Social media and political contention - challenges and opportunities for comparative research

ABSTRACT

In this special issue, the authors theoretically, methodologically, and empirically address challenges and opportunities associated with comparative social media analysis in political contention. Actors from civil society, media, and institutional politics use social media to coordinate, mobilise, and communicate, turning public online communication into an arena of conflict that offers researchers valuable windows of observation. In this introduction to the special issue, we systematise comparative perspectives on social media and political contention. We outline the traditional comparative dimensions of space, time, platform, and case; and suggest an approach for comparison within dimensions that are less dependent on the rapidly changing social media environment and more attuned to the interconnection between social media and political contention.

KEYWORDS:

• Social Media
• comparative research
• political contention
• activism
• digital communication

This special issue aims to advance comparative studies of social media analysis within the context of political contention, including demonstrations, protests, strikes, riots, civil disobedience, and insurrection. Recent contributions in political science and communication studies have seen growing attention paid to the analysis of social media data to understand how people worldwide resist injustice, disobey authority, and protest against repression. The homogeneity of social media data in both format and content promises “systematic comparisons for different types of actors, across multiple countries, and over time” (Barberá & Steinert-Threlkeld, 2020). However, there is a notable lack of theoretical deliberation on the reasons for engaging in such comparisons and of methodological reflection on the various challenges encountered. Rather than considering the diverse and evolving ways in which social media are used for political contestation by different actors in various political contexts, most studies are concerned (and thereby constrained) with a single country, single platform, or a single issue, topic, or organization (see Brändle, Eisele & Kulichkina, this issue). Such lacunae not only prevent the development of vigorous comparative methodologies and the advancement of empirical knowledge, but also hinder the opportunities for building generalizable theories that extend beyond specific cases of digitally-mediated political contention. This special issue addresses these shortcomings, theoretically, methodologically, and empirically, by tackling the challenges and opportunities associated with comparative social media analysis in political contention.

Overall, there are five main dimensions in comparative social media research: cross-spatial, cross-temporal, cross-functional, cross-media (or platforms) and cross-case comparisons (Zhao & Liu, 2020). Cross-spatial comparison, a dominant approach in comparative communication research, begins with “comparing two or more nations with respect to some common activity” (Edelstein, 1982, p. 14), contrasting different countries, territories, or groups of political systems in a manner that allows researchers to examine the object of investigation simultaneously. Cross-spatial comparison in social media research typically involves country-based comparisons and may include multilingual data sets (e.g., Boulianne & Ohme, 2022; Jenkins, 2019; Matassi & Boczkowski, 2021, 2023), specifically on “Cross-National and Regional Comparisons”). Cross-temporal (or longitudinal) comparison involves “two or more geographically or historically (spatially or temporally) defined systems,” where “the phenomena of scholarly interest” are seen as “embedded in a set of interrelations that are relatively coherent, patterned, comprehensive, distinct, and bounded” (Blumler, McLeod, & Rosengren, 1992, p. 7). Hallin and Mancini’s (2019) discussion on media systems as dynamic, rather than static, exemplifies cross-temporal comparison by acknowledging that macro-level units are continually evolving under societal transformation processes. Despite its value, cross-temporal comparison has been underutilized, primarily due to the scarcity of time-series data (but see, e.g., Zamponi & Daphi, 2014).

