An approach for merging quantitative-qualitative data in order to enlarge the evaluators' perspective and provide an enriched data base for evaluating elusive evaluation problems in school settings is described. Mini-case studies were conducted over a period of three years in 11 urban schools to examine the impact of an arts-in-the-schools program and identify the cluster of factors that impede or enhance successful implementation and program effects. The replicable model for conducting mini-case studies and analyzing within and across school data illustrates that quantitative and qualitative data are integrative and serve a confirmatory purpose. Both qualitative and quantitative data have inherent limitations, however; integration of the two will extend the scope of the data base. Integration of the two types can generate new variables. Abandoning the view of quantitative and qualitative data as dichotomous permits the development of synthesizing methods for discerning program effects in an organized, rational, scientific manner. (Author/GK)