Can Rigorous, Observation-based Teacher Evaluations Move the Needle on Student Achievement?

In the wake of the federal government’s 2010 Race to the Top initiative, states and local school districts have dramatically revised their teacher evaluation systems. These new systems incorporate more rigorous performance evaluation through the use of multiple measures of teacher performance (value-added measures based on student test scores and classroom observations based on standards-based rubrics) and multiple ratings categories aimed at differentiating teacher effectiveness. By the start of the 2014–15 school year, 78% of states and 85% of the largest 25 districts and DC revised and implemented new systems. California is among the few states that have not instituted statewide teacher evaluation reforms.

Despite the intense focus on using student test scores to measure teacher performance, the majority of teachers receive annual evaluation ratings based primarily on observations of classroom practice. So, even as states and districts have amended their evaluation systems, classroom observation remains the most important component of a teacher’s evaluation rating. This is particularly notable since upwards of 70% of teachers nationwide teach in nontested grades or subjects and, therefore, do not have student test score data from which to construct value-added scores.

At the time that states and districts were designing and implementing their new teacher evaluation systems, little evidence existed about how rigorous, process-oriented teacher evaluation affects student performance. Today, however, we can say with some confidence that teacher observations designed to capture the quality of within-classroom teacher–student interactions can improve teaching practice and, ultimately, student outcomes. In this study, our findings are based on a study of the Excellence in Teaching Project (EITP), an intensive teacher evaluation program implemented in Chicago Public Schools. The program involved classroom observation and principal–teacher conferencing as well as extensive training for principals on using the Danielson Framework for Teaching observation protocol.

With Lauren Sartain, a researcher at the University of Chicago Consortium on School Research, I examined the effect of the program on instructional practice and student achievement. The pilot began in 44 elementary schools in 2008–09, and scaled up to include an additional 48 elementary schools the next school year. Leveraging the experimental nature of the rollout, we found that treatment schools—those that participated in the EITP evaluation system in 2008–09—performed better in reading and math than the control schools at the end of the first year, though the math effects are not statistically significant. Further, this improvement persisted even after the control schools adopted the EITP evaluation program in 2009-10.

While our findings, which were recently published in Education Finance and Policy, make it clear that rigorous observations can lead to improved student achievement, they also indicate that certain conditions must be met in order to achieve these results.

First, observations of teacher instructional practice require a tremendous expenditure of time by the observers—usually principals—and the teachers themselves. Likewise, districts must commit significant resources to train observers and provide ongoing support. Simply put, if observers are not provided adequate supports, we shouldn’t expect to see meaningful changes in teacher practice and student achievement.

Second, the provision of adequate resources appears to be particularly important for lower-performing, higher-poverty schools. We found that these schools made significantly less progress than more advantaged schools. Presumably, to make gains in these schools would require more resources. And yet school leaders in such schools likely already do a significant amount of juggling of their time, with little capacity to train for and implement the time-intensive evaluation system.

From a policy perspective, the past 5 years have seen states and school districts make systematic and systemic changes to the ways that teachers are evaluated. As we found in Chicago, whether and how we support principals and school leaders responsible for observing teacher practice matters for the efficacy of teacher evaluation reforms. As California considers revisions to its teacher evaluation system, state policymakers should provide the necessary resources so that education professionals responsible for observing teacher practice can do so in ways that support the work of teachers toward improving student achievement.

The full study is in Matthew P. Steinberg and Lauren Sartain, “Does Teacher Evaluation Improve School Performance? Experimental Evidence from Chicago's Excellence in Teaching Project,” Education Finance and Policy Fall 2015, Vol. 10, No. 4, Pages 535-572