The use of statistics in the understanding and forecasting of political and social regional stability offers some important advantages and shows important promise as a policy aid in evaluating possible interventions, but it faces important challenges that need to be overcome in terms of (1) the data that are necessary to perform accurate assessments, and (2) the best means to choose the most effective methods for performing assessments. (By statistics, we mean the use of mathematical analysis to identify, quantify, and summarize relations between numerical values that represent real world observations.) Most importantly, statistical analysis allows the application of a well-used approach to the analysis of stability. This permits the analyst to put confidence intervals around the possible relationships between the possible causes analyzed and the outcome of instability, as well as the impact of possible interventions. This also allows the analyst to appropriately caveat the resulting analysis. On the other hand, a statistical approach requires that the information to be studied be entered into a spreadsheet format, as well as a numerical format. This may be relatively easy with certain phenomena such as inches of rain a year or the gross domestic product of a country, but it is much more challenging to translate such phenomena as discrimination or violence into numbers. Furthermore, certain kinds of phenomena of interest are difficult to study because of their rare occurrence or an almost complete lack of relevant data.
Asal, Victor, and Stephen Shellman. 2010. "Analyzing Political and Social Regional Stability with Statistics: Challenges and Opportunities." In Perspectives on Political and Social Regional Stability Impacted by Global Crises—A Social Science Context. Washington DC: White Papers, Strategic Multi-Layer Assessment and U.S. Army Corps of Engineers Research and Development Directorate, 196-202.