Predicting Local Violence in Liberia
Is it possible to predict where violence will occur?
Many postconflict countries suffer from high rates of crime, violence, and unrest, and these incidents of local violence threaten life and property and can escalate into larger scale disputes. But what if police and peacekeepers could rely on early warning systems to anticipate violence before it happens?
In response to this question, Christopher Blattman and a team of researchers built a statistical model based on data that Innovations for Poverty Action gathered over four years in the most conflict-prone areas of Liberia. They collected data in three waves from 242 towns and villages in three conflict-prone counties: Lofa, Nimba, and Grand Gedeh. They focused on the most destabilizing forms of local violence: violent strikes and protests, violent clashes between ethnic groups, murders, rapes, fights or assaults involving weapons, and extrajudicial punishments.
Researchers used the first two waves of survey data, from 2008 and 2010, and a variety of different statistical techniques to build models for predicting violence. They then used the models to generate predictions for where violence was most likely to occur two years later, in 2012. Then, in 2012, IPA collected data from the same 242 communities to see where violence had actually occurred, and researchers compared the models’ predictions to reality.
The model correctly predicted 88 percent of violence two years into the future, albeit at the expense of many incorrect predictions that violence would occur. The study also found that of 56 potential risk factors, only a handful consistently predicted violence over time—especially ethnic diversity and polarization. Violence was more likely to occur in communities that were larger, more diverse and more polarized. More surprisingly to researchers, violence was also more likely where multiple ethnic groups and religions were represented in local leadership (i.e., power-sharing). In fact, local-level power sharing was the single best predictor of violence in the best model. However, this finding is a correlation, not evidence that sharing power causes conflict.
It remains unclear whether these models would perform well in other time periods or settings. Nevertheless, the results suggest that relatively simple statistical models to predict violence may indeed be feasible. Peacebuilding researchers and practitioners should replicate similar exercises to identify which risk factors, if any, reliably predict violence across different time periods and settings. Replication will help develop fast, effective, and low-cost early warning systems for the future. Currently, Blattman and his team are applying the model to local-level conflict in Indonesia and Colombia, over longer periods of time, with many more municipalities, as well as with cross-national data.
Journal of Peace Research