The Magazine


Mar 9, 1998, Vol. 3, No. 25 • By CHESTER E. FINN JR. and MICHAEL J. PETRILLI
Widget tooltip
Single Page Print Larger Text Smaller Text Alerts

THE PRESIDENT HAS PROPOSED to shrink class sizes in the early grades by hiring 100,000 more teachers at federal expense. This is quintessential Clintonism -- a warm Labrador puppy of a policy notion, petted by teachers and parents alike, but destined to bite when it grows up.

There is precious little evidence that smaller classes help students -- achievement may even go down if the new teachers are mediocre -- but don't try telling this to voters. Smaller classes are a pollster's delight. The idea is so popular that many states and communities have jumped the gun. Indiana shrank its primary classes more than a decade ago. California's Pete Wilson was hailed when he said the state's surplus should be used for this purpose. Class-size reduction was part of the successful campaign platform of Virginia's new Republican governor, Jim Gilmore, who has promised 4,000 new teachers in the state over the course of his four-year term. Similar proposals await legislative action in Alabama, Delaware, New York, and many other jurisdictions.

Why this lemming-like rush off the class-size cliff? "Teachers are thrilled, parents are thrilled," explained a California elementary-school principal in response to the president's plan. Parents simply take for granted that smaller classes mean better education. Teachers cheer because their jobs get easier with fewer students per classroom. Unions get more members. Administrators get more staff. And most local school boards welcome any move by Uncle Sam to pay teacher salaries.

Congress will therefore likely end up saying yes. But it shouldn't. The administration's plan -- and others like it -- is bad for at least five reasons.

First, the conventional wisdom that students do better in smaller classes is flat wrong. After surveying all the relevant research, economist Eric Hanushek of the University of Rochester concludes that "there is little systematic gain from general reduction in class size." Besides, classes have been shrinking for decades -- today's national average of 22 kids per classroom is down from 30-plus in the 1950s -- with no commensurate gains in learning, although the cost has been immense. (No "reform" is more expensive than smaller classes.) The Asian lands that trounce us on international assessments have vastly larger classes, often 40 or 50 youngsters per teacher. Yes, there are one or two studies indicating that fewer kindergarten children in a classroom is linked with modest test-score gains. But put it this way: If smaller classes were a drug, the FDA would not let it onto the market. Additional experiments might be warranted, but no scientist would say that its efficacy has been proven.

There's a simple reason why small classes rarely learn more than big ones: Their teachers don't do anything differently. The same lessons, textbooks, and instructional methods are typically employed with 18 or 20 children as with 25 or 30. It's just that the teacher has fewer papers to grade and fewer parents to confer with. Getting any real achievement bounce from class shrinking hinges on teachers who know their stuff and use proven methods of instruction. Of course, knowledgeable and highly effective teachers would also fare well with classes of 30 or 35. Jaime Escalante, renowned as the " best teacher in America," packs his classroom every year with 30-plus " disadvantaged" teenagers and consistently produces scholars who pass the tough Advanced Placement calculus exam. But such teaching is not the norm in U.S. schools, and adding teachers to the rolls won't cause it to be. (Indeed, a federal program hellbent on raising achievement would probably do better by firing rather than hiring 100,000 teachers. Students would be in larger classes but with better teachers, who could be paid more with the salary moneys freed up by the lay-offs.)