The Creation of Muhajirat in America: Social Media as a Platform for Crafting Gender-Specific Interventions for the Domestic Radicalization of Women, 5 countries, 1970-2018 (ICPSR 37682)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Study Purpose View help for Study Purpose

This research centered around two broad goals: first, to inform the field of terrorism risk assessment specifically as it pertains to women; and second, to create a dynamic, multi-dimensional risk model for use by the intelligence community (IC) in monitoring at-risk individuals according to their fluctuating levels of threat over time. The study revolved around two specific research projects: first, to develop the Moral-Situational-Action Model of Extremist Violence (MSA-EV); second, to explore the mechanisms by which extremist propaganda both captures the attention of individuals and evokes in them emotional reactions which may facilitate the internalization of ideological content.

Researchers sought to address three short comings of terrorism-related research. First, terrorism studies lack primary sources based on direct observation and participation with subjects, while secondary sources with secondhand information on terrorist organizations is the norm. Second, it became apparent with the rise of movements such as the Islamic State (IS) that women played a larger role in terrorism than once thought, so more research on women was desired. Scholars noted that 550 of the 25,000 foreign fighters recruited to IS in 2015 were European migrant women, referred to as "muhajirat", giving rise to concerns that these and other individuals across America could become IS recruits. Third, observers have pointed to the rise of social media platforms as a recruiting tool for terrorists and the need for research to factor this into studies.

Study Design View help for Study Design

To acquire data for the MSA-EV, sources pertaining to subjects' backgrounds and social media presence were saved as Microsoft Word documents and PDFs and added to the Female Terror Database. Researchers utilized a Social Media Coding Form and the MSA-EV to record social media activities. Each subject was coded using the MSA Risk Formulation Coding Protocol (MSA-RFCP), which captures 496 variables. The MSA-RFCP encapsulates numerous sources including online activities, preferred images, personal writings, media reports, and interactions with others. Ten research assistants were trained for two days before coding participants, then coders were given feedback by a research manager.

Based on the totality of information, action roles were identified within their primary organization. Actions included plotting, support, traveling, recruiting, propagandizing, and direct action. These action roles were coded as: (1) never attempted but aspirational; (2) attempted but failed; and (3) attempted and successful. To assess risk levels, subjects were assigned to a single action role which seemed to indicate her highest level of involvement within attempted but failed and attempted and successful. Subjects were categorized into the following levels: Highest Risk Category (HRC), Medium Risk Category (MRC), and Lowest Risk Category (LRC). HRC included subjects involved in direct action, whether it was attempted but failed, or attempted and successful. MRC were those coded as plotters, recruiters, or propagandizers, without being involved in direct action, whether these actions were attempted but failed, or attempted and successful. LRC were coded as travelers, necessary agent of state building, and supporters, whether they were attempted but failed or attempted and successful, and they were not involved in MRC or HRC actions. Three categories of motivational catalysts within the MSA-EV were assessed: personal sense of morality, antisocial interests and tendencies, or developmental immaturity.

Researchers studied the effects of propaganda by examining eye gaze, pupil dilation, galvanic skin response (GSR), facial electromyography (FE), and self-reports as they viewed online material generated by jihadist, alt-right, and alt-left organizations. Physiological measures were collected using Tobii Pro X3-120 software, a laptop camera, and Shimmer3 physiological sensors. Tobii Pro X3-120 collects data from the four neuro-physiological devices approximately every eight milliseconds. During the experiment, participants viewed images on a computer in a tent-like structure with consistent ambient lighting for four sessions. The data was processed through iMotion software, transformed into Excel spreadsheets, and displayed as Areas of Interest (AOIs) with heat maps projected onto images.

Utilizing the data obtained from Tobii Pro X3-120, researchers were interested in potential correlations between eye gaze, participants' religious/political affiliation, and responses of perceived emotional reactions as obtained by the Lab Session Inventory (LSI). LSI is a two-page self-report questionnaire that participants filled out after every material-viewing session for a total of four times. The instrument captures three separate measures. The first measure uses Plutchik's emotion wheel diagram, a three-dimensional cone model of emotions, whereby the wheel is based on eight primary emotions, grouped into four pairs of polar opposites. Second, propaganda cognitions, which asks participants to identify techniques conveyed in each series that appear designed to change an individual's thinking concerning their view of the world and their participation in it. The third measure used the Self-Assessment Manikin to measure emotional valence, arousal, and dominance. Scores were entered on a scale of 1 through 9 for each of the three dimensions.

Sample View help for Sample

The study sample for the MSA-EV project consisted of 300 women from 45 countries chosen from a larger sample of 1,468 women who were involved in violent extremism. The 300 women had to meet the following four conditions: 1) the subject's legal identity was known; 2) the circumstances of the subject's involvement in violent extremism were clear; 3) the subject's involvement was significant enough to warrant criminal charges; 4) there was sufficient information in the subject's folder that gave the researcher a sense of her personality and decision-making process.

Originally, the study only included jihadist and far-right groups, but the focus expanded to include far-left groups for two reasons. First, researchers were concerned they would miss valuable information on individuals who were not considered alt-right or jihadist. Second, ideological overlap between jihadist, right-, and left-wing groups was more common than expected. Inclusion of women was originally confined to those from one of the Five Eyes countries (Australia, Canada, New Zealand, the United Kingdom, and United States), but was expanded to include women who weren't native to these countries but committed violent extremism there.

The propaganda research project sample consisted of 90 women ages 18 through 35 years, who self-identified as being conservative, liberal, or Muslim. Half of these women participated in a series of three hour-long online search sessions and focus groups. The other half underwent a single experimental session, as outlined above, measuring eye gaze, pupil dilation, GSR, and emotional facial recognition in response to neutral, alt-right, alt-left, and jihadist materials.

Universe View help for Universe

Women involved with extreme Muslim, alt-right, or alt-left organizations who committed acts of violence in Australia, Canada, New Zealand, the United Kingdom, or United States, from 1970 to 2018.

Unit(s) of Observation View help for Unit(s) of Observation

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Response Rates View help for Response Rates

Not applicable

Presence of Common Scales View help for Presence of Common Scales

none

Original Release Date View help for Original Release Date

2021-06-30

Weight View help for Weight

Not applicable