Locating Litter: An Exploratory Multilevel Analysis of the Spatial Patterns of Litter in Philadelphia

Abstract

The city of Philadelphia has long been criticized for the amount of litter strewn about its streets. To more fully understand this problem, a recent city initiative, the Clean PHL program, sent out individuals to rate the amount of litter on more than 28,000 blocks across the city. We used multilevel models to analyze these data to identify both address-level and block group–level correlates of increased litter ratings. The results show that being on an arterial or collector street, the late summer and early fall seasons, and exposure to food-serving businesses and vacant properties are linked with the odds of increased litter, whereas proximity to a park is linked with decreased litter. We also found that increased block group–level disadvantage is directly related to increased litter and serves to moderate the relationships between some address-level predictors and litter.