Predicting local violence
| Author | Robert A Blair,Christopher Blattman,Alexandra Hartman |
| DOI | 10.1177/0022343316684009 |
| Published date | 01 March 2017 |
| Date | 01 March 2017 |
| Subject Matter | Research Articles |
Predicting local violence: Evidence from
a panel survey in Liberia
Robert A Blair
Department of Political Science & Watson Institute for International and Public Affairs, Brown University
Christopher Blattman
Harris School of Public Policy, University of Chicago
Alexandra Hartman
Department of Political Science, University College London
Abstract
Riots, murders, lynchings, and other forms of local violence are costly to security forces and society at large.
Identifying risk factors and forecasting where local violence is most likely to occur should help allocate scarce
peacekeeping and policing resources. Most forecasting exercises of this kind rely on structural or event data,
but these have many limitations in the poorest and most war-torn states, where the need for prediction is
arguably most urgent. We adopt an alternative approach, applying machine learning techniques to original
panel survey data from Liberia to predict collective, interpersonal, and extrajudicial violence two years into the
future. We first train our models to predict 2010 local violence using 2008 risk factors, then generate forecasts
for 2012 before collecting new data. Our models achieve out-of-sample AUCs ranging from 0.65 to 0.74,
depending on our specification of the dependent variable. The models also draw our attention to risk factors
different from those typically emphasized in studies aimed at causal inference alone. For example, we find that
while ethnic heterogeneity and polarization are reliable predictors of local violence, adverse economic shocks
are not. Surprisingly, we also find that the risk of local violence is higher rather than lower in communities
where minority and majority ethnic groups share power. These counter-intuitive results illustrate the usefulness
of prediction for generating new stylized facts for future research to explain. Ours is one of just two attempts
to forecast local violence using survey data, and we conclude by discussing how our approach can be replicated
and extended as similar datasets proliferate.
Keywords
Africa, forecasting, machine learning, surveys, violence
Introduction
Riots, murders, lynchings, and other forms of local vio-
lence are an urgent concern for police and peacekeepers,
especially in weak and war-torn states. Local violence is
more common, and possibly even more costly, than war-
or terrorism-related violence (Fearon & Hoeffler, 2014).
Local violence can also shape, and be shaped by, national
conflict dynamics (Autesserre, 2010). Resources for pre-
vention are often scarce, and any information that helps
identify risk factors and predict where local violence is
most likely to occur should have large practical and,
potentially, theoretical returns.
Many studies have attempted to predict national-level
conflicts – for example, civil war, political instability
(Goldstone et al., 2010) or ‘irregular regime change’
Corresponding author:
robert_blair@brown.edu
Journal of Peace Research
2017, Vol. 54(2) 298–312
ªThe Author(s) 2017
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0022343316684009
journals.sagepub.com/home/jpr
To continue reading
Request your trial