摘要：This paper develops a measure of efficiency when data have been aggregated. Unlike the mostcommonly used efficiency measures, our estimator handles the heteroskedasticity created byaggregation appropriately. Our estimator is compared to estimators currently used to measureschool efficiency. Theoretical results are supported by a Monte Carlo experiment. Results showthat for samples containing small schools (sample average may be about 100 students per schoolbut sample includes several schools with about 30 students), the proposed aggregate dataestimator performs better than the commonly used OLS and only slightly worse than themultilevel estimator. Thus, when school officials are unable to gather multilevel or disaggregatedata, the aggregate data estimator proposed here should be used. When disaggregate data isavailable, standardizing the value-added estimator should be considered.
