Do Teacher Effectiveness Measures Exhibit Differential Validity? Evidence From the Measures of Effective Teaching Database

• In Event: 31.091 - Roundtable Session 10
  In Roundtable Session: 31.091-9 - Accountability: Issues of Measurement and Model Quality

Abstract

The current study uses an orthogonal regression technique proposed by Kane and Mroch (2010) to determine whether the relationship between effectiveness measures and value-added scores are different for initially low-, average-, and high-performing classrooms in the MET dataset. To do so, we examine the relationship between observation scores, student surveys, and teacher value-added scores on two reading and two mathematics assessments for initially high–achieving, moderately achieving and low-achieving elementary school classrooms in which students were randomly assigned to teachers.

Authors

• Megan E. Welsh, University of California - Davis

• Eva Yujia Li, Yale University

• Christopher H Rhoads, University of Connecticut