Distinguishing Violent and Non-violent Criminal Extremists Based on Risk and Protective Factors

This study explores whether extremists who commit violent or nonviolent crimes in pursuit of ideological goals can be differentiated based on the risk and protective factors they experience. Using a sample of (n = 499) violent and nonviolent criminal extremists from the Risk and Protective Factors Dataset, we conducted descriptive and classification analyses to examine fifty risk and protective factors of extremism. Our descriptive findings indicate that violent extremists demonstrate a higher frequency of risk factors and a lower frequency of protective factors than nonviolent criminal extremists. We then developed a machine learning model to investigate whether these factors could be used for classifying offenders as violent or non-violent extremists. We also explored which factors contributed most to the classification performance and find that age, blaming others for grievances, alienation, mental illness and impulsivity are especially important. We conclude by discussing our findings’ implications for theory and practice.