As a third dimension, Caramani (2020, p. 11; also see Esser & Hanitzsch, 2013) suggests functional (cross-organizational or cross-process) comparison (Burke & Şen, 2018). Examples include comparing social media choices in political activism (Burke & Şen, 2018) or analyzing the impact (i.e., “function”) of social media during different phases (e.g., episodic and “latent” or sustainable phases) of digital activism (Leong, Pan, Bahri, & Fauzi, 2019). Functional comparison offers a method for investigating the varied roles social media can play at different stages of political contention. The fourth dimension of comparison is cross-media (or, in some studies, “platforms”), which in this case includes comparisons of different social media platforms’ affordances (Chagas, Carreiro, Santos, & Popolin, 2022; Theocharis, Boulianne, Koc-Michalska, & Bimber, 2023) and contrasts between platforms or between social media and legacy media. An increasing number of cross-platform comparison studies begin to challenge the predominance of single-platform social media research (Özkula, Reilly, & Hayes, 2023; Pearce et al., 2020). These studies also broaden the scope of the investigation beyond text to include various modalities such as visual (e.g., McSwiney, Vaughan, Heft, & Hoffmann, 2021; Rogers, 2021), audio (Neumark, 2006), and multimodal combinations (Literat & Kligler-Vilenchik, 2021). Additionally, it is important to note our preference for using the term “media” rather than “platforms” in this comparative inquiry. This preference stems from the desire to encompass the dynamics of the hybrid media system (Chadwick, 2013), which involves interactions and complementarities between legacy and emerging media logics. Lastly, cross-case comparison investigates different topics of debates, providers, or user communities. For example, Theocharis, Lowe, Van Deth, and García-Albacete (2015) explore the Occupy Wall Street protesters in the United States and the indignant activists in Spain and Greece.

Overall, while there are some comparative studies based on the analysis of social media data (e.g., Larsson, 2015; Oz, Zheng, & Chen, 2018), they are surprisingly limited in the field of political contention and often focus on a comparison between two cases (e.g., Hellmueller, Lischka, & Humprecht, 2021; Mare, 2017; Qiu, Lin, & Leung, 2013). Comparative social media analysis seems problematic due to variations in communication culture (such as sharing behavior and platform usage), political systems, legislation (such as censorship), opportunity structures, social media affordances, user demographics, or access to data. Comparative approaches are further essential to understand the scope of transnational political contention, its diffusion and impact. Yet, we lack methodological, conceptual, and practical approaches for comparative social media analysis of political mobilization, protest, activism, engagement, and contestation (with exceptions, such as Caiani, 2019; Jost et al., 2018).

The challenge of comparatively analyzing protest and political mobilization in the digital age is being taken up by a new generation of social movement and civil society scholars with advanced methodological skills in using digital data sources. The implementation of comparative research designs requires new interdisciplinary alliances between scholars of comparative politics and media and communication studies, but also relies on computational skills, visual analytics, and legal expertise (e.g. in navigating different regulatory regimes). There is a noticeable shift in methods from case studies and often ethnographic research designs of social movement cultures (Tarrow, 2021) to quantitative methods involving large datasets explored by computational techniques. At this moment of departure toward new research designs in comparative social media studies, it is all the more important to look back at the legacy of comparative politics and its applications in the field of social movement studies and political protest. New comparative perspectives in contentious politics have been opened up with the intention of not only analyzing the impact of different forms of protest, but more broadly to encourage the crossing of various disciplinary, historical, or geographical boundaries that divide the field of contentious politics and to explain its broader societal and political resonance (McAdam, Tarrow, Tilly, Lichbach, & Zuckerman, 2009). We can think of comparative social media research as a boundary-crossing exercise that observes digital and social media political cultures, repertoires of contention and their impact over time, types of media and the shifting dynamics of actors, movements, publics, and audiences. Previous studies have primarily focused on how different types of social media facilitate the expression of political protest (Jost et al., 2018), and often assumed a soft form of technological determinism in the way platform affordances influence the political behavior of individuals or groups online (Ronzhyn, Cardenal, & Batlle Rubio, 2023). This perspective may leave us with a limited view on how political contention, in particular, can be studied through comparative social media analysis.

