Communities in Puerto Rico saw their populations shrink after Hurricane Maria in 2017. Of the archipelago’s 884 census tracts, 613 tracts (69.3%) experienced a net population loss with an average loss of 161 people. To understand the relationship between social vulnerability and post-disaster population loss, informed by theories of environmental migration, I compare a measure of social vulnerability in Puerto Rico to population change in each tract. This study also provides an opportunity to evaluate the validity of the Social Vulnerability Index (SVI) measure in Puerto Rico. Through six regression models, I find that the current 15-variable SVI significantly predicts greater population losses for more vulnerable areas in Puerto Rico, in which the most vulnerable tracts lost 70 more people when compared to tracts at the median. However, using factor analysis to create a revised 10-variable SVI produces an even larger effect size when predicting population loss, in which the most vulnerable tracts lost about 151 more people than the least vulnerable. These results suggest that a 10-variable SVI may have higher construct validity for the context of Puerto Rico and could serve as a foundation for a measure of vulnerability that better reflects local conditions and experiences with disaster. This is the first study to test the relationship between a social vulnerability index and post-disaster population change in Puerto Rico. These findings highlight the need for further investigation of the link between social vulnerability and post-disaster migration and underscore the importance of context-specific measures of social vulnerability.

M.A. in Sociology
University of Colorado Boulder

Committee Members

Lori Peek (Chair)
Fernando Riosmena
Kyle Thomas

Download Thesis