Dare We Compare?

How American Students Stack up Against the Competition

The results of the Trends in International Mathematics and Science Study were released on Dec. 15, detailing the performance of fourth- and eighth-graders around the globe. On the eve of the results’ publication, I predicted:

Japan, Singapore, South Korea, Taiwan and Hong Kong will be clustered at the top of the international heap. U.S. fourth-grade students will perform at about the average for industrialized nations, while U.S. eighth-grade students will be below the average for industrialized nations — possibly far below it.

Stay Engaged

Receive our weekly emails!

Here’s how it played out: Among eighth-graders, the top five nations in combined mathematics and science performance were Singapore, South Korea, Taiwan, Hong Kong and Japan. Among fourth-graders, the top four nations in combined mathematics and science performance were Singapore, Japan, Taiwan, and Hong Kong (Korea did not test students in the fourth grade).

How did the United States perform compared to other industrialized nations — that is, the top 40 nations in terms of per-capita income? [1] At the fourth-grade level, American students were nine points above the average in science and 11 points below it in math, putting them almost dead average overall. At the eighth-grade level, American students were four points below average in science and 24 points below average in math, putting them clearly, but not abysmally, below the rich-country average.

The weak point in my prognostication thus appears to be my suggestion that U.S. eighth-graders’ overall performance might be "far below" the average of industrialized nations.

But perhaps it actually was.

Many nations that typically outscore the United States in math and science at the eighth-grade level did not participate in TIMSS 2003. Those countries include France, Germany, Canada, Ireland, Finland, Switzerland, Iceland and Poland.

But while they skipped TIMSS 2003, they all participated in another test of mathematics and science: the 2003 Program on International Student Achievement (PISA). Tellingly, every one of those countries significantly outscored the United States on the PISA test. In math, Canada bested us by 49 points, while Finland outscored us by 61. In science, France and Switzerland beat us by 20 and 22 points, respectively. If all of these nations had participated in TIMSS 2003, it seems likely that U.S. performance at the eighth-grade level would have been considerably further below the average of industrialized nations than it already was.

* * *

One question is often raised in response to international test comparisons: Do these results really mean anything? In the past, international testing programs have been criticized on a variety of grounds. Two allegations, in particular, have been common: first, that other nations have not tested as large a percentage of their student population, and hence their scores have been inflated; and second, that our best students are among the world’s best, with our average being brought down by a large cohort of low-achievers.

Whatever the historical validity of such concerns, they are now, if anything, reversed. In science, the overall U.S. participation rate[2] at the eighth-grade was just 73 percent, the third-lowest of all 45 participating countries, and 11 points below the average participation rate of industrialized nations. In fact, the United States had the third-lowest overall participation rate for both grades in both subjects. Japan, Taiwan and Singapore all had participation percentages in the 90s.

How about our best and brightest? At the fourth-grade level, there is some truth to the idea that the best American students are among the best in the world. Looking only at the top 5 percent of test-takers, American fourth-graders beat the average of wealthy nations by 13 points. By the eighth grade, however, the tables have turned, with America’s brightest students falling 10 points behind their foreign peers.

If we carry this comparison to the final year of high school using the 1998 12th-grade TIMSS results (the most recent available), we discover that America’s top students placed last in combined science and math achievement among all the industrialized nations for which data were available. In both math and science, the gap between our best and the world’s best was substantially larger than the gap between our average performance and the average performance of other nations – not smaller, as many Americans believe.

The start of a new year is the ideal time to face reality. The notion that America’s public school problems are confined to inner cities, and that our wealthy suburbs produce world-beating high school graduates is a myth. It’s time we resolve to do better.


Andrew J. Coulson is senior fellow in education policy for the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the author and the Center are properly cited.

[1] My intent was to compare the United States to its economic and technological peers. The simplest and least arbitrary way of doing this was to consider the world’s leading nations by per-capita income. The top 40 seemed a good cut-off point, since narrowing the field would have eliminated advanced nations like New Zealand, while broadening it would have included less-advanced nations like Cyprus.

I obtained my data from the World Bank’s latest Gross National Income (GNI) per-capita table, adjusted for Purchasing Power Parity. Using the PPP allows a more accurate comparison of what people in various countries can actually afford to buy than simply adjusting GNI figures with prevailing currency exchange rates. Interestingly, the selection of the top 40 countries by PPP-adjusted GNI per capita actually excludes a number of countries — Korea, Estonia, Hungary, Latvia — that often beat the U.S. on international tests, including some of the latest TIMSS tests.

[2] This is the percentage of schools and students that were randomly selected for participation in the study that actually did participate.