As Michigan student achievement continues to fall behind a growing number of other states, it’s clear that Michigan needs to support teachers better to improve instruction. Developmental feedback, in the form of a well-crafted, annual teacher evaluation, is an important first step toward that goal. Echoing their peers in other states, many Michigan educators say helpful, routine evaluations and useful professional development have been rare for much of their careers.

In an effort to give teachers the feedback and training they need to improve, the Michigan legislature passed a law in 2009 requiring local school districts and charter schools to evaluate all teachers every year, taking into account how much students learned. Since then, districts and charter schools have worked to develop their own evaluation models, often struggling mightily to ensure that the complexity and difficulty of teachers’ work is taken into account.

Recognizing that struggle, the Michigan legislature returned to evaluation reform in 2011, creating the Michigan Council for Educator Effectiveness. The council of state-appointed education experts is charged with developing a statewide system of educator evaluation, including: Michigan’s first common definition of what effective teaching looks like and a statewide evaluation model that any district or charter school in the state can use if it chooses. In addition, for those jurisdictions that want to develop their own models, the council is developing a set of state standards that all districts and charters would have to meet to have their models approved.

Still, a reasonable person might ask, is all this state-level action really necessary? Aren’t our local districts and charter schools capable of deciding how to get their teachers the feedback and training they need to grow? To answer that question, the Education Trust–Midwest examined the teacher evaluation models now being used across Michigan.

The Education Trust–Midwest wanted to see how Michigan schools are responding to demands for developmental, technically sound — in other words, smart — evaluation practices that provide high-quality professional development and feedback for our state’s teachers. If we are to raise student achievement in our state, then we must do more to support and develop our educators’ capacity to teach at higher levels. With this in mind, we reviewed local evaluation models adopted by 28 Michigan districts and charter schools of different sizes and capacities across the state. We then asked, “Do they measure up?” To help answer this question, we looked to best practices according to national research, lessons learned by other states and districts, and practitioners’ recommendations. This report summarizes our conclusion, which is that most local models — despite the hard work that has gone into them — do not measure up to reasearch-based standards for smart evaluation.

Included among our findings from the Michigan district and charter school models we examined:

  • Almost 20 percent used checklist-style teacher observation tools with no opportunity for rich developmental feedback for teachers.
  • Almost half allowed, or did not explicitly prevent, tenured or experienced teachers to go unobserved for an entire school year.
  • Only 18 percent used the state’s standardized tests to measure individual teachers’ impact on student learning. State assessments are designed to measure how well students are learning Michigan’s curriculum to ensure all students are getting rigorous, high-quality instruction to prepare them for an extraordinarily competitive global economy. Neglecting to use these available assessments leaves Michigan parents with no confidence that their children are learning what they are supposed to learn in school.
  • None used a student growth or value-added model that was technically-sound enough to reliably gauge teachers’ impact on student learning. Such measures are needed to provide rich feedback to teachers — and actually protect them from arbitrary evaluations.
  • The majority, 61 percent, did not provide clear guidance to evaluators on how to combine the many measures of teaching performance into a final rating. This means administrators are more likely to produce unreliable or inaccurate final evaluation ratings — which may be risky for teachers, as these ratings will have a profound impact on their careers and futures.
  • No model created a master or mentor teacher status or training to empower highly effective teachers to become observers in the evaluation process, which would help local schools manage the increased workload that meaningful evaluation may create.

Many of the district and charter school leaders we spoke with say they’ve long needed guidance to improve teacher evaluations. In other words, some state action is essential to protect everyone’s best interests — that is good for teachers, students and administrators.

To that end, the Michigan Council for Educator Effectiveness (MCEE) is developing a teacher evaluation system for the state. In the coming months, the Michigan legislature will have the opportunity to adopt the council’s recommendations. This report uses the lessons learned from local evaluation models already being implemented across the state to inform the work of the council and share these lessons with educators across the state, who often do this hard work in isolation. We also recommend standards the state should adopt to ensure that even those districts and charter schools that opt out of the state system meet minimum criteria for smart evaluation.

It’s also important to note that, despite the flaws found in these local systems, district and charter leaders across Michigan say the newfound emphasis on evaluation is helping teachers grow. Indeed, progress in other states and pockets of evidence from within Michigan have convinced us that it is possible to give teachers the kind of developmental, supportive feedback and data they need to truly excel.


Published: November 29, 2012