Making evaluation really count

See all posts

A new study finds that a high caliber teacher evaluation system can produce a rather unexpected and measurable benefit. Teachers can significantly improve their instruction, and not just in the short-term, but make improvements that really stick.

Two researchers took advantage of Cincinnati's multi-year phase-in of its new teacher evaluation system to conduct an empirical test of the impact of teacher evaluations on student learning. They found that student learning increased slightly in the year a teacher is evaluated, but that the trend continues and even deepens with time—even though teachers are only evaluated once every 3 to 5 years in the district.

For example, a teacher whose students had typically scored in the 50th percentile on a math test before being evaluated was able to achieve math gains more in the 55th percentile range in the years after evaluation.

Further, contradicting a whole lot of evidence finding little growth in teacher effectiveness after the first few years on the job, teachers, no matter how long they had been teaching, got better. Best of all, teachers who scored in the lowest quartile on their evaluations showed the greatest improvements in their teaching.

The authors theorize that the high quality of Cincinnati's system produced these great results, and point to several of its key elements:

  • specific, prescriptive feedback on each point on which teachers are evaluated;

  • an easy-to-understand rubric in which "distinguished" performance is clearly separated from "basic;"

  • feedback from multiple sources, including both administrators and peers.

The price tag for the Cincy system is hefty, as much as $7,500 per teacher per evaluation. Perhaps well worth the price.