Class Size and School Size: Taking the Trade-Offs Seriously

Abstract

Class Size and School Size: Taking the Trade-Offs Seriously Douglas N. Harris Small classes and small schools appear to have educational benefits for students. In small classes, children experience fewer disruptions and receive more personal attention and individualized instruction. Likewise, in small schools, students appear to feel safer and less likely to get "lost in the crowd" and teachers are able to provide a more coherent curriculum. There is a large research literature, including many of the papers in this volume, suggesting that these qualities of the learning environment translate into actual gains in student achievement and attainment.1 But nearly all previous studies on the benefits of small classes and schools ignore the other element that is of equally great interest to administrators and policymakers—the costs. Resources that go to small classes and small schools cannot be used to buy laptops for teachers, raise teacher salaries, increase professional development, add pre-kindergarten programs, or purchase new textbooks. That does not mean, of course, that smaller settings are necessarily unwise, but it does mean that they entail the loss of some opportunities. The purpose of this paper is to provide evidence about the benefits and costs of smaller educational settings and thereby provide guidelines for developing school resource policies that maximize educational outcomes. Below, I refer to these as the efficient or "optimal" resources, though it is important to recognize that other research on this topic, such as the paper in this volume by Ready and Lee, uses the term "optimal" even when they are excluding costs and considering only the benefits of resources. [End Page 137] The first step in estimating optimal resources is to consider the actual costs of smaller classes and schools. Most of these are relatively easy to identify and measure. To reduce class size, schools must either hire more teachers or have existing teachers work longer hours; they also must reconfigure existing classroom space or provide additional space in portable structures or new buildings. In the case of small schools, facility costs also figure prominently, especially for items such as gymnasiums and offices, which generally have to be replicated in every school. Other costs that are more difficult to measure arise because small settings reduce opportunities for productive use of resources: for example, a teacher who is effective in teaching math may be less able to use that skill in a small school if he or she has to spend time teaching other subjects such as science. In large schools, teachers are better able to specialize. Interestingly, education researchers, including economists, pay little attention to these or any other costs, and in the case of small classes and schools, the few studies that do exist do not yield entirely consistent conclusions. Part of the problem, as I show below, is that there are different ways to combine costs and benefits, and some are better than others. Costs and benefits also vary across circumstances—urban versus rural schools, elementary versus high schools—so some of the apparent contradictions are just reflections of differences in what is optimal in different situations. Through a review of past studies and additional analysis, this paper draws some tentative conclusions about class size and school size, but it also suggests that most of the important questions require future study. The fact that there is no clear evidence supporting the value of small classes and small schools has not, however, prevented policymakers at all levels from devoting substantial resources, particularly to smaller classes. Table 1 shows that the pupil-teacher ratio has dropped from 22.3 to 16.2 since 1970.2 Adjusting for changes in the number of pupils, that represents a nearly 40 percent increase in the number of teachers. Also, note that pupil-teacher ratios in elementary schools, which historically have been larger than those in secondary schools, actually dropped below those in secondary schools for the first time in 2000 and have remained there since. School size also has changed, but not as much. Average school size has increased by about 9 percent since 1980, although that small percentage partly reflects the disproportionate increase in the number of special schools—vocational, alternative, and combined elementary-secondary...