Beyond LATE: Estimation of the Average Treatment Effect with an Instrumental Variable

Political scientists frequently use instrumental variables (IV) estimation to estimate the causal effect of an endogenous treatment variable. However, when the treatment effect is heterogeneous, this estimation strategy only recovers the local average treatment effect (LATE). The LATE is an average treatment effect (ATE) for a subset of the population: units that receive treatment if and only if they are induced by an exogenous IV. However, researchers may instead be interested in the ATE for the entire population of interest. In this article, we develop a simple reweighting method for estimating the ATE, shedding light on the identification challenge posed in moving from the LATE to the ATE. We apply our method to two published experiments in political science in which we demonstrate that the LATE has the potential to substantively differ from the ATE.

[1]  S. Naidu,et al.  Economic Determinants of Land Invasions , 2010, The Review of Economics and Statistics.

[2]  I K Fodor,et al.  A Survey of Dimension Reduction Techniques , 2002 .

[3]  J. Angrist,et al.  Extrapolate-Ing: External Validity and Overidentification in the Late Framework , 2010 .

[4]  D. Lazer,et al.  Estimating Treatment Effects in the Presence of Noncompliance and Nonresponse: The Generalized Endogenous Treatment Model , 2011, Political Analysis.

[5]  D. Rubin.  Should observational studies be designed to allow lack of balance in covariate distributions across treatment groups? , 2009 .

[6]  Donald B. Rubin,et al.  Bayesian Inference for Causal Effects: The Role of Randomization , 1978 .

[7]  D. Rubin,et al.  Addressing complications of intention-to-treat analysis in the combined presence of all-or-none treatment-noncompliance and subsequent missing outcomes , 1999 .

[8]  Model Averaging Methods for Weight Trimming in Generalized Linear Regression Models. , 2009, Journal of official statistics.

[9]  Joseph G. Altonji,et al.  The Effects of Family Characteristics on the Return to Education , 1996 .

[10]  Alberto Abadie.  Bootstrap Tests for Distributional Treatment Effects in Instrumental Variable Models , 2002 .

[11]  A. Deaton.  Instruments of Development: Randomization in the Tropics, and the Search for the Elusive Keys to Economic Development , 2009 .

[12]  Marie Davidian,et al.  Doubly robust estimation of causal effects. , 2011, American journal of epidemiology.

[13]  J. Angrist,et al.  Two-Stage Least Squares Estimation of Average Causal Effects in Models with Variable Treatment Intensity , 1995 .

[14]  Kevin Arceneaux,et al.  Who Is Mobilized to Vote? A Re-Analysis of 11 Field Experiments , 2009 .

[15]  Joshua D. Angrist,et al.  Identification of Causal Effects Using Instrumental Variables , 1993 .

[16]  D. Rubin,et al.  Assessing the effect of an influenza vaccine in an encouragement design. , 2000, Biostatistics.

[17]  M. Humphreys.  Bounds on least squares estimates of causal effects in the presence of heterogeneous assignment probabilities , 2009 .

[18]  Joseph W Hogan,et al.  Principal stratification with predictors of compliance for randomized trials with 2 active treatments. , 2008, Biostatistics.

[19]  Dean A. Follmann,et al.  On the Effect of Treatment among Would-Be Treatment Compliers: An Analysis of the Multiple Risk Factor Intervention Trial , 2000 .

[20]  D. Rubin,et al.  Bayesian inference for causal effects in randomized experiments with noncompliance , 1997 .

[21]  James J Heckman,et al.  Comparing IV with Structural Models: What Simple IV Can and Cannot Identify , 2009, Journal of econometrics.

[22]  Lesley Schulte,et al.  Bowling Alone: The Collapse and Revival of American Community , 2009 .

[23]  Marshall M Joffe,et al.  Weighting in instrumental variables and G‐estimation , 2003, Statistics in medicine.

[24]  G. Imbens,et al.  Better Late than Nothing: Some Comments on Deaton (2009) and Heckman and Urzua (2009) , 2009 .

[25]  Joshua D. Angrist,et al.  Mostly Harmless Econometrics: An Empiricist's Companion , 2008 .

[26]  A. Wald.  The Fitting of Straight Lines if Both Variables are Subject to Error , 1940 .

[27]  S. Geman,et al.  Nonparametric Maximum Likelihood Estimation by the Method of Sieves , 1982 .

[28]  Steven J. Rosenstone,et al.  Mobilization, participation, and democracy in America , 1993 .

[29]  David W. Nickerson,et al.  Getting Out the Vote in Local Elections: Results from Six Door-to-Door Canvassing Experiments , 2003 .

[30]  Marshall M Joffe,et al.  The compliance score as a regressor in randomized trials. , 2003, Biostatistics.

[31]  David Card,et al.  WORKING PAPER SERIES BETTER LATE THAN NOTHING : SOME COMMENTS ON DEATON ( 2009 ) AND HECKMAN AND URZUA , 2022 .

[32]  R. Little,et al.  Inference for the Complier-Average Causal Effect From Longitudinal Data Subject to Noncompliance and Missing Data, With Application to a Job Training Assessment for the Unemployed , 2001 .

[33]  David W. Nickerson,et al.  Comparing Negative and Positive Campaign Messages , 2010 .

[34]  D. Horvitz,et al.  A Generalization of Sampling Without Replacement from a Finite Universe , 1952 .