect Effects.” Department of Government, Harvard University. Unpublished manuscript, Mimeo.

10.1017/S0003055409090066

King, 1994, Designing Social Inquiry, 10.1515/9781400821211

10.1007/978-1-4615-0569-3_3

10.1037/a0014207

10.2307/2952075

10.1214/ss/1177012031

10.1037/0022-3514.89.6.845

10.1093/pan/mpl013

10.1097/EDE.0b013e31818ef366

Imai K. , and Tingley D. . N.d. “A Statistical Method for Empirical Testing of Competing Theories.” American Journal of Political Science. Forthcoming.

10.1097/EDE.0b013e31818f69ce

10.1002/sim.3493

10.1097/01.ede.0000208475.99429.2d

10.3102/1076998607307239

10.1126/science.1157627

10.2307/1952717

MacKinnon, 2008, Introduction to Statistical Mediation Analysis

10.2307/2960149

10.1037/h0037350

10.1111/j.1467-985X.2007.00527.x

10.1017/S0022381609990910

10.2307/2669312

10.1080/01621459.1996.10476902

10.2307/2939040

10.2307/1905714

10.1111/j.1540-5907.2007.00274.x

10.1037/0022-3514.81.6.973

10.1177/0146167207301028

10.1257/jep.9.2.85

Rosenbaum, 2002, Covariance Adjustment in Randomized Experiments and Observational Studies: Rejoinder, Statistical Science, 17, 321, 10.1214/ss/1042727942

Shadish, 2001, Experimental and Quasi-Experimental Designs for Generalized Causal Inference

Manski, 2007, Identification for Prediction and Decision

Rosenbaum, 2002, Covariance Adjustment in Randomized Experiments and Observational Studies (with Discussion), Statistical Science, 17, 286, 10.1214/ss/1042727942

10.2307/1953807

10.2307/2585672

10.1257/0002828042002561

10.1214/08-AOS613

10.1080/01621459.1986.10478354

Pearl J. N.d. “The Causal Mediation Formula: A Guide to the Assessment of Pathways and Mechanisms.” Prevention Science. Forthcoming.

10.1037/a0020761

Druckman, 2005, Media Matter: How Newspapers and Television News Cover Campaigns and Influence Voters, American Political Science Review, 22, 463

10.1037/0022-3514.51.6.1173

Pearl, 2001, Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence, 411

Imai K. , Keele L. , Tingley D. , and Yamamoto T. . 2011. “Replication Data for: Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies.” The Dataverse Network. hdl:1902.1/16467 (accessed September 1, 2011).

Thông tin
Thông tin xuất bản

American Political Science Review

Tập 105 Số 4

765-789

Thông tin tác giả