Judging the success of school reform requires an interpretative context in which to judge whether effects obtained are large enough to be important or so small as to be a disappointment. The logic of school reform suggests two frameworks with which to judge the importance of effects. One is the size of the existing achievement gaps between important groups in society. The other is the size of gaps in mean achievement among schools (adjusted for student characteristics). NAEP data is used to demonstrate that in national data, gaps which appear large by one standard may appear small by the other. We argue that the most appropriate framework for judging reform effects is the national distribution of school effects.