Getting Teacher Observations Right: A Balance of Design, Implementation, and Use

Ever since the federal Race to the Top competition and Elementary & Secondary Education Act (ESEA) waivers began incentivizing states to develop more rigorous, multi-measure teacher evaluation systems, much of the surrounding debate has focused on the inclusion of student learning growth as one of the measures within these systems.  Meanwhile, classroom observation measures—on which the bulk of most teachers’ final evaluation ratings are based—have received much less attention. Recently, however, several education policy and research groups have recommended improvements to the design of observation measures. Observation design is clearly important—but is it sufficient to ensure that classroom observations positively impact teacher quality and student learning?

TNTP’s “Fixing Classroom Observations” report is the most recent call to rethink how new observation measures are designed. The report advocates for improving teacher observation scoring rubrics by: 1) ensuring they assess whether appropriate-level content is being taught and 2) focusing on a handful of skills students should be demonstrating or outcomes they should be accomplishing during a lesson, rather than on which strategies teachers are using to try to elicit student skills and outcomes.  To inform teacher development, TNTP recommends that each student-centered objective still be paired with a list of potential strategies that teachers could employ to achieve it. The organization plans to release a prototype of such a rubric early next year for public comment.

By changing rubrics in these ways, TNTP contends that observations will improve in several ways, including:

• Better integrating grade-level Common Core State Standards (CCSS) into observation rubrics;
• Reducing administrative burden for observers, given fewer observation items to rate;
• Better supporting teacher development, by focusing observation and supports on attaining a small list of desired student outcomes instead of myriad instructional strategies; and
• Increasing accuracy of observation ratings.

Some might wonder whether states and districts should consider making changes while many schools are still getting comfortable with the observation measures currently in place. But TNTP says four actions can be taken now without major disruption to current systems: 1) ensuring that lesson content carries significant weight in the observation rating; 2) consolidating observation items that are redundant; 3) eliminating items that can’t be directly observed in a classroom visit (e.g., collegiality); and 4) giving observers—and their managers—data on the  quality and quantity of feedback they provide to teachers, and eventually factoring these data into observers’ own performance evaluations.

While evaluation design is important, implementation is even more so.

The first of these near-term actions is immensely important to ensure that all students are being taught the content they need to master at their grade level, and one that has been missing from most discussions about aligning teacher evaluations and CCSS. The second and third actions just seem like common sense. But it’s the goal of the fourth action—to focus on how observations are implemented and used—that I’m afraid still isn’t getting enough attention. As TNTP’s report acknowledges, while evaluation design is important, implementation is even more so—and thus, “rubrics are only as effective as the observers who use them and the systems that support them.” Which begs the question, can classroom observations—regardless of how well they’re designed –be used to effectively evaluate and improve teachers’ performance if there aren’t incentives for evaluators to perform observations with fidelity and use the results to guide instructional changes?

Practitioners, state policy advisors, and “education reformers”—who often don’t see eye-to-eye—seem to agree that observers (who are most often school principals) should be well-qualified to do both of these things. But, to date, most state and local policymakers have not taken steps to ensure that that they are. Some states require observers to pass a certification test (e.g., Kentucky) and a few others mandate that observers be held accountable for delivering constructive feedback and development plans (e.g., North Carolina). But many states leave observer training up to individual school districts—perhaps not surprising given that 21 states require districts to design their own evaluation systems, and play only a small role in implementation—regardless of district capacity. And many districts are more focused on ensuring that observers complete observations and submit ratings than that they provide feedback to help improve instruction.

How can we encourage more states and districts to pay attention to the important issues of teacher observation implementation and use? The answers aren’t readily available, and perhaps this is why many have opted to focus on aspects of observation measures (like design) that seem easier to fix. But I’ll throw out a few ideas. As the U.S. Department of Education continues its ESEA waiver monitoring process, observer quality could be an area that it requests information about and provides feedback on as it scrutinizes states’ implementation of teacher and principal evaluation systems. And if there is a next round of ESEA waiver renewals, it’s an area worth considering putting some muscle behind.