Micro-Level Crime Trajectories Across Urban Contexts: A Multi-City Longitudinal Study in the U.S.

Abstract Research on micro-level crime trajectories has largely focused on large metropolitan areas, leaving limited understanding of whether the same neighborhood-level processes operate across diverse urban contexts. Addressing this gap, this study examines street-segment crime trajectories across a variety of U.S. cities over the most recent decade to assess how neighborhood structures and city contexts shape crime change. Official violent and property crime data from 297,199 street segments in 12 U.S. cities between 2010 and 2018 are analyzed across different city contexts including city size, population growth, and employment. Group-based trajectory modeling is used to identity distinct longitudinal crime patterns at the street segment level, and predictors of group membership are estimated jointly with the trajectory model. The results reveal that concentrated disadvantage is associated with higher crime trajectories for both violent and property crime, but its effect is weakest in small cities and considerably stronger in medium and large ones, and for property crime the association reverses direction in small cities. Residential stability emerges as a consistently protective factor in small and growing cities, where street segments are more likely to follow crime-free trajectories and less likely to follow high-crime trajectories as residential stability increases, compared to those in larger and stagnant cities. The findings indicate that broader city conditions shape micro-level crime trajectories in a complex way. It highlights the importance of incorporating city-specific structural conditions into crime prevention and policy strategies.