There is little evidence that smaller classes help students, says education expert Chester E. Finn Jr., and reducing class size may even hurt student achievement if the new teachers are mediocre. Yet President Clinton has proposed shrinking classes in the early grades to 18 students per teacher by hiring 100,000 more teachers at federal expense for seven years.

After reviewing the relevant research, economist Eric Hanuskek of the University of Rochester concluded "there is little systematic gain from general reduction in class size."

• Class size has been shrinking for decades -- the national average is now 22 kids per classroom, down from more than 30 in the 1950s -- at immense cost, but with no comparable gain in achievement.

• In fact, the Asian countries that trounce the U.S. on international education assessments have vastly larger classes, often 40 or 50 per teachers.

• And in California, when Gov. Pete Wilson shrank class sizes, veteran teachers left inner-city schools in droves, lured by higher pay and easier working conditions in suburban schools that suddenly had openings.

One or two studies that suggest fewer kindergarten children in a classroom is linked with modest test-score gains, says Finn; but more research is necessary before it can be said its efficacy has been proven.

Alternatively, Finn suggests the $12 billion in new federal spending Clinton proposes would be better spent to fund $4,000 scholarships for 425,000 low-income students for seven years. Or it could be used to improve teaching by providing a $4,500 college tuition grant for every one of the nation's 2.7 million teachers.

That would be useful, Finn points out, because the Department of Education reports that 36 percent of public-school teachers of academic subjects neither majored nor minored in their main teaching field.

Source: Chester E. Finn Jr. (president, Thomas B. Fordham Foundation) and Michael J. Petrilli (Hudson Institute), "The Elixir of Class Size," Weekly Standard, March 9, 1998.