December 13, 2005

Triangulation on accountability?

On December 4, Andrew Rotherham begged Democrats to take a strong stand on accountability ... [or miss] good chances to legitimately attack the Republicans on education and build collateral credibility on other issues. He explained what he terms the lack of such a stand on too many years of inhaling NEA laughing gas about the law while forgetting who the real constituents are. This argument is similar to those of a USA Today editorial, which Rotherham cites.

One side of this argument is a crude reliance on a limited set of polling about attitudes towards one version of accountability. The USA Today editorial makes that point explicitly, while Rotherham addressed it more subtly. As an historian, I'll dun both of them for violating the fundamental rule of reading polls: The attitudes of Americans on any important policy question are more subtle, contradictory, and malleable than any poll can measure. The latest Gallup/PDK poll on education has results at variance with these claims (much like the 2004 Americans for Better Education poll came up with different results from that of the NEA), which is not to say that the September 2005 PDK poll is perfect but that overgeneralizing is unwarranted. I'm not sure what the motivation is of either the USA Today editorial board or Rotherham in this, but the namecalling is unproductive and distasteful.

Full disclosure: I'm a member of the United Faculty of Florida, whose state affiliate is the Florida Education Association and thus is jointly affiliated with the NEA and AFT at the national level. I'm moderately active at the higher-ed level in Florida and see a small part of internal debates. I don't always agree with my state affiliate or the nationals, but they're relatively democratic with a small d, which means that discussion can be quite heated if civil in the arenas I've seen.

The more important points about policy and politics are below the cut-line.


The more serious argument is that the high-stakes-testing version of accountability is not only effective, not only promotes equity in education, but is the only strategy to do so, and anything other than that is just plain wimpy on civil rights. That's a strong statement, in part because it implies that the burden of proof lies with those who disagreement with a particular version of accountability.

But there are several fundamental problems any serious researcher should have about this claim, and the first one is whether there is a clear consensus about the empirical evidence here (on the effectiveness of high-stakes accountability). As the editor of an education policy journal, I'm not only officially agnostic about the empirical claims of high-stakes testing proponents and opponents but in favor of active disagreement about this (as contentious articles on this point boosts readership for the thus-far-revenueless online, free journal—we're talking reputation, not cash). Fortunately for journals if unfortunately for policymakers and pundits who crave certainty, there's widespread, well-founded disagreement—even acrimony—over the empirical evidence.

So policymakers are really making policies and changing practices without much consensus guidance from research. No surprise there, says my cynical side, but on a more pragmatic approach policymakers often must make changes without evidence about consequences, because the consequences of inaction would be greater. So there is some abstract justification for treating the citizens as guinea pigs. Each generation of children comprise a group of guinea pigs for that era's fads. Some of the fads are serendipitous and some are just foolish.

So if high-stakes testing has no research consensus behind it, but we can't dismiss it as entirely without merit (even if some arguments in its favor make me wince), then how should we categorize it? As an historian, I see the last 20 years of accountability rhetoric and policy as a period of experimentation. More than 30 years ago, an astute observer of a 1972 conference of Southern state legislators noted that his or her (almost certainly his) informants didn't know what they really wanted in accountability, but they were sure that administrators didn't know what it should be either, and they didn't trust the administrators. (The report of this anonymous observer appears in both the papers of Governor Rubin Askew in Tallahassee and also former Florida House Speaker T. Terrell Sessums at the University of South Florida, if my memory serves.) Over the past 33 years, then, we've seen a variety of attempts to get a handle on accountability, loosely speaking making sure that schools fulfill some implied social contract in return for public funding.

Yet, somehow, you generally don't see much analysis of accountability efforts as experimental. I've read lots of huffing and puffing about values—much of it well-intentioned, some of it thoughtful, some of it full of warm nitrogen—and a lot of assumptions about what high-stakes testing is actually doing. But often enough, after reading either newspaper stories or internet-released reports, I feel more hyperventilated than enlightened. There is more complexity in the ever-changing world of schools-under-pressure than a single article (such as one that I'll be releasing in early January) or a book (such as Linda McNeil's Contradictions of School Reform) can capture. This is good news for thoughtful researchers: more to keep us busy! It's not so good news for those who want certainty in their lives. To the latter, I say: tough!

And, for the record, I have no objections to combining political analysis with school reform. Askew did that brilliantly in the early 1970s for K-12, arguing that the key issue for education in Florida at the time was equality for all poor and disadvantaged children. I draw the line, though, at ad hominem attacks, for a few reasons: they distort reality, they're distasteful, and they suppress open debate. You can have a vigorous debate civilly, and I expect it. To those who are advocates of high-stakes accountability or critics of it, I challenge you to suppress your snarling. Try gazing in the mirror when doing it: It really doesn't become you.

Listen to this article
Posted in Education policy on December 13, 2005 7:01 PM |