Exploring the impact and process of merit pay for public school teachers

Merit pay for public school teachers has been one of the most controversial educational policies in the US. Built on the free market principle, it is designed to reward teachers according to their performance and hence improve the quality of education they provide to students. Nevertheless, merit pay experiments seldom appear to be successful in schools. Notwithstanding the controversies and amount of dollars invested, there have been few systematic studies on the impact of such policies on student achievement and the conclusions come largely from reviews of the staying power of merit pay systems. This dissertation examines the impact of merit pay on student achievement with a three-prong approach, using both quantitative and qualitative methods. The first design, using national data from the Schools and Staffing Survey (1987--88), looks at the effect of merit pay on student college application rate and graduation rate. The second design, using panel data from Fairfax County Public Schools (FCPS), Virginia (1989--1996), examines how merit pay plans affect students' standardized test scores. The results suggest that merit pay has little, and sometimes even negative effect on student achievement. In addition, the study verifies that students' socio-economic status, class size and teacher education, each in its own way, have significant impact on student performance while the influence from teacher experience is often insignificant. The third prong of the dissertation is a case study of the merit pay experiment in the FCPS. It aims to uncover the "black box" of policy design and implementation by tracing the actions and interactions of major players in the process. The case study finds that the rise and fall of merit pay does not depend on whether it achieves its intended outcomes, but rather the changing political and economic support of such policy. The study provides insights on why merit pay doesn't work. The marginal productivity theory that merit pay is based on is flawed and cannot be applied to schools where outcomes rely on multiple factors. Implementation of merit pay plans that depend on excessive measurement and considerable financial investment also creates resource problems that make most programs short-lived. Finally, cyclical returns of merit pay simply reflect the frustration that effective means to improve student achievement remain illusive to educators and policymakers. The study discusses the implications of merit pay on recent developments in teacher compensation such as the national board certification and school-based incentives. While it indicates some of the promising aspects of the school-based incentives, it cautions that compensation reform is not a magic elixir that will cure all of education's ills and that any compensation reform must be accompanied with other reforms in instruction and management.