Authors: Matthew Laird*,
Topics: Population Geography, Spatial Analysis & Modeling
Keywords: Population, Spatial Analysis
Session Type: Virtual Paper
Start / End Time: 1:30 PM / 2:45 PM
Room: Virtual 32
Presentation File: No File Uploaded
Central to endeavors to achieve social justice is the alleviation of material and non-material deprivations. As measures of deprivation are used to both target these efforts and measure their success, developing accurate and theoretically sound measurements is paramount. By relying solely on income, using arbitrary and dated thresholds, and failing to consider how spatial as well as social contexts influence deprivation, the official poverty measure paradigm in the United States fails to meet this challenge. We propose a computationally driven multidimensional poverty measure as a solution. Using publicly available micro-data from the American Community Survey defining deprivation across health, education, housing, economic security, and neighborhood context, we employ k-prototypes clustering to develop groupings of persons. These groupings assess poverty without the requirement for a-priori thresholds or cutoffs to define the deprived and instead allow for statistically driven demarcation. Through its use of micro-data we are able to assess how typologies of deprivation vary across space and within geographic boundaries allowing for specific targeting of issues disproportionately effecting communities. Considerations are made to understand how marginalized groups experience this expanded definition of deprivation through demographic decomposition.This approach advances computational practices in service of aiding the disadvantaged in its application of novel statistical approaches to the study of deprivation and deepens our understanding of place based deprivation.