Right Topic, Right Source? Source Diversity and Balance in Right-Wing Alternative News Content Across Topics

ABSTRACT This article investigates how the hybrid nature of right-wing alternative news media striving for journalistic legitimacy and partisan credibility plays out on source and topical diversity and balance in article content. The article draws on a sample of 1000 randomly selected articles published by 20 right-wing alternative online news media from six countries (the US, the UK, Germany, Austria, Denmark, and Sweden) from March 2019 to February 2020 (i.e., in “routine” pre-COVID-19 times). The results show that most of the alternative media outlets in the sample cover relatively broad topical spectra. More specifically, US and UK media primarily focus on politics and policy issues, whereas Scandinavian media are more heavily geared toward societal issues and crime coverage. Overall, right-wing alternative news content is characterized by a variety of partisan and non-partisan sources. However, core partisan topic areas, such as politics and mass media, are more likely to include partisan and especially right-wing sources. Often, with respect to these topics, right-wing sources are evaluated positively, and left-wing sources are evaluated negatively. Finally, right-wing and non-right-wing sources often appear in separate articles rather than in direct confrontation.


Introduction
Right-wing alternative news sites constitute "hybrid 'infopolitical' organizations" (Yang 2020): organizations with fluid repertoires, that can function as political organizations in one situation and as news organizations in another.In doing so, they transgress the boundaries between journalism and activism, merging journalistic logics and a mimicry of legacy media's practices and routines with outright partisanship and political movement orientation (Atkinson et al. 2021;Mayerhöffer and Heft 2022).These characteristics affect not only their self-understanding and self-presentation but also their published content, which has been shown to feature mainstream media criticism, anti-elitism, and other partisan themes (Schwaiger 2021).Consequently, studies on alternative media's referencing practices have so far predominantly analyzed these practices in relation to specific partisan topics, such as migration and social affairs (Buyens and van Aelst 2021;Freudenthaler and Wessler 2022), media criticism (Figenschou and Ihlebaek 2019), and truth claims (Schwaiger and Eisenegger 2021).
While such studies are insightful, they are less suited to answer the question of how the hybrid nature of alternative news media outlets, striving for both (quasi-)journalistic legitimacy and partisan credibility, plays out on the nature and variety of their referenced sources across different topics in news coverage.While these news sites' political ideologies may play out most strongly in their referencing practices for core partisan themes, other topics may be more conducive to conveying an image of more normalized and professionalized news coverage by adhering to established journalistic norms of a diverse and balanced inclusion of sources.To gain a better understanding of how the hybrid nature of right-wing alternative news media affects their news output beyond those topics where most overt partisanship can be expected, we study their referencing practices across different topics and assess these against an expectation of a varied and balanced inclusion of sources, and thus against established journalistic norms.We understand diversity as the presence of different types of sources and views in news content, whereas balance denotes the question of how different sources and views are presented in relation to each other.We therefore assess the source composition on the aggregate level of overall alternative news output, as well as on the level of the individual article.
The question of diversity and balance in alternative news coverage is by no means only of relevance for right-wing partisan news outlets.Previous research has however indicated that partisan referencing practices are particularly prevalent in right-wing alternative news coverage, whereas left-wing alternative media have been shown to focus particularly on the inclusion of civil society actors as sources (Buyens and van Aelst 2021;Freudenthaler and Wessler 2022).Therefore, this article studies diversity and balance in right-wing alternative news media's referencing practices through a crosstopical perspective, drawing on a sample of 1000 articles published by a diverse set of 20 right-wing alternative news outlets from six countries (the US, the UK, Germany, Austria, Denmark, and Sweden).We understand (right-wing) alternative news media as media that represent "a proclaimed and/or (self-)perceived corrective, opposing the overall tendency of public discourse emanating from what is perceived as the dominant mainstream media in a given system" (Holt, Figenschou, and Frischlich 2019, 862).As hyperpartisan actors, these media exhibit core characteristics of the political far-right, such as nativism, authoritarianism, and populism (Mudde 2007).This article contributes to a growing body of literature on the degree and nature of the potential mainstreaming of alternative news media that centers on the question of how these media mimic established journalistic practices to convey the appearance of "real" journalism while maintaining credibility as a (partisan) alternative to established news media.

