The analysis consider Boston, Chicago, Los Angeles, Miami, New York City, and San Diego. The school systems provided information regarding their definition of dropouts, the procedures for collecting the necessary information to determine dropout statistics, and the method used in determining the dropout rate. Findings showed that no single definition exists for a dropout. The six districts varied in the ways that they identified and counted dropouts, in the formulas used to calculate their dropout rates, and in the context in which data are gathered and reported. The integrity of the data is also a major concern, particularly because of the controversies and political aspects of the issues. Results also showed that students who are overage when they enter high school are far more likely to drop out than are their classmates of normal entering age. Thus, school policies of promotion and retention must be carefully examined for their negative effect on dropout propensity and with a careful analysis of their positive educational effects. Results also suggest that the current emphasis on accountability of school managers and teachers both directs attention to specific aspects of school operations and creates the incentive to misreport relevant data. Greater consistency in definitions and inclusion of demographic and performance data are needed. 60 footnotes.