Abstract

http://ssrn.com/abstract=2305788
 
 

References (34)



 


 



Double Sampling for Missing Outcome Data in Randomized Experiments


Peter M. Aronow


Yale University - Department of Political Science

Alan Gerber


Yale University - Department of Political Science

Donald P. Green


Columbia University - Department of Political Science

Holger L. Kern


Florida State University - Department of Political Science

August 4, 2013


Abstract:     
Missing outcome data plague many randomized experiments. Common solutions such as selection, imputation, and reweighting models rest on untestable assumptions that may not be credible. We propose a measurement and estimation strategy for dealing with missing outcome data that makes minimal assumptions but still yields substantively informative bounds on the average treatment effect. Our approach is based on a combination of the double sampling design and non-parametric worst case bounds. We develop an R package to calculate the ex ante optimal design when experimenters expect some fraction of outcome data to be missing. Across a range of scenarios, double sampling is shown to reduce the width of the confidence interval for the identification region for the average treatment effect.

Number of Pages in PDF File: 20

working papers series





Download This Paper

Date posted: August 5, 2013 ; Last revised: August 10, 2013

Suggested Citation

Aronow, Peter M. and Gerber, Alan and Green, Donald P. and Kern, Holger L., Double Sampling for Missing Outcome Data in Randomized Experiments (August 4, 2013). Available at SSRN: http://ssrn.com/abstract=2305788 or http://dx.doi.org/10.2139/ssrn.2305788

Contact Information

Peter Michael Aronow
Yale University - Department of Political Science ( email )
P.O. Box 208301
New Haven, CT 06520-8269
United States
Alan Gerber (Contact Author)
Yale University - Department of Political Science ( email )
Box 208269
New Haven, DC 06520-8269
United States
203-432-5232 (Phone)
Donald P. Green
Columbia University - Department of Political Science ( email )
7th Floor, International Affairs Bldg.
420 W. 118th Street
New York, NY 10027
United States
Holger L. Kern
Florida State University - Department of Political Science ( email )
Tallahassee, FL 30306
United States
Feedback to SSRN


Paper statistics
Abstract Views: 591
Downloads: 104
Download Rank: 155,128
References:  34
Paper comments
No comments have been made on this paper

© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright   Contact Us
This page was processed by apollo2 in 0.329 seconds