At the head of every successful school is a strong, savvy principal who hires, supports and retains good teachers. But the system for evaluating principals’ effectiveness is weak, writes Andrew Rotherham in Time.

Principal-evaluation methods vary widely — from observations to more formal assessments involving input from teachers — but are frequently not meaningful in terms of consequences. In fact, although less attention is focused on principals’ unions than on teachers’ unions, in many places labor agreements make it as difficult to fire low-performing principals as it is to remove teachers.

. . . And if that’s not disheartening enough, consider the report released last month by New Leaders for New Schools, a national non-profit that trains principals to work in challenging schools, which concluded that “most principal evaluation systems tend to focus too much on the wrong things, lack clear performance standards, and lack rigor in both their design and attention to implementation.”

Principals’ powers are limited. Often, seniority rights prevent the principal from hiring the teachers he thinks will be most effective. Despite multi-million-dollar school budgets, the principal may control as little as $60,000, earmarked for supplies, field trips and such, concludes Paul Hill, who leads the Center for Reinventing Public Education at the University of Washington.

(Good principals) skirt or subvert personnel rules, figure out how to circumvent budget rules or raise additional funds and look the other way while teachers do things that are technically against various policies but in the best interest of students. Hill argues, however, that these kinds of deft, evasive maneuvers make it all the more difficult to assess their productivity — in terms of dollars spent compared to gains in student performance — relative to others.

The New Leaders report recommends basing principal evaluations primarily on student outcomes and holding central-office administrators accountable for principals’ effectiveness.

If accountability is good for teachers, it’s good for principals. But it’s not easy to figure out how to measure effectiveness and how to attract principals who are leaders, not just paper-pushers.