Good Schools VI / Multiple Measures of Teacher Performance: What Does It Mean? How Is It Implemented?

Meeting Overview and Purpose

Eugenia Kemble, Executive Director, Albert Shanker Institute

How Should Teacher Effectiveness Be Evaluated? Creating a Strong Teacher Evaluation System Using Multiple Measures That Are Fair, Relevant and Reliable

Linda Darling Hammond, Charles E. Ducommon Professor of Education, Stanford University; member, Albert Shanker Institute Board of Directors (via teleconference)

The quest to define and measure teacher effectiveness has sparked useful research on many different fronts, using different means to gauge various important outcomes. How should we think about assembling a strong body of evidence demonstrating effective teaching practice as teachers move through their careers – at induction; to achieve tenure; as an ongoing evaluation procedure; and as a way to identify the “master teachers,” mentors, and peer evaluators who can help sustain systemic improvement? And how do we address the underlying problems with our school system which limit the value of these measures – lack of a rich, common student curriculum; inequities in resource allocations; the summer learning slump and other differences in out-of-school educational experiences, etc.

Keynote and Dinner

Randi Weingarten, President, Albert Shanker Institute and American Federation of Teachers

This is an opportunity for relaxed, cross-district and team discussions. Teams from each district are strongly encouraged to use some of this time to identify their reform priorities in regard to teacher evaluation and discuss the kinds of information that they might want to obtain (both from experts and from peers) in order to move forward.

Measures of Student Progress

Robert Linn, Co-Director, National Center for Research on Evaluation, Standards, and Student Testing (CRESST); Distinguished Professor Emeritus of Education, University of Colorado at Boulder

This session tackled the tricky issue of incorporating student achievement measures into evaluations of teacher performance. What are the strengths and limitations of using student assessment data, as they are currently gathered and reported? What is the effect of the lack of a common student curriculum on which to base assessments? What is the best that can be hoped for from the forthcoming assessments aligned to the new Common Core Standards? How does this relate to value added measures? What other less controversial measures – graduation rates, end-of-course exams, term papers, portfolios of students’ work – should also be considered? What are the strengths and weaknesses of the various options? And which are the best existing systems (and systems under development) that might serve as useful models?

Respondents:
Andres Alonso, Chief Executive Officer, Baltimore City Public Schools
Anna Brown, Director for Assessment and Performance Management Empowering Effective Teachers Grant, Hillsborough County Public Schools

Measures of Teaching Practice

Raymond Pecheone, Director, Performance Assessment for California Teachers (PACT); Executive Director, Stanford Center for Assessment, Learning, & Equity (SCALE)

This session will delved more deeply into the multiple measures of teaching practice – observation protocols, tests of pedagogical content knowledge, portfolios of teachers’ and students’ work. Which are the most relevant and reliable for evaluating teachers? How does the lack of a common student curriculum limit our ability to judge teaching quality? What are the strengths and weaknesses of the-available options? And which are the best existing systems (and systems under development) that might serve as useful models?

Respondent:
Dan Montgomery, President, Illinois Federation of Teachers

Measures of Teaching Conditions

Eric Hirsch, Chief Officer, External Affairs, New Teacher Center

According to one recent study, variations in school conditions may account for much as 25 percent of teacher effects on student learning. While it may seem obvious that teacher working conditions affect student learning, how can they be measured and accounted for in teacher evaluation? What are the possible sources of data – parent surveys, teacher surveys, principal surveys, student surveys, school discipline incident reports, school demographics, school funding – and which have been found to be correlated with variations in student achievement? What are the strengths and weaknesses of the instruments and of the research behind them? How can these (often preliminary) data best be used, both to increase the accuracy of teacher evaluation systems and to tailor and improve the instructional supports available to teachers?

Respondent:
Clifford Janey, Superintendent, Newark Public Schools; member, Albert Shanker Institute Board of Directors

Putting It All Together

Linda Darling Hammond, Charles E. Ducommon Professor of Education, Stanford University; Albert Shanker Institute Board of Directors

In the quest to devise a more effective teacher evaluation system, there is an inherent tension between the need for relevancy and the need for accuracy. How much should we rely on student test scores and other “objective” measures, which most experts consider to be inadequate measures of student learning? How do we weight respected measures of teaching practice which have not yet been correlated with improvements in student performance? This session will pull together what we have learned about the relevancy and accuracy of different types of measures across the categories, as well as the potential benefits and drawbacks to giving more and less weight to each of these, both individually and in combination with one another.

Facilitator:
Marcia Reback, President, Rhode Island Federation of Teachers and Health Professionals