A Spatio-Temporal Contextual Approach for Evaluating the Unfolding Landscapes of U.S Population Loss

Gainbi Park, Newcastle University
Rachel Franklin, Newcastle University
Eric Seymour, Rutgers University

In recent decades, the United States has experienced substantial changes in population size, composition, and distribution. As growth has slowed, population loss has taken on increased salience, presenting an emerging social, economic, and political challenge and highlighting a need for measures that capture the reality of disparate population change trajectories. In this paper, we demonstrate the value of incorporating both spatial and temporal context—that is, previous history of population loss as well as neighboring county population change—for understanding how similar observed levels of population change may mask substantially different experiences. Specifically, we use sequencing and clustering algorithms, paired with contemporary and historical U.S. county-level population data (1950–2020), to evaluate varying pathways of loss and growth and to develop a typology of change trajectories. Results indicate 8 distinctive pathways of population change, ranging from emerging localized loss to constant growth. Each pathway is characterized by a particular geography (e.g., “persistent loss” counties dominate the landscape of the Great Plains region), reinforcing the notion of a spatially polarized U.S. demographic experience. A comparison of population composition and components of change also reveals notable differences in demographic characteristics across different types of loss and growth counties.

Keywords: Spatial Demography, Geographic Information Systems (GIS), Population and Development, Census data

See paper.