The political contention of LGBTQ+ communities in the digital age - state of the art, limitations, and opportunities for comparative research

ABSTRACT This paper develops an analytical framework for comparative research on political contention in the digital age and, building upon it, provides a literature review of social media research related to LGBTQ+ political contention. So far, we lack systematic insights into the literature on digitally-mediated LGBTQ+ political contention and its potential for comparative research. Applying scoping literature review, we focus on the key comparative dimensions such as political context, social media, and knowledge production. The results provide an overview of the state of the art, limitations, and opportunities while also developing an agenda for future comparative research. This paper supports the visibility of LGBTQ+ issues in a still hostile public debate for both LGBTQ+ activists and researchers.


Introduction
Social media have become essential tools for activism by and for marginalized groups such as LGBTQ+, allowing innovative and diverse repertoires of contention.However, there is only little comparative research about digital LGBTQ+ contention, suggesting a need for theoretical development in the field.Comparative research is crucial to evaluating strategies of social movements and identifying common challenges and successful practices for achieving change.Comparative research of digital LGBTQ+ contention, in particular, advances our understanding of the promises and pitfalls of social media for political activism under conditions of strong societal polarization, where marginalized communities often need to operate in adversarial political contexts.
Addressing this lacuna in the literature, our scoping review is guided by the following research question: What comparative dimensions can we use to develop an analytical framework for studying political contention in the digital age?And to what extent does the existing literature examine and focus on these dimensions?The aim of our study is twofold.Firstly, we argue that for a comprehensive analytical framework, we not only need to consider contention as such but also the constraints and opportunities for research on contention.Accordingly, in addition to the actual knowledge produced by researchers, we include the process of knowledge production as a distinct analytic dimension to be considered in the comparative study of digital LGBTQ+ contention.In a second step, we provide a scoping review of social media research on LGBTQ+ political contention in the digital age to take stock of existing literature.
Inspired by extant literature reviews on digital activism (e.g., Boulianne, 2015;Boulianne & Theocharis, 2020;Skoric, Zhu, Goh, & Pang, 2016), we draw a comprehensive sample of the relevant research from EBSCO, World of Science, and SCOPUS, resulting in 87 published and peerreviewed social science journal articles in English over the last 20 years.To analyze this literature, we determine categories for comparison, such as forms of political contention, regional focus, applied methodology, analyzed social media platforms, topics, or specific functions of social media.This categorization process allows us to take stock of the dominant foci and identify research gaps, limitations, and opportunities emerging in this evolving, highly interdisciplinary sub-field.Based on this scoping exercise, we identify a lack of diversity in terms of methodological approaches and regional focus in the literature characterized by single-case studies, primarily focused on US-based social media platforms and featuring a multitude of terms for social media-based activism.
We find that the unfulfilled potential of comparative research in this literature also lies in its limited regional scope in terms of how scientific knowledge on digital LGBTQ+ contention is produced (knowledge production) and in how people engage in digital LGBTQ+ contention (the produced knowledge about contention).In addition, we need more resource-intensive research based on cross-country and cross-platform comparisons, considering both the distinct political context and social media platforms as comparative dimensions.
By identifying the unfulfilled potential of comparative research and conceptualizing it alongside two axes of "produced knowledge" and "knowledge production," our paper contributes to building a solid foundation for analyzing digitallymediated political contention across different constellations and identifying existing knowledge gaps on digital LGBTQ+ contention.Furthermore, the paper increases visibility and advances the theoretical discussion on the digital LGBTQ+ contention by providing options for more systematic concept definitions and identifying methodological challenges.Finally, our paper also highlights inequalities and political constraints for researchers to varying degrees in the processes of global knowledge production.

An analytical framework for comparative research
Comparative research often serves as the basis for developing theories and hypotheses about social phenomena (e.g., Esser & Vliegenthart, 2017;Mancini & Hallin, 2012).A lack of comparative research can, therefore, be an indicator of the under-theorization of a field.The ultimate goal of comparison is going beyond the context of one case to explain "differences and similarities between objects of analysis and relations between objects against the backdrop of their contextual conditions" (Esser & Vliegenthart, 2017, p. 4).Comparative analysis is, therefore, ultimately about the theorizing of context (Mancini & Hallin, 2012, p. 515).
In this paper, we focus specifically on analyzing existing literature on digital LGBTQ+ contention to contribute to fostering comparative research in this field.We use the umbrella term LGBTQ+ to refer to lesbian, gay, bisexual, transgender, queer, or other non-straight, noncisgender identities or the diverse community of individuals with such identities. 1 Despite recent progress toward greater acceptance and equality for LGBTQ+ individuals in many parts of the world, discrimination against the community persists on different levels, ranging from laws and policies to hate crimes and microaggressions (Nadal, 2019).Experiencing multiple forms of discrimination leads to greater mental distress, especially among those who do not use social media to seek support in the community (Marciano & Antebi-Gruszka, 2022).
This article aims to assess the state of this field and highlight future avenues of comparative research to help build a solid theoretical foundation.First, we need to understand the constraints and opportunities of digital LGBTQ+ contention (first axis) as contextual dimensions.The contextual dimensions are (a) the demarcated political or cultural contexts that influence, incentivize, or constrain political activism in some way, as well as (b) the specific affordances of social media platforms that shape different forms of political contention.Second, we also take into account the two levels of action (second axis) on which we focus, namely (a) the process of knowledge production, i.e., the ways research on digital LGBTQ+ contention can be conducted, and (b) the "doing" of contention, i.e., engaging in contention.Our focus on the knowledge production process regarding LGBTQ+ political contention is a further, yet often ignored comparative layer because it considers the additional constraints on LGBTQ+ research through political persecution, economic resources, and the mental and physical abuses LGBTQ+ community members and their supporters face (e.g., Galpin, 2022).The final section summarizes our suggested framework based on these two axes and illustrates how they relate to each other, providing research avenues for future comparative research beyond political systems across contexts and platforms.

