Wanna Get Away? RD Identification Away from the Cutoff - Working Paper 18662

Sep 15, 2026

In the canonical regression discontinuity (RD) design for applicants facing an award or admissions cutoff, causal effects are nonparametrically identified for individuals near the cutoff. The paper examines identification away from the cutoff for inframarginal applicants, where estimating treatment effects requires stronger assumptions. It proposes an identification strategy using dependent-variable predictors other than the running variable, assuming the running variable is ignorable given these predictors. Evidence from Boston exam school applicants shows estimates that are precise and robust, and the approach extends to fuzzy RD designs to identify compliers’ effects away from the cutoff.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages49

Reading Time2 h 3 min

AccessFree to download

Abstract

Introduction

Regression discontinuity framework and local identification

Boston exam schools as an application

Motivation for identification away from the cutoff

NBER WORKING PAPER SERIES WANNA GET AWAY? RD IDENTIFICATION AWAY FROM THE CUTOFF Joshua Angrist Miikka Rokkanen Working Paper 18662 [http://www.nber.org/papers/w18662](http://www.nber.org/papers/w18662) NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 December 2012 Our thanks to Parag Pathak for many helpful discussions and comments, and to seminar participants at Berkeley, CREST, and Stanford for helpful comments. Thanks also go to Peter Hull for expert research assistance. Angrist gratefully acknowledges funding from the Institute for Education Sciences. The views expressed here are those of the authors alone and do not necessarily reflect the views of the National Bureau of Economic Research or The Institute for Education Sciences. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications. © 2012 by Joshua Angrist and Miikka Rokkanen. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. Wanna Get Away? RD Identification Away from the Cutoff Joshua Angrist and Miikka Rokkanen NBER Working Paper No. 18662 December 2012 JEL No. C26,C31,C36,I21,I24,I28,J24 ABSTRACT In the canonical regression discontinuity (RD) design for applicants who face an award or admissions cutoff, causal effects are nonparametrically identified for those near the cutoff. The impact of treatment on inframarginal applicants is also of interest, but identification of such effects requires stronger assumptions than are required for identification at the cutoff. This paper discusses RD identification away from the cutoff. Our identification strategy exploits the availability of dependent variable predictors other than the running variable. Conditional on these predictors, the running variable is assumed to be ignorable. This identification strategy is illustrated with data on applicants to Boston exam schools. Functional-form-based extrapolation generates unsatisfying results in this context, either noisy or not very robust. By contrast, identification based on RD-specific conditional independence assumptions produces reasonably precise and surprisingly robust estimates of the effects of exam school attendance on inframarginal applicants. These estimates suggest that the causal effects of exam school attendance for 9th grade applicants with running variable values well away from admissions cutoffs differ little from those for applicants with values that put them on the margin of acceptance. An extension to fuzzy designs is shown to identify causal effects for compliers away from the cutoff. Joshua Angrist Department of Economics MIT, E52-353 50 Memorial Drive Cambridge, MA 02142-1347 and NBER [angrist@mit.edu](angrist@mit.edu) Miikka Rokkanen Department of Economics MIT􀀏 (􀀘􀀕􀀐􀀖􀀜􀀔 50 Memorial Drive Cambridge, MA 02142-1347 [rokkanen@mit.edu](rokkanen@mit.edu) Both the tie-breaking experiment and the regression-discontinuity analysis are particularly subject to the external validity limitation of selection-X interaction in that theeﬀect has been demonstrated only for a very narrow band of talent, i.e., only for those at the cutting score... Broader generalizations involve the extrapolation of the below-X ﬁt across the entire range of X values, and at each greater degree of extrapolation, the number of plausible rival hypotheses becomes greater. – Donald T. Campbell and Julian Stanley (1963; Experimental and Quasi-Experimental Designs for Research) 1 Introduction In a regression discontinuity (RD) framework, treatment status changes discontinuously as a function of an underlying covariate, commonly known as the running variable. Provided conditional mean functions for potential outcomes given th

FAQs

What does the paper study about regression discontinuity (RD) identification?

It studies how RD identification can extend beyond the immediate neighborhood of the admissions cutoff, focusing on effects for inframarginal applicants.

Why is identification away from the cutoff more demanding than identification at the cutoff?

Because estimating treatment effects for inframarginal units needs stronger assumptions than those required for nonparametric identification at the cutoff.

What identification strategy does the paper propose?

It exploits predictors of the dependent variable besides the running variable, and assumes the running variable is ignorable conditional on these predictors.

Sep 15, 2026

Uploaded by Alex Sinclair