The Cartography of Opportunity: Spatial Data Science for Equitable Urban Policy

Elijah Knaap

As evidence on the contextual effects of place upon individual outcomes has become increasingly solid over time, so too have urban policies and programs designed to connect underserved people with access to spatial opportunity. To this end, many attempts have been made to quantify the geography of opportunity and quite literally plot it on a map by combining evidence from studies on neighborhood effects with rapidly expanding spatial data resources and GIS technology. Recently, these opportunity maps have not only become increasingly common but their preparation has been encouraged and facilitated by the US Department of Housing and Urban Development (HUD). On one hand, the increasing prominence of opportunity mapping is a useful and important step forward for equity planning. Maps are powerful means of displaying the concept of opportunity and its variation across space. On the other hand, the institutionalization of opportunity mapping portends a need to examine critically the foundations that underlie the construction of opportunity metrics, the display of maps, and the application of these techniques in public policy. A closer look at the conceptual foundations and analytical methods that underlie these exercises offers important lessons not just for the practice of opportunity mapping but also for the practice of equity planning in general. In the following essay, I examine the practice of opportunity mapping from both theoretical and methodological perspectives, highlighting several weaknesses of the common methods. Following, I outline an improved theoretical framework based on Galster’s (2012) categorization of the mechanisms of neighborhood effects. Using data from the Baltimore metropolitan region, I then use confirmatory factor analysis to specify a measurement model that verifies the construct validity of the proposed theoretical framework. The model provides estimates of four latent variables that may be conceived as the essential dimensions of spatial opportunity: Social-Interactive, Environmental, Geographic, and Institutional. Finally, I develop a neighborhood typology by applying an unsupervised machine learning algorithm to the four dimensions of opportunity. The results suggest that the practice of opportunity mapping can be improved substantially through (1) a better connection to the empirical literature on neighborhood effects, (2) a multivariate statistical framework, and (3) more direct relevance to public policy interventions.