Two levels of constraints and opportunities
Based on Tilly (2008) and Tarrow (1996), we define political contention as diverse forms of collective activity enabled through opportunity structures on the part of claimants or their representatives, exercised through suitable repertoires, with an aim to interact with the object and bring about change.Based on this definition, political contention can take different forms depending on the context and available repertoires that can change over time.
Importantly, political contention is strongly context-dependent as the repertoires of contention are influenced by different contextual dimensions because contentious actors do not "work in a vacuum" (Meyer, 2017, 1).To develop our comparative framework, we focus on two contextual dimensions that can enable and/or constrain digital contention: the political context and the social media context.In this way, we go beyond the traditional focus of political opportunity structures on institutionalized politics, including other forms of opportunity structures, such as gendered opportunity structures and digital opportunity structures, as well as related constraints, the latter to which we turn further below: Regarding opportunities, the political context, first, includes traditional, institutionalized politics and laws that provide political opportunities for activists operating in it, thus the structures in which political contention occurs.Tilly (1978) recognized and identified many forms of contention -the "repertoires of contention" -such as protests, demonstrations, riots, or strikes, and noted that forms change and new ones appear responding to changing opportunity structures, thus highlighting the relationship of political opportunities and the repertoires of contention from which activists choose.Tilly's work helps assess how changes in institutionalized politics shape the availability of contention forms.For example, concerning LGBTQ+ contention, the Rainbow Maps (ILGA-Europe, 2023) provide comparative dimensions on the strictly institutionalized context from rights protection over civil liberties to intersectional issues such as asylum regulations.Second, following Tarrow (1996, p. 874), "collective activity on the part of claimants -or those who claim to represent them - [rely] at least in part on noninstitutional forms of interaction with elites, opponents, or the state", thus expanding our understanding of political opportunity structures beyond institutionalized politics.Third, we suggest going beyond the focus on "politics" in opportunity structures.Looking at gender struggles, McCammon, Campbell, Granberg, andMowery (2001, 2007) further added to the concept of political opportunity structures other forms of opportunity structures, such as "gendered opportunity structures".Arguing that in the case of the suffragette movement in the US, "the broad context in which those movements operate [must be considered], including political and other social dynamics that can affect movement success" (McCammon, Campbell, Granberg, & Mowery, 2001, p. 50).They showed how the success of the U.S. Women's Jury Movements cannot be explained by political opportunity structures alone but by a constellation of other structures, including gendered opportunity structures.Examples of gendered opportunity structures are the increasing number of women in professions such as politics, higher education, or the legal profession, a change in the relation between men and women that then led to other opportunity structures, such as discursive ones, for women to speak up (McCammon, Newman, Muse, & Terrell, 2007, p. 734).The point here is that political opportunity structures alone are insufficient to address the complexities of the broader political context.Indeed, they seem to be mostly understood as "haves" or "have nots".
So far, we have focused on opportunities.However, our understanding of the political context goes beyond this and includes the consideration of constraints.Depending on the scope of rights protection and civil liberties, these can also work as institutional constraints.For example, we consider the lack of civil liberties for LGBTQ+ people not only as a lack of an opportunity structure but as a direct constraint on people's options for making their voices heard.While political opportunities are taken for granted in democratic contexts, they might not similarly exist in authoritarian ones.They might also be marginal in democratic contexts, like the EU, where antidiscrimination laws and human rights are an element for identity construction but subverted by anti-gender legislation, especially on the national level (Slootmaeckers, 2020).
Furthermore, even in comparative indices like the Rainbow Map, the lived experiences of LGBTQ + people are not considered.Also, in democratic contexts, hostility to LGBTQ+ people is a severe constraint to being heard in public debates and reduces the emergence of political opportunity structures.For example, research on gender policy progress (Krizsán & Roggeband, 2018) suggests that "oppositional gendered structures'' need to be considered as an explaining factor for success or failure, thus allowing for the possibility to consider how constraints affect LGBTQ+ contention.In this sense, for our framework, we not only focus on whether political opportunity structures are present in a certain context but also consider the broader political context ranging from inclusive to hostile for LGBTQ+ people.Thus, this comparative dimension takes into account a variety of (non-) institutional constellations, which shape opportunities or work as constraints across different political systems.