State of the Art: Topical and Source Diversity in Alternative News Content
Research on the content of alternative news media-and, specifically, on the topics they cover-is scarce but emerging (Müller and Freudenthaler 2022).Some studies have examined the topics covered by alternative news media in specific contexts, such as on individual platforms (Knüpfer, Schwemmer, and Heft 2023) or during events like national elections (Kaiser, Rauchfleisch, and Bourassa 2020;Sandberg and Ihlebaek 2019), the refugee crisis (von Nordheim, Müller, and Scheppe 2019), or the COVID-19 pandemic (Boberg et al. 2020).Other studies have focused on these outlets' discursive strategies and frame repertoires, again with a focus on distinct topics or limited to particular outlets (Klawier, Prochazka, and Schweiger 2022;Ylä-Anttila, Bauvois, and Pyrhönen 2019).A third set of studies has delved into specific traits of alternative media content, such as media criticism (Cushion, McDowell-Naylor, and Thomas 2021;Figenschou and Ihlebaek 2019;Holt 2019;Roberts and Wahl-Jorgensen 2020), emotionality (Ihlebaek and Holter 2021;Tuomola and Wahl-Jorgensen 2022), and the recontextualization of cited content (Haanshuus and Ihlebaek 2021).These studies have provided important insights into what constitutes alternative news content today but have not addressed the source and topical diversity of right-wing alternative news media in a more general sense.
Research that approached alternative news media's topical diversity across the entire news output has pointed to significant heterogeneity between outlets.Many outlets focus on a wide range of topics, with a common denominator being their vast coverage of crime and immigration-related issues.In a cross-country study of 70 right-wing news sites, Heft et al. (2020) found that these sites represented variations of "alternatives" manifested, for example, in different topic categories, funding strategies, and differences in organizational transparency.However, this study captured topic categories by analyzing website sections and tags and not their actual articles.The study implied that journalistic normalization and partisan legitimization may vary not only across outlets but also concerning the topics at stake.Müller and Freudenthaler (2022) divided nine Germanlanguage outlets into two groups based on the topics they cover.The first group resembled traditional news media and covered diverse topics, including sports and business, whereas the second group focused on core right-wing populist topics, such as Islam, immigration, and migrant crime.Similarly, Mayerhöffer (2021) found that Danish right-wing alternative media could be divided into outlets focusing predominantly on core right-wing topics, such as migrant crime and critical coverage of left-wing politicians, and outlets covering a broad topical range, which is achieved through stories citing other news articles and the reissuing of press releases.In a study of nine American right-wing outlets, Kaiser, Rauchfleisch, and Bourassa (2020) showed that, although these outlets were somewhat similar in terms of their coverage of abortion, immigration, and Islam, the most far right of them differed in the degree to which they focused on other topics as well.Nygaard's (2019) qualitative analysis of Scandinavian alternative media outlets showed that while each had its distinct thematic focus, such as politics, integration, or economics, they resembled each other in terms of the substantial weight they gave to stories about crime committed by migrants.
