Workshop Report: Expert Discussion – Protest Event Analysis in the Corona Pandemic
Author: Elisabeth Oertel
On April 21-22, 2022, international experts met in Dresden to discuss developments in protest movement activities during the pandemic. The workshop, “Protest Event Analysis in the Corona Pandemic,” focused on the anti-democratic orientations found in some of the protests and the methodological challenges of empirically studying them. Funded by the Deutsch Polnische Wissenschaftsstiftung (DPWS), the workshop was organized by the Hannah-Arendt-Institute for Totalitarianism Studies (HAIT) at the TU Dresden in cooperation with the Institute for the Study of Protest and Social Movements (ipb) and the Institute of Philosophy and Sociology of the Polish Academy of Sciences (IFiS PAN). Scientific support for the workshop and coder training for early career researchers was provided by Dr Michael Neuber from the Center for Technology and Society (ZTG) at the Technical University Berlin.
Protest Event Analysis: methodological challenges and new ideas
The Corona-Pandemic is more than a health crisis. In the last two years, protest movements with an anti-democratic orientation have formed throughout Europe. In an international cooperation with IFiS PAN and ipb, HAIT launched a new research project on November 1, 2021, funded by DPWS. The project’s objective is a comparative analysis of protests against the measures to contain the Covid pandemic (“Corona protests”) in Poland and Germany (focusing on the federal states of Saxony and Baden-Württemberg). Alongside an investigation of protest events, the project’s scope extends to methodical advancements in Protest Event Analysis (PEA). On April 22, 2022, the project team gathered with guest researchers for a workshop and expert discussion regarding Protest-Event-Analysis in the Corona Pandemic.
PEA aims to shed light on the frequency and characteristics of protest events by using different methods of content analysis. To do so, various source materials are evaluated and prepared for an analysis that uses statistical methods. For a long time, PEA relied mainly on newspaper articles as empirical material – in the meantime, digital sources such as social media content are also used. In many cases, PEA is used to study longer time periods spanning years or even decades, making it possible to identify protest waves and observe the influence of contextual factors. Additionally, the methods of PEA provide an ideal basis for comparative studies on a regional, national or international level and allow for the discovery of peculiarities connected with specific topics of protest.
Even without the extraordinary context of the Covid crisis, researchers in PEA are confronted with several challenges. Reports in news articles are often biased, be it regarding the number of protest participants, partisanship, or ideological affiliations, down to the very basic matter of what protest events are reported about at all. For example, stark differences in the frequency of the coverage of protests can be seen when comparing regional and national newspapers. In terms of PEA’s epistemic interest, this selectivity is important given the large number of protests involving only double-digit numbers of people. Reports on gathering sizes also tend to vary greatly, being exaggerated by organizers on a regular basis.
The analysis of “Corona protests” is further complicated by context-specific factors. A large portion of communication between organizers and participants took place in the digital sphere, such as via Facebook or Telegram Messenger. Event announcements and mobilization efforts can, of course, be retrieved from these platforms, although their explanatory power is only limited. During the lockdowns, some protest events have been restricted or even prohibited by local authorities. In consequence, a multitude of demonstrations has not been registered in the first place, making it difficult and time-consuming to find out which events have actually taken place. Furthermore, in cases of recurring protest events with low participant numbers, journalists may lose interest and disregard them altogether. Due to the right-wing (often extremist) and anti-democratic nature of parts of the protest scene, the usage of digital media is also affected by so-called “de-platforming,” meaning the exclusion of groups from social media platforms and the deletion of their contents.
Workshop proceedings: Facing the challenges
With these and other challenges that PEA is facing in the age of Covid in mind, the workshop discussed ideas and solutions to address them. The meeting was opened by Prof Sebastian Haunss (University of Bremen), who introduced the guests to the operating principles and reliability of machine learning as an instrument of PEA. Such machine learning projects hope to reduce the time and effort of manual coding by testing and training models to identify those parts in a body of data pertaining to protest events and – ideally – to extract their information. The models “practice” this process on training datasets made up of material that addresses only a few exemplary cases. An ideal final product would therefore be a model that is able to identify material similar in content but differing in origin or context.
Subsequently, Prof Wolf J. Schünemann (University of Hildesheim) presented the project EPINetz (Exploration of Political Information Networks) conducted by an interdisciplinary team of researchers at the universities of Heidelberg and Hildesheim. It focuses on testing new methods of network exploration with the aim of visualizing political debates in a structured and systematic manner. The researchers continuously collect data from different online platforms to be used, for instance, in analyses of agent networks, clusters of co-occurring terms, or evaluations of the dynamic changes in political debates over time. In practice, researchers can find out whether networks of political parties overlap or are the same on different platforms or which groups of users prefer which hashtags, to name but a few examples.
Shifting the focus from text-based material to audio-visual data, Dr Henry Rammelt (SNSPA Bucharest) presented his ongoing project on live-streamed protest events. Although the perspective of the people behind the camera is always subjective and therefore distorted, the protest streams nevertheless offer valuable insights into the course and composition of the protest events – especially when they are coded following a standardized questionnaire. Given the problems related to participant observation in times of high contagion, illegal assemblies, and ethical concerns, the approach seems to hold great potential for the analysis of protest events.
Finally, principal investigator Prof Isabelle-Christine Panreck and protest researchers Dr Piotr Kocyba (ipb) and Dr Michael Neuber presented first results of their research on corona protests in Poland and Germany. The presentation gave a first look into the application of new technological approaches in PEA. Two discussants, Prof Dieter Rucht and Dr Simon Teune (both ipb), opened a fruitful debate on the future of PEA as well as on the specific case of protests during the pandemic.
Funded by the German-Polish Science Foundation
Gefördert aus Mitteln der Deutsch-Polnischen Wissenschaftsstiftung
Projekt wspierany przez Polsko-Niemiecką Fundację na rzecz Nauki
Zitiervorschlag: Elisabeth Oertel, “Workshop Report: Expert Discussion – Protest Event Analysis in the Corona Pandemic“. In: Denken ohne Geländer. Der Blog des Hannah-Arendt-Institut für Totalitarismusforschung e.V. vom 15.06.2022: https://haitblog.hypotheses.org/3144
Bild 1: © Maximilian Kreter
Diesen Blogbeitrag zitieren
haitblog (2022, 15. Juni). Workshop Report: Expert Discussion – Protest Event Analysis in the Corona Pandemic. Denken ohne Geländer. Abgerufen am 24. Februar 2024, von https://doi.org/10.58079/pf1i