Principals are believed to have great influence on school climate, instructional practices, and student outcomes, but legitimate questions remain about how to measure their effectiveness.
In a new working paper from the Annenberg Institute, researchers analyze data from Tennessee, New York, and Oregon to understand if recent studies are accurately quantifying the impact of school leadership on student outcomes.
The authors–Brendan Bartanen of the University of Virginia, Aliza N. Husain of Pivot Learning, and David D. Liebowitz of the University of Oregon–found that such value-added models, a common method of analyzing principal performance, often attribute effectiveness to factors outside a principal’s control. After explicitly accounting for these dynamic factors, this paper shows that little to none of the variation in student performance is explained by differences between principals.
Ultimately, the authors conclude that value-added models do not provide valuable information about principal effectiveness and recommend states and districts avoid relying on test scores or attendance when evaluating leadership impact. They suggest a more useful tool might be the ratings that principals receive from their supervisors and from surveys of teachers.