You go, I stay: intrahousehold evacuation behavior upon a disaster

B-Tier
Journal: World Development
Year: 2025
Volume: 192
Issue: C

Authors (4)

Maombi, Elias Ndatabaye (not in RePEc) Lunanga, Elie (not in RePEc) Stoop, Nik (Universiteit Antwerpen) Verpoorten, Marijke (Universiteit Antwerpen)

Score contribution per author:

0.503 = (α=2.01 / 4 authors) × 1.0x B-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

Low-income countries are disproportionately affected by disasters, a situation that will worsen with global warming. Evacuation is an effective strategy to reduce the burden of disasters. Existing evacuation plans are however primarily based on studies conducted in high-income countries, ignoring contextual factors of low-income countries, such as large families with many children, low car ownership and high crime rates. We argue that these contextual factors give rise to partial evacuation, going against the long-held assumption in evacuation studies that households evacuate as a unit. To demonstrate this empirically, we study the evacuation behavior of almost 4,000 individuals from 500 households in Goma, Democratic Republic of Congo, after the 2021 Nyiragongo volcano eruption. We rely on a structured survey and statistical analysis, complemented with narratives from open-ended interviews. Almost a third of households partially evacuated, leaving some members behind. Traditional gender roles largely determined who stayed behind or evacuated. Able women were more likely to evacuate, mostly on foot, to accompany children and the least mobile to safety, while able men and household heads were more likely to stay behind to protect property against looting. Our findings highlight the need to consider intra-household dynamics in evacuation behavior and design evacuation policies tailored to the specific context of low-income countries.

Technical Details

RePEc Handle
repec:eee:wdevel:v:192:y:2025:i:c:s0305750x25001172
Journal Field
Development
Author Count
4
Added to Database
2026-01-29