Value-Added Estimates of Teacher Effectiveness: Measurement, Uses, and Limitations

Abstract

Since the turn of the 21st century, an abundant body of research has demonstrated that teachers meaningfully contribute to their students’ learning but that teachers vary widely in their effectiveness. Measures of teachers’ “value added” to student achievement have become common, and sometimes controversial, tools for researchers and policymakers hoping to identify and differentiate teachers’ individual contributions to student learning. Value-added measures aim to identify how much more a given teacher’s students learn than what would be expected based on how much other, similar students learn with other teachers. The question of how to measure value added without substantial measurement error and without incorrectly capturing other factors outside of teachers’ control is complex and sometime illusory, and the advantages and drawbacks to any particular method of estimating teachers’ value added depend on the specific context and purpose for their use. Traditionally, researchers have calculated value-added scores only for the subset of teachers with students in tested grades and subjects—a relatively small proportion of the teaching force, in a narrow set of the many domains on which teachers may influence their students. More recently, researchers have created value-added estimates for a range of other student outcomes, including measures of students’ engagement and social-emotional learning such as attendance and behavioral incidences, which may be available for more teachers. Overall, teacher value-added measures can be useful tools for understanding and improving teaching and learning, but they have substantial limitations for many uses and contexts.