Regarding the use of sources and hyperlinks by right-wing alternative news media, studies have similarly found variety between sites.Overall, alternative news media frequently refer to a wide range of different actors in their coverage (e.g., Frischlich, Klapproth, and Brinkschulte 2020;Mayerhöffer and Heft 2022).In some instances, these media have even been found to include greater numbers of different actors than their mainstream counterparts, though this could be attributed to differences in article lengths (Buyens and van Aelst 2021).Studies have shown that other media make up a substantial share of sources in right-wing alternative news media (Mayerhöffer and Heft 2022).More specifically, traditional mass media appear as essential reference points for alternative news media's reporting (Haller and Holt 2019;Heft et al. 2021;Kaiser, Rauchfleisch, and Bourassa 2020), which is seen as an attempt to mimic the citation practices of legacy media (Atkinson et al. 2021;Ryfe, Mensing, and Kelley 2016).Besides media sources, Buyens and van Aelst (2021) showed that left-wing alternative news media include more civil society actors, whereas right-wing alternative media give more room to right-wing politicians and parties.
To date, only one study has analyzed alternative media's topical and source diversity in conjunction.Comparing different types of alternative news outlets to mainstream outlets, Freudenthaler and Wessler (2022) found only small differences regarding speaker diversity but found higher topic diversity in mainstream media.Whereas some alternative media follow the topical agenda of the mainstream media, others feature more independent agendas.However, Freudenthaler and Wessler (2022) did not conduct a cross-topical analysis but focused on articles concerning migration and refugees.Moreover, they did not analyze whether the composition of speakers varied between individual (sub)topics.
Despite the growing body of literature addressing topic and source diversity in alternative news media content (Ihlebaek et al. 2022), we still have little knowledge about whether alternative news media use a particular set of sources to cover particular topics.Such a perspective not only contributes to a deeper understanding of both topic and source diversity in alternative news content but also sheds further light on the hybridity of alternative news media.Rather than investigating the hybrid profile of specific news outlets, we focus on how the aim to strive for journalistic legitimacy and partisan credibility plays out differently, depending on which topics are covered.

Theoretical Framework: Journalistic Diversity and Balance
In the making of news content, journalists consistently follow a variety of regulative rules that, in twentieth and twenty-first-century (Western) professional journalism, have been defined by (ritualized) adherence to objectivity (Anderson and Schudson 2019).Moreover, this adherence has been especially manifest in attempts to provide a balanced account of different viewpoints in news coverage (Entman 1989).Schudson (2001) has dubbed journalistic objectivity a moral ideal stemming from needs for social cohesion (e.g., defining what is [not] journalism) as well as needs for social control (e.g., passing on journalistic norms to new generations).Guided by this moral ideal, journalists have strived to "represent fairly each leading side" in their reporting (Schudson 2001, 150) to show that they avoid taking sides.Along the same lines, Tuchman (1972, 666) has explained journalistic objectivity in news content as journalists showing both sides of the story and a "diversity of views" in the news.However, according to Tuchman, this adherence is explainable not as an ideal but a strategic ritual employed to deflect potential criticism and claim professionalism.Likewise, Ryfe (2006) has explained the reason behind journalists' adherence to news rules, such as writing balanced stories, with their need to be recognized as journalists by themselves and their colleagues.Thus, regardless of whether displaying diversity and balance in news content is based on a moral ideal of objectivity, this adherence is a cornerstone of Western professional journalism and is linked to what "the public" understands as trustworthy and credible journalism (Knudsen et al. 2022).
In broad terms, we can understand diversity as the presence of all relevant differences in the news reflecting society as a whole (Loecherbach et al. 2020;McQuail 1992) and balance as the equal representation of the conflicting sides of a dispute (Benham 2020;Entman 1989).Thus, diversity has to do with the presence of different sources and views in news content, while balance refers to how these views are treated.In reality, a perfect diversity of and balance between sources is rarely achieved or even strived for, given the limited time and financial resources that news organizations must operate under, as well as the fact that it is difficult to precisely define how diversity and balance should be achieved in practice (Schudson 2003;Stavitsky and Dvorkin 2008).
