But the mechanics of quantifying good teaching are tricky. How can districts discern who the best – and worst – teachers are? Are test scores reliable? Is observation too subjective? How does something as subtle and nuanced as teaching a roomful of individuals – a job that is arguably more art than science – get reduced to a score?
Most everyone agrees: There's no single foolproof measure of a teacher. Standardized tests give one indication of what students are learning. Observations – when the observer is trained well and looking for specific details – can offer more nuance.
But how do you evaluate the essence of a teacher who inspires a true love of learning and problem solving versus one who just gets students to master certain concepts? And what about all the factors a teacher has no control over, such as family life, poverty, or a student who is having a bad day when it's time to take a test?
There is consensus that the status quo doesn't work. The majority of districts in the United States are like the one in which Newman found herself always "perfect," or close to it. Principals come in for a required observation every few years, from which the teachers get little feedback, and nearly all teachers are considered good.
A landmark 2009 study of 12 districts in four states by TNTP found that more than 99 percent of teachers were rated satisfactory. The study confirmed what many in the education field already knew: Traditional "evaluations" gave very little useful feedback to teachers and their administrators.
So far, the current mantra in redesigning evaluations is "multiple measures."