blank.gif (51 bytes) Education
Does Reducing
Class Size Help?

by Lance Izumi

Here in Nevada, both major party candidates for governor are attempting to demonstrate their support for education by touting their backing for more and smaller classrooms.
The reality, as the following shows, is that smaller classes are just an old, tired shibboleth that keeps the state from focusing on real education reforms.

f all proposed school reforms, class-size reduction probably has the broadest support. Parents, teachers and politicians can't seem to praise the idea enough. But is it effective?

A recent study suggests that smaller classes don't necessarily mean greater student achievement. The real difference is in the teaching.

Still, with such widespread support, it's no surprise that states are generously funding class size-reduction programs. California, for example, spent $1.5 billion to shrink class sizes in kindergarten through third grade in 1997-98. That's a half-billion-dollar increase over the previous fiscal year.

California also plans to lock in funding in future years through a November ballot initiative that will permanently keep class sizes at their reduced levels in grades K-3.

Even with that kind of commitment, the question remains whether reducing class size is really an effective way to improve student performance.

University of Rochester Professor Eric Hanushek, perhaps the nation's top education economist, says no. Hanushek found little if any correlation between smaller class sizes and higher achievement.

Pupil-teacher ratios fell nationally by 35 percent from 1950 to 1995. But performance indicators, such as scores on the National Assessment of Educational Progress, haven't gone up.

After surveying 277 studies that attempted to correlate pupil-teacher ratio and student achievement, Hanushek concluded: "There is little reason to believe that smaller class sizes systematically yield higher student achievement. While some studies point in that direction, an almost equal number point in the opposite direction."

Data from overseas also offer no encouragement for backers of smaller classes. When Hanushek looked at the relationship between student performance and pupil-teacher ratios using six international tests in math and science given between 1960 and 1990, he actually found some correlation between larger class sizes and improved performance.

Tennessee's vaunted class-size-reduction experiment, which California and other state officials cite to support their efforts, doesn't hold up under scrutiny, either. Called Project Star, Tennessee's program reduced some K-3 classes to 15 students from around 23 in regular-sized classes. Tennessee then collected data on students randomly assigned to classes with reduced sizes and those assigned to regular-sized classes.

Hanushek found that even though the Tennessee program did improve the performance of kindergarten students, there was no increase in student performance from the first grade on up.

The big difference between California's efforts and Tennessee's experiment is that 15-to-1 ratio. "The California program was designed to move classes down to around that of the regular-sized classes in the Tennessee experiment," Hanushek noted. "No evidence from [the Tennessee experiment] relates to the likely effects of such a policy change [as opposed to reducing classes to the ratio of 15-to-1]."

His final analysis: "Across-the-board policies of class size reductions, such as those enacted in 1996 for elementary education through grade three across … Californa, are unlikely to have a beneficial effect on overall student performance."

Teacher quality, not class size, makes the most difference.

While California’s proposed class-size reduction measure includes a competency test for newly hired teachers, it contains no such test for current teachers. The ballot initiative’s backers feared that too many veteran teachers would fail the test. Still, an incompetent teacher is just as bad whether he or she is in front of a class of 15 or a class of 30. Decreasing class size is easy. The hard part is raising teacher competence, requiring students to meet rigorous standards and allowing more parental choice. It’s the hard alternatives that will produce the result everybody wants: greater student achievement.  NJ

Lance T. Izumi is co-director of the Pacific Research Institute's Center for School Reform in Sacramento.


Journal front | Search | Comment | Sponsors