Troubling and Promising Findings in NCTQ Scan of State Teacher Evaluation Policies

Of the 40 states and D.C. that require student achievement as a factor in teacher evaluation, fewer than half have an explicit policy for using student achievement measures to evaluate teachers of untested grades and subjects, according to a new report from the National Council on Teacher Quality. That includes, in most cases, PreK-3^rd grade teachers. (Read my report An Ocean of Unknowns for more on what policymakers need to know about evaluating pre-K and early grade teachers.) Nearly every one of the 41 states, within the next year or so, will fully implement its teacher evaluation system.

Prompted in large part by the Obama Administration’s Race to the Top grant program and No Child Left Behind waivers, states have been rapidly overhauling their teacher evaluation systems. But what new requirements did NCTQ find state lawmakers setting for teacher evaluation systems?

• Only 10 states—Alaska, California, Idaho, Iowa, Montana, Nebraska, New Hampshire, North Dakota, Texas, and Vermont—do not explicitly require the use of student achievement data in evaluating teachers.
• In 20 states, student achievement or growth is weighted as the most significant factor in teacher evaluation; nine of those states have yet to define “significant.”
• In 28 states, multiple measures of student growth and achievement are required; 20 states allow or require student learning objectives (SLOs) as one of the measures.
• Thirty states require standardized states tests to be used as a measure of student growth when available.
• Only 15 states require some unannounced observations and fewer than half of states require that teachers receive feedback on their evaluation.
• Eight states require student surveys as part of teacher eval; three (Alaska, Connecticut, and Iowa) require peer surveys.

Also worth noting is that only 12 states require districts to adopt a statewide evaluation model. Of the 37 states with no required model, districts in only 15 states must have their evaluation system approved. Some states offer a recommended state model. But the fact that in the majority of states, districts are responsible for figuring out evaluation design and implementation on their own raises some concerns. Many districts, especially the small districts, have limited capacity to do this. It’s important that states provide ample technical assistance and guidance to the districts that need. Of course, states have capacity issues of their own.

To be effective, evaluation systems need to be fair, relevant, and helpful. They need to be good at identifying, rewarding, and supporting effective teachers, as well as identifying and helping struggling teachers to improve or leave the classroom. How do states stack up when it comes to how these new evaluation results are used?

• Just 20 states require that professional development be aligned with evaluation results for all teachers, not just those deemed “needs improvement” or “ineffective.” (Even the best of teachers can improve and should have access to PD specific to their needs.)
• Only 13 states and Washington, D.C. require evaluator certification; only D.C., Indiana, New Mexico, and New York require evaluators to be effective teachers; and only five states require multiple observers.
• In 23 states, districts can fire teachers for poor evaluations, and in six states, evaluation results matter for compensation decisions. Nineteen states consider teacher evaluation ratings in tenure decisions.
• Only Florida, Illinois, and Tennessee require that student teachers be assigned to classrooms with teachers who have been rated as “effective.”

So what are the policies affecting teachers in the untested grades and subjects?

Most states are still figuring out what to do here. NCTQ makes an important recommendation, stating that “nontested grades and subjects cannot be an afterthought.”After all, in most states about 70 percent of teachers fall into this group. Some states are requiring or allowing student learning objectives. (More on SLOs here.) Some states are identifying new assessments. Some states allow the use of schoolwide measures or a measure from a tested grade to be used to evaluate these teachers. For instance, part of a kindergarten teacher’s evaluation could be based on how well 4^th-graders performed on the state’s math and reading test. NCTQ recommends states use caution when using schoolwide measures of growth for individual teachers. We, here at the Early Ed Initiative, advise against using this kind of measure as the sole measure of growth for an early grade teacher. It is much more appropriate, and valuable, for these teachers to be rated based on an alternate classroom measure.

NCTQ also criticizes some states for lowering standards for teachers of untested grades and subjects. The weighting of measures should be fair and well-planned. Student achievement should not count less for some teachers and more for others because they teach a subject that is tested by the state. Additionally, a teacher’s student achievement rating should be based on the range of learning and development each teacher is responsible for. And in the early grades, that means much more than basic reading and math skills – social-emotional skills or students’ ability to persist at a challenging task, for example. (The last thing we want is teachers with ratings of “ineffective” flocking to grades with less rigorous measures.)

Just five years ago, the landscape looked vastly different. In 2009, NCTQ found that no states allowed evaluation ratings to be tied to tenure and just 15 required that student learning be considered. States have come a long way in passing policies to make systems more rigorous, but there is a lot more work to do on implementation, making systems more meaningful, and promoting the systems’ ability to foster improvement in teacher instruction and student learning. (Recent reports from New York showing most teachers still rated as “effective.”)

It’s not enough to simply make student learning, unannounced classroom observations or students surveys pieces of teacher evaluation. Changes need to be purposeful, and soundly implemented with appropriate support systems – like professional development based on teacher evaluation and rigorous training for teacher observers. And time needs to be allotted to work out the inevitable kinks before stakes are attached. Without strong implementation and ongoing review, the intended improvements to teacher evaluation systems will fall flat.”