News

The 74 Debate: Huffman vs. Barnum on Duncan’s Legacy, Testing Backlash and the Future of Reform

By | January 5, 2016

Photo: Getty Images
The below transcript is an e-mail exchange between New America Foundation fellow and former Tennessee Education Commissioner Kevin Huffman and The Seventy Four’s Matt Barnum, discussing the resignation of Arne Duncan, the backlash to testing and the education legacy of President Obama. What, if anything, should the Obama administration have done differently? Was reform too fast — or too slow? What are the lessons for reformers going forward? (Messages were sent over the span of two weeks in late December, and have been lightly edited for clarity.)
Matt Barnum: Arne Duncan is out and so is No Child Left Behind. What to make of the lessons from his tenure? Let me take a pass, starting with the trickiest of subjects: testing.
Although at times the Department of Education has skirted blame, there is a good deal of evidence that Duncan's focus on teacher evaluations created a proliferation of new tests. The federal government pushed to evaluate all teachers — from gym to kindergarten — by some measure of student achievement, but because many teachers didn't have tests aligned with their grades or subjects, states scrambled to fill that void. Where tests weren't created, teachers were sometimes evaluated by test scores of subjects or students they didn't teach.
Rightly or wrongly all this contributed to a fierce political backlash from parents, unions, and Republicans. Duncan tried to contain the anger by promising to cut back on testing, but Pandora’s box had already opened.
The literal lesson here is clear to me: Don't expect, at least in the short term, to be able to evaluate all teachers by student achievement. More broadly, it's a reminder that ‘should’ does not imply ‘can.’ The push for test-based evaluation was driven by an almost moral outrage around the idea that there was a limited connection between student outcomes and education policy, particularly teacher personnel practices. Fair enough, and to be clear, I believe testing matters, accountability is crucial, and value-added measures have a role in teacher evaluation. But for teachers in non-tested grades and subjects we haven't really gotten from ‘should’ to 'can' — doing so has proven technically challenging and politically disastrous.
What do you think? How did you address the testing issue in Tennessee, and was is successful? What lessons should we take away from this?


Kevin Huffman: Broadly speaking, the feds get too much blame for everything in education. I’ve found that a sizable number of local education leaders have a reflexive index finger that points to the next guy whenever any unpopular decision is questioned (in my case, they sometimes pointed the middle finger too). Arne was on the receiving end of a lot of fingers.
I would first challenge the premise of the over-testing question. I don’t think that most of the anti-testing crowd actually is responding to an increase in volume of tests. Put another way, the new tests created — to measure the so-called non-tested grades and subjects — are not the genesis of the anti-testing push. Most of the anti-testing push, in my view, comes from two groups: First, people who think that schools are over-emphasizing the general math/reading tests and, second, people who hate Common Core and therefore think that if we didn’t have aligned tests, maybe their schools wouldn’t teach it.
Now, I do think that tying tests — even the existing ones — to teacher evaluations ratcheted up the sense of pressure that teachers felt about the tests. And I think some teachers passed some of that anxiety on to their students. And I think some teachers picked the counter-productive path of using more test prep which frustrated families and didn’t improve test scores. And I think some teachers were nervous about a switch to new tests that were likely to be harder for students. And I think some parents were already mad about Common Core.
Then, throw in a well-intentioned but often poorly-executed move among local schools and districts to give regular no-stakes benchmark assessments. And remember that when parents asked about these tests, frequently school leaders blamed the districts and district leaders blamed the states or the feds. It all leads parents to just scratch their heads and feel frustrated.
In Tennessee, we used and still use test scores as part of teacher evaluation (35% for tested subjects, but using a three-year average, so each year’s tests count a little over ten percent). The main driving philosophy on non-tested grades and subjects was this: don’t give tests for the purpose of evaluating teachers. Only give them if measuring student learning in that subject is itself worth doing. It didn’t solve everything, and we have pushback against the use of school-wide results (which count for 25% of a teacher’s evaluation for certain grades and subjects) but I think it’s a solid approach.
Back to the initial question though. I get frustrated with folks criticizing Duncan and the Department of Education for being too aggressive on the use of teacher evaluation in their waivers because I would argue that he wasn't aggressive enough. The Department allowed states to repeatedly punt their evaluation timelines, they rarely cracked down on the slow-movers, and they generally let the whole process drag on for years. I think a better approach would have been to limit waivers to the 10-12 states that were moving quickly on evaluation and standards, and let everyone else hang out under NCLB. Then we would have had more genuine proof points and we would have had more pressure to reauthorize earlier.
I can’t smack Arne for going too slow if everyone else is hassling him for going too fast!
P.S. Tennessee and DC were the first state and district to fully implement teacher evaluation using test results. Tennessee and DC were the fastest improving state and district on NAEP between 2011 and 2015. Go figure.


