from Renewing America

Education: Do International Test Scores Matter?

A preschooler on his first day of school in Kuala Lumpur, Malaysia (Bazuki Muhammad/Courtesy Reuters).
A preschooler on his first day of school in Kuala Lumpur, Malaysia (Bazuki Muhammad/Courtesy Reuters).

May 8, 2017
8:00 am (EST)

A preschooler on his first day of school in Kuala Lumpur, Malaysia (Bazuki Muhammad/Courtesy Reuters).
A preschooler on his first day of school in Kuala Lumpur, Malaysia (Bazuki Muhammad/Courtesy Reuters).
Blog Post
Blog posts represent the views of CFR fellows and staff and not those of CFR, which takes no institutional positions.

More on:


We’ve all seen the headlines: American students are far from stellar performers on international tests. Whether it’s the OECD’s oft-cited gold-standard PISA test on math and reading, the TIMSS test on math and science, or the PIRLS test on reading, American students consistently score in the middle-of-the-pack among their peers in the rich world. Arne Duncan, the Secretary of Education, has called such mediocre results a “wake-up call” for urgent action to reform the U.S. education system. The suggested reforms are often inspired by the highest-performers on international tests, including Singapore, South Korea, Japan, Shanghai, and Finland.

But how much currency should we give international test scores?

According to the work of Eric Hanushek of Stanford University, they matter a very great deal; if Americans were testing as high on PISA as Finns, the cumulative gains to the U.S. economy could add up to $100 trillion over eighty years. He also argues that math skills—where U.S. students perform worse than reading—are most important for a nation’s productivity in a modern economy.

There are certainly benefits to comparing the quality of human capital across the world. To the extent that international labor markets transcend borders, there is more direct competition for the same jobs. Test scores may be a more sophisticated way to measure human capital than the more conventional way, which has been the number of years spent in school.

Even if we accept the current array of international tests as a good marker of education quality, questions linger about how useful the international comparisons are. The top-performing countries tend to be small and homogenous. Massachusetts and Minnesota, which are demographically similar to top-performers, also test at similarly lofty levels internationally. Asian top-performers tend to favor a high-stakes and unforgiving testing culture that many in the United States are loathe to adopt.

The United States has never been world-leader on international tests, going back to the 1960s. Yet the United States continues to have among the most productive workers and dynamic economies in the world. Japan, whose students have traditionally done well on math and science tests, has an economy that has been in the doldrums since the 1990s.

Perhaps the international tests themselves are missing the mark and testing the wrong things. Keith Baker, a retired researcher from the Department of Education, ran an analysis of international test scores from the 1960s against a whole set of country performance indicators in the intervening years, from GDP growth to individual productivity, to democratic environment, creativity, and livability. His findings are striking and counterintuitive; the countries with the highest test scores in the 1960s ended up with the lowest scores of national “success” decades later. It is well established in psychological research that an especially high IQ or being a high school valedictorian is not necessarily correlated with life success later on. In China, there is growing awareness of troubled “high-testing, low-ability” adults, who could not thrive professionally as adults but did extraordinarily well on the grueling high school examination test known as the Gaokao that determines college entry and, many believe, a student’s life prospects. James Heckman of the University of Chicago has done ground-breaking research showing that some of the most valuable skills learned in high school are less academic and more akin to personality traits, such as “stick-to-it-iveness,” optimism, and grit in the face of setbacks.

U.S. students may not score better than their East Asian peers in math and science, but surveys indicate they have higher levels of self-confidence—a trait that has its downsides, but that also correlates strongly with entrepreneurship. Countries that come out on top with the PISA test also tend to have low levels of confidence and low levels of entrepreneurship. The United States devotes more money and student time to extracurriculars and less to in-class instruction. Many point to the reversed priorities as a reason for U.S. students’ middling test scores, but it may also be a reason why the U.S. education system produces adults who are more confident in their abilities to start and run a business.

In the future, confidence may be every bit as important if not more important than math skills. Thomas Friedman has written extensively about his new equation for individual success in a highly competitive global economy: IQVivek Wadhwa, who writes regularly about what makes Silicon Valley tick, adds that “the independence and social skills American children develop give them a huge advantage when they join the workforce. They learn to experiment, challenge norms, and take risks. They can think for themselves and they can innovate.”

Before the United States takes big steps to emulate high-scoring East Asian countries, it should take note of how those countries perceive their own education systems. When international test scores are released, their press often overlooks the high performance, focusing instead on the low confidence of the test takers. And in a telling move, East Asian countries are looking for ways to tone down the high-stakes testing in favor of more extracurriculars and a well-rounded student experience.

While it is important to guard against undue complacency about U.S. test scores, we should also guard against jumping to conclusions about whether the test scores represent a crisis of U.S. education.

Correction: A previous version of this blog stated that, according to Eric Hanushek's research, "if Americans were testing as high on PISA as Finns, the U.S. economy could be $100 trillion bigger eighty years down the road." This is incorrect. The $100 trillion figure is actually the cumulative difference in GDP over eighty years.

More on: