Policy Implications of Different Approaches to Describing the Accuracy of School and Classroom Environment Measures

• In Event: Policy Issues Related to Instructional Practices and School Environments

Abstract

Measuring school and classroom environments has become central in an international effort to develop policy that can stimulate the development of conditions that support student achievement and teacher retention. Formulating successful policies necessitates accurate and reliable methods for measuring these environmental variables. This paper contrasts two widely used reliability coefficients and shows how in certain circumstances, they may overestimate reliability and lead to spurious conclusions about relationships between school climate and policy-relevant outcomes. Additionally, this paper shows how those coefficients cannot answer many questions that are of interest to policy makers. This paper proposes investigating reliability under the broader, more flexible framework of generalizability theory.

Author

• Jonathan Schweig, University of California - Los Angeles