If we approach journalistic diversity and balance as a ritualized practice to establish journalistic credibility and showcase professionalism rather than as a (mere) expression of the normative ideal of journalistic objectivity, the study of news diversity also becomes relevant for news media characterized by explicit non-adherence to objectivity on normative grounds.Research on alternative media and expectations of the diversity they bring about has changed considerably throughout the years.The alternative media concept was initially linked to left-wing outlets typically seen as progressive, democratizing challengers of a hegemonic media mainstream (e.g., Atton 2002;Downing 2001;2010;Fuchs 2010;Harcup 2012).By giving voice to minorities and marginalized groups, they were expected to increase the diversity of views in the media landscape overall.With the proliferation of "alt-right" news outlets (Atton 2006;Haller, Holt, and de La Brosse 2019), research on alternative media has over the last decade shifted to focus on outlets situated on the political right-wing (e.g., Figenschou and Ihlebaek 2019;Holt 2016;Nygaard 2019).These right-wing alternative media still perceive themselves as representatives of viewpoints neglected by a mainstream (Holt, Figenschou, and Frischlich 2019), aiming to contribute to what Buyens and van Aelst (2021, 5) call the "external content diversity of the news environment" while altering the expectation of what kind of diversity their content entails.Beyond external diversity, partisan, non-objective newsmaking is not per se freed of a standard of news diversity but may still commit "to fulfill the public's right to know through accurate and comprehensive reporting" (Meyers 2020, 180).We thus argue that alternative news media may at least partly seek to mimic and adopt practices originally conceived as practices of objective reporting to be seen as professional news outlets and to convey an image of responsible partisan news coverage.

Measuring Diversity and Balance in Relation to Sources and Topics
Defining and measuring what makes news content diverse and balanced is a prominent but difficult question in journalism research (Joris et al. 2020).Although diversity in news coverage is usually assessed from two angles-diversity of sources and topics-this has been measured in countless ways (Baden and Springer 2017;Freudenthaler and Wessler 2022).First, the diversity of sources concerns the presence of referenced actors and speakers with different societal backgrounds (Benson 2009), from different fields of society (Masini and van Aelst 2017), or with different political affiliations and leanings (Benson 2005).Studies measuring topical diversity have similarly taken different points of departure, ranging from specific issues (Day and Golan 2005;Nord 2013) to broader themes and subjects (Carpenter 2010;Kaiser, Rauchfleisch, and Bourassa 2020;Rodgers et al. 2004) and even broader editorial categories (McCombs 1987).Moreover, balance in news coverage has been studied through different aspects such as visibility, tone, and issue balance (Hopmann, Aelst, and Legnante 2011;Lewis and Cushion 2019).Visibility concerns how much coverage certain (especially political) sources receive compared to their counterpart(s).Tone concerns the evaluations or so-called latent favorability of different actors in the coverage.Issue balance concerns the coverage of actors in the context of topics they usually emphasize themselves, meaning how (often) they get to talk about issues that are expected to align with their agendas.
