Abstract
These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release, but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.
The study aimed to develop and empirically test a dynamic risk assessment model of radicalization process characteristics of homegrown terrorists inspired by Al Qaeda's ideology. The New York Police Department (NYPD) model developed by Mitchell D. Silber and Arvin Bhatt was chosen as the basis for creating a typology of overt and detectable indicators of individual behaviors widely thought to be associated with extremism. Specific behavioral cues associated with each stage of radicalization were coded and used to estimate the sequencing of behaviors and the duration of the average radicalization trajectory. Out of 331 homegrown American Jihadists (Group A), 135 were selected for further examination of their radicalization (Group B). Data were collected from public records ranging from social media postings by the offenders themselves to evidence introduced in the adjudication of the offenses for which the offenders were incarcerated. Life histories were compiled for Group B, whose detailed biographies were used to chart the timelines of their radicalization trajectories.
The collection includes an Excel file which contains one data table for Group A (10 variables, n=331) and two data tables for Group B (32 variables, n=135 and 5 variables, n=135, respectively). An accompanying codebook file details the variables in these tables. There is also a document with approximately 1 page narratives for each of the 135 individuals in Group B. A file containing a key indicating the names of the subjects is not available with this collection.
The objective of the research was to outline a general and dynamic model for assessing the radicalization trajectory of violent extremists.
Researchers identified 331 individuals who met the study criteria (Group A). From this group, 135 individuals (Group B) were selected for more detailed study.
Detailed forensic biographies were collected for each of the 135 subjects in Group B. The
biographies were essentially case summaries compiled from court documents, online
communications posted by the terrorist offenders, and investigations into their activities
conducted by the United States government and news media. Media interviews with family and friends or
local community members, including Imams and mosque officials, were also drawn on when the
information was considered reliable. Each individual was assigned a case number and a row in a
spreadsheet, and biographical timelines were constructed.
Dates for the first evidence of behavioral cues for behaviors consistent with extremism
were inferred from the documentation, and the dates entered into a spreadsheet. This allowed the
timelines for the individuals' radicalization process to be charted.
Coders looked to infer dates for when an offender was first seen to engage in a particular
behavior. Following a codebook developed to enumerate specific cues associated with each of four stages of extremism (pre-radicalization, detachment, peer immersion and steps to actions), the researchers combed through the primary material and
entered into a spreadsheet the first date at which a particular cue indicative of increasing
militancy was noted. The onset of radicalization was inferred from the narrative records using a battery of
indicators generally thought to be associated with early-stage extremism.
The data table for Group A contains 331 cases and 10 variables. The variables include age, gender, ethno-national origin, religious conversion status, education, criminality before radicalization, year of radicalization, year of action, and an indicator as to whether the individual is in Group B.
The data table for Group B contains 135 cases and 32 variables. The variables include multiple binary variables indicating such information as whether an individual's radicalization took place online or through real-life peers, whether an individual has a confirmed diagnosed mental illness, and whether the individual was radicalized before or after 2010. There are also several dates related to when certain events may have occurred such as arrests, conversion, lifestyle changes, domestic physical training, and date of criminal action.
A third data table has 135 cases and 5 variables. The variables were developed for estimating the radicalization timelines for subjects in Group B when no precise information was available to indicate specific dates.
The qualitative narratives include information related to the following:
life and family;
criminal history;
path to extremism;
radical beliefs and motivation;
plot and action;
internet and social media;
bibliography of sources used to create the narrative;
A file containing a key indicating the names of the subjects is not available with this collection.