by David Safier
This is an interesting observation which, like all observations about education, needs to be stored away and evaluated rather than being accepted outright. That being said . . .
As part of a Gates study, thousands of students evaluated their teachers in confidential questionnaires. The students' evaluations tended to agree with the more data-driven evaluations of teachers' abilities to boost their students' achievement.
To the extent this is true, it means students aren't fooled. They know when their teachers are getting the job done and when they're not. That makes sense, especially with students at the high school level. These young folks see five teachers a year, day in and day out, and they have at least 8 years of prior experience with all the teachers they had in elementary school and middle school. They have a reasonable basis for comparison. So they are in a good position to say, "Compared to my other teachers, Ms. Jones is doing a good job in the following areas, but Mr. Smith sucks as a teacher on 9 criteria out of 10."
It also says to me, students of good teachers generally show positive results on the data-driven analyses that involve testing and similar criteria. I don't want to give testing a good rap, but it's a fact that good teachers tend to raise their students' achievement, even the achievement narrowly measured on standardized tests, more effectively than poor teachers. That tends to be true even when the good teachers don't spend a lot of time teaching to the test.
I can see a number of problems if this kind of evaluation were institutionalized, most of which come from teachers and students gaming the system, so I'm not advocating for its general use. But as a spot evaluation tool, it can be valuable.
Here's the most promising news for me. One of statements for students to evaluate was, “We spend a lot of time in this class practicing for the state test.” Those classes where students agreed made smaller gains on the state tests than when they disagreed. I have a funny feeling I shouldn't reach the conclusion I want to reach, that teaching to the test doesn't work. But it's tempting.