Moreover, social media research highly adapts to the evolving landscape of digital and social media platforms and the changing access to data available for analysis. This field of research is subject to increasing inequalities and different opportunities resulting from social media data being rather a by-product of large corporations’ activities than a result of social scientists’ rigorous research design (Walker, Mercea, & Bastos, 2019). Scholars have highlighted that companies like X, Meta, and Alphabet follow capitalist business models aimed at generating ad revenues for their shareholders (Poell, van Dijck, Burgess, & Marwick, 2018). While such a business model based on generating interactions and promoting visibility may in itself breed tensions between democratic norms of moderation and rationality and the shrill, anger-driven voices that seem to dominate the digital public sphere (Gerbaudo, 2018), it also imposes a practical barrier for researchers in terms of data governance. While much academic research has relied on the loophole of Application Programming Interfaces to collect social media data, this often required scholars to pose as developers, requiring skills not traditionally taught at social science departments to first access the stream of information and second, process vast amounts of data not initially designed for academic research but for marketing purposes. In the absence of a binding legal framework for accessing social media data for the public interest, the volatility of access and attempts to monetize such access (most notoriously by X) have created significant obstacles for comparative research. Questions such as who has the programming skills to access data, who has the processing power to run sophisticated computational analyses, and who has the financial resources to purchase social media data render inequalities in researchers’ assets and inhibitions ever more salient. Comparative research often requires data from multiple social media platforms, historical data over extended periods, and data across different countries and/or languages. This exacerbates a gap often obscured by publications resulting from projects that do possess the necessary resources, producing visibility only for a limited number of issues.

The authors in this special issue show that comparative social media analysis is feasible (even for small-budget projects), but it typically requires collaboration among researchers across countries and disciplines and funding through international research grants. Taken together, the contributions are based on social media analyses of political contention from a comparative perspective. They allow us to move beyond the above-mentioned dimensions of comparative social media research in general and advance its application to understand political contention in particular.

The articles in this special issue

Verena Brändle, Olga Eisele, and Aytalina Kulichkina conduct a systematic literature review of social media research concerning political contention of LGBTQ+ movements. They develop a theoretical framework embedded into the theoretical concept of opportunity structures (Tarrow, 1996; Tilly, 2008) to identify opportunities and constraints for social movements that can be comparatively assessed in social media research. The framework considers contextual dimensions (cultural, political, as well as social media affordances), and knowledge production and doing of contention as two levels of action. The authors conclude by arguing that the potential for comparative research in LGBTQ+ movements is unfulfilled. In terms of knowledge production they find that there is a lack of mixed methods approaches combining qualitative and quantitative research; theories, hypotheses and concepts resulting from qualitative work are not exploited in quantitative empirical work; and there is a low number of collaborations across countries. Moreover, opportunity structures within national contexts add constraints due to the risk researchers still face when studying these topics in certain national contexts (such as China or Hungary). Identifying these variations in opportunity structures, they argue, could be a first step. Finally, they suggest that the differences in social media usage as well as the changes in access to social media data create additional challenges. Yet, we may address these by introducing comparative dimensions that are less dependent on social media platforms but instead, allow for more continuity and comparison across various country contexts and platforms. Building on theories grounded in social movement studies, conscientious politics, and/or media and communication studies with less focus on specific platforms may be a first step toward developing dimensions that enable comparative social media analyses that are less dependent on data accessibility and data structure as provided by social media platforms.

In a methodologically innovative design, Nicolò Pennucci combines semi-structured interviews with key activists and social media managers and quantitative content analysis of movements’ Facebook pages. This design is used to study the “Sardines” movement in Italy and the Anti-Brexit campaign in the United Kingdom, showing that social media is a vital tool for the construction and reinforcement of a collective identity, even in times of personalized contentious politics. Doing so, Pennucci generates insights from the backstage dimension of coordination and decision making (through interviews), and from the frontstage of public communication and deliberation (through content analysis). Pennucci discusses the way social movements engage in reactive identity construction via social media, when faced with populist parties in power. While much scholarly attention has focused on the way, populist actors, either as challengers or from a position of power, have harnessed the affordances of social media platforms, this research sheds light on the often overlooked aspect of how an anti-populist civil society reacts to populism “from above.” The two parallel processes of political identification and political counter strategy are venues of internal and external struggle over narratives and meanings, especially for social movements that act not as defenders of liberal democracy against (right-wing) populism, but at the same time negotiate their own counter-hegemonic ideas of state and society. At the same time, country-specific contextual factors can explain differences. In Italy the response strategies to populist power focus on the mobilization for street-based activism through digital interactions, whereas in the UK, activists adopted a strategy of lobbying parliament to adopt a supranational and inclusive understanding of citizenship. As this study shows, cross-spatial social media analysis gives us insights into the various national contexts political contention takes place in and how they may affect the formation and expression of collective identify through social media.

