2016 International Conference on Social Media & Society (#SMSociety

• Cornelius Puschmann, Alexander von Humboldt Institute for Internet and Society, Germany
• Julian Ausserhofer, Alexander von Humboldt Institute for Internet and Society, Germany
• Markus Hametner, Der Standard, Austria
• Noura Maan, Der Standard, Austria

Background:

In the past decade, Europe has witnessed the birth of many right­wing protest movements, such as Pegida and the Identitarian Movement. This interdisciplinary project, jointly conducted by communication scholars and journalists at a leading European daily newspaper, explores the networks and messages that characterize these movements of populist outrage over a period of nine months to better understand their interrelations. It achieves its aim through systematic social media and web data analysis of public communication, combined with expert interviews and on­site research.

Objective:

Our central objective is to map the relations of right­wing movements in Europe with a focus on German­speaking countries. Such movements are characterized by their opposition to immigration, European integration, and perceived ‘islamization’. Their organizational structures have commonalities with grass­roots civic movements and rely strongly on social media for organization and communication. Since summer 2015, their outrage has had significant impact on public discourse in Europe.

The project’s objectives are summarized by three research questions

1. Howdonewright­wingmovementscommunicatewitheachotherandtheir followers?

2. Canconnectionsbeidentifiedbetweenthesemovementsandpoliticalparties?

3. Howistheearly­stagecommunicationofselectedgroupsstructured?Whatare

   factors in their communication that let these groups endure?

Methods:

The project takes place over a period of nine months, with successive stages of data collection, analysis, and presentation in different formats. The project combines multiple methods:

• Network analysis, to discover relations among actors, relations between actors and media sources, relations between actors and issues (Rogers, 2013),
• Quantitative (manual) content analysis: We will invest approximately 200 coder hours to code messages, user profiles and websites sampled from the larger volume of material analyzed,
• Supervised machine learning to extrapolate functional categories from structural properties of messages, profiles, and websites, based on human coding decisions (Grimmer & Stewart, 2013, Scharkow, 2013),
• On­site research with focus on in­depth interviews with experts and members of the groups and the observation of physical communication infrastructures.

Results:

The products of our research will both be disseminated through the mass media and in scholarly publications. These products will take on the form of interactive networks, statistical data, maps and stories. To­date, we have collected:

• 150.000 tweets from 50.000 users under the hashtags # pegida and # nopegida, as well as the search term p egida
• 389 wall posts, 54.000 comments and 180.000 likes from the main Pegida Facebook page.
• This has been achieved using DMI TCAT for the collection of Twitter data (Borra & Rieder, 2014), and the use of Rfacebook (Barberá, 2015) for Facebook data. We plan to collect further data from these platforms relying on curated lists of accounts, and to also include YouTube and Instagram.

Future Work:

When the conference takes place, our research will have progressed to an intermediate stage, at which we will be able to present early qualitative (e.g. individual cases) and quantitative (macroscopic relations between actors) results. In addition to the outcome of the project as such, we will also be able to report on the collaboration between academic research and journalism on this vital issue at the interface of public communication and scholarly knowledge.

References:

Borra, E., & Rieder, B. (2014). Programmed method: developing a toolset for capturing

and analyzing tweets. A slib Journal of Information Management, 6 6( 3), 262–278.

Grimmer, J., & Stewart, B. M. (2013). Text as data: The promise and pitfalls of automatic content analysis methods for political texts. P olitical Analysis, 2 1( 3), 267–297. doi:10.1093/pan/mps028

Rogers, R. A. (2013). D igital Methods. Cambridge, MA: MIT Press.

Scharkow, M. (2013). Thematic content analysis using supervised machine learning: An empirical evaluation using German online news. Q uality & Quantity, 4 7( 2), 761–773. doi:10.1007/s11135­011­9545­7