Moving beyond “Does it reduce crime”: The curvilinear and moderating effects of greenspace

Abstract

Purpose: Previous studies have primarily examined linear relationships between greenspace and crime, overlooking the possibility that greenspace may have both crime-reducing and crime-producing effects in neighborhoods. Additionally, the ways in which the effects of greenspace systematically vary by other neighborhood characteristics remain understudied. This study addresses these gaps by analyzing the relationship between greenspace, crime, and neighborhood characteristics. Methods: We conducted a block group analysis in Cleveland, Ohio, examining crime, greenspace, sociodemographic characteristics, and built environment features. Zero-inflated negative binomial regression models were used to assess both the main and moderating effects of greenspace on crime. Results: The findings indicate that an index of greenspace has U-shaped effects on all forms of crime, whereas tree canopy—a specific type of greenspace—exhibits U-shaped effects on property crime only. Moderation analyses show that the crime-reducing effects of greenspace and tree canopy are strongest in neighborhoods characterized by ethnic heterogeneity and mixed land use, respectively. Conclusion: These results highlight the nuanced relationship between greenspace and crime, emphasizing that while greenspace can reduce crime under certain conditions, very high levels of greenspace may have unintended crime-producing effects. Importantly, the benefits of greenspace are maximized in neighborhoods that, in theory, face challenges regulating crime.