Furthermore, news diversity and balance can be assessed at different levels: at the level of the entire news environment, the individual news outlet, or the individual news item (Buyens and van Aelst 2021).For this study, we assess the question of overall source diversity at the level of overall news output across several right-wing alternative media in six countries and analyze source balance at the level of individual news articles.We have focused specifically on the question of partisan source diversity and balance, that is, the question of whether right-wing alternative news content contains right-wing, non-partisan, and left-wing sources and to what degree and in what kinds of topics these sources are presented in a neutral, balanced way.Our research questions (RQ) are:

Design, Data, Methods
To provide general insights into the topics and partisan referencing practices of rightwing alternative online news media, we relied on a diverse sample of 20 news sites that vary according to their (a) degree of professionalization regarding their website appearance, (b) audience reach, (c) genesis, and (d) country context.We followed the tendency measure proposed by Heft et al. (2020), which uses the particular categories through which these media present their content on their homepages to differentiate between right-wing alternative news media with a more normalized, conventional website appearance and those with a more overtly alternative, partisan appearance (e.g., through categories particular to right-wing ideology, such as Immigration, Freedom of Opinion, or Mainstream Media), including ten sites of each type (a).These sites differed concerning their size, comprising sites with larger and smaller audience reach, which has been determined by the rank a website has in the respective country, using traffic analytics data provided by Similarweb 1 (b).The sample also reflected diversity, as it comprised digital native news media and alternative news media with print editions, as well as news media with ties to specific social and political movements or political entrepreneurs, as established through prior research (e.g., Askanius and Mylonas 2015;Dreesen and Krasselt 2022;Mayerhöffer 2021;Mayerhöffer and Heft 2022;Wasilewski 2019) (c).Finally, the selected sites originate from six countries which have been shown to represent both countries with a larger and smaller right-wing online news infrastructure across different (Western) political and media contexts (Heft et al. 2020): Germany, Austria, Denmark, Sweden, the UK, and the US.We selected four cases for each country with a larger alternative right-wing online news infrastructure (Germany, Austria, Sweden, and the US) and two cases for each country where this infrastructure is less extensive and less heterogeneous (Denmark and the UK) (d).With this diverse selection of sites that are differently embedded in their specific media and political contexts, we aimed to provide general insights minimally driven by organizational or country-specific particularities.Table A1 in the appendix provides details of the included outlets.
We drew a random sample of 1000 articles (50 articles per news site) published between March 2019 and February 2020, using the Media Cloud 2 database as an access point to the sites' reporting, which underwent content analysis based on a standardized codebook. 3At the article level, we categorized each article's main overall theme.The classification scheme is based on an existing instrument for capturing topics in media texts (Wessler et al. 2008) that has been further adapted to the study subject.The coding was based on the headline and the first paragraph of the article and was hierarchically organized, differentiating between broader topic fields and specific topics within those fields.Topics were coded as specifically as possible and aggregated into broader topic fields during the analysis.If the article touched upon several issues, coders had to determine the most important topic, based on volume and intensity.We distinguished between politics (e.g., election and party politics), policy issues (e.g., economic, security, and immigration policy), international policy (e.g., European policy), military/conflict (e.g., military interventions and terrorism), law and order (e.g., court decisions, intelligence services, and crime), society (e.g., migration as a societal issue, religion, and health), economy, science and technology, leisure, and media and culture.
The second unit of analysis was the reference level and included two reference types: hyperlinks and sources.For hyperlinks, we collected all static in-text hyperlinks-that is, all hyperlinks in the article's running text, including (sub)headings and (with restrictions) imagery and videos.Sources were defined as non-hyperlinked speech acts and included direct (i.e., directly quoted) and indirect (i.e., paraphrased and reported) speech by clearly discernible actors in an article's running text.Up to a maximum of ten references per article were coded manually.We categorized the actor types-differentiating between several types of media organizations, political and administrative actors, civil society actors, and so on-and partisanship of the referenced actors.Here, we distinguished broadly between right-wing partisan actors, left-wing partisan actors, and actors with no discernible partisanship.The categorization of partisanship was based on the actors' selfdescriptions on websites and social media platforms, and on additional desktop research.Keywords indicative of a right-wing partisan leaning included conservative, nationalist, and right-wing, in their country-specific expressions.Terms coded as indicative of a leftwing political leaning included progressive, feminist, socialist, and liberal (country-specific for the US).Established news media's party leanings or editorial stances were also used as indications for partisanship, if they were determined in prior research, e.g., on pressparty-parallelism (e.g., Eilders 2004;Newman et al. 2017).Organizations and institutions like courts, universities, and public service media were coded as non-partisan by definition, even though they may be perceived as "partisan" by right-wing audiences.All in all, this means that the coding of partisanship was not limited to political actors in a narrow sense but was applied to all referenced actors in our sample.
