Quality of Evaluations Draws New Attention As Stimulus Aid Flows
The nation’s oft-criticized systems for evaluating the quality of its educator workforce are poised to receive increased scrutiny, thanks to an Obama administration plan to require school districts to disclose how many teachers perform well or poorly.
Although nearly every state requires districts to evaluate teachers, the instruments are typically designed locally. And as both policy experts and some union leaders attest, they are frequently of poor quality, not based on standards of good teaching, and incapable of rendering fine-grained, fair judgments about teacher performance.
Policy experts widely view the U.S. Department of Education initiative, which is part of the implementation of the federal economic-stimulus package’s aid to education, as an attempt to collect baseline data on teacher evaluations and to promote an overhaul of those systems. It could also, they say, pave the way for shifting the “highly qualified” teacher provisions of the No Child Left Behind Act from a focus on paper credentials to one on outcomes.
“To me, it means that there’s real recognition at the highest level that ... if we’re going to do some of the other human-capital reforms on the table, like trying to link compensation and tenure decisions to performance, then evaluation is kind of the linchpin,” said Raegen T. Miller, the associate director for education research at the Washington-based Center for American Progress, a Democratic-leaning think tank.
But, experts add, the initiative’s success will depend on the administration’s follow-up steps—including the metrics the Education Department sets for reporting evaluation data, and what steps it expects states and districts to take with the resulting data.
“I think this is a big black hole, and I’m not sure how [the federal government] is going to do it,” said Julia Koppich, a San Francisco-based consultant on teacher issues. “This is very difficult, high-stakes work.”
The guidance given to states earlier this month by the Education Department on how to spend the first round of education aid—$32.5 billion—under the American Recovery and Reinvestment Act requires that they report on the number and percentage of teachers and principals scoring at each level on local districts’ evaluation instruments. States must also disclose whether the evaluation tools take student performance into account. ("First Education Stimulus Aid Flows to States," April 8, 2009.)
The teacher-evaluation requirement falls under an “assurance” states must submit saying that they will take steps to improve the effectiveness of teachers and increase access by low-income and minority students to effective teachers. The department hasn’t yet set a deadline for states to begin reporting.
Districts use different methods to measure teacher performance. The results are used—with varying degrees of success—to provide teachers with feedback for improvement and for formal accountability decisions.
Observers rate teacher performance on a checklist of input-based factors, typically summing up a teacher’s performance as either “satisfactory” or “unsatisfactory.”
The system uses measures describing teacher performance at four or five escalating levels of performance.
Peer Review and Assistance
Principals assign struggling teachers to “consulting” teachers for a specified period of intense support. Teachers who do not improve are recommended for dismissal or nonrenewal—depending on the system—by a committee of both administrators and teachers. Cincinnati and Toledo, Ohio, and Montgomery County, Md., are examples.
This model links students’ year-to-year test scores to teachers, and attempts to screen out other factors, such as family-income level, to determine teachers’ contributions to achievement growth. It is used informally in some Ohio districts and in New York City.
Multiple measures of performance are included. Schools in the Teacher Advancement Program use value-added data in conjunction with performance-based observation.
In a letter sent to governors with the guidance, Secretary of Education Arne Duncan clarified that states need only submit an assurance that they plan to report. To receive the second round of stimulus funds, $16 billion, they must show that they have developed the capacity to file such reports.
The lack of detail on what districts and states should do with the data they eventually gather and report has raised some skepticism within the education policy world.
“The best possible scenario is that, in fact, there is a lack of information on teacher evaluations, and this reporting creates a tidal wave of support to get something done here,” said Andrew J. Rotherham, publisher of the Washington-based think tank Education Sector.
But the publishing of performance results does not always guarantee action by state and local policymakers, he cautioned. President George W. Bush’s administration held such a theory about the test-score reporting requirements in the early days of the NCLB law, and many states and districts chose to obscure the data rather than act on it, he said.
“I just think it’s a tough lift,” said Mr. Rotherham, who served as an education aide to President Bill Clinton. “The systems look the way they do because of politics.”
The creator of a widely adopted teacher-evaluation system said that any data generated from the reporting requirements will be hard to parse accurately without contextual information.
“I don’t think that the proportion of teachers rated at each of the levels is, actually, particularly important in itself. The teachers might all be really good,” wrote Charlotte Danielson in an e-mail. “But ... it may reflect a symptom of something else, such that there really aren’t any standards, and that site administrators can’t tell the difference” in teacher performance.
Districts moving to develop better systems are likely to confront several hot-button choices, including two that drive much of teachers’ concern about evaluations: who conducts them, and whether they include objective measures of student performance, such as test scores.
Teachers’ unions have long argued that many principals are not trained to make such judgments, and thus cannot do so fairly and uniformly. More recently, union leaders have objected to the linking of student test scores to individual teachers, both for technical reasons and out of fear that such links would encourage “teaching to the test.”
“You have on one side of the pendulum evaluation done exclusively through principal observation, and on the other side done through the lens of individual test scores, which is equally flawed,” said Randi Weingarten, the head of the 1.4 million-member American Federation of Teachers.
Ms. Weingarten said she hopes the stimulus requirements will spur the creation of “far more robust, more accurate, more appropriate, multiple-measured instruments that can be collectively bargained and used in districts throughout the country.”
Vol. 28, Issue 29, Page 8Published in Print: April 22, 2009, as Quality of Evaluations Draws New Attention As Stimulus Aid Flows