Reducing class size is a reform that is popular with teachers, parents, and the education establishment, but policymakers need more solid information about the costs and benefits of other reform options before they commit billions more dollars to across-the-board class-size reduction. So conclude four respected education researchers in a careful six-page overview of the evidence on class size that appears in last month's Scientific American. Hundreds of studies of the effect of class size reduction on student achievement have been inconclusive, but fresh analyses of data from the Project STAR experiment in Tennessee and new data from California and the SAGE study in Wisconsin suggest that students (particularly minority students) can benefit from very small classes in the early grades, though how large the benefit is and how long it lasts are debated. Even if class size does make a difference, however, the researchers argue that California's effort to reduce class size in grades K-3 statewide is a poorly designed policy that will consume billions of dollars ($5 billion so far) and produce miniscule gains while exacerbating the problem that California's urban districts already have finding qualified teachers. Policymakers ought to consider other options like attracting and hiring better teachers, the analysts suggest. Sidebars accompanying the article examine why it is that smaller classes might benefit children, given that teachers who are assigned smaller classes tend not to change their teaching styles or spend much more time with individual pupils, and how Japan manages with 40 students per teacher in elementary school classes. "Does Class Size Matter?" by Ronald Ehrenberg, Dominic Brewer, Adam Gamoran, and Douglas Willms, Scientific American, November 2001 (not available online). A longer version of the article appeared in Psychological Science and the Public Interest in May 2001.
