Search results for "regression discontinuity design"
showing 4 items of 4 documents
Low-risk isn’t no-risk: Perinatal treatments and the health of low-income newborns
2019
We investigate the effects of perinatal medical treatments on low-income newborns who are classified as low-risk. A policy rule in The Netherlands states that low-risk deliveries before week 37 should be supervised by physicians and later deliveries only by midwives with no physician present. This creates large discontinuities in the probability of receiving medical interventions only physicians are allowed to perform. Using a regression discontinuity design, we find that babies born slightly before the week-37 cutoff are significantly less likely to die than babies born slightly later. Our data suggest that physician supervision of birth reduces the likelihood of adverse events such as fet…
Effectiveness of Private and Public High Schools: Evidence from Finland
2019
Abstract A number of papers have compared the effectiveness of private and public schools in different institutional settings. However, most of these studies are observational and do not utilize experimental or quasi-experimental design to evaluate the value-added or the effectiveness of private schools in comparison to public schools. This study focuses on private and public high schools in Helsinki, the capital city of Finland. We use two different methods to compare private and public schools, value-added estimation and regression discontinuity design (RDD). Although based on somewhat different assumptions, both methods allow us to evaluate the causal effect of private schools on the exi…
Does Regression Discontinuity Design Work? Evidence from Random Election Outcomes
2014
We use data for 198121 candidates and 1351 random election outcomes to estimate the effect of incumbency status on future electoral success. We find no evidence of incumbency advantage using data on randomized elections. In contrast, regression discontinuity design, using optimal bandwidths, produces a positive and significant incumbency effect. Using even narrower bandwidths aligns the results with those obtained using the randomized elections. So does the bias-correction of Calonico et al. (forthcoming). Standard validity tests are not useful in detecting the problems with the optimal bandwidths. The appropriate bandwidth seems narrower in larger elections and is thus context specific.
When does Regression discontinuity design work? Evidence from random election outcomes
2018
We use elections data in which a large number of ties in vote counts between candidates are resolved via a lottery to study the personal incumbency advantage. We benchmark non‐experimental regression discontinuity design (RDD) estimates against the estimate produced by this experiment that takes place exactly at the cutoff. The experimental estimate suggests that there is no personal incumbency advantage. In contrast, conventional local polynomial RDD estimates suggest a moderate and statistically significant effect. Bias‐corrected RDD estimates that apply robust inference are, however, in line with the experimental estimate. Therefore, state‐of‐the‐art implementation of RDD can meet the re…