Authors: Kexin Chen*,
Topics: Geographic Information Science and Systems
Keywords: Human mobility, partisan difference, visualization, COVID-19, public health
Session Type: Virtual Paper
Start / End Time: 3:05 PM / 4:20 PM
Room: Virtual 14
Presentation File: No File Uploaded
Since the announcement of the COVID-19 pandemic, social distancing has been promoted and required as one of the most important non-pharmaceutical interventions intended to mitigate the spread of COVID-19 across the United States. This paper employs heatmap, multidimensional scaling and k-means clustering techniques to effectively visualize state-level human mobility patterns during a 30-day period after stay-at-home order, and presents the level of similarity of mobility patterns with respect to state partisanship. Partisan differences are observed in how people change their mobility behavior. We found that states that have the same partisanship tend to have more similar mobility patterns. Moreover, democratic states show more overall reduction in mobility during the 30-day period after stay-at-home order. The visualization of mobility data shows the effectiveness of dimensionality reduction techniques that convert high-dimensional time series data into a 2-dimensional plane for better understanding the clustering pattern of adherence level to stay-at-home social distancing mandates.