| Author/Presenter |
Dan Black (Syracuse University) |
| Co-authors |
Jose Galdo (Syracuse University) |
| |
Jeffrey Smith (University of Michigan) |
| Title |
Evaluating the Regression Discontinuity Design Using Experimental Data |
| Abstract |
The regression discontinuity (RD) design has recently become a standard method for identifying causal effects for policy interventions. We use an unusual "tie breaking" experiment, the Kentucky Working Profiling and Reemployment Services, to investigate selection bias in the RD approach. Two features characterize this program. First, the treatment (reemployment services) is assigned as a discontinuous function of a profiling variable (expected benefit receipt duration), which allows the identification of an experimental sample and two alternative non-experimental groups. Second, we deal with a discontinuity frontier rather than a discontinuity point, which allows the identification of marginal average treatment effects over a wide range of the support of the discontinuous variable. Using a variety of multivariate parametric and nonparametric kernel estimators, we estimate the bias with respect to the benchmark experimental estimates. In general, we find that the RD estimates are sensitive to the sample used in the estimations, the outcome of interest, and the econometric models. |
|