Arne Duncan at Harvard's Ed School

Arne Duncan gave a GREAT speech at Harvard's Ed School on Monday highlighting how misguided some of the fights are in the ed reform space and how there's room for collaboration and compromise (the full text of the speech is at the end of this email, and the audio of a related interview with him is posted at:www.gse.harvard.edu/news-impact/2012/02/the-arne-duncan-interview/):

Now, if a curious visitor from another country plunked down in the midst of our education debates, he would likely find this new generation of gap-closing schools to be very exciting news. He would find them a wonderful testament to the power of outstanding teachers, great principals, and strong community partners to transform the life chances of children.

But in fact the response of some in the U.S. education establishment to schools that produce dramatic gains in student learning has been much more critical, even dismissive.

That curious visitor would be puzzled by those who respond to successful no-excuses schools by making excuses for why they don't really matter.

Of course, no one should object to understanding the limitations and strengths of this new research on gap-closing schools. But the skeptics of successful schools have jumped from critique to critique, none of which have found much confirmation in rigorous research.

It is telling that advocates wedded to the idea that school achievement is simply a reflection of poverty seem determined to diminish the value of great teachers and great schools. That disrespects the hard work, talent, and tremendous commitment of the teachers and principals at these schools, who dedicate their lives to working with disadvantaged children because they know they can make that special connection that changes children's lives.

…Now, the second, false choice that I want to talk about today is the debate over whether teacher evaluation should include measures of student achievement and growth.

Again, I reject the idea that this should be an either-or debate. Critics of standardized testing make a lot of good points. It is absolutely true that many of today's tests are flawed. They don't measure critical thinking across a range of content areas. They are not always aligned to college and career-ready standards. They don't always accurately measure individual student growth.

And they certainly don't measure qualities of great teaching that we know make a difference—things like classroom management, teamwork, collaboration, and individualized instruction. They don't measure the invaluable ability to inspire a love of learning.

As I have said, over and over again, teacher evaluation should never be based only on test scores. It should always include multiple measures, like principal observation or peer review, student work, student surveys, and parent feedback.

…Still, the shortcomings of today's tests don't mean that we should simply abandon the use of standardized testing in schools and teacher evaluation.

In the last decade, I have talked to literally thousands of teachers and school leaders. I have yet to speak to one who thinks teacher evaluation in America works well today.

Let me be clear: Teacher evaluation today is largely broken and dysfunctional. No one can say who the great teachers are, how teachers in the middle can improve, or which teachers should be dismissed if they fail to improve, even after receiving help and support.

California has 300,000 teachers. It's top 10 percent of teachers—30,000 teachers—are world-class teachers and some of the best in the world. Its bottom 10 percent of teachers should probably not be in the classroom. But today, no one knows who is in which category.

Again, we have to ask the compared-to-what question. Is an evaluation system that uses at least some measure of student achievement and growth, even if imperfect, preferable to an evaluation system that takes no account of student learning? I've learned a lot in Washington. But I was literally stunned when I discovered that several states had laws on the books that actually prohibited using student achievement in teacher evaluation. Think about how crazy that is—and what a perverse signal that sends about the entire teaching profession. Thanks in part to Race to the Top, those laws are now all gone.

The use of value-added analysis to measure student growth is still very much a work in progress. But it is, with all its imperfections, a big improvement over a system that takes no account of student growth in the classroom.

Thanks to groundbreaking research by Raj Chetty and John Friedman here at Harvard and their colleague at Columbia, Jonah Rockoff, we know now that the long-term impact of good teachers on students in adulthood is profound. Their study was not about good teachers creating short-term bumps in test scores; it demonstrated how teachers, for better or worse, literally altered the trajectory of their pupils' lives.

Their analysis of the long-term impact that teachers had on 2.5 million children found that simply replacing a teacher in the bottom five percent for advancing student growth with an average teacher would increase the students' lifetime income in that classroom by more than $250,000.

And improvements in teacher quality also significantly reduce the chance of having a child while a teenager and increase college matriculation. Want to increase earnings potential, decrease poverty, and reduce teen pregnancy? Then please spent a lot of time thinking how to attract, retain, and reward great teachers, particularly in disadvantaged communities.