[論文レビュー] Socioeconomic disparities in mobility behavior during the COVID-19 pandemic in developing countries

Lorenzo Lucchini, Ollin Langle-Chimal+7 more|arXiv (Cornell University)|May 11, 2023

Human Mobility and Location-Based Analysis被引用数 9

ひとこと要約

本論文は6つの中所得国にわたる携帯電話のGPSデータを分析し、COVID-19に対する移動行動の社会経済的格差が一貫して現れることを明らかにし、高所得地域の居住者は低所得地域の居住者より自己隔離、移動、通勤停止がより可能であった。

ABSTRACT

Mobile phone data have played a key role in quantifying human mobility during the COVID-19 pandemic. Existing studies on mobility patterns have primarily focused on regional aggregates in high-income countries, obfuscating the accentuated impact of the pandemic on the most vulnerable populations. Leveraging geolocation data from mobile-phone users and population census for 6 middle-income countries across 3 continents between March and December 2020, we uncovered common disparities in the behavioral response to the pandemic across socioeconomic groups. Users living in low-wealth neighborhoods were less likely to respond by self-isolating, relocating to rural areas, or refraining from commuting to work. The gap in the behavioral responses between socioeconomic groups persisted during the entire observation period. Among users living in low-wealth neighborhoods, those who commute to work in high-wealth neighborhoods pre-pandemic were particularly at risk of experiencing economic stress, facing both the reduction in economic activity in the high-wealth neighborhood and being more likely to be affected by public transport closures due to their longer commute distances. While confinement policies were predominantly country-wide, these results suggest that, when data to identify vulnerable individuals are not readily available, GPS-based analytics could help design targeted place-based policies to aid the most vulnerable.

研究の動機と目的

• Quantify how mobility behaviors during COVID-19 differed by neighborhood wealth across six middle-income countries.
• Infer home, work, and other locations from GPS trajectories to study time use.
• Assess how wealth proxies from census data relate to changes in isolation, relocation, and commuting.
• Examine how policy restrictions (e.g., public transport closures) differentially affected mobility by wealth.
• Discuss implications for place-based policies and targeted support during pandemics.

提案手法

• Use GPS location data from 281 million users across Brazil, Colombia, Indonesia, Mexico, Philippines, and South Africa during 2020.
• Infer location types (home, workplace, other) via spatiotemporal clustering of stop events with 5-minute and 25-meter thresholds.
• Assign a wealth proxy to each user from census-based wealth indices of their primary home administrative unit.
• Restrict analysis to frequent users in the five most populated cities per country (46% of users).
• Model mobility changes as functions of global and local COVID-19 incidence and five policy types using panel regression with fixed effects.
• Control for multicollinearity and validate robustness across time windows and country-level exclusions.

実験結果

リサーチクエスチョン

• RQ1 Do mobility responses to COVID-19 differ across socioeconomic groups in developing countries?
• RQ2 How does neighborhood wealth influence self-isolation, relocation to rural areas, and work commuting during the pandemic?
• RQ3 What is the role of place-based wealth in mediating responses to containment policies?
• RQ4 Are observed disparities consistent across multiple middle-income countries and continents?

主な発見

• High-wealth neighborhood residents were more likely to self-isolate at home than low-wealth residents, with 252% vs 141% increases relative to pre-pandemic levels.
• High-wealth individuals were 2.9 times more likely to relocate to rural areas than low-wealth individuals during the early pandemic period.
• High-wealth groups were 1.3 times more likely to stop commuting to work than low-wealth groups; gaps persisted throughout 2020.
• Among low-wealth commuters, those who previously worked in high-wealth neighborhoods were 89% more likely to stop commuting than those who worked in low-wealth neighborhoods.
• Public transport closures significantly reduced commuting for low-wealth workers who had previously commuted to high-wealth neighborhoods (coefficient 0.10, 95% CI [0.05, 0.15]); no significant effect for those working in low-wealth neighborhoods.
• Gaps in mobility responses persisted across the observation period, underscoring a need for place-based policy interventions when individual-level targeting is limited.