Did We Get Anything Out Of NCLB Accountability?

We're in the midst of the 20th anniversary of No Child Left Behind, a legislative offspring of bipartisan consensus that has itself been left behind by virtually everybody. The bill was proposed in March of 2001, spent the rest of the year wending its way through the process, to be signed into law at the very beginning of 2002. So pretty much any time this year is fair game for a 20 year retrospective on this truly crappy law. 

Cue Deven Carlson offering a quick six-page take on the law for the right-tilted, free-market-loving American Enterprise Institute- "Holding Accountability Accountable: Taking Stock of the Past 20 Years."

Carlson leads with some good nostalgia about how wrong NCLB proponents were about the influence the law would have. He doesn't talk about the usual swing of the accountability pendulum, and he probably doesn't give A Nation At Risk enough credit for helping lay the groundwork for the unholy marriage of accountability and "reform" of public ed. But he offers a decent analysis of a handful of "successes" and "failures" for NCLB, and it's worth a look just to see what reform types believe about the NCLB fiasco.

First, the "successes."

Shifting focus from inputs to outputs. In principle, I agree that it's bad practice to simply throw money and schools without looking at what's happening. But too many reformsters moved from "let's not just measure money spent" to "the amount of money spent doesn't matter." And the Cult of Outputs immediately ran into a huge problem--we don't have any good way to measure most of the really important outputs of education. In fact, the whole input-output model (Input a piece of sheet metal and the assembly line will output a toaster) doesn't really fit the process of becoming a more educated human being. 

Like many reformsters, Carlson argues that the standards movement pushed outputs, but I disagree. Standards (what stuff will we be teaching) are about inputs. But you get into real trouble if, as NCLB did, you decide to tie the standards to bad tests, declaring in effect that you will only have standards set to things your bad standardized test can measure. NCLB "pushed student outcomes to the forefront of policy debates," except that in this case, "student outcomes" just means scores on a single narrow math and reading multiple choice test. This also led to the upside down school, where students were not there to have their needs met, but to generate the scores the school wanted and the government demanded.

Carlson writes:

And although the focus on student outcomes hasn't been without drawbacks, on balance it has been a positive development.

The first part of that sentence is a huge understatement; the second part is just wrong.

Shining a light on different student subgroups. Carlson argues that before NCLB, we didn't have information about disparities of race, ethnicity, disability, economic status, etc. I'm not convinced that's true; I don't know of anybody who looked at disparate results of the Big Standardized Test disaggregated results and said, "Woah, I had no idea." Of course, some of that awareness can be traced to awareness of disparate inputs (like, the differences between $$ spent on wealthy schools vs. poor schools) and inputs were now verboten. The more particular argument about the new NCLB-fueled "awareness" is that while we may have known about disparities in US education, we needed some kind of shiny data that could be used to convince policymakers; Carlson seems to be hinting at that here.

However. Here's the language that Carlson uses to describe what NCLB did. It "illuminated the outcomes of different groups. It allowed "for a better grasp." It "led to a clear-eyed understanding" and "such illuminations hit particularly hard." Carlson's description hits hard on the idea of being able to better see the disparities in US education, but he has absolutely nothing to say about what actions grew out of all this illumination. Under NCLB, it wasn't allowed to talk about the possible contributing factors for the disparity, and policymaker's new clear-eyed understanding consistently failed to lead to any actual action. 

What good does it do to shine a light on an issue if policymakers then say, "Yup, there it is. Somebody ought to do something about that. Probably those teachers." That was one of the central problems of NCLB. Problems were illuminated and policymakers did nothing. The great wave of accountability was for teachers--and not for anybody else.

Developing data systems. Carlson thinks that now that we have all these data (because numbers are magic) we have all sorts of insight. But the data is by and large results from the lousy Big Standardized Tests. Garbage in, garbage out. And ed reform's increase of the grasp of Big Data is nothing to brag about. 

So where does Carlson think they went wrong?

