Is the Urban World Small? The Evidence for Small World Structure in Urban Networks

The initial definition of small world networks triggered a rush among network scientists, working in a variety of fields and with data from many different contexts, to identify and document empirical examples of small world networks. Researchers studying urban networks - networks of cities and networks in cities - have also participated in this exercise, but because their work took place in a variety of disciplines, no definitive answer has emerged to the question: Is the urban world small? I answer this question through a systematic review of the evidence for small world structure in 172 urban networks. I find that although authors often claim urban networks are small world (71.5%), such claims are rarely grounded in a formal index or guided by a specific decision rule, and may overestimate the ubiquity of this structure. However, existing indices of small worldliness offer promising options for summarizing the extent to which an urban network is small world. I conclude with recommendations that urban network researchers make use of these indices and begin conceptualizing small worldliness as a continuous, rather than binary, characteristic.