No Child Left Behind: Measuring Progress

"As yesterday's positive report card shows, childrens do learn when standards are high and results are measured."— President Bush, New York, Sept. 26, 2007

Alright, so we'll start with a cheap shot. But, following that, a serious question. Yesterday's NYT published an article on the most recent round of national standardized test scores, which may or may not prove the success (or lack of success) of the No Child Left Behind act:

America’s public school students are doing significantly better in math since the federal No Child Left Behind law took effect in 2002, but gains in reading achievement have been marginal, with performance declining among eighth graders, according to results of nationwide reading and math tests released today.

The results also showed that the nation has made only incremental progress in narrowing the historic gaps in achievement between white and minority students, a fundamental goal of the federal law.

The tests, known as the National Assessment of Educational Progress and administered by the Department of Education, will be carefully scrutinized by lawmakers and educators debating whether to reauthorize the law this year, and if so, what changes to make.

They offer ammunition to both sides of the issue: the business leaders and other groups who support the law’s renewal, and the teachers’ unions and groups who say the law’s emphasis on standardized testing hurts schools.

So, given this ambiguous assessment of the impact of a public policy, and given the increasingly bizarre political environment surrounding education reform (no sane person can deny that we need it, but it's hard to figure who's got the right intentions in these battles...), can anyone point me in the direction of any meaningful evaluations of No Child Left Behind, or other efforts, national or local, at education reform? And more broadly, any interesting discussions on how the impact of educational policy (or techniques) can be effectively measured?