Arne Duncan And Dennis Van Roekel: Together On Teacher Preparation, But Not On Tests

When Secretary of Education Arne Duncan presented the Obama administration's reforms to teacher training programs before the D.C.-based think tank Education Sector last Friday, he was joined by an unlikely partner: Dennis Van Roekel, president of the National Education Association.

The National Education Association, the largest teacher's union in the country, has warred with the Obama administration in the past, going as far as adopting a resolution this summer that took on the title, "13 Things We Hate About Arne Duncan."

But Van Roekel appeared by Duncan's side on Friday, along with Teach for America President Wendy Kopp. "This plan is a useful tool in helping to ensure that candidates entering the profession from any pathway meet the same high and rigorous standards," Van Roekel said in a statement.

The Education Department's teacher preparation package seeks to alter the training process by basing the ratings of teachers' colleges on outcomes of graduates and their students, creating a new scholarship grant program and diverting funding toward minority-serving schools. Most controversially, the package would require the use of student tests, described by the DoE's report on teacher preparation as "multiple, valid measures of student achievement to reliably ascertain growth associated with graduates of preparation programs."

As Education Week notes, little in the proposal is new, as almost all of it is from the Education Department's fiscal year 2012 budget proposal. Friday's unified front, though, masked the fundamental difference in the approach that Duncan and Van Roekel take regarding the role of student tests in measuring teachers' performance. While Van Roekel would rather teachers be measured by exams that assess teaching practices, Duncan wants the exams to track how much teachers helps their students improve.

"They're making a very big assumption when they assume that a test that measures student learning also measures my contribution to that," Van Roekel, a former Arizona teacher, told The Huffington Post.

These two philosophies come to a head in a classroom like John Bierbaum's, who teaches social studies in Normal, Ill. "I feel really strongly that as a teacher, I should be judged based on a standard," Bierbaum said. "But people also have to understand that the students I have are extremely diverse, and that I can't move them all the same way. Those factors are working against me."

This summer, the NEA adapted new teacher evaluation guidelines that, for the first time, took student performance into account.

"We did not say to what extent [the tests should count in evaluations]," Van Roekel said. "They must be valid measures."

While data-driven reformers lauded the guidelines as a big step for the NEA, Van Roekel said he wouldn't want new teacher evaluations to use standardized exams that are already in place.

The ideal student exams for Van Roekel, he said, "are not based on an individual [student] test provided to an individual teacher." When asked for an example, he pointed to the National Board for Professional Teaching standards exam, which rates portfolios of student work and videos of teachers in action. "It's not like the tests we use in No Child Left Behind, that wouldn't be it," Van Roekel said.

When asked about current exams, Duncan conceded that they have flaws, but also highlighted their utility. "Are they measuring some things? Yes. Are they doing it perfectly? Of course not," Duncan said. While they are only one measure of teaching, he said, he still finds the information they provide to be useful.

Stanford University's Eric Hanushek, an expert on teacher quality, said the distinction between the two approaches is significant. "Once [like Van Roekel] you start trying to measure how they do it, it suggests that you know the technology of teaching," Hanushek said. "Most of the time, it would not allow for the fact that you and I do things differently."

Randi Weingarten, president of the American Federation of Teachers, does not support the plan. "We were surprised that a principal recommendation of the report was to judge the effectiveness of a teacher preparation program by, among other things, the test scores of students being taught by its graduates," Weingarten said in a statement. "At the same time that the validity of using standardized tests as the ultimate measure of performance is being widely questioned, the U.S. Department of Education appears to be putting its foot on the accelerator."

David Nungaray, a member of Teach for America in San Antonio, said he feels that tests as they currently exist are a somewhat accurate reading of his teaching skills. "It gives me a picture of where students grew," he said. "But doesn't give a full picture of what I did." Few advocate judging teachers solely based on test scores.

At the end of the day, Van Roekel and Duncan found common ground in upending teacher preparation programs, which currently leave three fifths of teachers feeling unprepared for the realities of the classroom, according to a recent survey cited in the Education Department report -- despite the fact that states have only identified 37 of 1,400 such programs nationwide as under-performing.

"I don't think those two approaches are necessarily mutually exclusive, which is why you see them standing together," said Tim Knowles, director of the University of Chicago's Urban Education Institute. "With the right assessments, not just an end-of-year standardized test, you could build a full picture of whether a teacher is effective or not."