Investigators completed a series of studies of data reported in Tennessee's 1988-89 school district cards, and compared these with 1990-91 school district report cards. The relationships among 15 school district variables and relationships between each variable and student outcomes were determined. The 1990-91 report cards contain results of the new Tennessee Comprehensive Assessment Program. Several conclusions from the previous study were reinforced by the current one. Policymakers at all levels need to consider that few of the individual inputs commonly associated with student achievement have much impact on student performance. With the exception of student attendance (and perhaps per pupil expenditure), treatment of any isolated variable will have little effect. It is urged that consideration be given to collecting, reporting, and analyzing data on such things as school organization, school culture, student motivation, parental involvement, instructional methodologies, curriculum features, and other factors. Building-level results appear more useful than do district-level results, and report cards appear to be only as good as the assessments used to determine student performance. Ten tables and 1 figure summarize findings, and seven appendixes contain seven additional tables. (SLD)