Exploring the Potential Role of Staff Surveys in School Leader Evaluation

The Mid-Atlantic Regional Educational Laboratory partnered with the District of Columbia Public Schools (DCPS) to explore the potential use of teacher surveys in school leader evaluation. The DCPS evaluation system, like many others, currently consists of two components: an assessment on how well a school performs on a set of student achievement metrics (such as proficiency on standardized tests) and an assessment by a supervisor of the principal's leadership across multiple domains. Incorporating teacher surveys could provide an additional perspective on principals' leadership and performance. Examining data from two teacher surveys that DCPS has used (Panorama and Insight), the study found that:

Most of the domains and scales from the teacher surveys can meaningfully differentiate among schools.

School averages of (Insight) survey domains are more similar for adjacent years in which a school had the same leader than when the school leader changed, suggesting principals have effects on these measures.

When a school gets a new principal, (Insight) survey domain scores change more than student proficiency rates, suggesting (consistent with theory) that principals may affect teachers more immediately than they affect student outcomes. All but two of the teacher survey domains and scales have small or moderate correlations with supervisors' ratings of principals' leadership, with values ranging from 0.26 to 0.46. Teacher survey measures are most correlated with three of the domains assessed by DCPS principal supervisors: Instruction, Talent, and School Culture.

Overall, our findings suggest that it could be useful for DCPS to use elements of teacher surveys to bring in teachers' perspectives on principals' leadership related to instruction, talent, and school culture. Other districts may also wish to consider employing teacher surveys to gain an additional perspective on principals from staff who interact with the principal every day.