Matt Barnum: Your points are well taken, but my view is that although the feds may often get unfairly blamed generally, in the particular case of testing, pointing a finger at the Department of Education is warranted, even if there is blame to go around.
I'm wondering, though, how does trying to preempt a political backlash play into things? Sure some of it was inevitable, but I don't think all of it was. The pushback against the testing aspects of NCLB didn't reach this level, as far as I can tell. It was a combination of teacher evaluation and Common Core that drove the movement.
Duncan emphasized that he didn't want to play politics and focused exclusively on what's right for kids. That's admirable, and no one doubts his authenticity. But I think reformers should be careful in ignoring political dynamics. The new ESSA is a direct result of the politics of the Obama/Duncan administration — do you think it's good for kids? Better politics might have produced a better federal law.
Are there any lessons here for reformers about thinking about how politics drive policy and vice-versa?
P.S. I don't buy into your NAEP analysis — that's misNAEPery, as some call it. We can't look at raw test scores to judge specific policies. If we did, shouldn't you be condemning Duncan for the recent drop in scores? That said, there is certainly reason to be optimistic about aspects of DC's evaluation system.


Kevin Huffman: The question of politics in the context of reform is tricky, at both the federal and the state level. The biggest challenge I ran into: on the one hand, literally everything I did was deemed “too fast.” Sometimes I felt like if I had suggested a ten year rollout of elementary school art standards, it would have been “too fast.” This makes one cynical, needless to say.

On the other hand, sometimes the juice isn't worth the squeeze, to get a little Tennessee on you. Something might be the right policy in isolation, and it might be the right thing for kids, but it might be overkill when layered on top of lots of changes. The overkill moves people from pushback to angry opposition. And the reality is: It is impossible to get this exactly right, and impossible to know in the moment, in large part because there is a vocal contingent of folks in education who will complain about literally every change. Sussing out the difference between general change-aversion, and genuine indigestion is not easy, to say the least.
Looking at Common Core, conservatives have settled on the “federal overreach” theory of opposition. That’s total historical revisionism. Most of the right wing pushback on Common Core wasn’t about overreach, it was about insane conspiracy theories; retina scans and sex education and United Nations Agenda 21. The overreach argument was a tiny part of the anti-Common Core momentum in most states, and has been settled on as a face-saving way to talk about Common Core opposition now that it has become socially unacceptable to support the standards in Republican circles.
On the left, the combo of evaluation and Common Core assessments no doubt took the unions from staunch supporters of the standards into anti-test (and in some cases anti-standards) agitators. Is that Duncan’s fault for layering them together in Race to the Top and the waivers? Maybe. It’s instructive, though, that New York was a Race to the Top state, so it voluntarily — with the union — agreed to all the things that later combined into a massive opt-out moment. Nobody made New York do evaluations or change standards, and if it was such an obviously terrible idea, I’m not sure why states and their unions agreed in the first place. (And no, I don’t think it was desperation for money — the total grant amounted to less than 1% of New York’s education spend).
But yes, one of my lessons learned over the last four years: just because you are right about something, doesn’t mean you have to do it. Politics matter.
P.S. Oh come on with the misNAEP-ery! I don’t claim NAEP proves evaluation works. But looking at Tennessee and DC absolutely undercuts the “evaluation made everything go to hell in a handbasket” argument. And frankly, if we can’t look at outliers on NAEP in gains and losses and try to draw some lessons from them, let’s stop giving the damn test!


