April 22, 2009

Margins for error in policy

We're operating without water in my building for today and tomorrow, at least until the lab reports come back after the drop in pressure overnight. Faculty and staff know, so they'll get to know their colleagues in nearby buildings. I feel sorry for students who run to the right (usual) place and then hope to dip into the bathroom right before class without knowing in advance that the bathroom is unavailable; I hope they all give themselves a little extra margin for error in timing.

I wish the same for policymakers, that they give themselves and their desired/favored policies some margin for error. A policy that falls apart without perfection is a doomed policy, and while everyone understands this, it's sometimes hard to put in place. Here's the practical difference between the stimulus package and Geithner's management of financial policy: the stimulus package can do a lot of good even if implementation is imperfect. I may not get my desired high-speed rail line going from Tampa to New York or Chicago, but someone will get jobs, take the money and spend it, and thus help replace the demand we're losing in this downturn. I'm much more concerned about Geithner's management of the financial mess and the resuscitation of credit markets, because I think there is far less margin for error without nasty consequences (either a waste of money or ineffectiveness).

To some extent, the discussion of the difference between assurances and firm enforcement in federal education policy is an issue of margin for error. I'm an historian, so I can go back to the prehistory of the 1960s (for all you young wonkish types) and Gary Orfield's first book, The Reconstruction of Southern Education (1969). As Orfield explains, Southern states resisted and tried to work their way around the requirements of the Civil Rights Act's Title VI (nondiscrimination), and it took some years for the federal government to make a bona fide threat of enforcement before school districts would desegregate in response to demands by the Office of Civil Rights. Someone who expected instant enforcement would have been sorely disappointed. Someone who expected that lawyers and the federal government would have to push and push hard for several years to begin the ball rolling -- and really rolling -- would have been more realistic.

There's isn't very good language for talking about this with education policy other than the vague terms implementation and transition. Some fields do have practical terms, though. For example, in meteorology the term I have heard tossed around with regard to hurricane forecasting is the "path of least regret." That means that if the choice for hurricane forecasters is between alerting people to evacuate when there is a definite chance it's unnecessary or failing to alert people to evacuate when they might really need to, the forecasters see evacuation or other preparations as the path of least regret. That term does not mean that forecasters always pick the path of least regret, but the language allows them to discuss choices in a clear fashion.

I'll be clear: while I agree with some of the mandates in ARRA money, I've already gone on the record as skeptical of unique student-teacher record linkage. But because regulation is a fact of life in both state and federal education policy, I think it's important to step back occasionally and think about the broader issues involved. That's part of why I've come to think that accountability systems need have a positive defense that teachers and schools can use, because it can allow systems and individuals to manage risk in ways that benefit students. But I'm getting ahead of myself. Basic takeaway: all regulatory policies need some margin for error.

Listen to this article
Tags: desegregation, stimulus
Posted in The academic life on April 22, 2009 11:22 AM |