AIR’s Gary Phillips has previously shown how U.S. states compare with countries around the world in math and science achievement—generally
not well at all. In this new analysis, using NAEP, TIMSS, and PIRLS
data, he demonstrates how widely discrepant are the academic expectations of
these jurisdictions. In fact, for math, the expectations gap between
states, at a whopping four grade-levels, is double the nation’s
black-white achievement gap. For Fordham regulars, this story is old
hat—our 2007 study, The Proficiency Illusion,
also found state cut scores to vary greatly—and other research has
shown most of them to be much lower than NAEP’s “proficiency” cut-off.
But Phillips also goes on to suggest a new “benchmark method” of
standard setting (which, he hints, could be used for determining the cut
scores on new Common Core assessments). This method links state-based
performance-level descriptors to those for international assessments,
assuring both national and international comparability of state
proficiency levels. In an America seeking to regain its international
edge, this benchmarking idea is a good first step.
Gary W. Phillips, “International Benchmarking: State Education Performance Standards,” (Washington, DC: American Institutes for Research, October 2010).