Sifting Statistical Significance From the Artifact of Regression- Discontinuity Design

摘要: When the covariate of an evaluation study using regression analysis is falhbly measured, statistical test program effectiveness biased. In programs where target population "below average, " bias tends to suppress beneficial effects program, while raising them in designed for those "above average. The magnitude this calculated, and a correction method derived illustrated. many cases, these biases will make practical differences assessment, lending further support general distrust nonexperimental methods. Nonetheless, our approach improves reliability nonexperimen tal methods by correcting one potential source their bias.