The purpose of this paper is to report on the methodology, findings, and implications of two meta-analyses of class size. It begins with an explanation of a graph that indicates smaller classes result in increased pupil achievement. Next, summaries are provided of meta-analyses 1 and 2, followed by an extensive critique. The paper concludes that smaller classes can have positive effects on pupil achievement in reading and mathematics in the early primary grades and for low-ability, economically disadvantaged, and socially disadvantaged pupils. Conversely, larger classes may be justified in areas in which pupil achievement is affected little and where facilities and personnel may be used more efficiently. Much research is needed, the paper says, on the effects of class size on both pupil achievement and nonachievement processes. A need also exists for research concerning efficient class sizes and teaching techniques in other subject and skill areas. (Author/LD)