Turning to the social media context, it has considerably changed and expanded the opportunities for political contention.In the same vein as gendered opportunity structures cannot be grasped by political opportunity structures (McCammon et al., 2001, p. 51), social media provides pro-and anti-LGBTQ+ activists with practical tools of contention, owing to the generally ungated, unedited access to public space which, for better or for worse, offers a broad range of mobilization tools and ways to communicate and connect.Regarding opportunities, we consider social media affordances as digital opportunity structures, which often function beyond the traditional arena where political opportunity structures are found.Social media has changed how people organize, mobilize and identify, affecting the forms of contentious actions (Bennett & Segerberg, 2013).However, there is still a lack of clarity about different forms of social media activism, which Theocharis, Boulianne, Koc-Michalska, and Bimber (2022, p. 5) refer to as a "classification problem," manifesting in various categorizations of digital activism.For example, based on a systematic literature review focusing on business and organization literature, George and Leidner (2019) have identified "ten representative digital activism activities: clicktivism, meta voicing, assertion, e-funding, political consumerism, digital petitions, botivism, data activism, exposure, and hacktivism." Nevertheless, social media affordances provide users with flexibility in how action is communicated and organized by offering tools that not only add to but even supersede traditional repertoires of contention (George & Leidner, 2019).Moreover, the increased availability of digital action repertoires has simultaneously strengthened and challenged social movement organizations' values while expanding activists' interactions and modes of engagement (Selander & Jarvenpaa, 2016).Indeed, social media have become a powerful tool for expressing political claims, participating in political discussions, and engaging in LGBTQ+ movements for visibility and equality in democratic systems (e.g., Becker & Copeland, 2016) and authoritarian systems (e.g., Yang, 2019).They are especially important for marginalized social groups, previously underrepresented or invisible in traditional media and political processes (see Gil de Zúñiga, Jung, & Valenzuela, 2012).
However, social media also create a favorable environment for anti-LGBTQ+ actors who discriminate and even commit hate crimes toward LGBTQ+ people (e.g., Galpin, 2022;Reichelmann et al., 2021), increasing the risk of psychopathological symptoms and increased suicidality (Haas et al., 2010).As opportunity structures for antigender mobilization, they work as constraints for pro-LGBTQ+ movements to promote inclusive and open societies.For digital LGBTQ+ activism, online hostility can quickly pose a powerful constraint, which needs to be considered when conducting comparative research.For example, hostile online environments against LGBTQ+ can be sourced by institutional political actors to promote support for anti-LGBTQ+ policies.These constraints are specific to marginalized communities, such as LGBTQ+ communities, in that their members become silenced as targets of hostility attacksand increase the chance of institutional discrimination.Furthermore, activism is under constant risk by regimes to crack down on social media through monitoring, censorship, and blackouts (e.g., Comunello & Anzera, 2012).Recent and particularly poignant examples are the protests in Iran for women's rights and security, as well as protests in several Chinese cities against prolonged COVID-19 lockdowns.
Overall, an appropriate comparative framework needs to be general enough to consider digital activism in both opportunities and constraints.However, it also needs to be precise enough in spelling out what kinds of opportunities and constraints shape digitally mediated LGBTQ+ activism across political and social media contexts.We acknowledge that the political context is not always neatly separable from the social media context, nor should it be.As discussed above, despite the contextual dimensions for comparison, i.e., political and social media, presenting at times overlapping arenas, the social media context has left a distinct mark on the repertoires of contentious politics.Where digital opportunity structures and constraints exist, both pro-and anti-LGBTQ+ activists have novel forms of activism at their disposal.This creates a diverse set of opportunities and constraints across different political systems, platforms, communities, regions or cultures.Therefore, the political and social media contexts, with their opportunities and constraints, serve as separate comparative context dimensions, influencing both how digital LGBTQ+ contention can take place and how it can be researched, as we discuss in the next section.

