More on incentives

I am seeing an increasing number of articles pointing out the growing gap between results of state tests and those on the federal NAEP tests. A number of states report a much higher percentage of students proficient in math and reading than the federal test finds for students in that state.

The explanation, of course, is that the federal No Child Left Behind requires that all children in schools be proficient in reading and math within a few years. But in a bow to states-rights conservatives, it left the testing and the definition of proficiency to the states. The predominant incentive on the states is to set the standards low so that their schools will appear to be turning out proficient students.

One answer would be national tests, similar to those in most other countries. Educationally, this seems like the best solution. Mathematics, for instance, does not change when a graduate crosses the state line. Even though the advocates of national testing seem to be increasing, I doubt that the political will exists for this. Much of the left will oppose it because they hate testing, while many on the right will oppose national testing because it implies national standards.

A possible compromise might be to leave states to determine their own tests but use NAEP results to recalibrate the proficiency levels. For example if 40% of the students in a state were proficient by NAEP standards, the state score corresponding to the 40th percentile would be the cut-off for proficiency.