Setting unrealistic goals and expectations. Well, yes. Politicians set an impossible goal of 100% student proficiency by 2014, with the ever-increasing goals set to become unattainable shortly after many politicians left office. But hey--it was all okay, because the education law was due to be rewritten and reauthorized before then, allowing politicians to stop the train before it hit the wall. Instead, Congress dithered and the Obama/Duncan administration got to use the looming deadline disaster as leverage to get states to sign on for the new set of reforms. Oopsies.

Carlson correctly notes that baking the unattainable goals into the law guaranteed that it would ruin public support. It certainly guaranteed from the very first moment that teachers would know it was not a serious attempt to improve education, but simply political grandstanding. Ten years later, those unachievable goals became demoralizing as well; by the early 2010s, there were only two types of schools in this country--those that were failing and those that were lying. Carlson correctly notes that parents saw a disconnect between how they viewed their school and how the government viewed it, and decided mostly that it was the government that was wrong. 

What Carlson doesn't address is the why. Why would policymakers choose such an option that was so clearly a dumb? There was more at play than, as I said above, the belief that they could stop the machinery before it started to chew schools up.

For one, the 100% NCLB goal gave proponents a nuclear option in debate. In those days, if you tried to bring up some of the challenges or obstacles to 100% proficiency, NCLB supporters simply asked, "And which children do you want to leave behind?" The program came with a rhetorical tool for painting all opponents as child-haters.

For others, the inevitable failure rate of public schools was a feature, not a bug. Nothing provides more support for the modern school choice privatization movement than a tool for painting public schools as failing, and NCLB guaranteed that all public schools would, eventually, be labeled as failures. For charter and voucher fans, it was a marketing dream. For opponents of teachers unions, it was a golden opportunity to gather ammunition. Teachers said, "It's not fair to judge us by a system that is a bad measure and is designed for ultimate failure." Opponents shot back (and still do) that teachers and their unions were just afraid of accountability because they didn't want to have their sloth and incompetence revealed.

Narrowly focusing on reading and math test results. Carlson gets this exactly right. The test-centric system signaled what NCLB truly valued, and schools twisted themselves into ugly nots trying to give NCLB what it asked for--scores on a bad multiple choice math and reading test. Curricula were narrowed, students lost breadth in their education, and test prep reigned supreme. The emphasis on high stakes testing is the signature policy of the last twenty years. It has provided little real accountability, and has twisted education out of shape in the process. But hey-- it generates lots of numbers and spreadsheets and data.

Federal control without flexibility. NCLB was in large part about federal politicians and bureaucrats looking at the 1990s and saying, "Well, you wouldn't let us nudge you into doing what we wanted, so now we're just taking control." And it contained a bunch of (usually) unstated assumptions about why, which Carlson unpacks very neatly:

The NCLB accountability system's inflexibility highlights its motivating assumption; Educators weren't trying hard enough, and threatening to punish their schools would make them work harder. It's a mistake to make policy based on assumptions that question educators' motives and efforts.

NCLB assumed that teachers were the problem, and built its system based on that insulting and ill-informed notion. Race to the Top and RttT Lite (waivers) doubled down on that notion. Twenty years later, it's getting harder to find people who want to fill teaching jobs. 

Carlson's grasp of what went wrong is pretty good, but he doesn't really admit that those failure far overshadow any possible gains from the policy. He's worried that this will taint the notion of accountability; I feel certain that the accountability pendulum will continue to swing back and forth as it always has. 

He hopes that accountability fans will learn from this. I feel confident that, mostly, they won't, in part because a universal system for universal accountability is an impossible target. Any accountability system has to be able to explain-- accountability to whom, for what, measured in what way, in order to accomplish what. Not having a good answer for all of those questions guarantees a flawed system that eventually collapses on its own fractured base.

Teachers and schools should be accountable. So should policymakers, politicians, and educrats. And anybody who claims they have an easy way to do it is selling something.