Bill & Melinda Gates Foundation

Measuring Effective Teaching: A Potential for Change

January 05, 2012

Why should anyone assume that improving teacher evaluation and feedback could greatly improve K-12 education? After all, the basic form of teacher evaluation has changed little in a generation.  Most teachers have experienced the conventional “drop-in” classroom observation, a typically harmless and accepted part of school culture that results in satisfactory ratings for 99 percent or more of teachers.

From this perspective, it doesn’t seem like a particularly powerful lever for supporting teaching and learning.

Yet we know that existing teacher evaluation and feedback systems are not capturing large differences among teachers that can have crucial consequences for their students.

To appreciate the potential for change, you can look at the most recent findings from the Measures of Effective Teaching (MET) project . The most extensive study of its kind, MET is a research collaboration that analyzes data from nearly 3,000 U.S. public school teachers’ classrooms.

The MET data emphatically supports the conclusion that teaching practices matter.

Among the findings released last year, MET reported that students with top-quartile teachers learn far more than their peers with bottom-quartile teachers. Preliminary results estimated that in a single school year effective teaching practices can add “months” of learning -- as much as 7½ additional months in math and over a year of additional learning in writing.

In the second wave of findings, released in January 2012, MET examines classroom practice through the lens of several nationally recognized teaching observation instruments.

MET partners, along with education services firms Teachscape and Educational Testing Service (ETS), designed a rater training and certification system that produced highly reliable scores. MET’s findings suggest that the classroom practices of the majority of teachers, as many as 85 percent, are remarkably similar.

That is, most teachers employ a mix of strategies that are described by the observation instruments as either “basic” or “proficient”. The remaining 15 percent are roughly split into those whose practices are described as “accomplished”, and those whose practices are described as “harmful”. 

While some may find it disappointing that even more advanced feedback and evaluation systems still result in the same outcome -- a satisfactory rating for the majority of teachers -- what MET has found is more nuanced and hopeful.  

The MET data emphatically supports the conclusion that teaching practices matter and even small differences in practice are associated with small differences in student performance gains.

We estimated that if the teachers judged to be part of the 85 percent whose practice is described as “basic” or “proficient” raised their practice even minimally, nearly all students would receive instruction at or above the current system average.

It is not too optimistic to envision a system that supports teachers in continually improving their practice over time.

We know that teachers want better feedback. We know that teachers view their success in terms of their students’ success. Yet today, nearly all research shows the same trend – beginning teachers improve quickly in their initial years, but hit a plateau by their fourth or fifth year of practice.

Indeed, it is hard to improve what has not been adequately described or reliably measured. A good feedback and evaluation system clearly describes effective practices and how these practices are linked to student progress, for example.

Of course, an ill-designed or poorly administered feedback and evaluation system will not serve teachers or their students well. Many districts and states have undertaken courageous work to re-envision these systems. The MET project is not an attempt to identify the “one best system;” rather, it provides research insights and guidance for both design and implementation.

MET examines how to combine multiple measures, including classroom observations, student surveys, and student-achievement gains to get a more holistic view of teaching. The value of a multiple measures system is not that there are more measures, but that the measures work together to describe the aspects of teaching most important to student performance in ways that any single measure could not. Improvement does not result from measurement, however, but from the actions of teachers and administrators who value continued improvement.

An evaluation system will support teachers and improve student outcomes if it is built upon an understanding of what practices lead to greater student success, an assessment of the state of current practice, a plan to support teacher growth, and a way for the system and individual teachers to track progress for all students.

Check out this video - and listen to what teachers have to say!

blog comments powered by Disqus