Identification and classification of radical and extremist actors on Telegram

• Applicants: Dr. Annett Heft; Dr. Pablo Jost
• Subject Area: Communication Sciences
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 543762082

In the recent past, right-wing radical and extremist actors, in particular, have succeeded in politicizing along new lines of social conflict that go beyond established ideological camps. Digital platforms are becoming increasingly relevant for both mobilization and the acquisition of resources. The messenger platform Telegram, in particular, has become very popular here. For political communication research, Telegram offers valuable data access for the analysis of non-institutionalized radical activism. Nevertheless, when analyzing the digital communication and organizational networking of these protest groups, researchers are confronted with specific challenges in (1) the systematic definition, sampling, and collection of relevant actors (identification) and (2) the concrete classification of relevant actors and content according to ideological, typological and other content-related characteristics (classification). The research project aims to contribute to the reliability and validity of the identification and classification of heterogeneous actor groups on digital platforms. To this end, public Telegram channels of far-right actors and actors associated with them as well as their messages on Telegram will be examined in a period between 2019 and 2023. The design includes various identification, classification and simulation studies based on manual and computational methods and answers the following questions: How do different methods of identification and classification of radical and extremist actors on Telegram differ in terms of reliability and validity? How do different identification and classification decisions influence the recorded actor constellations and content on Telegram? This involves a) a systematization of strategies for identifying, sampling, and classifying radical and extremist actors in digital communication, b) the identification of an approximate population of this (German-speaking) target population using network analysis methods, c) the classification of actors using manual and automated methods with various specifications for collecting actor characteristics. In studies on d) evaluation, the effects of various identification and classification decisions are simulated, and their impact on the actor constellations and content is evaluated. Through e) documentation and archiving, we ensure the long-term accessibility and reuse of the findings by means of a reference data set and derive best practices, which should contribute to the informed identification and classification of relevant actor groups by civil society and international research within the framework of f) knowledge transfer.

• DFG Programme: Research Grants