RAPID: Pinpointing Expected Covid-19 Related Voter Turnout Problems

Voting is a fundamental democratic activity. And yet, the process is vulnerable to disruption. Natural disasters such as hurricanes, and those created by humans, such as the 9/11 terrorist attack pose serious impediments to citizens' ability to vote. Hurricane Sandy, for example, affected twenty States during an election. When disasters occur during voting seasons, special investments must be directed into affected areas to ensure that voters can cast their ballots. Such investments may be as simple as information campaigns focused on particular areas or more hands-on such as providing an auxiliary transportation supply. However, all these measures require monetary, volunteer, and government commitments, and thereby advance notice.

The Covid virus is the latest disaster that threatens an election. As with hurricanes, some US regions are more affected than others. Similar obstacles to voting may be expected: Polling places may need to be moved, voting hours may need to change, and disaster related economic harm to the population may lower motivation. This
project creates a data driven early warning system that discovers regions threatened by disaster related voter turnout collapse. The tool aims to help all stake holders take localized action towards mitigating problems in time. The system will continuously monitor many diverse online sources such as Google searches, National Conference of State Legislatures (NCSL) data, and social media trends. While each source by itself may not be wholly reliable, their composite signals will help the system decide where difficulties are in the making. The results of this work are initially aimed to support the 2020 election. But the tool will serve for elections in the future.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.