Join the conversation

...about what is working in our public schools.

Teacher Evaluations: Where Do We Go From Here?

obriena's picture

Teacher evaluation is an extremely hot topic these days. Districts from Houston to DC  – and states from Indiana to Washington – are experimenting with new ways to measure teacher performance. But, as Dr. Laura Goe (principal investigator for the National Comprehensive Center for Teacher Quality and research scientist in the Performance Research Group at ETS) pointed out recently at the Learning First Alliance’s annual Leadership Council meeting, “Policy is way ahead of the research in teacher evaluation measures and models.”

In reviewing the research to date on teacher evaluations, Dr. Goe reminded us that we don’t really know which evaluation model and/or combination of measures will identify effective teachers. And what stuck out to me from her presentation: She, a very accomplished researcher on this issue, doesn’t promote any of the models of teacher evaluation currently in use.

Of course, as she pointed out, it is hard to develop or support an evaluation that can measure whether an educator is effective, given that we don’t have a good definition* of what an effective teacher is. As she said, “Much of the research on teacher effectiveness doesn’t define effectiveness at all, though it is often assumed to be teachers’ contribution to student achievement.”

That is certainly the theory behind value-added modeling (VAM), one of the most widely promoted – and debated – measures touted for teacher evaluations. (For those unfamiliar, value-added modeling attempts to measure a teacher’s contribution to student learning in a given year by comparing students’ test scores in that year to the scores of those same students the previous year.)

Dr. Goe pointed out many of the challenges associated with VAM. For example, these models really measure classroom – not teacher – effects. They do not shed light on why a particular teacher’s students score the way they do (for example, is a teacher focusing instruction narrowly on test content?) – information that is critically important if we hope to truly impact teaching practice at a large scale. And they have not been shown to be consistent over time – teachers' ratings fluctuate greatly from one year to the next, adding important considerations to questions such as, “What is the consequence of one bad evaluation?”

Even assuming one supports the idea of VAM, there are practical constraints with these models. Research suggests that about 69% of teachers can’t be accurately assessed with VAMs or other growth models, including teachers in subject areas that are not tested with annual standardized tests and teachers in grade levels where no prior test scores are available (such as lower elementary). There are also questions about the validity of measuring special education and ELL teachers with VAMs.

Of course, identifying concerns about VAM and other proposed measures for evaluating teachers does not mean that one is happy about the current state of teacher evaluations. No one finds the current state of most teacher evaluation systems acceptable. But where do we go from here?

Dr. Goe did not have an answer. But she did offer suggestions for how districts and states interested in reforming teacher evaluations could proceed, including:

  • Engaging stakeholders (teachers, administrators, parents, school board members, union representatives, business leaders, etc.) in decision-making processes early and often
  • Conserving resources by joining forces with others in this endeavor
  • Considering whether human resources and capacity are sufficient to ensure fidelity of implementation
  • Evaluating the process and data each year and make needed adjustments

To me, these seem a common-sense approach to any new project. Yet in many states and districts that are being compelled by federal incentives and state legislation to move forward with new teacher evaluation systems, these steps can get skipped. Hopefully, decision makers will begin to realize this, and slow down the development of these systems. The process shouldn’t be rushed. We don’t want to replace one failed evaluation system with one that is just as bad.

 

*In part because of a lack of a universal definition of an effective teacher, she (with some colleagues) developed one back in 2008. It involves five components:

  1. Have high expectations for all students and help students learn, as measured by value-added or alternative measures.
  2. Contribute to positive academic, attitudinal, and social outcomes for students, such as regular attendance, on-time promotion to the next grade, on-time graduation, self-efficacy, and cooperative behavior.
  3. Use diverse resources to plan and structure engaging learning opportunities; monitor student progress formatively, adapting instruction as needed; and evaluate learning using multiple sources of evidence.
  4. Contribute to the development of classrooms and schools that value diversity and civic-mindedness.
  5. Collaborate with other teachers, administrators, parents, and education professionals to ensure student success, particularly the success of students with special needs and those at high risk for failure.

It's interesting how

It's interesting how complicated those five components are. A few of them contain multiple embedded components. The prominent placement of the term "value-added measures" is also puzzling, given the serious limitations of the concept and the measurement of it (partially detailed above).

I do like the idea of engaging everyone in the conversation rather than having important decisions made behind closed doors. The involvement of teachers in the process and the implementation is one of the key principles in a teacher evaluation policy report that I helped produce with the network Accomplished California Teachers; the report can be found at our website: http://acteachers.org

"All I need to do is outrun

"All I need to do is outrun you." That is the punch line from the old joke about two hikers that observed an angry bear approaching. One hiker kneels down to change his hiking boots for running shoes. His partner laughs and says, "You can't outrun a bear!" His reply, " I know, all I need to do is outrun you." This is the scenario we are creating in schools with standardized teacher evaluation systems. The result will not increase student achievement, the result will be for teachers to compete against one another. Teachers will opt for easier teaching assignments and "better test taking" students to raise their performance score. The consensus in school reform and increasing achievement, is that improvement comes from collaboration, sharing practices and group problem solving. I fear current state evaluation systems will drive teachers apart rather than urge them to work together. I am putting on my running shoes!

The most important

The most important consideration in teacher evaluation, both for improvement purposes and for personnel decisions, is the use of multiple methods of teaching evaluation involving multiple sources of data. The central element is the assessment of the quality of classroom instruction.