This post serves as a short appendix for a forthcoming, online article on the cross-national historical dynamics of rioting.
Here is the citation for the dataset. The event data is based on news reports.
Banks, Arthur S., and Kenneth A. Wilson. Cross-National Time-Series Data Archive. Jerusalem, Israel: Databanks International, 2017. https://www.cntsdata.com.
A common problem with cross-national historical data in political science is that countries enter and exit the international system of states. So global averages could potentially be misleading, especially if newer states are for some reason more or less conducive to rioting. For the main article, I report from the more general and complete sample of countries, but I also checked for this possible problem empirically. It turns out the story does not change substantially, whether you allow or disallow states to exit/enter the sample.
The 57 countries in the sample from 1919 to 2016 are:
Afghanistan, Albania, Argentina, Australia, Austria, Belgium, Bolivia, Brazil, Bulgaria, Canada, Chile, Colombia, Costa Rica, Cuba, Denmark, Dominican Republic, Ecuador, El Salvador, Estonia, Finland, France, Germany, Greece, Guatemala, Haiti, Honduras, Hungary, Iran, Italy, Japan, Latvia, Liberia, Lithuania, Luxembourg, Mexico, Nepal, Netherlands, New Zealand, Nicaragua, Norway, Panama, Paraguay, Peru, Poland, Portugal, Romania, Saudi Arabia, South Africa, Spain, Sweden, Switzerland, Thailand, Turkey, United Kingdom, United States, Uruguay, Venezuela