Finally, we classified the evaluation of each reference in a way that captured whether the reference was explicitly evaluated when introduced in the article text (positive/supportive, negative/dismissive/delegitimizing, or neutral).Evaluation thus only pertains to the particular context in which the source was cited or referenced, i.e., not to the overall tone of the article or a general evaluation of the source.A positive/ supportive evaluation needed to contain an explicit affirmative formulation such as "as the source rightly points out".Expressions such as "contradictory reasoning", "lacks insight" or "primitive slam" were indicative for a negative/dismissive/delegitimizing evaluation.References without an explicit evaluation were coded as neutral/neither positive or negative.This provides us with a further means to assess the extent to which the news sites adhere to the established journalistic norm of neutral reporting.

Topical Diversity in Right-Wing Alternative News Media
Assessing the topical diversity of the right-wing alternative news media's reporting in our sample (RQ1a) shows that this reporting covers a relatively broad thematic spectrum, including the full range of political, economic, societal and cultural themes (Table 1).In general, reporting is dominated by political themes, such as news on the political processes and actions of parties and individual politicians and news on elections and the political system (total: 28.3%).Societal issues constitute the second most relevant theme of coverage (18.9%), comprising aspects of migration, integration, and minorities; the full range of societal discussions surrounding nationalism, feminism, and racism; protest activities; and other social problems.Reporting on law and order themes, such as court decisions, crime, and other police news is also frequent (11.9%), corroborating the findings of earlier studies on the main themes of right-wing alternative news.
The analysis also reveals particularities of the English-speaking countries (the US and UK), the German-speaking countries (Germany and Austria) and the Scandinavian countries (Denmark and Sweden) in our study: The coverage of the US and UK sites in our sample is significantly stronger driven by politics (Pearson Chi 2 , p < .01). 4 The US sites in particular report more frequently on elections, parties, and politicians (p < .01)than the media from the other countries, possibly a reflection of the US majoritarian political system and high levels of polarization in both the political and media sphere (Levendusky 2013).As opposed to this, the German media in our sample focus more on science, technology and environment related topics (p < .01)and in particular on migration and minorities (p < .001),which neatly reflects the country's far right's topical agenda (Backes 2018;Heft et al. 2022) that migration topics have long dominated.Overall, the German and Austrian sites are more evenly split in their attention on politics and societal topics.
The Scandinavian sites show a similar pattern in devoting significantly less attention to reporting on politics (p < .01)but instead focusing on topics of law and order (SWE 17%, p < .05;DEN 22%, p < .01),which in Denmark even exceed the attention on general politics.The Swedish sites, in addition, report more frequently on military conflict (p < .05).This aligns with research on the country's far right's issue focus which seems generally more determined by cultural aspects of immigration, that are framed as Islamisation, terrorism and a security threat (Heft et al. 2022) and corroborates findings that show that in Scandinavia, alternative news media are heavily focused on reporting on migrant crime (Mayerhöffer 2021).
Beyond these country particularities, these findings also highlight a more general aspect to our deep dive into source diversity and balance in right-wing alternative news content across topics, namely that all sampled news sites devote attention to multiple topic areas.While the individual news sites differ somewhat in the compositions of their individual "news buffets"-as shown in Figure 1-none of them is a single-issue site or overtly dominated by a single topic.Thus, in the following, we abstract from individual sites to a right-wing alternative media coverage.We acknowledge that this approach will necessarily neglect site-and country-specific particularities that are important to study in their own right.Yet, this generalization allows us to focus on the topics they cover as the main comparative dimension and provides a deeper understanding of source diversity and balance across diverse topical contexts.

