Have We Identified Effective Teachers? Validating Measures of Effective Teaching Using Random Assignment

To develop, reward, and retain great teachers, school systems first must know how to identify them. We designed the Measures of Effective Teaching (MET) project to test replicable methods for identifying effective teachers. In past reports, we described three approaches to measuring different aspects of teaching: student
surveys, classroom observations, and a teacher’s track record of student achievement gains on state tests (Kane and Staiger, 2010 & 2012). In those analyses, we could only test each measure’s ability to predict student achievement gains non-experimentally, using statistical methods to control for student background differences. For this report, we put the measures to a more definitive and final test. First, we used the data collected during 2009–10 to build a composite measure of teaching effectiveness, combining all three measures to predict a teacher’s impact on another group of students. Then, during 2010–11, we randomly assigned a classroom of students to each teacher and tracked his or her students’ achievement. We compared the predicted student outcomes to the actual differences that emerged by the end of the 2010–11 academic year.