Your Education Road Map

Politics K-12®

ESSA. Congress. State chiefs. School spending. Elections. Education Week reporters keep watch on education policy and politics in the nation’s capital and in the states. Read more from this blog.


The Twists and Turns of Arne Duncan’s Testing Talk

By Alyson Klein — February 20, 2015 4 min read
  • Save to favorites
  • Print

U.S. Secretary of Education Arne Duncan—and President Barack Obama—have been fine-tuning their case against a Republican bill to rewrite the No Child Left Behind Act that’s slated to pass the U.S. House of Representatives next week.

One of their big beefs with the legislation? It doesn’t include any language to cut back on testing.

The difficulty is ... Duncan is also one of the loudest voices clamoring in favor of keeping NCLB’s testing schedule, which calls for statewide assessments in grades 3 through 8 and once in high school.

To be sure, the policy here—keep the NCLB law’s schedule while trimming back local and state tests to the bare minimum—isn’t exactly contradictory. But the rhetoric is twisty and probably pretty tough for a parent or teacher who isn’t hanging on the administration’s every word to grasp.

And it’s even tougher to wrap your mind around what where Duncan started out on testing and where he is now, at least rhetorically. After all, standardized tests have formed the backbone of nearly every major Obama K-12 initiative, from teacher evaluation, to school improvement grants to Race to the Top in its various forms to teacher preparation.

Check out some flashbacks:

April 2009: Early on his tenure, Duncan addressed the testing question, in speaking to the Education Writers Association. And, while he’s sympathetic to teachers’ concerns, he sure doesn’t sound like he thinks there’s too much testing:

We know that test scores don't tell you everything about students or teachers ... but they do tell you something ... and until we come up with better measures ways to measure achievement ... we must use what we have."

June 2009: The administration ended up giving a big boost in its Race to the Top competition to states that were willing to include test score data as one component of teacher evaluations. In fact, states that prohibited linking student test scores to individual teachers were ineligible to win a grant.

Here’s how Duncan explained his reasoning, in a speech to the Institute of Education Sciences:

To somehow suggest that we should not link student achievement and teacher effectiveness is like suggesting we judge a sports team without looking at the box score. It's like saying, since standardized tests are not perfect, eliminate testing until they are. I think that's simply ridiculous. We need to monitor progress. We need to know what is and is not working and why."

June 2009: The administration also set aside $360 million in Race to the Top money to help states craft tests aligned to the Common Core State Standards. (That eventually led to the tests developed by two consortia of states.)

Here’s Duncan’s sales pitch on that concept, in speaking to the Governors Education Summit. These will be better tests, he tells them:

We need tests that go beyond multiple choice, and we know that these kinds of tests are expensive to develop. It will cost way too much if each state is doing this on its own. Collaboration makes it possible for this to happen quickly and affordably. Now, again, some people may claim that a commonly created test is a threat to state control, but let's remember who is in charge. You are. You will create these tests. You will drive the process. You will call the shots."

June 2011: In speaking to the National Forum, a non-profit organization’s, Annual Schools to Watch Conference, about successful middle schools, Duncan said that testing often can actually be a good thing:

What is most striking about the higher-performing middle schools was that they saw data and the frequent use of assessments as a blessing, not as a burden."

July 2011: Later that summer, he talked up the power of the testing consortia again, in speaking to the National Board of Professional Teaching Standards:

In the 21st century, we shouldn't be guessing whether or not a teacher is impacting student learning—we should know—and while we know that the current generation of tests are far from perfect, a new generation is in development that will be better."

June 2013: Duncan said in a speech at the National Press Club in Washington that most folks, beyond the Beltway, are actually in favor of standardized tests:

In the real world, most people are not against meaningful testing. They know we need some kind of test to know if kids are actually learning and to hold everyone accountable, including students themselves."

June 2014: Then, around 2014, we start to notice a clear shift—Duncan talks about how local testing adds to teachers’ burden:

A lot of teachers are frustrated—and maybe that's a gentle word—about plans that will hold them accountable for tests attached to new, career- and college-ready standards before they feel they have mastered them. ... I know that there are places where testing has gotten out of hand. ... But too many school systems have allowed unnecessary and redundant tests to be layered on ..."

•In August 2014, Duncan told states that they could hit the pause button on tying teacher evaluation to test scores for an additional school year. In a blog post explaining the change-up, he talked about how testing—and test preparation—take up too much time. (Great analysis by Catherine Gewertz here.)

And in a commentary piece for the Washington Post, published in October 2014, he explained his thinking further:

Many have expressed concern about low-quality and redundant tests. And in some places, tests and preparation for them dominate the culture of schools, causing undue stress. ... I strongly believe in using high-quality assessments, including annual tests, as one (but only one) part of how adults improve instruction and hold themselves responsible for students' progress."

February 2015: In criticizing Republican proposals to rewrite the NCLB law, the current version of the Elementary and Secondary Education Act, both Duncan and Obama note that the proposals don’t include any sort of language to limit testing. In a Feb. 14 radio address, Obama said that passing a strong ESEA reauthorization “means cutting testing down to the bare minimum required to make sure parents and teachers know how our kids and schools are doing from year to year, and relative to schools statewide.”

And earlier in the month, in speaking to members of the Parent Association in Annapolis, Md., Duncan said that, under the GOP bill, “There would be no cap on testing ... We want to hold the line on accountability and testing, but we want to set some common-sense limits there.”

Meanwhile, just a few weeks earlier, Duncan gave his biggest speech in years on NCLB and made it clear he wants to keep the testing schedule. “All students need to take annual, statewide assessments that are aligned to their teacher’s classroom instruction in reading and math in grades 3 through 8, and once in high school.”

So essentially, Obama and Duncan are criticizing Republicans for not imposing limits on testing in the same breath that they are calling for keeping annual, statewide assessments. It’s not contradictory, per se, but it’s pretty darn confusing, for the casual observer if not the wonk. (I feel for Duncan’s speechwriters. It cannot be fun to parse that policy. )

What’s more, the administration isn’t exactly calling for testing “caps” as Duncan told folks in Maryland. It’s supporting an effort by states and districts to take a close look at the number of state and local tests they require, and encouraging them to pare back the ones that are redundant or just plain unhelpful. The administration likes a bill by Sen. Tammy Baldwin, D-Wis., and Rep. Suzanne Bonamici, D-Ore., that would allow states to use federal resources to do such an this “audit"—and it’s true that it’s not in the House bill yet. But, to my mind, encouraging local and state limits is not the same thing as setting actual caps.

The bottom line: Clearly, Duncan’s messaging has shifted over the years, from making tests better and richer and using that data for everything from teacher evaluation to program improvement, to cutting tests down to the bare minimum needed to get a clear picture of student achievement for accountability purposes.

Is that an evolution in thinking, rhetorical gymnastics ... or actual flip-flopping? Hit up the comments section.

Shout-out: A big thanks to our stellar Library Intern Maya Riser-Kositsky who combed through pages of speeches and op-eds to help with this post.

Related Tags: