Association for Public Policy Analysis & Management

Since the passage of the federal No Child Left Behind Act over a decade ago, the primary lever that American policymakers have used to improve school performance is “accountability.” Accountability is understood to mean a regime of regular standardized testing of students coupled with public reporting of school-level results and sanctions for schools and staff for failure to achieve test-score targets. Accountability in the form of high-stakes testing is now pervasive, even as its efficacy is hotly contested. We argue that the policy debate can be informed by an extensive literature from behavioral science on accountability (Lerner & Tetlock, 1999). That literature makes clear, first, that accountability comes in many forms that activate different mechanisms, and second, that accountability can produce positive or negative effects, depending on the accountability type, the decision context, and the nature of the task. We discuss the implications of the behavioral literature for schooling, where the mix of public and private purposes suggests the need for accountability to multiple constituencies, including public officials, parents, and students. We conclude that an effective accountability regime will involve (1) multiple forms of accountability; (2) multiple measures of educational practice and educational outcomes; and (3) feedback mechanisms to promote the improvement of practice. Moreover, a multi-pronged accountability approach should specifically increase the use of professional accountability, which has historically been underutilized in schools.