Two levels of action
The second axis of our comparative framework concerns two levels of action.The first, more obvious one, is the dimension of the produced knowledge about contention, i.e, the focus on how contention is "done:" As we have discussed above, contention interacts with the constraints and opportunities of the political and social media contexts, and this interaction shapes how people can engage in digital contention.In this section, however, we add another level of action, the dimension of knowledge production.We described this as the interaction between the constraints and opportunities of the political and social media contexts with the ways in which research about digital LGBTQ+ contention can be conducted.In other words, we argue that for a comprehensive analytical comparative framework, we need to consider how researchers on digital LGBTQ+ contention are enabled and challenged in their processes of uncovering dynamics, patterns, or explanations of political contention embedded in such contexts.
Knowledge production is challenged or enabled by different political contexts where academic freedom and institutional autonomy of universities cannot be taken for granted (e.g., Hünler, 2022).Depending on the specific characteristics of the political climate, for example, research could be unethical if it helps uncover identities and becomes complicit in state oppression tactics or increases the chances of online harassment.Therefore, for researchers working on digital LGBTQ+ contention, their work is associated with individual risks, which often has an influence on what to research and what not -and so on the overall patterns of knowledge production.But also in "safer places" where university research is conducted more freely, the massive amount of online harassment, trolling and worse does not only target people engaged in activism but also those researching it (Galpin, 2022).
Members of the LGBTQ+ community are still marginalized and even legally persecuted in many countries (OHCHR, 2023).Such problems contribute to increased risks of legal or political persecution for researchers on LGBTQ+ digital activism in these contexts.For example, the expected high number of studies in democratic contexts does not have to imply a higher research interest in the region but more likely points to the consequences of becoming visible as a researcher on LGBTQ+ activism and the risks that come with it in contexts where living or even supporting LGBTQ+ identities comes with persecution or even execution.Therefore, we need to consider political opportunities and constraints substantially concerning both LGBTQ+ digital activism and the process of knowledge production on digital LGBTQ+ contention.
Concerning the aspect of social media, a comparative framework also needs to consider the political economy of social media platforms.First, research resources, political persecution, and access to social media enable and constrain research on LGBTQ+ issues particularly.Furthermore, academic knowledge production "displays systematic boundaries and mechanisms of exclusion across and between countries and regions of the world" (Collyer, 2016, p. 57).For example, with increasing technological capacities, some authoritarian states regularly engage in technical internet and social media shutdowns, a practice that distinguishes them from the practices of liberal-democratic states where legal bans are more typical (e.g., Mare, 2020).Second, monopolization and market concentration further constraints on researchers in the global South of knowledge production since most publishers are located in the global North (Collyer, 2016, p. 61).
These unequal structures of academic knowledge production through publishing date back to pre-social media times (Canagarajah, 1996).Social media research is still challenging given that social media data, while public, belong to private enterprises such as Meta, which are more likely to provide/allow access to their data if the respective research engages with a company-defined issue or topic.Other examples include the recent changes in the accessibility of Twitter (now X) data or the blocking of TikTok in several countries in areas of political-public arenas due to data privacy issues.

