Testing, Accountability, and School Improvement

Abstract

The provision of public schooling in the United States has primarily been the states’ responsibility, but states generally lack the capacity to manage day-to-day school operations. Thus, states delegate responsibility to districts while maintaining some oversight. Forms of oversight include regulations and political and market-based accountability. However, these can only do so much in holding schools accountable for providing high-quality schooling. Administrative accountability based on student outcomes and school process measures presents an alternative to complement other accountability mechanisms. Standardized measures of performance used for administrative accountability can better align curriculum with state standards, improve quality, and signal the skills that society wishes for students to build. However, they can be counterproductive if they are not reliable, valid, or comprehensive. We suggest in this article that no measure is perfect and that the usefulness of test-based accountability depends on whether the measures enhance educational opportunities and reflect shared goals with reliability, validity, and comprehensiveness.