'Dark Germany': Hidden Patterns of Participation in Online Far-Right Protests Against Refugee Housing

Sebastian Schelter, Jérôme Kunegis

TL;DR

This paper provides a quantitative look at the 2015 surge of anti-refugee protest movements on German Facebook. By analyzing over a million interactions, the researchers found that while the movement is geographically fragmented and lacks a central "giant" community, its participants are deeply unified by a strong shared affiliation with far-right political parties like the AfD and NPD.

What Problem Does This Paper Address?

In 2015, Germany saw a massive influx of refugees, sparking a wave of local "anti-refugee housing" protest movements. These groups often organized under decentralized Facebook banners like “Nein zum Heim” (No to the shelter).

Because these protests were hyper-local and spread across hundreds of individual pages, they were difficult for researchers to track as a single cohesive movement. The authors aim to shine a light on this "Dark Germany"—the hidden patterns of how these users interact, what triggers their activity, and how closely they are actually connected to established far-right political organizations.

Methods

The researchers employed a multi-pronged computational approach to analyze the social media landscape:

  • Data Mining: They performed exhaustive searches on Facebook to identify 136 specific protest pages, crawling all public posts, comments, and likes from the year 2015.
  • Topic Modeling: To understand what was being discussed, they used a variant of Latent Dirichlet Allocation (LDA) called Topics over Time (ToT). This allowed them to see not just the themes (like crime or housing costs) but how those themes waxed and waned in response to news events.
  • Sentiment Analysis: Using the GermanPolarityClues dictionary, they performed part-of-speech tagging on comments to calculate normalized positive and negative polarity scores over the course of the year.
  • Network Analysis: They constructed a "user co-like network," where edges were created between users who liked the same posts. They also calculated an affiliation metric to measure the overlap between protest page users and followers of major German political parties.

Datasets

The study is based on a custom-crawled dataset from 2015 consisting of:

  • 136 Facebook protest pages tagged with geographic coordinates.
  • 46,880 posts and over 1 million interactions (comments and likes).
  • 209,822 unique users.
  • The authors also leveraged a secondary dataset from their previous work containing interaction data from the Facebook pages of German political parties (AfD, NPD, CDU, SPD, etc.) for comparison.

Key Insights and Ideas

  • The "Merkel Effect": Activity didn't grow linearly; it was highly reactive. A massive spike in posts, negative sentiment, and new user attraction occurred in September 2015, immediately following Chancellor Angela Merkel’s decision to admit Syrian refugees.
  • Geographic Skew: Despite being a "nationwide" issue, 77% of the protest pages were located in Eastern Germany, reflecting long-standing political and regional divides in the country.
  • The Illusion of Community: Surprisingly, the movement lacks a "Giant Connected Component." In most social networks, a single large cluster connects almost everyone. Here, the largest component only covered 57.5% of users. This suggests the movement is a "long tail" of isolated local bubbles rather than one big digital family.
  • The AfD-NPD Overlap: One of the most striking findings is the political affiliation. Users of these protest pages were roughly ten times more likely to interact with the far-right AfD (0.45 median affiliation) or the extremist NPD (0.41) than with mainstream parties like the CDU or SPD (~0.04). This suggests that even if the AfD claims a "conservative" mantle, its digital grassroots are inextricably linked to radical protest circles.

Key Results

  • Temporal Peaks: Discussion volume regarding "Europe" and "Refugee Policy" peaked sharply in late 2015, coinciding with the peak of the refugee crisis.
  • Negative Dominance: Sentiment analysis showed that negative speech dominated the comments throughout the entire year, with a significant "anger peak" in September.
  • Low Growth Acceleration: While the pages maintained a constant stream of new users, they failed to "go viral" in a way that accelerated growth over time; instead, they relied on a core group of "continuing users" who were highly correlated with the volume of posts.
  • Network Fragmentation: There was a very low correlation (-0.19) between geographic distance and shared users, meaning people weren't necessarily collaborating with neighboring towns; they were mostly active in their own local silos.

Limitations and Future Work

The authors acknowledge several constraints imposed by the Facebook platform and their methodology:

  • Data Access: Due to API limitations, they could not access shared posts, page-level likes, or private profile information, which might have provided a deeper look into user demographics.
  • Lack of Baseline: The study lacks a "control group" of non-controversial or left-wing movements to determine if these patterns (like the lack of a giant component) are unique to the far-right or common across all decentralized Facebook movements.
  • Future Directions: The researchers plan to move beyond dictionary-based sentiment analysis to conduct a deeper textual analysis of comments to measure "controversy" and the specific rhetoric used in user-to-user interactions.

Summary updated

Abstract

The political discourse in Western European countries such as Germany has recently seen a resurgence of the topic of refugees, fueled by an influx of refugees from various Middle Eastern and African countries. Even though the topic of refugees evidently plays a large role in online and offline politics of the affected countries, the fact that protests against refugees stem from the right-wight political spectrum has lead to corresponding media to be shared in a decentralized fashion, making an analysis of the underlying social and mediatic networks difficult. In order to contribute to the analysis of these processes, we present a quantitative study of the social media activities of a contemporary nationwide protest movement against local refugee housing in Germany, which organizes itself via dedicated Facebook pages per city. We analyse data from 136 such protest pages in 2015, containing more than 46,000 posts and more than one million interactions by more than 200,000 users. In order to learn about the patterns of communication and interaction among users of far-right social media sites and pages, we investigate the temporal characteristics of the social media activities of this protest movement, as well as the connectedness of the interactions of its participants. We find several activity metrics such as the number of posts issued, discussion volume about crime and housing costs, negative polarity in comments, and user engagement to peak in late 2015, coinciding with chancellor Angela Merkel’s much criticized decision of September 2015 to temporarily admit the entry of Syrian refugees to Germany. Furthermore, our evidence suggests a low degree of direct connectedness of participants in this movement, (i.a., indicated by a lack of geographical collaboration patterns), yet we encounter a strong affiliation of the pages’ user base with far-right political parties.
BibTeX
@inproceedings{schelter2017dark,
  author = {Sebastian Schelter and Jérôme Kunegis},
  title = {{'Dark Germany': Hidden Patterns of Participation in Online Far-Right Protests Against Refugee Housing}},
  booktitle = {SocInfo},
  year = {2017},
  doi = {10.1007/978-3-319-67217-5\_17},
}

Loading related content...

Figure