The future effect of backloaded NCLB models demonstrated in Calif.

There is a day of reconning coming for states that have backloaded their NCLB targets.

As 2014 approaches, more and more successful states will experience the same kinds of decline seen recently in Massachusetts - while their students continue to improve.

The law known as No Child Left Behind (NCLB), enacted in 2002, set an ambitious goal: that across the nation, every state would test students annually in reading and math, and that the number of students scoring at the level of "proficient" or higher would rise each year, until all students reached proficiency in the year 2014.

Towards that end, each state developed its own assessment tests and identified a rate of adequate yearly progress (AYP) towards full proficiency by 2014. Schools that do not meet AYP can face sanctions ranging from being identified as a school needing improvement to (after five years) being subject to corrective action and restructuring--even a complete reorganization or takeover of the school.

One of the challenges of meeting AYP is that schools, districts and states must report not only a rise in total scores, but also progress in the scores of subgroups of students including minority students, English language learners (ELL) and students with disabilities.

"The result is that the lowest-performing subgroup will ultimately determine the proficiency of a school, district or state," says Rich Cardullo, one of the authors of a paper published in the Sept. 26 issue of Science magazine, which analyzes testing data from California's elementary schools.

Researchers at the University of California, Riverside (UCR) used state assessment data reported for the school years 2002-2003 through 2006-2007 to project the growth in student proficiency through 2014. Data was drawn from more than 4,900 California elementary schools.

The researchers used three different growth models (represented by the blue, grey and green lines) to project average annual growth in proficiency for mathematics (solid lines) and English language arts (dotted lines).

Models are plotted out to 2014 to illustrate that the available data (through 2007) does not indicate the accelerated growth in proficiency required to meet legislated goals. California's benchmarks for adequate yearly progress (AYP) under No Child Left Behind are shown in the red lines.

More information on this research appears in the Sept. 26, 2008, edition of Science magazine.

Credit: University of California, Riverside by way of Schools Matter.

Kinda like the way Kentucky Landed Toyota

This from Kevin Carey at The Quick and the Ed:

Lawless Policymaking

...Secretary of Education Margaret Spellings announced that she would allow 10 states to change what they do with schools that fail to make "adequately yearly progress" under NCLB.

...This is not, in the strictest sense of the word, legal.

The black-letter law is pretty clear: If you want federal money, you have to implement an accountability system that works as follows.

There are no sections or subparagraphs that say The Secretary of Education may at her discretion alter or ignore the previous subparagraphs if people seem to agree they're not written well and Congress doesn't get around to reauthorizing the law on schedule.

But this is nothing new; Sec. Spelling did the same thing with a "growth model" pilot project a few years ago, which allowed states to rate schools based on year-to-year improvement, rather than absolute levels of performance.

Sec. Spellings can do this for a simple reason: nobody objects.

She's using what amounts to an extra-legal, consensus-driven process of amending the law without going through the whole hassle of introducing bills, havings votes, getting lobbied, etc. etc.

The check on this method is that anyone with standing can derail it simply by saying so: If Senator Kennedy, Representative Miller, Representative Boehner, or any of the major interest groups hated the idea, it wouldn't be happening. But since the
only real objections have been "it doesn't go far enough," the process goes ahead. It's actually a pretty efficient when you think about it....