Types of Sources Across Main Topics
To the question of what types of sources are referenced in right-wing alternative news media (RQ1b), our analysis shows that they reference a wide range of actors throughout their coverage (Table 2).No single source type dominates, and each source type makes up at least 9% of all sources.However, when all media actors are grouped together-mainstream, alternative, and other-media represent almost 40% of all sources referenced.In addition, (self-)references to a news sites' own website or social media profiles amount to 17%.Thus, without delving into specific topics, more than half of all sources are media, meaning that the overall coverage relies heavily on external and internal media sources.This pattern corroborates research findings showing that media sources-and in particular legacy media-are a prime source for connecting to broader news environments and for gaining, amongst others, credible and cost-saving content for alternative news medias' reporting (Heft et al. 2021;Mayerhöffer and Heft 2022).While all topics contain a substantial share of media sources, both legacy and alternative media sources are significantly more frequently used in cultural topics and reporting on the (mass) media (p < .001),which might be more than a reflection of the topic itself.These issues are known as some of the most contested and fertile fields for right-wing alternative news media's mass media criticism (Figenschou and Ihlebaek 2019).Self-references are significantly more prominent in articles on e.g., (national) politics or leisure, while they appear less frequently e.g., in articles on international policy or law and order (p < .05),both topic fields with high provision of easily citable external sources to sustain coverage.The same applies to reporting on military conflict, in which legacy media and political sources are overrepresented (p < .01).Articles on law and order have a particular high share of other sources (p < .001),partly due to a high number of police sources.Political sources, such as politicians, parties, and ministries, are significantly more often quoted or linked to in articles about politics and policy as well as military conflict (p < .001).

Partisan Balance (Source Level)
Splitting the sources into partisan groups allows us to discern the extent to which the right-wing alternative news media in our sample reference partisan and non-partisan  Note: Total percentages may not add up to 100 due to rounding.Source types across topic fields: x 2 = 938.675,df = 60, p < .001. a n represents the actual number of cases in the sample, and the data have been weighted to equalize country-specific differences in distributions.
sources differently across distinct topic areas (RQ2).We find that, in general, the rightwing alternative news media most often refer to external sources with no discernible partisanship, amounting to 43% of all sources.Right-wing external sources make up the second-largest group in the context of partisanship, accounting for around 26% of all sources, while around 14% of the sources are characterized by left-wing partisanship.Finally, in 17% of references, the sites refer and link to their own previous coverage and thus automatically reinforce their partisan reporting through self-references.
In the following, we focus the analysis on external sources that the media include beyond their own reporting (for the values including self-references, see table A1 in the appendix).As shown in Figure 2, there are notable differences in the use of partisan sources across topical fields, with some topics including significantly more external rightwing sources than others, namely Politics and cultural (Culture, Art, Media) topics (p < .05).This pattern might reflect strategic alliances with other right-wing actors pursued in issue fields at the core of right-wing criticism, such as reporting on mass media, echoing previous findings on the topics of these media (e.g., Kaiser, Rauchfleisch, and Bourassa 2020;Mayerhöffer 2021;Müller and Freudenthaler 2022;Nygaard 2019).Although right-wing external sources outweigh their left-wing counterparts across the board, the latter are present in all topics and thus far from neglected in the coverage.As shown in Figure 2, the biggest share of left-wing sources is found in the coverage of political issues (Politics, Policy Areas), which contains a noteworthy number of sources with liberal, socialist, or other left-leaning traits.However, how these sources are treated in the articles-whether they are evaluated negatively, neutrally, or positivelyoften differs from how right-wing sources are treated.
Generally, left-leaning sources mostly receive either neutral (62%) or negative (36%) evaluations while positive evaluations are scarse (2%).In contrast, right-wing external sources are far more often treated neutrally (85%) and less frequently evaluated negatively (7%) or positively (8%, see Table A2 in the appendix).Concerning all external sources, Figure 2 shows not only the partisanship of sources in different topics but also how these sources are evaluated depending on the topic, specifically how often leftwing sources are evaluated negatively and how often right-wing sources are evaluated positively.This highlights to what extent and in which contexts these outlets, which are all situated on the right side of the ideological spectrum, display the most partisan practices in explicitly evaluating the sources they cite.
