The focus of this paper is on the methodology used in analyzing studies on the effects of school desegregation on black achievement. The paper also addresses a number of substantive issues, including the overall effectiveness of school desegregation, the impact of type of achievement (math or reading), and time of desegregation (early or later grades). In the first section, a discussion of the advantages and disadvantages of meta-analysis is presented. The complications are explored of using the method to analyze the research literature on the effectiveness of school desegregation when that literature is almost totally composed of quasi-experimental or weaker research designs. In the second section, the adjustments that were made in the meta-analysis method for the purposes of the National Institute of Education (NIE) panel's study on the effects of school desegregation are described. The procedures and criteria used for including the 31 studies finally chosen for analysis by the author are described in the third section, as well as the criteria used by NIE in narrowing the selection down to 19 studies for consideration by the panel. In section 4, the results of analyses of both sets of studies are given. An appendix lists the studies used in both analyses. (CMG)