Tracking changes in resilience and recovery after natural hazards: Insights from a high-frequency mobile-phone panel survey

• Published in: Global environmental change 2020-05, Vol.62, p.102053-15, Article 102053
• Main Authors: Jones, Lindsey, Ballon, Paola
• Format: Article
• Language: English
• Subjects: Aftermath; Annual variations; Building design; Cell phones; Cellular telephones; Economic conditions; Evaluation; Exposure; Flooding; Floods; Households; Innovations; Measurement; Mobile phones; Natural disasters; Polls & surveys; Reliance; Resilience; Socioeconomic factors; Tracking
• ISSN: 0959-3780; 1872-9495
• DOI: 10.1016/j.gloenvcha.2020.102053

•Self-evaluated levels of resilience fluctuate considerably on intra-annual timescales.•Resilience scores drop significantly for three months after flooding before rebounding and slowly converging after one year.•Female-headed households are hit hardest in the aftermath of the floods.•Combining subjective-evaluations with mobile phone surveys allows for new insights into tracking resilience. Knowing how resilience changes in the aftermath of a shock is crucial to the targeting of effective humanitarian responses. Yet, heavy reliance on face-to-face household surveys often means that post-disaster evaluations of resilience are costly, time-consuming and difficult to coordinate. As a result, most quantitative assessments are either carried out via one-off snapshots or by combining surveys conducted years apart. Doing so severely restricts our understanding of the temporal dynamics of resilience, particularly as it relates to inter- and intra- annual fluctuations. In this paper we examine how household's resilience to multi-hazard risk changes over time. To do so we combine two novel approaches. Firstly, we use a high-frequency mobile-phone panel survey to conduct remote interviews in Eastern Myanmar. Surveys took place every six weeks over a one-year period. Secondly, we adapt a self-evaluated subjective measure of resilience to allow it to be administered via mobile phone. Shortly after the first survey was conducted, monsoonal flooding affected the site, allowing for the effects of flood exposure on resilience to be compared over time. Our findings reveal how self-evaluated levels of resilience fluctuate considerably over the course of a year. To probe the effects of the monsoon floods, we compare resilience scores between households directly and indirectly affected by flooding. Scores drop sharply for the first three months amongst directly affected households, before slowly converging up to a year later. We also compare the effects of flood exposure on different socio-economic groups, revealing how female-headed households are particularly affected in the aftermath of flooding. Insights from the study highlight the dangers of using one-off resilience surveys to measure resilience, and underscore the need for development actors to account for shorter-term changes in the design of resilience-building interventions. Lastly, our findings showcase the potential of methodological innovations in addressing many of the resource, time and logistical constraints o