Collective identity is also a cornerstone of Aurora Perego and Katia Pilati’s contribution to this special issue, which applies a relational perspective to the study of within- and across-field interactions in LGBTQIA* actors in a cross-spatial comparison of Milan and Madrid. Organizational identity, in this study, plays an intermediary role, that depending on contexts and opportunities, can incentivize actors to focus on bonding ties within a field of similar actors, or focus on bridging ties, that bridge connections to actors from different collective action fields. Perego and Pilati follow a tradition of social movement studies that compares social movement across localities (or cross-spatial), while at the same time seeking to understand them through their digital, public communication via Facebook. Ties are thus operationalized as different forms of interaction through the affordances of social media, allowing for three different networks to emerge from either mentioning, sharing, or event-promoting. Using both descriptive and inferential techniques of social network analysis, Perego and Pilati add a cross-temporal temporal layer to compare five different points in time, thus accounting for variation in context and opportunity not only across cities, but also across time. The latter dimension is crucial against the backdrop of the changing politics in both countries and the rise of populist political parties from both left and right. Finding that LGBTQIA* organizations indeed stand together, the study advances our understanding of social medias’ role for progressive movements under different contextual conditions. The research presents an example of fruitful employment of social media analysis across two dimensions (temporal and spatial) embedded into a framework of social movement studies to include the specific opportunity structures of movements.

Ofra Klein, Hans-Joerg Trenz and Nadine Hesse theoretically ground their work in the field of political conflict and develop an analytical tool that enhances comparative, visual analyses across hashtags to assess variation in degrees of conflict. More specifically, they investigate visual communication on Instagram during the COVID-19 pandemic in Europe comparing Germany, Poland, and The Netherlands in a content analysis across six hashtags (two per country case) and with images including post texts as their unit of analysis. They argue that hashtags can be associated with and generate various conflicts within a country which can be a challenge to research designs including a comparison across countries. To overcome such limitations, they propose a framework that goes beyond conflict and relates to policies, actors, and values. Along these three dimensions they analytically differentiate between deep and regular conflicts (e.g., lockdown rules vs. questioning of scientific truths), level of polarization, level of conflictual framing in an image, and the general tonality (e.g., urgency) expressed in an image. They find that hashtag publics are not necessarily restricted to specific country contexts but can be used as markers of a debate with a heterogeneity of participants who both oppose or support a cause. They point to the challenge of conducting research restricted to the native language within each country, as much of the social media communication concerning political conflicts is structured along English language hashtags. Moreover, content moderation of controversial posts by social media platforms and the fluidity and ephemerality of social media content more generally, poses challenges to sampling for a systematic comparison. They conclude by arguing that visual analysis of Instagram posts allows for a comparative assessment of the degree of political conflict across countries. However, their study also points toward the specificity of comparative social media analysis. Debates around political conflicts as they play out in visual communication on Instagram (and other social media platforms) are not restricted to a country, language context, and/or hashtag but are more dynamic and conversely, using hashtag-publics as a starting point and comparing across such publics may be a fruitful avenue for future research. Doing so, they add a dimension of cross-spatial comparison that combines insights from social media research in the specific context of contentious public spheres.