Introducing an analytical framework for comparative research
Drawing on the above-discussed dimensions, Table 1 presents our comparative analytical framework organized along two axes.The first axis describes the constraints and opportunities as contextual dimensions: the political context and the social media platforms context.The second axis refers to the types of action: the dimensions of knowledge production and produced knowledge about contention.Considering how both political and social media contexts not only influence how contention takes place but also how contention can be researched provides a more inclusive and comprehensive approach for comparative social media research, especially for LGBTQ+ contention, which unfortunately still comes with considerable individual risks for both LGBTQ+ people engaged in contention as well as researchers working on LGBTQ+ contention, including the possibility that these roles overlap.

Search strategy and selection criteria
To identify the relevant literature, we searched for peer-reviewed articles published in academic journals from January 2000 to May 2022 that study political contention revolving around LGBTQ+ issues on social media.We chose January 2000 as the starting point to include scholarship concerning "the early years" in the history of social media (boyd & Ellison, 2007, p. 214).To our knowledge, one of the first articles on social media was published in 2003 (Adamic, Buyukkokten, & Adar, 2003); hence, enlarging our timespan to the year 2000 was deemed beneficial to capture any possible earlier work.Similarly, we chose May 2022 as the endpoint to cover the most recent articles relevant to our study.encountered limitations regarding resources and especially language skills, we settled for publications in English as common denominators.We address this issue in our findings section and the conclusion regarding knowledge production.We used the keyword "LGBT*" to capture related literature; using the asterisk ensured that all term extensions were found, too.Accordingly, manuscripts mentioning the term at least once were sampled (including related terms taken into account by algorithms), including research on LGBTQ+ community activism, activism contesting the existence and rights of LGBTQ+ people, as well as campaigns of third parties.We decided to include only this term in our search due to the extremely diverse and fast-changing terminology in this area (see, e.g., Lee, Ylioja, Lackey, & Huy, 2016, for a discussion of this issue rooted in health research).Including this admittedly crude proxy in a catch-all approach ensured some continuity and coverage of LGBTQ+ in a general sense.It is also true that some publications not referring to LGBT* at least once in the title, abstract, or keywords (depending on databases' algorithms) are left out, e.g., literature on specific topics such as transgender, gay, or lesbian.In that sense, our review aims to provide a rather general assessment of the field without diving deeper into specific branches.Our study should, therefore, be understood as a first step to better understanding the state of the art and assessing its potential regarding comparative research.
We combined "LGBT*" with terms that capture different forms of political contention according to our definition, such as "protest", "activism", "social movement", and so on.These were used together with terms that describe social media or correspond to various social media sites such as "social media", "SNS", "Facebook", "Weibo", and more.We defined the search mode as Boolean or phrase and restricted our search to the articles' abstracts.The complete search strings and the detailed search strategy are presented in Appendix A.
Removing exact duplicates from the search results yielded N = 87 potentially relevant journal articles.Out of these, 86 articles were successfully retrieved through Vienna University Library, and one was provided by the author after contacting them on ResearchGate.The abstracts of all articles were then screened for inclusion independently by all three authors.The articles were regarded as relevant if they empirically or theoretically examined political contention revolving around LGBTQ + issues and involving social media.We also screened the full texts of the articles deemed irrelevant to ensure their adequate exclusion.Any disagreement was resolved by consensus.As a result, we excluded eight and included 79 articles in the corpus.