Matt Barnum: I think we're going to have to agree to disagree about NAEP. I'll put my lot in with the many smart researchers who have warned against using raw NAEP data to draw inferences about specific policies.
Let's turn to Common Core though. You say that the opposition to it is largely built on half-truths and conspiracy theories. I think that's true to some extent. (Take a look for instance at the hilarious, but sad way the standards were discussed at the Conservative Political Action Conference, or a piece I wrote about the fight against them in New Hampshire.)
But let's also put a critical mirror up to the pro-Common Core side here: If the achilles heel of opponents is misinformation, the problem for supporters is cliches and over-promises. I hear all the time about how Common Core creates 'high standards' for 'college and career readiness.' Who can be against that? Nobody — and that's the problem. I worry that some supporters of the standards have eschewed substantive discussion for political bromides.
In some ways this is related to another problem with Common Core: its swift, under-discussed adoption. An anecdote: When I was a teacher from 2010–2012 I had no idea what the Common Core was. I have no recollection of hearing it discussed, and I was someone who tried to keep up with education policy issues.
This may not be Common Core supporters' fault — certainly the media deserves some blame — but I do think there was a sense among backers of, "Let's get this adopted fast before 'special interests' can mess with it." I would say this approach had benefits — the vast majority of states signed on to the standards and are still using them — but also costs, in the form of a weak political constituency to defend Common Core from attacks.
Do you think the creation/adoption of the standards should have been done differently? What should reformers learn from the backlash to them?


Kevin Huffman: I completely agree that many Common Core supporters are the kings of empty platitudes. As with all things education, in a lot of places, the Common Core transition has involved the trading in of old acronyms, truisms and bromides for new ones. I’m also generally skeptical that new standards will make significant positive difference in the absence of quality training on the front side and accountability on the back side, and having blown part one, many places now seem keen to blow part two.
I disagree, though, that states jammed the standards through for political expediency.
Honestly, most standards discussions are boring and happen all the time in public meetings by state boards or commissions. Tennessee changed reading and math standards a few years before adopting Common Core with little fanfare. During the Common Core transition, we also changed social studies standards and CTE standards, with public hearings and opportunity for input, but with little drama. Common Core in most places was actually discussed more than most standards changes. But given that there wasn’t massive pushback in the 2008-2011 era, the whole thing just chugged along the way all standards changes do. It wasn’t a conspiracy – it was business as usual.
In my view, it’s wrong to blame proponents of the standards for this. They were pretty straight forward about the changes, and the processes in most places were open to public input. I blame opponents of the standards who somehow had no objections during multiple readings and opportunities for public input, but then lost their minds years after the decisions had been made.
It seems wrong to put the onus on proponents of change to jump up and down and get people’s attention. And in the case of the Common Core, given that the biggest early opposition energy came from strange conspiracy theories, I’m hard pressed to know what proponents should have done differently. Put out a press release saying, “We are proposing some mundane changes to math pedagogy, but some people might believe that we are using retina scans to capture your children’s data for our giant government warehouses. Therefore, we would like to invite everyone to look closely at these new standards for signs of scanning, data collection, or Islamic indoctrination.”?
I do think one piece that was significantly overlooked in the whole Common Core exercise, though: states review and revise standards on a regular basis (every six years in Tennessee). Even under normal circumstances, many states would have reviewed Common Core standards and made some small changes based on lessons learned over time — and the changes would have been different from state to state. Therefore, the standards were inevitably going to drift away from commonality even under thoughtful implementation.
As for the media, well, it only shows interest in what is being taught if adults are fighting about it. We put in a whole new slate of high quality social studies standards, including great civil rights standards, but we only got coverage when some people became outraged that world history includes Muslims.


