Working Paper: NBER ID: w30803
Authors: Joshua Angrist; Peter Hull; Christopher R. Walters
Abstract: Many personal and policy decisions turn on perceptions of school effectiveness, defined here as the causal effect of attendance at a particular school or set of schools on student test scores and other outcomes. Widely-disseminated school ratings frameworks compare average student achievement across schools, but uncontrolled differences in means may owe more to selection bias than to causal effects. Such selection problems have motivated a wave of econometric innovation that uses elements of random and quasi-experimental variation to measure school effectiveness. This chapter reviews these empirical strategies, highlighting solved problems and open questions. Empirical examples are used throughout.
Keywords: school effectiveness; causal inference; econometric methods; lottery-based analysis
JEL Codes: C11; C26; I20; I21; I24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| school attendance (I21) | student achievement (I24) |
| KIPP charter schools attendance (I21) | test scores (C52) |
| lottery assignments (H27) | school attendance (I21) |
| school attendance (I21) | test scores (C52) |