This preview shows page 1. Sign up to view the full content.
Unformatted text preview: s generally regarded as having the greatest internal validity of all
o Its external validity is more limited, since the estimated treatment effect is local
to the discontinuity.
o But many instrumental variables (IV) estimators are also thought to estimate
Local Average Treatment Effect (LATE) Fortin – Econ 560 Lecture 0 Regression Discontinuities Design (RDD) Basic Idea: Sometimes whether something happens to you or not depends on your
‘score’ on a particular variable e.g you get a scholarship if you get above a certain mark in an exam, you get given remedial education if you get below a certain level, a policy is implemented if it gets more than 50% of the vote in a ballot, your sentence for a criminal offence is higher if you are above a certain age (an
‘adult’) your welfare assistance payment increase if you are above a certain age All these are potential applications of the ‘regression discontinuity’ design RDDs abound once you looked for them because program resources often
allocated based on a formula with a cut-off structure The idea was introduced in psychology by Thistlethwaite and Campbell (1960), but
not widely used until the 1990s T&C were asking: What is the impact of National Merit Award on students’ success
in obtaining additional college scholarships and their career aspirations? Fortin – Econ 560 Lecture 0 RDD: Award given to all students achieving a minimum score on a scholarship
exam Assignment rule: Score ≥ Min Score Award;
Score < Min Score No Award T&C noted that we could learn about the impact of award receipt for persons near
the cut-off. Under certain comparability conditions, assignment near the cut-off can be seen as
behaving as if random.
o Treatment group = “just above” cut-off and received award
o Control group = “just below” cut-off and did not receive award In a RDD, the assignment to the treatment is determined, either completely (sharp
design) or partly (fuzzy design), by the value of an assignment (or forcing) variable
(the covariate X) being on either side of a...
View Full Document
This document was uploaded on 02/26/2014 for the course ECON 560 at The University of British Columbia.
- Fall '13