As we reported earlier this week, the standoff continues between our state and the feds over the use of student test scores in teacher evaluations.
The U.S. Department of Education continues to insist that test scores should play some role in teacher evaluations. Washington lawmakers have refused to require school districts to do so and, as a result, lost the state’s waiver from the requirements of the federal No Child Left Behind Act.
This week, the feds refused Washington’s request to get back a piece of that waiver — the part that would have saved schools from having to send letters home saying they have failed — as most other schools in the nation have failed — to ensure that all students were proficient in reading and math this year.
So what about the substance of the argument? Are test scores a valid indicator of a teacher’s effectiveness?
People on both sides can point to research that supports their views.
Here are a few of the latest reports and studies, which raise a lot of questions:
1. In May, two researchers looked at whether the way some districts are using test scores to judge teacher performance is a good gauge of teaching quality. They found, to their surprise, that it doesn’t seem to be.
Morgan Polikoff and Andrew Porter looked at data about 327 fourth- and eighth-grade teachers in six school districts. They looked at the ratings teachers received from an analysis of their students’ test scores, often called value-added measures. Then they looked at other measures of quality, including how well teachers cover the material they were supposed to teach under their state’s learning standards.
Their bottom line: “What we’re left with is this concern that state tests really aren’t picking up the things that we think of as being good teaching,” Polikoff said in a video that accompanied the release of the study, which appeared in Educational Evaluation and Policy Analysis, a peer-reviewed journal of the American Educational Research Association.
“What these results, among other results, call into question is the sort of fixed and formulaic approach to teacher evaluation that is being promoted in a lot states right now,” he said.
(Their research was sponsored by the Bill & Melinda Gates Foundation, which is also the main sponsor of Education Lab.)
2. In April, the American Statistical Association urged caution in the use of value-added measures, saying most studies find that individual teachers account for, at most, 14 percent in the variation in test scores.
“The majority of the variation in test scores is attributable to factors outside of the teacher’s control such as student and family background, poverty, curriculum, and unmeasured influences,” the association wrote.
3. Also in May, the Brookings Institution concluded that judging teacher quality by observing them in the classroom may not be any better than looking at student test scores — and may be worse.
They found that teachers who start the year with academically strong students tend to get better evaluations than teachers with weaker students.
“We should not tolerate a system that makes it hard for a teacher who doesn’t have top students to get a top rating,” the authors wrote.