Abstract
Nearly 38 million people in the United States live in poverty. The Census Bureau's official poverty measure significantly undercounts poverty as it solely focuses on a minimum food diet and fails to account for the geographic variations in living costs. This article offers a geographically adaptive framework for combining multidimensional poverty indicators and modeling the locally adjusted costs of food, housing and utilities, healthcare, childcare, transportation, taxes, and other necessities to assess poverty and geographic inequality across neighborhoods and population sub-groups. We employ a co-design approach for developing the poverty assessment framework and evaluating the results with end users to ensure that communities can build trust and a sense of ownership that enhances the usability and actionability of poverty data. The datasets, quantitative frameworks, and algorithms were woven into an interactive geospatial dashboard toolkit for seamlessly integrating, cleaning, standardizing, and visually communicating the poverty metrics with a broad range of users. Results from this paper advance spatial data analyses and reproducible spatial model-building methods that enable researchers to gain higher resolution, context-specific, and geographically dynamic knowledge of poverty and inequalities.
