GEORGE F. WILL
Duncan's school of wisdom
"Since 1995 the average mathematics score for fourth-graders jumped 11 points. At this rate we catch up with Singapore in a little over 80 years . . . assuming they don't improve."
- Norman R. Augustine,
retired CEO of Lockheed Martin
What America needs, says one American parent, is more parents who resemble South Korean parents. Education Secretary Arne Duncan, 46, a father of a third-grader and a first-grader, recalls the answer Barack Obama got when he asked South Korean President Lee Myung-bak, "What is the biggest education challenge you have?" Lee answered: "Parents are too demanding." They want their children to start learning English in first rather than second grade. Only 25 percent of U.S. elementary schools offer any foreign-language instruction.
Too many American parents, Duncan says, have "cognitive dissonance" concerning primary and secondary schools: They think their children's schools are fine, and that schools that are not fine are irredeemable. This, Duncan says, is a recipe for "stasis" and "insidious paralysis." He attempts to impart motion by puncturing complacency and picturing the payoff from excellence.
He notes that 75 percent of young Americans would be unable to enlist in the military for reasons physical (usually obesity), moral (criminal records) or academic (no high school diploma). A quarter of all ninth-graders will not graduate in four years. Among the 34 Organization for Economic Cooperation and Development nations, only four (Mexico, Spain, Turkey and New Zealand) have dropout rates higher than America's, whose 15-year-olds ranked 23rd in math and 25th in science in 2006. Canadians that age were more than a school year ahead of their American counterparts; Koreans and Finns were up to two years ahead. Within America, the achievement gaps separating white students from blacks and Hispanics portend (according to a McKinsey & Co. study) "the economic equivalent of a permanent national recession."
Another study suggests that a modest improvement (from a current average of about 500 to 525) over 20 years in an international student assessment of 15-year-olds in the OECD nations - improvement in reading, math and science literacy - would mean a $115 trillion increase in these nations' aggregate GDP. Of that, $41 trillion would accrue to America. McKinsey calculated that if American students matched those in Finland, America's economy would have been 9 to 16 percent larger in 2008 - between $1.3 trillion and $2.3 trillion.
For now? "We go where the smart people are," says Howard High of Intel Corp. "Now our business operations are two-thirds in the United States and one-third overseas. But that ratio will flip over [in] the next 10 years." Annual federal funding of research in mathematics, engineering and the physical sciences is equal to the increase in America's health-care costs every nine weeks. A National Academy of Sciences (NAS) report says that in 2000, more foreign students than American students were studying engineering and the physical sciences in U.S. graduate schools.
Familiar recipes for improvement are dubious. "Many high-performing education systems, especially in Asia," Duncan says, "have substantially larger classes than the United States." In South Korea, secondary-school classes average about 36 students; in Japan, 33; in America, 25.
Duncan knows that Americans are uneasy about any national education standards that might emanate from a Washington they distrust, but he insists that it is irrational to have 50 different goal posts. Perhaps, but 50 different approaches might yield a few that are truly superior. Unfortunately, the response to that is this:
Allowing states to define academic proficiencies, while federal policy gives financial rewards for achieving those proficiencies, produces perverse incentives. The NAS report says that in New York, the percentage of eighth-graders reaching the state's proficiency standard increased dramatically, from 59 percent to 80 percent, between 2007 and 2009. Yet the eighth-graders' scores on a national math test "remained virtually unchanged."
Conclusion? The state defined proficiency down. Solution? Penalize that. Regarding grades K through 12, federal education policy - if such there must be - should permit, indeed encourage, 50 laboratories of educational experimentation. Federal policy should be confined to providing financial rewards contingent on improvements confirmed by national metrics - Duncan's single goal post.
The Education Department sits at the foot of Capitol Hill, where many new legislators consider "federal education policy" a constitutional oxymoron. They have a point. They might, however, decide that the changes Duncan proposes - on balance, greater state flexibility in meeting national goals - make him the Obama administration's redeeming feature.