Skip Navigation
Evaluation Studies of the National Center for Education Evaluation and Regional Assistance

Impact Evaluation of Teacher and Leader Performance Evaluation Systems

Contractor: American Institutes for Research, Instructional Research Group, University of Virginia, Teachstone, Discovery Education Association, Teachscape, The Danielson Group

Background/Research Questions:

Educator performance evaluation systems are a potential tool for improving student achievement through increasing the effectiveness of the teacher and principal workforce. The Every Student Succeeds Act of 2015 allows the use of Title II Part A funds for the development and support of evaluation systems. The current study is designed to examine the implementation of a package of performance evaluation system components and the impact of their use for formative purposes. These are components that states and districts might include in their evaluation systems. The evaluation system components include measures of student achievement growth, classroom observations and measures of principal leadership. Based on these measures, teachers, leaders and districts received timely and constructive feedback on teacher and principal performance. The components were implemented by the study's implementation team in a subset of schools in a sample of districts that did not already have an evaluation system similar to that being studied.

The study addresses the following key research questions:

  • What is the impact of the performance evaluation system on student achievement? On teachers' classroom practices?
  • What are districts' and educators' experiences with implementation?

Design:

Eight districts participated in the study. Within each district, approximately 15 schools were randomly assigned to receive the study's treatment during 2012–2013 and 2013–2014 or to participate only in the district's usual performance evaluation system during the same time period. In treatment schools, each year, teachers received four rounds of classroom observations and feedback sessions, based on one of two observation rubrics (the Framework for Teaching or the CLASS). District preference determined which rubric was used. Teachers also received information about their students' achievement growth in math and/or reading using a statistical model known as value-added. Principals in study schools received feedback on their performance based on the VAL-ED, which was administered twice per study year. The VAL-ED is an assessment of principals' leadership practices based on survey responses from the principal, their teachers, and their district supervisor. The study looks at the extent to which the measures were implemented as intended and whether the ratings from the measures reliably distinguish performance. It examines the extent to which educators' experiences and perceptions regarding performance evaluation differed between treatment and control schools. And, it examines impacts on teachers' classroom experiences, principals' leadership practices, and students' achievement.

Cost/Duration: $21,523,477 over 6 years (September 2011–September 2017)

Current Status: A report on the study's first year of implementation was released on November 2, 2016 (http://ies.ed.go/ncee/pubs/20174004/). A report on impacts on educator practices and student achievement as well as implementation during the study's second year is expected in Fall 2017. The report will be announced on http://ies.ed.gov/ncee/

Key Findings:

The study's report on the first year of implementation found that:

  • The study's performance measures were implemented generally as planned. Teachers and principals received multiple rounds of ratings and feedback on their practices. However, fewer principals and teachers accessed their value-added reports than the study intended.
  • Both classroom observation and student growth measures differentiated teacher performance, although observation scores were mostly at the upper end of the scale. Overall, observation scores varied across teachers, and both value-added scores and average classroom observation scores over the year had sufficient reliability to capture performance differences among some teachers.
  • The principal leadership measure differentiated performance, but principal self-ratings, teachers' ratings of the principal, and the principal's supervisor's ratings of the principal often differed.
  • Both teachers and principals in schools selected to implement the intervention reported receiving more feedback on their performance than did their counterparts. For example, teachers and principals in intervention schools reported spending more total time in performance feedback sessions across the year than teachers and principals in the control schools.