Confirmatory program evaluation is discussed as a method for conducting theory-driven evaluations. It is an impact assessment that emphasizes the explication and testing of a priori program theories within a broad assessment of program effectiveness. Greater emphasis is given to identifying the causal mechanisms or active ingredients of intervention effects. Confirmatory evaluation approaches are applied to the field of early childhood intervention. Although significant progress in understanding the effects of early childhood interventions has occurred over the last four decades, questions remain about the causal mechanisms of change, who benefits most from which program components, and the reliability of effects for large-scale programs. Examples from the Chicago Longitudinal Study are highlighted to show how confirmatory evaluation can help validate the effects of social interventions. Studies of the Chicago Child-Parent Centers are described to emphasize how the causal criteria of coherence, specificity, and within- and between-study consistency can strengthen causal inference and generalizability.