October 25, 2005

Fixing NCLB?

In Jenny D.'s blog is a question about fixing NCLB, which has started a bunch of pontificating. So I added my own as a comment and then realized I needed to bombard my 0.3737365 readers with it. In essence, major policy questions about NCLB are about two things: the shape of accountability and the federal-state relationship. Elsewhere, I've described my concerns about Florida's particular system and accountability in general, but it's been a few years, and this is a good excuse for some rethinking. Warning: this is not really about NCLB but about my cynical observations regarding accountability in general and federal-state relationships.


  1. Accountability should focus on three questions: are we expecting the right things from students, are students meeting those expectations, and what needs to change so the answers to the first two questions are yes? Any system that doesn't focus the attention of teachers on those questions is bound to fail.
  2. I'm convinced that any high-stakes system will distort teaching in some ways. It may provide incentives for some schools and systems, but there will also be puerile attempts to triage students, teach to the test, and game the system with student rolls and other things. Any system that does not take this distortion into account is bound to fail in some critical ways, and anyone who pretends that teaching to the test and gaming the system doesn't exist is channeling former FEMA head Mike Brown. There is currently no systematic way of gauging distortion except by observation and shrewd guessing about schools as institutions.
  3. There is a fundamental tension between gauging the achievement status of students and their growth. As Jennifer Hochschild and Nathan Scovronick point out in Chapter 4 of The American Dream and the Public Schools, emphasizing one over the other is dangerous. But it's not a matter of figuring out the right formula, as some propose by combining status with growth measures. It's a political problem. Let's imagine that we make very explicit the desire to close the achievement gap. "You, there! Disproportionately white, middle-class suburbs! We want the achievement of your students to grow over the next 5 years, but only a little, so everyone else in your area can catch up with you!" What do you think the response would be, if we phrased things that clearly?

Federal-state relationships

  1. The federal government has been most influential in shaping local behavior when acting as a guarantor of civil rights in some way. The wisest thing is to shrink the federal government's role in NCLB to that and a supporter of research, rather than trying to make the Secretary of Education into a national superintendent. On the other hand, unlike David Tyack and Larry Cuban (most prominently in their 1995 book Tinkering toward Utopia), I don't believe that top-down efforts are sui generis flawed. Their picture is colored by top-down accountability efforts, I suspect, and they would have to change the answer if asked to apply their reasoning to civil-rights laws.
  2. The Bush administration is only the most crass administration in hiding and manipulating data to serve petty political purposes in education. Don't expect it to end in 2009, even if other presidents don't try to buy journalists.
  3. Research and dissemination efforts should be funded solely on peer review, and primarily field-initiated. Over the last 15 years, centers and other less-competitive contests have dominated grants and contracts on the research side, and it's just shameful. It would be a wise thing if the federal government stopped funding anything claiming to be a national center for about 5-10 years and put the whole funding scoop into peer-reviewed, field-initiated studies.

"And so ends my catechism..." Listen to this article
Posted in Education policy on October 25, 2005 6:18 PM |