Using Facebook Advertising Data to Estimate Out-Migration From Puerto Rico After Hurricane Maria
Wednesday, April 24th, 2019
12:30 – 1:30 pm
Natural disasters such as hurricanes can cause substantial population out-migration. However, the magnitude of population movements is difficult to estimate using only traditional sources of migration data. We utilize data obtained from Facebook’s advertising platform to estimate out-migration from Puerto Rico in the months after Hurricane Maria. We find evidence to indicate a 17.0% increase in the number of Puerto Rican migrants present in the US over the period October 2017 to January 2018. States with the biggest increases were Florida, New York and Pennsylvania, and there were disproportionately larger increases in the 15-30 age groups and for men compared to women. Additionally, we find evidence of subsequent return migration to Puerto Rico over the period January 2018 to March 2018. These results illustrate the power of complementing social media and traditional data to monitor demographic indicators over time, particularly after a shock, such as a natural disaster, to understand large changes in population characteristics.
Kivan Polimis is a data scientist with experience in both the public and private sector building creative solutions with natural language processing and software development. Previously, he was a postdoctoral fellow at Bocconi University’s Centre for Research on Social Dynamics and Public Policy and a postdoctoral affiliate with Bocconi’s Institute for Data Science and Analytics. Prior to Bocconi, he worked for Microsoft, the Department of Veteran Affairs, and healthcare startups. Kivan received degrees in Sociology with a focus on Demography from Princeton (BA), the University of North Carolina at Chapel Hill (MA), and the University of Washington (PhD).
For more information about Kivan, visit his website here.