The recent news that Washington state legislators voted down a bill that would require statewide tests to be used—in some locally determined amount—as part of teacher and principal evaluations has three major implications:
1. Washington likely won’t lose its current waiver, which expires at the end of this school year. But based on current policy the Administration would likely look unfavorably on renewing Washington’s waiver for next year unless they can meet one of the only hard and fast rules in the NCLB waiver initiative (Disc. I worked on waivers during my time at the U.S. Department of Education). The waivers require states to create teacher and principal evaluation systems that include student growth in “significant” part. The Obama Administration has never defined “significant” and that’s not at issue here. What is at issue is that the definition for “student growth” dictates that for teachers in tested grades and subjects (and for principals), student growth on statewide assessments must be included. The Administration never specified a percentage or weighting of statewide assessments, but Washington state lawmakers aren’t even willing to promise that they will be included at all.
2. The chasm between the research on value-added models for teachers and the public debate over them is wide and growing. Washington State Senator Rosemary McAuliffe, the ranking Democrat on the education committee, said she voted no because she heard from parents, teachers, and school boards registering complaints that the state tests are not designed to measure student growth.
McAuliffe’s concern, while common, doesn’t stand up to research. In fact, an under-reported result from the MET Project is that current state tests can be used to identify effective teachers. Assessments that require higher-order thinking skills will likely to be better at differentiating teachers, but even the current low-level tests that states are using are valuable in identifying effective teachers. The MET Project found that effective teachers are good at raising student achievement on low-level state tests and more cognitively challenging, open-response type assessments. It’s not only that—research has linked teacher value-added scores on the low-level tests that states were giving in the 1990s to long-term outcomes like teen pregnancy, college attendance, and employment.
3. The waivers are built on a series of promises and plans. Washington’s approved waiver includes the statement that, “State assessment data…in tested grades and subjects will be used when available. Growth data on state assessments will be available September 2012.” That was 2012 and it’s now two years later, but even if you give the state a pass on timing, the substance of that passage flies in the face of Washington’s current stance—that the inclusion of statewide assessments is entirely left to local decisions. Most states are living up to the promises in their waiver, but Washington over-promised in this case, and failure to fix it may force them back under No Child Left Behind.