[Re-posted from Mark Kleiman, with permission]
I'm on a listserv embracing a bunch of real journalists and a bunch of bloggers, academics, activists, and think-tankers, representing a pretty good spread of Blue opinion. (Please, don't tell Mickey Kaus; you might hurt his feelings.)
Two threads on that list come together in a way I haven't seen commented on before.
One thread is about the extent to which the Information Age dictates basic changes in social policy. Crudely speaking, those who are politically centrist tend to think that the Information Age changes a lot, and that the thinking behind many established programs dear to liberals is Industrial Age thinking.
The other thread on the same listserv is about No Child Left Behind. Supporters of NCLB think it puts pressure on schools to actually teach somebody something, and especially to stop failing poor and otherwise disadvantaged kids. Critics focus on the problems: the tests measure to narrow a spectrum of capacities, measure them too infrequently, measure them badly, measure them in ways that aren't robust to "teaching the test," and lead to a soul-deadening rote-learning atmosphere not only in the failing schools where it may be better than the aimless chaos it replaced but also in schools whose demographics ought to make them aspire to do better. I'm definitely on the critical side, but the supporters certainly have a point that the system that NCLB challenged was failing badly and not obviously getting better.
Now in the broader political world, I think it's fair to say that support for NCLB is a typical centrist, New Democratic, technocratic idea, while opposition tends to be among people who consider themselves unreconstructed liberals/progressives.
But note that the (non-Luddite) critique of NCLB is precisely that it uses an Industrial Age model of what it means to manage by measurement. The spirit of NCLB is the spirit of Taylorite "scientific management," with time-and-motion experts finding the One Best Way of doing each task, workers treated as interchangeable parts whose only role is to do their job the One Best Way and never think about it (think: Open Court), and quality assurance done by inspecting each part as it comes off the line by comparison with a set of rigid yes-or-no criteria and then counting the number of failing parts. It's as if W. Edwards Deming had never invented statistical QA and Douglas McGregor had never adapted the Maslovian needs hierarchy into the Theory Y approach to encouraging good, creative performance by employees.
(I sometimes think that the worst thing about NCLB is the aid and comfort it gives to the Luddites in the educational establishment who use "Our function is much too sacred to measure" as a defense against demands that the system start producing value for money, especially when it comes to teaching kids who don't get an educational leg up at home. But if I were in a grade-school classroom, I'd probably find other aspects of it even worse.)
Of all the social-service functions, education probably has the most to gain from using Information Age technology. The teacher-stands-by-the-blackboard-and-talks-to-thirty-students model of schooling is doomed by the Baumol Cost Disease, which is why real per-student costs of schooling have been skyrocketing and teacher quality has been falling. Putting that system in an Industrial Age measurement straitjacket has to be precisely the wrong thing to do.