It's time to stop holding schools hostage to the overly rigid and often counterproductive demands of the federal No Child Left Behind Act, which was supposed to have been rewritten four years ago. More and more schools — many of them good or at least improving — are being labeled failures and are facing severe sanctions as the 2014 deadline approaches, when the law requires schools to make 100% of their students proficient in reading and math. A frustrated Obama administration, which has tried in vain to persuade Congress to overhaul the act, is now pursuing a workaround. But the plan advanced by Secretary of Education Arne Duncan sacrifices some of the best features of the law in an effort to fix the worst ones.
Duncan has announced that he will grant states waivers from selected No Child Left Behind mandates. That would certainly be taking administrative powers to a new level — the department wouldn't be interpreting the education reform law; it would be countermanding parts of it. Yet Duncan has little choice. The law, as he pointed out, doesn't differentiate between schools at which a modest percentage of students are failing and those at which almost all are; nor does it give any credit to schools making real, measurable improvements that don't happen to match the specific targets set for each subcategory of students (black, Latino, white, poor, in need of special education and so forth).
Yet some of the potential requirements for states to obtain a waiver, as vaguely outlined by Duncan last week, have their own problems. The Education secretary appears only partly concerned about devising more reasonable measures of school progress. He also is using the waivers, and the threat of sanctions under the law as it exists, to impose his own agenda for school reform. As he draws up the specific requirements, Duncan should keep in mind why the 2001 act passed with strong bipartisan support.
Before No Child Left Behind, scant attention was paid to what was happening to low-income students, especially low-income black and Latino students, who attended schools with higher proportions of underqualified teachers. A shocking number of them reached high school unable to read so much as Dr. Seuss books. The federal government provides extra money for schools with large numbers of low-income students, but with those funds spent on unproven programs that didn't seem to improve outcomes, Congress rightly felt it was getting a bad deal for the money.
So the law hasn't been an utter failure. If nothing else, its yearly testing requirement showed the nation just how badly disadvantaged students were faring. The law also was appropriately focused on results, not techniques. Schools could employ whatever educational method they thought would be useful, as long as it worked. And achievement has improved at least somewhat in many schools.
Yet No Child Left Behind was riddled with problems from the start. Its poorly conceived formula for measuring progress unintentionally gave schools reason to ignore the lowest-achieving students. Instruction lost depth even in high-achieving schools, as teachers focused on raising scores on multiple-choice tests in reading and math. At the same time, states were left to adopt their own standards for proficiency; those that set the bar low made themselves look good on their own tests. California set its standards fairly high.
All of this must be fixed. But Duncan's solution goes far beyond fixing a broken accountability system; he is prescribing the educational changes he wants to see states make in order to receive a waiver. Those include tying teacher evaluations to standardized-test results, adopting certain curricula and reorganizing low-performing schools by firing much of their faculties.
There's nothing necessarily wrong with these reforms, most of which are worth trying at least. But there is little evidence that any of them will significantly raise achievement. Duncan's approach recalls the days before the federal school-reform law, when well-intentioned but possibly ineffective programs were valued over accountability.
Fortunately, Duncan mentioned overhauling the measurement system as well. One change should be to judge schools by individual students' growth year by year, rather than holding schools to some arbitrary measure of student proficiency. And although standardized test scores are one important achievement indicator, student progress also should be measured by dropout rates, college attendance and portfolios of work that go beyond penciling in bubbles on a multiple-choice test.
Duncan is unfortunately stuck with doing Congress' work. As he prepares detailed requirements for obtaining waivers, he should concentrate on defining reasonable goals for schools. Then he should leave it to states and their school districts to decide how to go about meeting them.