With the obvious caveat that in politics and policymaking there is no minute like the last minute for getting things done, it does look like NCLB reauthorization is on life-support for ’07. The Miller-McKeon “discussion draft” in the House of Representatives (dubbed the “destruction draft” by more than one wag) has caused more dissension than agreement and collateral damage in both chambers as well as on the campaign trail. And, even McKeon himself is backing away now. Hell, Ed Week’s Hoff can’t even find the draft bill these days…Meanwhile, the Senate is hard at work but it’s hard to see much happening there for the rest of the year with everything else going on. Update: That was just made official. And, the NEA is on the warpath and really leaning on all involved pols, and that hardly helps.
I suspect that there could be a burst of activity next year, perhaps after the primaries but before Congress gets locked down for the year with presidential politics. But the odds are still long. And, at some point, you have to assume the Administration will pull the plug on reauthorization from their perspective. If you pass a law you want to be around long enough to regulate on it.
Meanwhile, less No Child market volatility than some had hoped. The first big returns from the Department of Education’s “growth model” pilot* are coming back and, surprise!, in a lot of places it’s making little difference. Why? Well, because the standards are still pretty modest in most places, No Child already had one growth component in it — the safe harbor provisions, there is a lot of state evasion, and schools with less than 20 percent of kids at grade level — and there are too many of those — will be identified by almost any meaningful accountability system. So as a result of all that right now most the schools that No Child is identifying as persistently under-performing are, well, persistently under-performing. Bottom line: If you want to weaken NCLB you’ll need a more diversified portfolio than just investing in growth…Ed Daily’s Stephen Sawchuk did a nice write-up of some of the data you should try to get your hands on, but I’d be violating copyrights if I posted it here. As more states report data the picture could change, but that’s the deal now.
Also worth noting that you can figure out the questions about which schools would/would not make “adequate yearly progress” by just modeling based on the data. You don’t need a pilot for that; you need a computer and a spreadsheet. The more interesting question is the behavioral one: What changes when you use growth models instead of the current NCLB status model? So far it seems, not a whole lot. Why? Perhaps because there is a much more robust industry in talking about school improvement efforts and fighting against school improvement, than actually, you know, doing the work!
Over at his blog Charlie Barone has been blogging up a storm. He’s all over the “bubble kids” issue. Two thoughts on that. First, the new study from Ed Next provides some much-needed perspective. This, along with other empty calorie strategies, is a problem in some places though hardly as widespread as critics claim. In fact, if teachers and schools were this agile with data in the first place, well, we wouldn’t have some of the other problems we see! Second, the dilemma built into No Child Left Behind is that the floor rises each year or every couple of years (this paper explains all that). So focusing on “bubble kids” to the extent it happens is a time-limited strategy anyway. But, the rising floor “raises” a different and not illegitimate concern: That the targets quickly become out of reach for schools that are trying to improve and are actually making progress.
That’s why some people want growth models…and they’re not wrong so long as the standards remain stringent for what constitutes a growth model. But, like national standards, vouchers, etc…growth models won’t alleviate the basic challenge in school improvement: Improving teaching and learning, in other words, school improvement. A simple yet incredibly challenging issue.