As Bush prods Congress to reauthorize No Child Left Behind he has rightfully taken a tremendous amount of heat for his seemingly well-intentioned but poorly-executed education law. As NCLB has given the federal government a larger role in education than it has ever had before, stakeholders including state and local policymakers, teachers and education advocates have called for Congress to "get out of the classroom." But most seem to agree that while the federal government should not be setting a national curriculum or mandating teaching methods, it does have a key role to play in setting standards for student achievement. So it's interesting that while federal policymakers seem all too happy to take a hands-on role in schools where they shouldn't, they've declined to do so where they should.
Learning standards are best set at the federal level, but NCLB allows states to set their own. The definition of "proficient" should not be any different in Mississippi than in Michigan. A uniform standard of proficiency should be set by the federal government so that students in different states are not held to different expectations. We should not tolerate low-performing schools in a state with low standards simply because they meet the local criteria for proficiency, while low-performing schools in a state with high standards stick out like sore thumbs and receive extra attention from stakeholders until they improve.
By tying funding to test scores, NCLB has given states a perverse incentive to lower their standards so that politicians and educators can dishonestly pat themselves on the back when tests show their students meet local standards for proficiency. (Lowering standards is often coupled with making tests easier to the point that they are no longer a valid measure of student achievement.) Deflating standards deflates student achievement because school districts align their instructional programs with their respective state standards. One needs look no further than Mississippi where in 2005, 89 percent of fourth graders scored "proficient" on the state's reading test, while just 18 percent scored at "proficient" on the NAEP reading test, considered by some a "national report card."
Some scholars and researchers such have charged that NAEP should not be considered the "gold standard" by which to compare state assessment tests because it has a different "measurement function" than state tests. I will leave it to the experts to determine the merits of using NAEP as a national standard. But regardless of whether they agree on using NAEP or some other measurement vehicle, policymakers must develop uniform, rigorous national standards so school districts know that the same is expected of them as is of every other district in the country.
While we should have the same expectations for every student and school district, we must understand that not every student and school district start at the same place and some will perform better than others. Setting standards at the national level will highlight these disparities rather than hide them as watered-down state standards currently do. Furthermore, national standards will allow us to address these inequalities by enabling policymakers and educators to study the practices of high-performing districts so that they might be implemented in lower-performing ones.
As Diane Ravitch wrote in last week's New York Times , "In our federal system, each level of government should do what it does best." The federal government should set standards and collect information about how students are measuring up to these standards. But it should do so for the purpose of assessing student achievement in different schools in order to help states and local districts in their role of improving student performance, not to punish failing schools. That way local officials can do what they do best - design reforms tailored to the needs of particular schools and students - not inflate test scores to win votes and federal funding.