This paper explores what researchers have meant when they talk and write about the effects of schooling on the educational performances and educational development of children. First, a literature review illustrates both the lack of consensus about theory and methodology and some reasons for the inconsistent conclusions about school effects. A conclusion is that past research on school effects has devoted too much attention to establishing statistical relationships between underconceptualized measures of school characteristics and student outcomes. Instead, frameworks interpreting the multileveled effects of schooling on student outcomes must be developed. In particular, research is needed that asks the questions: (1) How do schooling processes and structures interact with race and gender to create different learning and opportunity structures?; and (2) To what extent can different school performance patterns be attributed to such interactions? If the interactions among social context, group culture, and individual identification are important, research that ignores them distorts an understanding of the educational process. The most complete understanding of any educational phenomenon will combine multilevel theoretical frameworks with data from a variety of sources. (Contains 44 references.) (LMI)