Data collection process and items
We constructed and refined the data extraction form based on the research purpose and information in the screened abstracts and articles.From each article, we obtained data across four categories: meta-level data, methodology, object and findings, and comparison.The meta-level data included the year of publication, title, journal, field, DOI number, and name(s), country, and university affiliation(s) of the author(s).The methodology level included information on data collection, analysis, method type, sample, and timespan.The object and findings dimension comprised the form of political contention (e.g., protest, campaigning), what function social media was ascribed (e.g., mobilization, information diffusion), locality, social media platform(s), political ideology (i.e., pro or anti-LGBTQ+ contention), group(s) or actor(s), and issue in focus.In addition, we also coded the extent to which the article focused on LGBTQ+-related aspects and political contention (scope).Finally, the comparison level included any comparative dimensions present in the article (if they were present), i.e., a comparison between political contention forms, localities, languages, platforms, modalities, over time, ideologies, involved groups, or methodologies.A detailed description of each variable is presented in Appendix B.
First, an intercoder-reliability check was conducted on a random sample of 10 articles between all three authors.We calculated the percentage agreement across three dimensions, excluding the meta-level data, since the databases provided it.The percentage agreement for the scope of focus on LGBTQ+ in the article was 0.85, for the scope of focus on social media at 0.78, and for the presence of a comparison in the analysis at 1.We discussed the results extensively and identified valid coding.Subsequently, we worked independently on extracting data from 33-34 articles each.These articles included independent studies from the corpus and a random sample of 12 repeating articles for the second reliability test, which yielded very similar results: The percentage agreement for the scope of focus on LGBTQ+ was 0.89, for the scope of focus on social media at 0.75, and for the presence of a comparison in the analysis at 0.86.Again, results were discussed extensively; given the comparably small scale of the coding, the variable on the scope of focus on social media was rechecked to ensure reliability.For a clearer picture, the two variables were recoded to "low" (1, 2), "medium" (3), and "high" (4, 5).
For the analysis, we relied on our sample's descriptive statistics and frequency distributions, as well as on more qualitative category-building for the string variables.In this way, we were able to investigate our sample most accurately to support the analytical conceptualization.

Knowledge production
Starting with some basic information on the dataset, we included 79 articles in the analysis; 59 articles were coded as containing empirical research; 9 articles were labeled as providing a literature review or a historical study; 8 articles were found to provide theoretical or conceptual considerations.Of the 59 empirical analyses, the majority (61.4%) was based on a qualitative methodology; 22.8% used quantitative methods, and 15.8% relied on a mixed methods design.The interest in the role of social media in the political contention of LGBTQ+ issues has clearly increased while not linearly so.Some of the peaks in the number of publications, such as in 2016, however, could also be an issue of a publication backlog.Therefore, we can only interpret individual peaks as an overall increase of interest in the field over the entire time frame.Figure 1 shows the distribution of articles included in the analysis over time and by methodology.
Qualitative methodologies often relied on interviews, some form of discourse analysis based on inductive coding or grounded theory, or took an ethnographic approach.Some of these studies used social media platforms as recruitment channels, i.e., by disseminating a call or an advertisement for participating in an online survey.Studies using quantitative methodologies were based on surveys, experiments, and quantitative content analysis, including automated content analysis.Obtained data were then statistically analyzed.Only nine articles employed a mixed methods design.Mixed methods studies often used (descriptive) statistics derived from surveys to complement qualitative analyses or engaged in multimodal analyses combining quantitative and qualitative content or text analysis.For example, La Rocca and Rinaldi ( 2020) used text and visual analysis to approach the multimodal aspect of social media analysis.Phadke and Mitra (2021) were among the few using a computational method and combining it with in-depth qualitative text analysis.
Generally, given the few mixed-method approaches, these examples remain exceptions, suggesting that empirical research on digital LGBTQ+ contention published in English is still in its infancy when it comes to bridging quantitative and qualitative methods.Furthermore, singlecase studies dominate this research field.There is consequently a wealth of empirical material to be used for hypothesis building and a need for large datasets.
Most of the journals in which articles were published were listed 2 in the category Social Sciences (71), followed by Arts & Humanities (26), Psychology (15), Medicine (13), Business, Management & Accounting (9), Computer Science (7), Economics (2), Earth and Planetary Sciences (1), Biochemistry, Genetics and Molecular Biology (1), Chemical Engineering (1), Immunology and Microbiology (1) and Decision Sciences (1).This overview mirrors a great diversity of perspectives from which research has addressed the role of social media in digital LGBTQ+ contention, ranging from the role of social media in political activism, mental health, or health education for marginalized groups and high-risk groups.
The researchers involved in authoring studies were, in most cases, affiliated with a university in the United States (37), followed by the United Kingdom (8), and Australia (7).While our focus on English-written publications creates a natural bias, these numbers also highlight the regional monopolization of academic publishing in the global, anglophone North and likely marginalization (Canagarajah, 1996).Monopolization of academic publishing in the global North favors Englishwritten publications (Collyer, 2016, p. 64).Only five articles are written by researchers based in different countries.This pattern suggests that collaborative, cross-country research in our sample is low.
Large funders, such as the EU, only in the last years have explicitly made gender equality and inclusion a criterion for proposals, either in research design or institutional setup across beneficiaries: The EU Horizon 2020 program was the first framework program defining gender equality as a criterion in research and research design (European Commission, 2023), without, however, explicitly mentioning LGBTQ+ equality.Resources, paired with difficulties in accessing data and high personal risks for researchers based in illiberal contexts, might explain the lower number of articles on LGBTQ+ political contention coming from non-Western institutions.The identified patterns also highlight the difficulty of comparative research across languages and underscore the inequalities of countries in terms of the concentration of the publishing market, the commodification, and the standardization of publishing (Collyer, 2016, pp. 60-61).

