I had the chance late yesterday to speak to Randi Weingarten, the American Federation of Teachers’ president, about President Obama’s education agenda-setting speech. We talked mainly about Obama’s contention that there is “no excuse” for districts to keep severely underperforming teachers in the classroom. (To me, that was newsier than the talk of performance pay, which dates way back to the campaign.)
“The point he’s making, that there needs to be better evaluation systems, is spot on,” Weingarten told me. “The reason we see default to individual student test scores [to judge teachers] is because of the lack of reliable teacher-evaluation systems across the country.”
As I expected, Weingarten brought forth the Peer Assistance and Review program, used in various permutations in Toledo, Ohio, and elsewhere, as a possible evaluation model. One of the benefits of PAR, she said, is that it doesn’t rely on test scores or solely on a principal’s observation for judging teacher performance; rather, it is part of a system of multiple measures.
(It’s important to keep in mind that in some instances, peer review is used to determine whether a teacher gets tenure, and in other cases, it’s part of the dismissal process.)
I asked Weingarten whether she thought test scores had a place as one such measure in evaluations. She didn’t explicitly say no, but her response indicated skepticism. “Where they’re used, they tend to become the predominant component, and it doesn’t account for all the other factors related to teacher performance,” she said.
The tests themselves have never been evaluated for their appropriateness in judging teachers, she added.
Interesting stuff, and sure to be on the burner as the Obama agenda moves forward.