A new study by the American Institutes for research compares the performance of 11 large U.S. districts to that of countries participating in the international mathematics test known as “TIMSS 2003.” As with earlier international comparisons, American kids do better the less time they have spent in school.

At the 4th grade, the earliest one tested, three of the 11 U.S. districts (Charlotte, Austin, and San Diego) score above the average of OECD countries participating in the test. (The OECD is a group of 30 or so nations, most of which are wealthy and industrialized, but a few of which are less wealthy transitional economies). By the 8th grade, the top two large U.S. districts (Charlotte and Austin) included in the report scored at the overall average of the participating OECD countries.

But the above results overstate the U.S. districts’ achievement. That’s because many industrialized countries that typically outscore us (France, Germany, Canada, Ireland, Finland, Switzerland, Iceland and Poland) did not participate in the TIMSS 2003 test. When the U.S. is compared specifically to other wealthy nations, it peforms worse than the AIR report will lead readers to believe. Finally, U.S. performance continues to deteriorate as students progress through high school, and so the absence of high-school test results also gives an inflated impression of relative U.S. performance.

In a nutshell, even two of the top large school districts in America can barely tread water internationally, when compared to students in other industrialized nations.