Is it possible to truly measure teacher effectiveness? To help educators develop pedagogically? To ensure students are provided the very best classroom instruction possible? Well, that depends on how you define effectiveness and on what tools one uses. If we rely strictly on teacher observation data, the predictability of teacher effectiveness is limited in terms of student performance on state or other assessment measures. However, if we combine teacher observation with student survey data and previous value-added state test results, then we have a more vigorous measure with good reliability and predictability. At least that is what the Measures of Effective Teaching (MET) Project is finding in their comprehensive study.
The Met Project is an exciting, unique study funded by the Bill and Melinda Gates Foundation seeking to define the measures necessary to accurately and reliably describe effective teaching. Unique in the variety of indicators used (five classroom observation instruments, student surveys (Tripod Survey), and value-added state test results); unique in scale (3,000 teachers, 22,500 observation scores, 900 trained observers, and 44,500 student surveys and supplemental assessments); and unique in variety of student outcomes (math and state assessment gains, Balanced Assessment in Math, Stanford 9 Reading Test, and student-reported outcomes such as effort and enjoyment in class). Findings from the MET Project may provide districts and state education departments leverage for education reform throughout the nation.
Each of the Project’s measures has different strengths and weaknesses, but combined, they have a cocktail effect for defining teacher effectiveness. If we focus on student assessment results, value-added results offer the greatest predictive power on future state tests followed by student surveys. Classroom observation has the lowest predictive power of the three measures. However, classroom observation offers the highest potential for diagnosing teacher effectiveness and guiding teacher development. Student survey data has moderate diagnostic potential, with value-added measures offering little potential. In terms of reliability, all three measures have moderate to high ratings. However, the complexities of teacher observation require specific steps to realize strong reliability.
The MET Project is a must-read for school educators and administrators. Besides presenting a strong case for multiple measures in teacher effectiveness, it provides strategies for developing inter-rater consistency within schools and resources on student surveys. Our WSWHE BOCES Network Team has created a webcast on the MET Project along with supporting documents from content presented by Senior Regents Research Fellow Amy McIntosh at the January RTTT Network Team Training Institute. We know how powerfully important teachers are to student achievement, and the high stakes nature of Race to the Top and school reform ask that we measure teacher effectiveness carefully with validity and reliability. As the nation undergoes a transformation in curriculum through the Common Core State Standards, now more than ever we must look at multiple measures for defining teacher effectiveness.