The Accountability Illusion
Take away all the jargon, emotion, envy, confusion, and embarrassment and much of the No Child Left Behind debate comes down to this: Which schools are good, which are bad, and does NCLB do a decent job of telling the difference?
The short answer, provided by a major new study from Fordham and the Kingsbury Center at the Northwest Evaluation Association, is no, not by a mile.
The analysis is complex and the report is long but its premise is simple: Take a set of real schools, pretend that we can drag them across the map and drop them down in various states, and see how many would make "Adequate Yearly Progress" (AYP) in each place. If the U.S. had something akin to a shared notion of what it means to be a good or bad school, we wouldn't see a whole lot of variation.
Yet we found nearly the opposite. In a few of the 28 states we studied (e.g, Wisconsin, Arizona), almost all of the elementary schools in our sample made AYP, while in other jurisdictions (e.g., Massachusetts, Nevada), almost none did. Putting it bluntly, most of the schools in our sample would be considered failures in some states but perfectly okay, even praiseworthy, in others. These are the same exact schools, mind you. Same students. Same teachers. Same achievement. What's different--sometimes drastically different--are the arcane AYP rules that vary from state to state.
Such variation surely existed before NCLB. Does it matter that it exists now?
Yes, for three reasons. First, it surely demoralizes educators (and probably students, parents, taxpayers, etc.) to know that their very own schools, deemed "in need of improvement" under NCLB, would be considered acceptable, even praiseworthy, if located elsewhere. Such capriciousness not only breeds cynicism toward NCLB; it undermines the idea of accountability itself--and certainly retards efforts to revitalize horrific schools and boost low-performing pupils.
Second, what drives the state-to-state variation in AYP results isn't principled differences about what it means to be a good school. Instead, obscure, deep-in-the-bureaucracy decisions around "cut scores," "annual measurable objectives," "minimum n sizes," and "confidence intervals" create these discrepant and discordant outcomes. We'd mind less if the variation were based on things that truly matter: whether schools are judged on their progress over time, for example, rather than each year's absolute performance; whether they're helping all students make gains or just those below a fixed bar; whether determinations hinge solely on reading and math or include other core subjects like science and history; and so forth. Those would be legitimate reasons for discrepancies, issues worth arguing over. But that's not what we're seeing here. Rather, we witness state education departments going through subtle machinations to create outcomes that they judge sensible, or at least politically saleable.
Third, NCLB created the trappings of a national accountability system. Variation around school ratings was okay when states also set the penalties (if any) for schools that didn't make the grade. But now every state operates under a federal mandate to offer "public school choice" and "supplemental services" (tutoring) to children stuck in "failing schools," and under further mandates to overhaul change-resistant schools. They're told in which year to intervene in which way. This cascade of sanctions is the same everywhere in America. The man in the street surely believes that it's a uniform accountability system. Yet we see that it's not. All those sanctions and interventions, uniform as they are, are triggered by AYP systems that couldn't be more different. At best, this is a disconnect. At worst, it's chaos.
What to do? Some imply that NCLB might be "repealed." Not likely. Another option would be to nationalize and standardize everything. Perhaps that's not as unthinkable as it might once have been, given the astounding growth of federal authority these past few months. We could move toward a national school system. Arne Duncan would determine each year which of America's 100,000 public schools makes the grade and what to do about those that don't.
But that isn't what we'd recommend. That approach would push Uncle Sam ever deeper into the impossible task of running schools and turning around those that fail.
We picture a very different approach to NCLB 2.0. Uncle Sam should create incentives for states to sign on to common standards (such as those now under development by the National Governors Association and other groups) and tests. Ensure that the tests are rigorous and comprehensive. Publish their results for every school in the country, with data sliced every which way--by race, income, disability status, progress over time, etc. And then stop.
That's right, stop. That's the end of Uncle Sam's role under NCLB 2.0.
Return to a regimen where each state then decides what to deduce from those test data and what to do about schools whose results don't satisfy the state's definition of success. Civil rights groups and others (e.g. Education Trust) that don't like the states' judgments can create their own school ratings, using the same uniform national data, accessible and transparent to all. So could private organizations such as GreatSchools.net. We could reopen the debate about what it means to be a good or bad school. And then it would be up to states and communities to do something about schools that aren't making the grade.
Such "Reform Realism" foregoes the utopian rhetoric of the NCLB era. It admits that Uncle Sam cannot actually ensure that every American child will get a world-class education. But what this strategy would do is ensure greater transparency around student results--something our new study shows is hard to come by--based on assessments that are rigorous, credible and uniform throughout the land. And it would reinforce the idea that states remain responsible for K-12 education and must make decisions that their own citizens will accept. Best of all, it would end the gamesmanship that has characterized the federal-state relationship since Congress passed NCLB.
"Study: Academic standards vary across states," by Libby Quaid, Associated Press, February 19, 2009
blog comments powered by Disqus
About the Editor
Michael J. Petrilli
Executive Vice President
Mike Petrilli is one of the nation's foremost education analysts. As executive vice president of the Thomas B. Fordham Institute, he oversees the organization's research projects and publications and contributes to the Flypaper blog and weekly Education Gadfly newsletter.
May 16, 2013
Sign Up for updates from the Thomas B. Fordham Institute
- Core Knowledge Blog
- Daniel Willingham: Science and Education Blog
- Education Next Blog
- Getting Smart
- Gotham Schools
- Jay P. Greene
- Joanne Jacobs
- NACSA's Chartering Quality
- National Journal Education Blog
- NCTQ Pretty Darn Quick
- NCTQ Teacher Quality Bulletin
- Ohio Education Gadfly
- Politics K-12
- Quick and the Ed
- Rick Hess Straight Up
- The Corner
- The Hechinger Report
- Top Performers