Stefan Wallaschek, Kavyanjali Kaushik and Monika Eigmüller focus on the contested field of EU politics. The authors suggest analyzing actor constellations in conflict as a single indicator measurement to enable systematic cross-spatial comparison of Germany, Poland, and Spain, and cross-media comparison of print newspapers and Twitter. In other words, they propose a comparative research design that controls for country variation and selects different legacy and social media outlets. The findings indicate the need to move beyond traditional comparative research designs of political conflict when dealing with strongly interlinked political debates, such as those in the EU. The politicization of EU issues is often driven by Eurosceptic and populist actors who aim to reclaim national sovereignty from perceived EU supremacy and challenge EU institutions with democratic demands from citizens. Meanwhile, the EU has taken on the role of protector of democracy and fundamental values, and has started to sanction rule of law violations by some member states. The crisis concerning the rule of law in the EU is becoming more and more of an internal struggle over democracy and fundamental values. The authors focus on the dispute over independence of the judiciary, which arose following judicial reforms in Poland. It is expected that conflicts regarding EU fundamental values will increasingly drive the Europeanisation of national debates. However, it is uncertain whether these debates are still primarily amplified by legacy media and traditional journalism in member states, or if value conflicts are now being driven by social media and bottom-up dynamics of transnational political conflict. The dynamics of conflict in these cases are driven by processes of cross-country and cross-media diffusion, where conflicts around democratic values become a driving force behind the larger integration project. The involvement of the EU in the conflict is important to protect and promote democratic values. While concerns about democratic backsliding in the EU are raised by both legacy media and social media entrepreneurs, it is worth noting that discussions about the future of democracy are mainly domestic and politicization in national media is driven by national (often critical and Eurosceptic) actors. Twitter debates are no exception to this rule, as they often highlight Euro-critical voices and their particularistic agendas.

Matthias Hoffmann and Christina Neumayer adopt a temporal dimension to investigate the interaction patterns of movement parties with established parties on social media across six European countries. In a cross-spatial and cross-temporal comparison they integrate concepts from the lifecycles of social movements and episodes of contention. Their work addresses an emerging yet less-studied issue in the field of political contention and party politics: the positioning and repositioning of movement parties such as Alternativet (Denmark), United Kingdom Independence Party (UK), and Alternative für Deutschland (Germany) emerging in the decade in institutional politics. The study combines the comparative perspective across six European countries with a temporal dimension over a period from 2010 to 2021. By longitudinally examining movement parties’ official social media communication regarding institutional positioning and referencing patterns, the authors identify three variations in the trajectories of movement parties in the polity as (a) toward electoral insignificance, (b) toward a permanent challenger in the polity, and (c) toward incumbency. The temporal dimension is valuable in this comparative study to uncover that movement parties are “not necessarily transitioning further to established parties” but “remain unstable organizations” as outsiders in the institutional arena. The comparative dimension further pinpoints country-specific contextual factors like electoral systems and political actors that produce the patterns and variations of positioning and repositioning in the polity arena. Taken together, this study exemplifies how the temporal dimension can complement the comparative perspective to challenge the ideal model of linear development of movement parties from positioning in the protest arena to institutional positioning in the polity arena.

Annett Heft, Kilian Buehling, Xixuan Zhang, Dominik Schindler and Miriam Milzner examine conspiracy-related debates and highlight the impact of social media on contentious politics. In the absence of journalistic filters, political opinions and news are diffused through a hybrid and interconnected digital sphere. Their cross-media comparison shows that effects across platforms are measured through user-driven exchanges of content, such as liking and sharing and through the activity of protest actors to strategically place their content on multiple platforms. Acknowledging the hybrid nature of the social media conspiracy sphere presents a challenge for comparative research in several ways. Cross-platform research designs depend on equivalent data sampling, but the types of data and access options vary widely across platforms. At the same time, researchers must consider the vast differences in the way platforms address distinct audiences, allow for various forms of user interaction, and the emergence of sub-cultural milieus, as in the case of conspiracy-related content. The design of cross-platform comparisons of the diffusion of contentious politics may not necessarily be compatible with the design of platform-comparative studies of the different spaces for the salience of contentious actions and the expression of discontent. To understand the potential of social media platforms in generating conspiracy theories, it is important to distinguish it from understanding how conspiracies spread across social media. Heft et al. highlight the limitations of valid data collection, but offer various approaches for actor- and content-specific sampling strategies to trace conspiracy-related conflicts. Both actor and content-specific searches using keywords are possible on open-access platforms such as Reddit, public Facebook pages, or YouTube. On hybrid communication infrastructures, such as Telegram, where content-based searches across the platform are not possible, identifying chat-groups or channels that were used for the diffusion of conspiracy-related content yielded valid results for comparison. Comparative research designs of conspiracy-related conflicts also face a time-related problem of sampling, as relevant content diminishes over time. The contestation dynamics of social media involve a struggle over censorship between producers, providers, and users. To conduct research in this area, it is necessary to rely on life or instant sampling strategies, which may not be equally available across all platforms. The authors suggest retrospective data collection through private archives, but this approach is dependent on the willingness of data owners to cooperate with researchers. This research shows that the challenge of comparative research ultimately requires increased cooperation within the social media research community. This is necessary for practical solutions such as data sharing, promoting technological innovation in sampling and collective data analysis, and developing content-based approaches such as dictionaries that can be used for cross-platform, country, and time comparisons.

