NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Why High-order Polynomials Should not be Used in Regression Discontinuity Designs

Andrew Gelman, Guido Imbens

NBER Working Paper No. 20405
Issued in August 2014
NBER Program(s):Labor Studies

It is common in regression discontinuity analysis to control for high order (third, fourth, or higher) polynomials of the forcing variable. We argue that estimators for causal effects based on such methods can be misleading, and we recommend researchers do not use them, and instead use estimators based on local linear or quadratic polynomials or other smooth functions.

download in pdf format
   (382 K)

email paper

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w20405

Users who downloaded this paper also downloaded* these:
Card, Lee, Pei, and Weber w18564 Nonlinear Policy Rules and the Identification and Estimation of Causal Effects in a Generalized Regression Kink Design
Abadie, Athey, Imbens, and Wooldridge w20325 Finite Population Causal Standard Errors
Gelman and Imbens w19614 Why ask Why? Forward Causal Inference and Reverse Causal Questions
Imbens w19983 Instrumental Variables: An Econometrician's Perspective
Lee and Lemieux w14723 Regression Discontinuity Designs in Economics
 
Publications
Activities
Meetings
NBER Videos
Themes
Data
People
About

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us