Horseshoes, Hand Grenades, and Treatment Effects? Reassessing Bias in Nonexperimental Estimators
Working Paper 16
Publisher: Oakland, CA: Mathematica Policy Research
Mar 30, 2013
Nonexperimental methods, such as regression modeling or statistical matching, produce unbiased estimates if the underlying assumptions hold, but these assumptions are usually not testable. Most studies testing nonexperimental designs find that they fail to produce unbiased estimates, but these studies have examined weaker evaluation designs. This working paper addresses these limitations and finds the use of baseline data that are strongly predictive of the key outcome measures considerably reduces bias, but might not completely eliminate it.
Charter Schools: Are They Effective?
U.S. Department of Education
You may also like...
What Works for Whom? A Bayesian Approach to Channeling Big Data Streams for Public Program Evaluation
Progress and Challenges in Developing Tiered Quality Rating and Improvement Systems in the Round 1 Race to the Top—Early Learning Challenge States (In Focus Brief)
Additional Working Papers