The basic problem in designing change efforts and in assessing them is determining which processes or consequences count and by how much. Whether it is teaching or learning that needs to be changed has to be determined. So it is with class size. What is to be changed as well as how that change is to be measured needs to be determined. The key question in considering class size is whether it is the nature of the school as an organization--in which the smaller classes exist--that accounts for the larger share of any observed increase in student learning, or whether increased learning can be brought directly, simply by increasing the number of small classes, regardless of other aspects of the school organization. It is suggested that discussions and research of class size would be more useful if they dealt with some of the significant relationships that exist between a resource allocation plan and the social processes that constitute the environment of these smaller classes. In attempting to point out the significance of the school context in looking at the question of class size, the question of criteria of success is also dealt with. However, it is suggested that student learning is not the criterion of success. It is suggested that the social slogan "smaller classes equals better learning" is a useful aphorism whose utility has been demonstrated. Its current validity or future validity, without redefinition of terms is in doubt. (Author/AM)