Teaching to the Student: Charter School Effectiveness in Spite of Perverse Incentives

Abstract

Recent work has shown that Boston charter schools raise standardized test scores more than their traditional school counterparts. Critics of charter schools argue that charter schools create those achievement gains by focusing exclusively on test preparation, at the expense of deeper learning. In this paper, I test that critique by estimating the impact of charter school attendance on subscales of the Massachusetts Comprehensive Assessment System and examining them for evidence of score inflation. If charter schools are teaching to the test to a greater extent than their counterparts, one would expect to see higher scores on commonly tested standards, higher-stakes subjects, and frequently tested topics. Despite incentives to reallocate effort away from less frequently tested content to highly tested content, and to coach to item type, I find no evidence of this type of test preparation. Boston charter middle schools perform consistently across all standardized test subscales.