[论文解读] Uncovering socioeconomic gaps in mobility reduction during the COVID-19 pandemic using location data

Samuel P. Fraiberger, Pablo Astudillo+9 more|arXiv (Cornell University)|Jun 26, 2020

COVID-19 epidemiological studies参考文献 11被引用 36

一句话总结

本文分析来自哥伦比亚、墨西哥和印度尼西亚的GPS位置信息，以显示在COVID-19非药物干预措施期间，较富裕的个人比较贫困的个人降低出行活动的程度更大，凸显出移动性降低的社会经济差距。

ABSTRACT

Using smartphone location data from Colombia, Mexico, and Indonesia, we investigate how non-pharmaceutical policy interventions intended to mitigate the spread of the COVID-19 pandemic impact human mobility. In all three countries, we find that following the implementation of mobility restriction measures, human movement decreased substantially. Importantly, we also uncover large and persistent differences in mobility reduction between wealth groups: on average, users in the top decile of wealth reduced their mobility up to twice as much as users in the bottom decile. For decision-makers seeking to efficiently allocate resources to response efforts, these findings highlight that smartphone location data can be leveraged to tailor policies to the needs of specific socioeconomic groups, especially the most vulnerable.

研究动机与目标

• Quantify how non-pharmaceutical interventions affected human mobility in three developing countries.
• Combine spatio-temporal trajectories with high-granularity socioeconomic data to examine wealth-based mobility differences.
• Assess implications for policy design and resource allocation during pandemics in developing contexts.

提出的方法

• Use anonymized GPS location data from Cuebiq for Jan 1–May 7, 2020 across Colombia, Mexico, and Indonesia.
• Define home and work locations from nighttime and daytime activity to compute mobility indicators.
• Construct a wealth index at the administrative unit level via principal component analysis of asset ownership and service access.
• Measure mobility indicators such as time spent at home, share of commuters, number of neighborhoods visited, and maximum distance traveled.
• Normalize indicators and analyze differences between top and bottom wealth deciles across countries.

实验结果

研究问题

• RQ1 How did NPIs affect mobility in Colombia, Mexico, and Indonesia?
• RQ2 Do mobility reductions differ systematically across wealth groups within these countries?
• RQ3 What are the implications of any wealth-based mobility gaps for policy targeting and resource allocation during COVID-19?

主要发现

• Time spent at home rose after policy announcements, with top-decile wealth users increasing more than bottom-decile users.
• Share of commuters and the maximum distance traveled decreased substantially, with high-wealth users showing larger reductions than low-wealth users.
• The average number of neighborhoods visited declined significantly, with mobility reductions up to 50–100% larger for high-wealth users than for low-wealth users in some indicators.
• Across all three countries, high-wealth users reduced mobility roughly twice as much as low-wealth users in several measures.
• Mobility gaps persisted during mid-February to mid-April, indicating persistent socioeconomic disparities in response to NPIs.