To the Editor:
In regard to the article “Principal Appraisals Get a Remake” (March 6, 2013), the Institute for Educational Leadership joins a steady chorus of researchers who assert that evaluating principals on the basis of student test scores is psychometrically indefensible, despite attempts at developing value-added formulas. A working paper published by the University of California, Berkeley’s education school proposes a tripartite approach to principal evaluation, including:
• A survey instrument that uses feedback from multiple respondents who know a principal’s work firsthand and can report their perceptions objectively;
• A leadership accountability report card that identifies quantitative metrics that principals have substantial influence or control over and that are leading indicators for improved student outcomes, such as attendance, teacher assignments, and discipline data; and
• Evidence-based practice using a calibrated rubric and a revised cycle of inquiry that informs next steps using asset observations, short-term outcomes, and evidence.
Since leaders do not have direct influence over student achievement and mediate instructional influence with students through teachers, we find that the use of student test scores as a measurement of leadership effectiveness is neither fair nor useful. School leaders have primary responsibility for teacher working conditions and teacher motivation; they do not teach students.
Principal-leadership evaluation that uses student outcomes as a proxy for leadership effectiveness obscures our ability to understand which leadership actions lead to increased supports for achievement.
We take issue with the use of student test scores for leadership evaluation, and we expect that the members of the leadership-research-and-practice community will enter into a discussion of alternatives that are more useful and productive.
S. Kwesi Rollins
Director, Leadership Programs
Lynda Tredway
Senior Program Associate
Institute for Educational Leadership
Washington, D.C.
