It's amazing, the traction a 25-year-old study can have. In 1985, the Student/Teacher Achievement Ratio (STAR) project, conducted in Tennessee, concluded that smaller classes are optimal for student achievement, especially among minority students.

Since then, average class-size numbers nationwide have trended downward, from the twenties to the teens, with California and Florida going so far as to enact laws putting limits on the size that any class is allowed to be.

But as Justin Snider, a contributing editor at The Hechinger Report, points out, the results have been "disastrous." California's 20-student cap on K-3 classes resulted in an increased demand for teachers that was met, in part, by newly-minted teachers with emergency credentials as the state's top teachers fled city schools for new jobs in the suburbs.

And in Florida, which approved a plan in 2002 to limit class size at every grade level, the cost thus far has been $16 billion.

Which might make sense if student test scores had risen. But they haven't; they've stayed flat, according to Steven F. Wilson, a senior fellow at Education Sector. The problem, he writes in a new book, is that the STAR findings, which showed only modest gains for K-3 students in Tennessee, can't be broadly applied to K-12 students nationwide.

But both Snider and Wilson concur that the idea of smaller classes has universal appeal--and, therefore, widespread support. After all, a smaller class equals more individualized attention equals higher achievement, right?

Not necessarily. The Equity Project Charter School in New York City pays its teachers a starting annual salary of $125,000, a price it can afford because its highly effective teachers can handle classes of 30 students. Same goes for the high-performing KIPP charter schools, where classes are as big as 45 students. "Class size is not an issue if teachers know how to manage kids," explains KIPP co-founder David Levin.

And in three of the five top-performing countries of the world--Singapore, Korea and Japan--average class sizes are 30 or more.

So why, especially during trying economic times, funnel so much money into capping class sizes? "Investments in teacher quality," Snider writes, "would do much more than smaller classes to raise student achievement in the U.S."