This article proposes a better foundation for crime prevention projects based on four types of analysis: an analysis (local, regional, or national) of the extent and development of crime, an etiological analysis, an analysis of possible measures, and an empirical analysis of the effects of the measures applied. These analyses provide a way to greatly improve problem formulation, assessment of likely "causes," selection of the most efficient measures, and evaluation. The article concludes with examples of recent burglary research and suggestions of ways to bridge the gap between the theory and practice of crime prevention. 1 table and 51 references