Thank you


Sign up to receive updates from the Gates Foundation

Bill & Melinda Gates Foundation

Can Teacher Evaluation Make Teachers Better?

September 04, 2012

As students and teachers return to school this fall, more than seven out of every ten states and countless districts have now promised to modernize their approach to teacher evaluation. Many jurisdictions are well into the work of design or early implementation. Yet, today, very little is known about how revitalized evaluation might change teacher effort or skill.

To start filling that knowledge gap, we have been studying a decade’s worth of data on teacher performance from the public schools in Cincinnati, Ohio. In the late 1990s leaders from the local teachers’ union and school district collaborated to develop the Teacher Evaluation System. The program’s experienced peer evaluators, detailed scoring rubric, multiple classroom observations, and feedback conferences foreshadowed many of the features that have become widely popular today. Thus the experience of Cincinnati’s teachers can provide a glimpse of what the future could hold for teachers elsewhere.

The results are encouraging. The average mid-career teacher was much more successful at helping her students make learning gains after she had gone through the evaluation program. She was also a more effective teacher during the school year she was being evaluated.

How big were the changes? Imagine two average students taught by the same teacher in different school years, but who both began the year with the same level of math knowledge. The student taught by the teacher after she went through the evaluation process would score about 4.5 percentile points higher at the end of the year than the student taught by the teacher before she went through the evaluation. In rough terms, that is equivalent to getting an extra 3-4 months of school time.

These findings contrast sharply with most data on how teacher performance changes over the course of a career; most researchers have found that the effectiveness of individual teachers does not improve much after the first few years on the job. The teachers in our study were all in the middle of their careers, most had 10-20 years experience at the time of their evaluation. We also find that the post-evaluation improvements in performance were largest for teachers whose performance was weakest prior to evaluation.

Hillsborough County, Florida recently implemented a new evaluation system that is, in many respects, quite similar to Cincinnati’s, and the early anecdotal evidence is complementary. “When Melinda and I talked to the students in Hillsborough,” Bill Gates reported recently, “they said that their teachers are changing the way they teach because of the feedback they’re getting from their peer [evaluators]…and the students can feel the difference.”

In many school systems today, experienced teachers rarely receive individualized, specific information about their performance. This suggests that a lack of information on where a teacher could improve and how she might become more effective could be a substantial barrier to individual improvement for teachers who are past their first few years on the job. Well-designed evaluation, like the Cincinnati approach, might make teachers more successful because it helps provide such information along with the opportunity to discuss it with an experienced colleague.

You can read more about Cincinnati’s program, our research methods, and results in the current issue of Education Next. The full study, “The Effect of Evaluation on Teacher Performance,” will be published in the coming months in the American Economic Review.

blog comments powered by Disqus