The Mobilization Puzzle: How Individual, Group, and Situational Dynamics Produce Extremist Outcomes

This study introduces a new data resource, the Profiles of Individual Radicalization in the United States-Plots (PIRUS-Plots) dataset, that builds on previous NIJ investments in the PIRUS and Social Networks of American Radicals (SoNAR) datasets. PIRUS-Plots introduces new variables related to successful, failed, foiled, and nebulous extremist plots in the United States for 1,433 ideologically motivated crimes that occurred between 1990 and 2021. The new variables introduced in PIRUS-Plots cover:

1. Event-level details, such as the dates and locations of the plots, the type of plot (financial crime, property crime, low casualty plot, or mass casualty plot), the number of people involved in a plot, target information, weapon information, and casualties.
    
2. Preparatory actions taken by perpetrators, including variables measuring the presence of the Department of Homeland Security’s Nationwide Suspicious Activity Reporting (SAR) Initiative indicators.
    
3. Outcome details about the plots, including whether the plots were successful, failed, or foiled.
    
4. Law enforcement engagement details, including interdiction strategies and information about when and how law enforcement became aware of the plots.
    
5. Arrests and criminal proceedings, which contain information about the criminal charges, convictions, plea agreements, and, if applicable, raids related to plots in the database. The new PIRUS-Plots dataset is designed to be used in conjunction with PIRUS and SoNAR, and contains plot ID, subject ID, and network ID keys that enable querying all three datasets as a relational database.
    

Our primary motivation in designing PIRUS-Plots was to provide a robust set of variables across multiple domains (the event or plot-level, the subject or perpetrator level, and the social network level) so that distinct radicalization and mobilization pathways can be observed, modeled, and understood. We also provide disaggregated information about the type of plots that goes beyond the non-violent/violent distinction to include observations about whether a plot was strictly non-violent, a low casualty plot, or a mass casualty plot. Crucially, we also introduce a control category that is missing from most terrorism research: plots that involved nebulous threats but never resulted in actual mobilization. By disaggregating observations about plot types in this way, we enable scholars and practitioners to study and understand how risk factors, protective factors, social networks, and law enforcement strategies vary across different plot types, determining which plots ultimately succeed and which fail.  

View Publication