Much attention has been paid to designing and implementing more rigorous teacher evaluation systems in an effort to shine a light on and improve teaching quality. But while principals play a central role in promoting successful teaching and learning, this role has been sorely overlooked. As with the leadership of any organization, principals should be held accountable for—and supported in accelerating—the learning growth of their staff, as well as of their students. But little consensus yet exists on how best to design and implement school principal evaluation and support systems. A few recent reports offer insight into the current state of affairs and pave a path forward for future research in this area.
Last month, the Institute for Education Sciences (IES) released a final report on Pennsylvania’s principal evaluation system pilot for the 2013-2014 school year. The state’s evaluation system for principals is similar to that for its teachers: half is based on measures of student achievement and the other half is based on a supervisor’s assessment of practice. For the latter, supervisors use a framework to observe and rate principals on 20 leadership practices such as “uses data for informed decision-making” and “ensures a high-quality, high-performing staff.” The authors found that the tool is somewhat valid and reliable. Which is to say, the tool partially differentiates between principals with high and low impact on student achievement, and principals’ scores in one year tend to predict their scores in the next.
Despite this, the authors also found that nearly all principals (95%) received one of the top two of four performance ratings across all practices—cause for some skepticism as to the scores’ accuracy. These findings suggest that the high practice scores may have less to do with the tool’s integrity and more to do with supervisor training and culture. Principal supervisors may not be adequately trained to rate principals accurately using the leadership framework and grade leniently—at least in part—as a result. The authors suggest that Pennsylvania use survey results from other stakeholders (e.g., teachers) to corroborate the other measures. Like student surveys in teacher evaluation, teacher surveys in principal evaluation could serve as a “check” on the student achievement and practice components and provide principals with more formative feedback.
Perhaps unsurprisingly, other evidence exists that Pennsylvania’s system could benefit from evaluator training and teacher surveys. Late last year, the National Council on Teacher Quality (NCTQ) released its periodic report on state teacher evaluation policies and for the first time looked into states’ principal evaluation systems. While NCTQ found that 27 states require principals’ evaluators to be trained on how to interpret their leadership frameworks accurately and consistently, Pennsylvania is not among them. And while 23 states allow stakeholder surveys in principal evaluations, Pennsylvania does not. (Yet, of those 23 states that do, only nine explicitly specify teacher surveys, which may provide a more direct assessment of principals’ ability as instructionalleaders than parent or student surveys.)
Still, Pennsylvania’s tool could be contributing to the lack of variation in principal ratings as well. For instance, NCTQ found that few states have acknowledged the primary role of school principals in conducting teachers' evaluations and supporting their professional growth. In almost all states, principals serve as the main teacher evaluator and a key driver of teachers’ development. And yet, according to NCTQ, no state places a great emphasis on these job responsibilities, with the exception of one: New Jersey bases 20% of principals’ evaluations on responsibilities related to teacher evaluation including how well they provide feedback to and coach teachers, provide reliable and valid observation results, analyze evaluation data to target professional development, and help teachers set high-quality student growth goals.
But Pennsylvania leads states on other indicators that NCTQ tracked. For instance, the state is one of 19 that use “student achievement” as the preponderant factor in principal evaluation, consistent with its teacher evaluation system. Still, NCTQ does not dig into how the student achievement component is calculated and whether or not it is likely to be a fair or meaningful measure. While many states include student achievement as a component of principal evaluations, a recent report from the Wallace Foundation found that there is general disagreement on whether and how best to account for principals’ student learning impact.
What’s clear from both the NCTQ and Wallace reports is that no set of promising practices has emerged regarding principal evaluation systems, and states vary widely in their design and implementation. States could benefit greatly from further research on these systems, particularly under full implementation. While the IES study provides some evidence for the validity and reliability of Pennsylvania’s system, the results only applied to a small subset of principals in the state who were not yet subject to any consequences based on evaluation ratings. And, principals and their supervisors will likely respond differently to the system once stakes are applied.
As more states move forward with implementing principal evaluation systems, they could benefit from a compilation of promising strategies from the field, including what’s working well and what isn’t, along with the risks and opportunities of using certain measures. In particular, states could benefit from a deeper exploration of places that acknowledge the central role of principals in supporting teaching quality and driving teacher evaluation and support (as in New Jersey and possibly other states and some local educational agencies as well). We can’t expect teacher practice to improve until we ensure those tasked with evaluating and developing it are themselves supported and evaluated in meaningful ways."