Overview
The primary goal of this project is to examine the political consequences of Hurricane Harvey by studying post-Harvey migration patterns as an exogenous shock. The exogenous shock allows us to understand 1) variations in the decision of displaced individuals to move to different areas and how this affects their associated experience, and 2) how people?s political attitudes change in response to rapid demographic shifts in their communities. Specifically, we use publicly available data from Twitter and the Texas voter registration file. The ongoing post-Harvey migration provides a unique opportunity to examine the effect of social contact and political context in a case where these encounters could not have otherwise been anticipated. This work will build on results on the political attitudes of displaced individuals after Hurricane Katrina. As a result of New Orleans? unique racial geography, Katrina evacuees were more likely to be people of color and of lower socioeconomic status. The 2017 flooding in the greater Houston area, however, affected both rich and poor communities alike, as well as majority white and majority-minority communities. Examining the impact of these evacuees on the communities to which they relocate allows us to separately estimate the effects of evacuee proximity from those of changes in racial demographics.

Intellectual Merit:
The proposed research holds great significance. Beyond the immediate policy implications, this work will address a classic question in social science: the way in which local context affects political opinions and behaviors. Geographic residence is rarely randomly assigned, making it difficult to establish whether observed correlations are the result of contextual effects or individuals? self-selection into particular ideological and social communities. Similarly, social contact may have a positive effect because those willing to engage across social barriers are also the individuals who are more likely to change their minds on political issues. Our proposed research project teases apart these potentially confounding variables to truly understand their independent effects.

Broader Impacts:
Beyond its application to social science, our data and novel social media methods have the potential to provide new tools for first responders and emergency workers during natural disasters. When evacuation orders are given, first responders must expand their limited resources by knocking on doors to ensure that all residents have evacuated. In a disaster?s wake, emergency workers must quickly locate people who were unable or unwilling to evacuate. By further developing tools to use social media data to estimate where users are located, we can improve emergency service efficiency by providing first responders with new and effective ways to quickly identify areas with people in need of attention.