We define a pro-government militia as a group that:
- is identified by media sources as pro-government or sponsored by the government (national or sub-national),
- is identified as not part of the regular security forces,
- is armed and
- has some level of organisation.
We use local, regional and international media sources via Nexis to identify PGMs, primarily relying on English language news sources from around the world from LexisNexis. The sources from Nexis include a wide range of sources including transcripts from local radio stations, national newspapers, BBC World Summaries of local news, and all major international newspapers.
Advance notice of update PGMD version 2.0
We are currently working on updating the PGMD version 2.0. Version 2.0 will cover the years 1981 to 2014 and will include new variables on how PGMs were terminated, how they were created, as well as more specific information on ethnic membership and ethnic targets. You can view a DRAFT of the codebook for the updated PGMD v2.0 here. The release of PGMD version 2.0 is planned for 2019.
When using these data, please cite:
Carey, Sabine C., Neil J. Mitchell and Will Lowe. 2013. ‘States, the Security Sector, and the Monopoly of Violence: A New Database on Pro-Government Militias. Journal of Peace Research 50(2): 249-258,
and the specific dataset you are using.
PGMD Summary Information:
The PGMD summary information dataset uses the pro-government militia group as the unit of analysis and contains basic information about each group.
PGMD Country-Year Dataset:
The PGMD country-year dataset uses the country-year as unit of analysis. It aggregates information on presence and activity of pro-government militias as binary variables for each observation.
Update of the PGMD country-year dataset version 1.1-2017
This updated version 1.1-2017 contains new presence variables based on a new coding role for the termination date of a PGM. Please read our "Notes Version 1.1-2017" file for this new coding rule. The updated PGMD country-year dataset version 1.1-2017 includes both the old presence variables and the newly generated ones.