This dissertation proposes latent profile analysis (LPA) as a novel method to study the complex relationship between facilities and crime in urban settings

While some studies recognize context-dependent impacts of facilities, typically considering broader physical and social environmental factors, this study primarily considers the role of proximal facilities influencing these associations. Specifically, IT focuses on how facilities within close spatial proximity may interact in different ways, potentially amplifying or mitigating their criminogenic effects not observable when examined in isolation. While environmental criminology has begun to address facility interactions, empirical research on this front is limited in number and in scope. This study analyzes street segments in new york city (NYC) to examine how certain combinations of facilities affect crime patterns in urban setting. Using LPA can uncover sets of street segments with similar characteristics based on facility exposures, which holds potential to address the analytical limitations of variable-centered approach like multivariate regression as well as case-oriented approaches such as the conjunctive analysis of case configurations (CACC). This dissertation has three primary objectives: 1) classifying NYC streets using LPA based on facility exposure indicators to examine their association with crime; 2) conducting a comparative analysis with existing methods to highlight its unique contributions and limitations; and 3) examining how these street profiles' relationships with crime differ across the five boroughs of NYC. This study provides several points of theoretical, analytical, and practical significance for future research on urban crime. CA/NCF