Neighborhood assets as enablers of urban-suburban innovation capacities

Uncovering patterns and divides in Massachusetts and New York using traditional and API-mined data

Abstract

Large, open-source, and API-mined data have revolutionized the analysis of urban and suburban city neighborhoods with respect to their innovation capacities. Such data complements traditional census and government data collection methods and sources to help explain significant variations in invention activity across neighborhoods within a city. This paper combines diverse data sources to assess urban-suburban capacities and divides in the generation of patent-based innovation. First, it identifies patterns in innovation outcomes using USPTO patent records and illustrates spillovers across all ZIP codes in New York and Massachusetts and compares them with data from Google Maps to obtain a more granular and broader understanding of the concentration of innovation spaces. The findings reveal variations in the distribution of activity when comparing traditional patent records and API-mined data. Second, the study examines how economic, social, and spatial characteristics are associated with high levels of granted patents across neighborhoods in New York and Massachusetts. Drawing on a composite dataset assembled from multiple sources, the research interrogates the density of correlations between neighborhood variables and patent activity. Third, a comparative design spanning the two US states with distinct urban contexts deploys difference-in-means testing to systematically identify both shared patterns and divergences in how place-level conditions relate to high patent performance. Beyond the empirical findings, the study contributes a replicable methodology for assembling neighborhood asset indicators from heterogeneous data sources and offers a grounded interpretation of results for planning practice and innovation policy. The findings suggest that while traditional datasets provide long-term insights for formal study, alternative data-driven approaches reveal dynamic information that is important for understanding the evolving nature of innovation landscapes and informal innovation activities. Ultimately, this study advocates for a hybrid model that integrates both traditional and API-mined data paradigms to promote urban intelligence and equity. It advocates considering alternative innovation measures when designing innovation policies.