A consortium of researchers dedicated to improving the understanding of the human causes and consequences of terrorism

START researchers earn $2.6 million NSF grant to investigate social media’s ability to predict political instability

UMD team will model grievances and instability through global media

Supported by a $2.6 million award from the National Science Foundation, researchers from the University of Maryland’s National Consortium for the Study of Terrorism and Responses to Terrorism (START) will conduct an innovative study into whether grievances aired on social media can predict political instability and conflict.

Led by Gary LaFree, Director of START and principal investigator of the project, the research team will use social media and web media sources to analyze the impact of perceived grievances on levels of unrest within countries, with the goal of developing a more accurate, comprehensive and timely forecasting of political instability than has been previously possible.

The relationship between grievances and political instability has been a staple of social science research for more than a century, but has been hampered by a lack of individual level data that can be assessed in real time, according to the research team. For example, the events of “Arab Spring” caught both policymakers and academics by surprise--even though these events appear to have developed primarily out of grievances that built over decades of autocratic rule, widespread corruption and economic stagnation.

“For the first time in human history social media and the Internet provide the possibility of gathering individual-level data on human sentiments for much of the world,” said LaFree, professor of criminology and criminal justice. “Hundreds of millions of people around the world are now using social media to communicate, making this technology-enabled forum a major de facto platform for political participation, expression, advocacy and mobilization.”

The research team plans to triangulate perceptions of grievance across a multilingual archive of billions of tweets and hundreds of millions of news articles from across 100 plus countries and 800 plus searchable industry categories. The project aims to provide worldwide estimates of political conflict and instability. The social media data and web-based media data will be compared against traditional databases and measures to evaluate the relative strength of each method in terms of forecasting political instability. 

The University of Maryland research team – LaFree and Brandon Behlendorf (Criminology), David Backer and David Cunningham (Political Science), Jennifer Golbeck (Computational Linguistics), Brooke Liu (Communications), Paul Torrens (Geography) and Brian Wingenroth (START Center) – will also conduct case studies on regions of primary interest to national security. They’ll examine the Middle East, Africa, South Asia and Latin America—where having accurate estimates of where and when political instability and conflict are likely to emerge, could be especially valuable.

“This research could have major implications for U.S. security and foreign defense policies,” LaFree said. “A more timely and improved understanding of how grievances come to bear could assist the U.S. government in anticipating zones of unrest and potential areas of instability and conflict as they are emerging. It could allow the Department of State and other government agencies to be more proactive in implementing policies and partnerships that are likely to affect perceptions of grievance in countries before they become zones of unrest.”

The three-year project, “Computational Modeling of Grievances and Political Instability through Global Media,” will begin this fall. It is funded by the Integrated NSF Support Promoting Interdisciplinary Research and Education (INSPIRE) program, which supports bold interdisciplinary projects in all NSF-supported areas of science, engineering, and education research.