Produced knowledge
Regarding the actual contents of the research reviewed, almost 50% of the articles were coded as having a strong focus on social media and LGBTQ+ issues (38 of 79), overall supporting the validity of our search strategy.LGBTQ+ focus was generally more strongly emphasized (63 of 79) than social media (46 of 79), suggesting that the research community behind these publications was more interested in the specific LGBTQ+ issue and how social media influenced it than the other way around, thus comprehending social media as a means to the specific ends of the LGBTQ+ community.In terms of the position toward LGBTQ+ issues, 74.7% (n = 59) of the articles investigate pro-LGBTQ+ contention, 13.9% (n = 14) focus on neutral or ambivalent issues, and 10.1% (n = 8) on anti-LGBTQ+ movements.One article (1.3%) could not be classified.Our sample, therefore, encompasses primarily research on pro-LGBTQ+ contention.
As shown in Table 2, the regional focus of our sample (n = 79) concentrates strongly on the US (30.4%), followed by China (8.9%), unspecified (8.9%), and Australia (7.6%).Therefore, a large share of research focuses on locations where English is the dominant language, especially in the US, where many social media platforms are headquartered.The attention to China can be explained by especially repressive legal and social challenges for numerous LGBTQ+ individuals, which leads them to resist and fight for their rights in innovative ways with the help of social media (e.g., Chen, 2020;Shaw & Zhang, 2018).Furthermore, several large social media platforms are specific to the Chinese context (e.g., Sina Weibo and WeChat).
According to our definition, political contention can take diverse forms depending on the context and available repertoires that change over time.Correspondingly, our sampled articles focused on various forms of LGBTQ+ political contention in the digital age.To understand them more deeply, we created two categories that recorded the modality of political contention (online, offline, or mixed) and the exact wording used to describe it.As a result, we found 37 articles (47%) focusing exclusively on online forms, 28 articles (35%) studying mixed forms, and nine articles (11%) paying attention to solely offline forms of political contention.Only five studies from the sample did not focus on a specific political contention type.They instead explored other topics, such as deliberation, populism, and contraception awareness, or included activists from different movements as their participants.Although peripherally touching upon or implying contention, they did not focus on it per se.
Among the articles that did pay particular attention to political contention, we found diverse wordings used to describe its forms (n = 77).The most popular were social movement (n = 6), protest (n = 6), activism (n = 5), hashtag activism (n = 5), online campaign (n = 5), online activism (n = 4), and petition (n = 4).Less popular terminology that appeared twice in the sampled articles were queer activism, connective action, digital activism, occupation, citizen journalism, performance, campaign, and march.The remaining 62 phrasings occurred only once as a form of political contention in our sampled articles.All the terminology can be found in Figure 2 and the online repository.
Often, these wordings have a similar meaning, e.g., online campaign, hashtag activism, social media campaign, online social activism, and online activism, showing a need for more consistency and agreement in conceptualizing identical forms of political contention across different fields and epistemologies.It is clear, however, that repertoires of contention might differ due to specific sociocultural contexts and, therefore, should be named accordingly.For example, such forms of contention can include live mass gatherings, promotion and distribution of banned queer documentaries, activism with Chinese characteristics, or antisurveillance counter-conduct studied by different authors interested in political contention in repressive regimes.Such terminology reflects the available repertoires of contention and provides possibilities for comparison of instrumentally similar but contextually distinct forms of LGBTQ+ political contention in the digital age.
It is also important to look at the social media platforms studied the most regarding LGBTQ+ political contention.Related to the location, the social media platforms analyzed have a strong US focus (see Table 3).US-based social media platforms make up the majority (53%) of articles using or analyzing Facebook (23.0%),Twitter (17.5%), and YouTube (12.7%), with Instagram and others receiving less attention.In contrast, China-based social media platforms take up only around a fifth of the share (Sina Weibo 5.6%, WeChat 3.2%, Tencent QQ 0.8%, Youku 0.8%, Tudou 0.8%, Zhihu 1.6%).There is consequently a bias in the English-language literature towards US-based social media platforms.Hence, more research on less dominant platforms is needed to understand forms of digital LGBTQ+ political contention in diverse locations.
For the same purpose, it is essential to look into particular practices of claim-making via social media and with the help of social media.Articles in our sample were included in the analyses if they dealt with social media to at least some extent.As a result, in four cases, social media was used by researchers to contact and recruit LGBTQ+ activists as participants.However, in most cases (n = 72), the articles reflected on the purpose and meaning of social media for LGBTQ+ individuals.Most researchers who study LGBTQ+ political contention focused on enabling functions of social media contributing to the organization of movements, counterpublics formation, or awareness building.For example, Charles, Khursheed, Ferrer-Fons, and Allaste (2018) studied cases where social media was used for mobilization and collective identity formation, organization of protests, and generating support for campaigns and actions.Some researchers focused on the importance of social media as an alternative source of information  that can be used for resistance through shaping news narratives (Tortajada, Willem, Platero Mendez, & Araüna, 2021) and building counterpublics where counter-narratives to mainstream news can be spread and discussed (Nummi, Jennings, & Feagin, 2019).Other researchers paid more attention to the harmful activism where social media use aimed at the harassment of marginalized groups, trolling, and surveillance (Lingel, 2021), or spreading extremist propaganda, and recruiting new members to attend anti-LGBTQ+ rallies (Phadke & Mitra, 2021).Discrimination online and in general makes social media also a venue for anti-bullying campaigns used for raising awareness and suicide prevention (e.g., Grzanka & Mann, 2014;Jones, 2015).All these practices show the enriching potential of social media for emerging repertoires of contention in the digital age, which can also be approached comparatively.
Finally, we also coded if papers had a comparative dimension and to what extent these aligned with our analytical framework.All in all, around 20% of the articles in our sample (n = 14) had a comparative element that can be classified as constraints and opportunities for digital LGBTQ+ contention, i.e., under "Produced Knowledge" (see Table 1).Nine studies compared activist groups or movements.For example, Bonilla and Tillery (2020) compared nationalist, feminist, and LGBTQ+ frames in the Black Lives Matter movement.Kenix and Abikanlu (2019) compared Facebook messages of NGOs for LGBTQ+ refugees in the Sub-Saharan region.Other articles compared actions across ideologies (pro-and anti-LGBTQ+) (e.g., Nekmat & Ismail, 2019) or across locations (e.g., Pepin-Neff & Wynter, 2020).Some applied several comparative dimensions, like Duguay (2016), focusing on the modality and different platform affordances of Vine and Instagram.Overall, the concept of identity was often central to these studies; the influence of context was implicitly acknowledged (e.g., Charles et al., 2018) but not made explicit or distinguished by level.