Conclusion

Taken together, the authors who have contributed to this special issue have collectively reinvigorated the field of social movement studies, including agents of protest, coalitions and networks, repertoires of contention, changing opportunity structures (such as the increased digitization of political activism during the pandemic), and variations in protest across cultures, ideologies, countries, and regimes (from liberal to illiberal democracies). They have also tackled normative issues related to the assessment of the socio-political impact of protest and its democracy-disrupting (e.g. through disinformation) or democracy-enhancing effects. The research in this special issue employs a rich variety of methods, focusing on structures and interactions (e.g. through social network analysis), the production and diffusion of audiovisual and textual content (e.g. via qualitative or quantitative content analysis), and mixed-methods approaches that enrich social media data with other data sources (e.g. surveys or interviews). A largely untapped area in the comparative analysis of social movements and protest is user and audience studies, which are gaining relevance for the reception of protest and broader patterns of support and opposition among different user communities, audiences, or larger populations. The triangulation of different research methods, such as user interaction and comment analysis with public opinion research, is highlighted as a desirable approach to advance the comparative research agenda and gain new insights into the emergence, manifestation, diffusion, and reception of online political protest.

All contributions also point toward challenges specific to comparative social media research in the field of political contention. Common challenges include, such as reliance on social media platforms for data, inequalities in data access, variation in social media usage across national contexts, language barriers, and the rapidly changing development and culture around social media platforms over time. Yet, the contributions also point to challenges particular to the study of political contention such as varying opportunity structures, risks to activists and social movements in authoritarian contexts, differences between latent and active protest periods, life cycles of social movements, and the specificity of publics on social media platforms. As the contributions to this special issue show, there is no one-size-fits-all solution to these challenges, but they each suggest ways forward through their rigorous research designs. They collectively emphasize the need to develop theory within the realms of political communication, social movement studies, the politics of contention, and media and communication studies, alongside analytical and methodological approaches to comparative social media analysis. This approach aims to move toward more stable dimensions for comparison that are less dependent on the rapidly changing social media environment and more attuned to the interconnection between (social) media and political contention.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work is a collaboration across projects supported by the Volkswagen Foundation (ProDem: Protests and Democracy and ValCon: Value Conflicts in a Differentiated Europe) and The Independent Research Fund Denmark [grant number 1055–00011B].

Notes on contributors

Matthias Hoffmann

Matthias Hoffmann is a Postdoctoral Researcher at Babeș-Bolyai University, Cluj. His research focuses on relational approaches to collective action and social movements, on political parties and protest, and on digital communication

Jun Liu

Jun Liu is an Associate Professor at the Center for Tracking and Society at the University of Copenhagen, Denmark. His research spans political sociology, digital technologies, and comparative and computational social science. His latest work includes a monograph titled “Shifting Dynamics of Contention in the Digital Age,” published by Oxford University Press. He is currently leading a four-year research project funded by the Independent Research Fund Denmark Research Leader Scheme, focusing on a comparative study of digital technologies and political contention in the US, EU, and China (https://tech-in-movement.ku.dk/).

Christina Neumayer

Christina Neumayer is associate professor at the Department of Communication at University of Copenhagen. Her research is concerned with intersections of political contention, digital media, digital data and artificial intelligence.

Hans-Joerg Trenz

Hans-Jörg Trenz is Professor of Sociology of Culture and Communication at Scuola Normale Superiore in Florence. His main fields of interest are media and public sphere transformations, digital society, political communication and democracy.