Authors: Yago Martin Gonzalez*, University of South Carolina, Susan Lynn Cutter, Hazards and Vulnerability Research Institute, University of South Carolina
Topics: Hazards, Risks, and Disasters, Migration
Keywords: Social media, displacement, migration, Puerto Rico, Hurricane Maria
Session Type: Paper
Start / End Time: 1:10 PM / 2:50 PM
Room: Maryland A, Marriott, Lobby Level
Presentation File: No File Uploaded
Displaced and migrant populations are frequently elusive using traditional methods, especially in developing countries, and even more so during post-disaster and emergency situations. Some researchers have called for an increased collection of quantitative data to measure migration flows, which entails a search for innovative approaches. Many disciplines are increasingly using passive, user-generated geo-referenced data, but its application in migration/displacement studies is still scarce. In this study, we aim to test the quality and reliability of social media (Twitter) data for measuring the disruption on population flows in a post-disaster situation. Focusing in the first year after the 2017 hurricane season in Puerto Rico, which devastated the island with two consecutive major hurricanes —Hurricane Irma and Hurricane Maria, we explore the timing, destination, and return processes of displaced groups, as well as the extent of the effect on the arrival of non-residents to the island. We evaluate the suitability of Twitter for assessing post-disaster displacement and migration by comparison with recent studies utilizing other sources of passive user-generated geo-referenced data and surveys. Results show relative agreement with previous publications on post-Maria displacement and migration in Puerto Rico and support further inquiry in this line of research. The analysis demonstrates the potential of social media to study post-disaster population flows, but drawbacks advice caution in the interpretation of the results. This study might set the foundation for additional research in the field and propel new advancements through the incorporation of new data sources and/or methods that contribute complementing traditional approaches.