Ignored but not Forgotten: The Broken Windows Tipping Point and the Question of Functional Form

Published online on July 14, 2021

Abstract

Wilson and Kelling provide a simple instruction to address disorder: direct limited police resources to neighbourhoods at the tipping point. In doing so, they suggest that the disorder–crime relationship reflects a threshold effect. A proper test of the validity of broken windows theory should, therefore, accommodate nonlinearity. To this end, this study empirically examines the functional form of the relationship between physical disorder and violent crime rate in Detroit, MI, utilizing a dose–response propensity score methodology. This examination is aided by open-sourced data collected at the census block-group level. Findings support a nonlinear relationship, favouring a threshold effect with two caveats. Several directions for future research are explored, as well as implications for theory, practice and policy.