The unfulfilled potential for comparative research
Based on the findings of our analysis, we identify several gaps in the English literature on digital LGBTQ+ contention issues, which point to a so far unfulfilled potential for comparative research.Starting with the dimension of knowledge production, we find a general lack of methodological diversity across the "quantitative-qualitative divide:" Our findings show that mixed-method designs are rare but particularly attractive for comparative research designs.Furthermore, quantitative studies apply systematic comparisons more often than qualitative studies.Finally, although there is a wealth of in-depth empirical data from qualitative research that could be used for hypothesis-and theory-building, these data have not yet been exploited by comparative research, suggesting promising yet unexploited evidence available via more pluralistic and comparative research designs.
Moreover, our analysis points to a low number of research collaborations across countries.We consider this an unfulfilled potential for comparative research because cross-country research collaborations enable the pooling of (language) skills and opportunities for access to nationally regulated data, which are crucial for designing valid and robust comparative research.One explanation we have offered is that funding by larger institutions (e.g., the EU) has slowly but surely made the implementation of inclusivity strategies in staff and research excellence a requirement (e.g., the LGBTQ+ Equality Strategy 2020-2025).Systematic, rigorous comparison is resourceintensive, and more funding can help alleviate this challenge.
Other reasons are more obvious and relate to constraints and opportunities: In many countries (including some European countries like Hungary), involvement in LGBTQ+ issues, also research, is not only often publicly shamed but linked to high personal risk of the individual researcher (LaSala, Jenkins, Wheeler, & Fredriksen-Goldsen, 2008), forcing, e.g., researchers in China to develop strategies for engaging in social media activism nonetheless through, e.g., self-censorship (Cui, 2022).The almost complete absence of specific country contexts in our findings, such as Hungary, therefore, highlights the importance of our comparative framework to consider political and social media-related constraints, such as persecution of LGBTQ+ involvement or online hostility, having a silencing effect on research (see Galpin, 2022).Such constraints can directly limit understanding of LGBTQ+ political contention -knowledge that is crucial in promoting human/civil rights for the community.Likewise, we find a lack of empirical data for cross-country and cross-platform comparisons, which could be related to lower resources for LGBTQ+-focused research in certain contexts or higher risks.However, as our analysis suggests, there is a rich body of in-depth single case study research, which can serve as a starting point for comparative research.There is consequently unexploited material and an opportunity to understand better how the political context shapes political contention and activists' practices in the field of LGTBQ+ on social media.
Furthermore, our analysis confirms the difficulty of conceptualizing political contention since we do not find agreed-upon definitions in our sample.This circumstance might be owed to the various affordances social media platforms have and the dynamics with which new social media platforms emerge and gain popularity, e.g., TikTok, which does not occur in our sample.The categorization or translation of the diverse social media affordances into repertoires of contention would, however, allow comparison beyond national/cultural contexts.Therefore, there is considerable potential in research contributing to conceptual clarity about new forms of political contention.It can provide novel analytical categories in the field and contribute to the understanding of transnational mobilization for both pro-and anti-LGBTQ+ movements.
Regarding limitations of our study, we excluded conference proceedings, master's and doctoral theses, non-peer-reviewed books or book chapters, and unpublished studies that can result in a bias toward qualitative articles with clear and striking findings and quantitative studies with significant effects.In doing so, we followed standard procedures of how to sample for literature reviews.Furthermore, indexing of journals negatively affects the diversity of publishing and contributes to maintaining the inequality in knowledge production between the global North and the global South (Collyer, 2016) by excluding journals not indexed but not automatically of lesser quality.In this sense, we need to acknowledge that our literature review is limited in the way the knowledge production of the global North is, as dominant as it may be.
Our research is also limited by our choice of keywords for sampling.Although our keyword list is extensive, it is not comprehensive and with a tendency toward peer-reviewed, published social science articles.Future research should extend the list of keywords.In particular, although we have discussed this list with experts actively involved in the LGBTQ+ community, we might not have done justice to the diversity of descriptive terms.Our compromise was to keep the focus narrow and explicitly focused on the term "LGBTQ+."We see this as a small step only and hope that our keyword list is helpful for other researchers to build on.Furthermore, our focus on English articles -as the language we are most competent with -is a commonly used yet limited approach because it excludes knowledge communities that do not generally publish in English, despite its status as Academic "lingua franca."Our analysis is a first step that, we hope, can also provide a framework for non-English speaking knowledge communities with specific insights into different regions to be modified, built on, or criticized.
Despite these limitations, our framework offers a first step to consider the different structures of influence visible in both digital forms of political contention and the way it is currently researched in English-written publications.Our analytical framework, while developed for LGBTQ+-related research, can be applied to other research subjects as well and thus makes a general contribution to the field.We have identified various gaps in this literature, deriving from resources, country/region-specific expertise, political constraints, and risks, which are probable stumbling blocks for systematic data collection, concept building, and hypothesis testing, all of which are helpful and often needed in comparative research.We could also further confirm the harmful influences of the current inequalities of academic publishing between the global North and the global South.Besides social and political support, we require greater attention to digital contention for LGBTQ+ rights when it comes to providing resources through unconventional forms of funding and stratification beyond access through the global North.

Notes
1. Please see our specified keyword search for sampling articles in the section on methodology.2. Journal categories were determined according to scimagojr.com.Many journals were categorized in more than one field.The numbers presented here are based on the mention of a category for any given journal.3. Articles can refer to more than one location.The categories are, therefore, not mutually exclusive.Figure 2, however, includes only one location/country.4. Articles can refer to more than one social media platform.The categories are therefore not mutually exclusive.

Figure 1 .
Figure 1.Number of articles included in the analysis over time by methodology.

Table 1 .
Dimensions of comparison for researching/engaging in digital LGBTQ+ contention.
Interaction of social media affordances with the process of knowledge production, e.g., platform regulations, access to data, multimodality of social media, resources for analysis and analysis skills Interaction of social media affordances with the constraints and opportunities for LGBTQ+ contention, e.g., individual activist's range of actions to mobilize, diversity of affordances, but also risks of suppression, discrimination, and hate crimes due to the same affordances

Table 2 .
Share of countries/regions focused on in the sample 3 .

Table 3 .
Share of social media platforms used/referred to in the articles 4 .