Matt Barnum: I take your points. Selling Common Core before it was in schools would not have been easy, as parents and media may not have cared. I do think reformers made a mistake by not more quickly backing off linking teacher evaluations to Common Core–aligned tests. The Obama administration eventually allowed for this through waivers — but perhaps too late to separate the two in the public conception, and some states have not taken advantage of this flexibility. What we got was the worst of both worlds: a politically toxic linking of Common Core and testing without a meaningful increase in teacher accountability. New York for instance is home to some the greatest Common Core/testing controversy, perhaps in part because Governor Andrew Cuomo vetoed a bill his own administration had negotiated to hold teachers harmless from test results. The result? As I reported, even without the safety net, according the State Education Department, only one tenured teacher has ever been fired under the state evaluation system.
Let's turn to what the Obama/Duncan administration did not focus on. Two things I've heard some critics seize on are school integration and funding equity. Nowhere in Race to the Top or waivers did these two issues see the light of day. (To be fair, on funding Duncan and Obama championed the stimulus package to help schools during the recession. Duncan also has used the bully pulpit repeatedly the emphasize the importance of equitable funding)
When asked about this Duncan has said doing so simply wasn't politically feasible. I must say that I found the response pretty unconvincing, considering Duncan has so aggressively pursued an agenda without regard for political blowback. A more honest answer might have been simply that Duncan believes his teacher quality and choice-based reforms were more worthwhile.
I guess my question then is: Do you worry that reform policies are too narrow, and whether issues like integration and funding equity have been crowded out? Do you think the administration's priorities were right? Should reformers' emphasis shift in the post-Duncan, post-NCLB world?


Kevin Huffman: It’s always fascinating to watch things play out in present-time, and then come back five years later and watch the second guessing. I don’t remember all the clamoring for funding around desegregation or funding equity during the stimulus discussions. The way I remember it, the unions were incredibly slow on the draw in arguing for funding at all (literally, we would visit Congressional offices in 2008-09 and talk about massive impending layoffs and the staffers would say, “That’s weird — the NEA was just in here yesterday and they only wanted to complain about Michelle Rhee.”) And ultimately when Race To The Top  was $5 billion, and $95 billion headed out to states on a formula basis, the debate around Race To The Top priorities was not particularly robust.
In hindsight though… I still don’t think the grant priorities were that far off-base. (The standards stuff could have been written more broadly to head off federal overreach arguments). Money targeted at integration would have been really tough to pull off. For starters, support for segregated neighborhood schools is a shameful bipartisan tradition; I don’t see a lot of white liberals all that eager to break up the mortgage-as-tuition system currently assigning kids to schools.
It also would be difficult operationally to run a federal grant program targeted at school assignment. States have the traditional power to set academic standards and to create evaluation laws and protocols. But districts deal with their own student assignment. Making grants to states that require districts to change student assignment would either require broad new expansions of state power, or sign-on/opt-in possibilities that would be very difficult to manage and enforce. Race To The Top was premised on leveraging state powers, not district powers, and I just don’t think it would have worked in the student assignment realm.
Beyond Race To The Top, though, I completely agree that the reform world has been narrowed by over-focus on a few specific policies. Reform should be, plain and simple, about doing things that are demonstrably good for kids. And stopping things that are demonstrably bad — particularly for poor and minority kids. It is not supposed to be a small package of policies.
There are leaders like Kaya Henderson in DC who are “reformers” by background but spend their time and energy on a very broad range of initiatives. Kaya was thrilled to launch a program to ensure that 2nd graders in DC all learn how to ride a bike. DCPS has a big focus on bringing back arts and music. When you talk to her, you don’t hear standard-issue, pull-my-finger reform stuff. It’s complex and multi-faceted and kid-focused.
I’m pretty optimistic that you will see more interesting reform conversations that expand beyond the well-traveled policy paths of the last eight years. Reform started as an opposition movement — opposition to a status quo that has failed poor and minority children. Looking at the direction of the current election, I think reform is headed back to opposition status, and it may be a good thing. As the great Woody Hayes once said, “There’s nothing that cleanses the soul like getting the hell kicked out of you.”
Submit a Letter to the Editor