Politicians and educators have struggled with the idea of developing useful yardsticks for school effectiveness that were honest, accurate and easy to comprehend. The initial stages of this effort often ignored the varying degree to which the social and political factors affected schools in various districts. Schools were frequently measured by a small set of benchmarks that often had no universal applicability. In other words, benchmarks were being used to rank order districts and schools as if the schools were competing on a level playing field. The rank-ordering yardstick by itself is not sensitive to differences in a school’s environment, its practices, or to its achievement results that are not part of the formal state assessment program. To help alleviate some of the problems in viewing raw achievement results, the researchers at the
Rhode Island Department of Education (RIDE) and the
National Center on Public Education and Social Policy
(NCPE) at the University of Rhode Island have devised a statistical model that attempts to address some of the differences in school composition. The remainder of this document is devoted primarily to the understanding of the statistical model – one of the key components of
Information Works!
|