In this regard, the coverage of Politics is particularly interesting, as it has the largest number of partisan sources.Looking at those in more detail, we find the largest overall share of negative evaluations of left-wing sources (10%) and one of the largest shares of positive evaluations of right-wing sources (4%).The gap between how left-wing and right-wing sources are treated in political coverage is especially pronounced when the alternative news media cover parties and political candidates.Similarly, Culture, Art, Media contains a substantial share of positive evaluations of right-wing sources (5%), while half of the left-wing sources in this topic are explicitly criticized (amounting to 8% of all sources in this topic).These negative evaluations stem predominantly from how these sources are evaluated in the coverage of mass media.Likewise, in the coverage of Society-which includes issues such as migration, inequality, and gender-the sites explicitly criticize close to half of the left-wing sources, (equivalent to 5% of the sources in this topic) and occasionally support right-wing sources (2%), though not nearly as often as in the political or cultural coverage.In contrast, the coverage of scientific and environmental issues, such as climate change, displays a more balanced reference practice, with fewer negative evaluations of left-wing sources (3%) and not a single positive evaluation of right-wing sources.The environmental coverage thus has a somewhat equal distribution in terms of how many sources the outlets employ from each camp and how these sources are evaluated in the text.This contrasts with the right-leaning balance -both in amount and evaluation-in the coverage of politics, society, and media that contain core right-wing alternative news topics, underscoring the fact that there are indeed important differences in how these outlets use sources, depending on their partisanship and the topics they write about.

Partisan Balance (Article Level)
The balance of sources on a reference level shows that right-wing alternative news media use both left-wing and right-wing actors in all topics but tend to use more right-wing sources in core right-wing topics such as those centering on migration or cultural topics, which are then introduced in a more partisan way.As news readers-be it legacy or alternative media-do not read everything an outlet publishes all at once but instead rely on individual articles, this section investigates how partisan balance looks at the article level.In other words, it is not only the balance of sources in coverage as a whole or the coverage of entire topics that can be taken as an expression of journalistic balance and diversity but also, and perhaps even more importantly, how (often) different sources are included in individual articles and especially how often right-wing partisan and counter-partisan sources are referenced in a balanced way in individual articles across topics (RQ3).
As the outlets are all characterized as right-wing, and as the biggest partisan source group in all topics are right-wing sources, left-wing sources and sources without explicit partisanship are grouped together in Figure 3 to show how often right-wing sources outweigh other sources in the articles.More than half of all articles indeed feature a majority of non-right-wing sources, whereas close to one-fourth of the articles have mostly rightleaning sources.There are even some articles with a majority of left-wing sources; however, this is only the case in 76 articles (less than one-tenth).Similarly, in less than one-tenth of the articles, the outlets achieve an exact balance between right-wing and other sources.This shows that although right-wing sources account for the biggest partisan source group at the source level, they are by no means the dominating source type in all articles.
A news portfolio that combines articles that feature a majority of non-right-wing, and in particular neutral sources, with articles in which right-wing partisan sources take center stage may help right-wing alternative news sites to enhance their credibility and at the same time maintain a clear partisan profile.
However, as the previous findings have shown on a reference level, the balance is not even across different topics.At the article level, Figure 4 shows that right-wing sources are more often in the majority in articles about either political or cultural issues.In other words, compared to the outlets' overall coverage, readers are indeed more likely to encounter a majority of right-leaning sources when they read articles on topics that include core right-wing themes.Yet, even here, articles with a majority of right-wing sources are in the minority.
Furthermore, other sources seldom come close to the number of right-wing sources in articles where right-wing sources are in the majority.Instead, articles where right-wing sources are in the "majority" are often those in which right-wing sources are in fact the only sources present and thus not in any way confronted or counterweighted by other sources (Figure 5).Likewise, articles with a "majority" of non-partisan and left-wing  sources often contain no right-wing sources at all.Therefore, although the right-wing alternative news media often employ sources that are discernibly not right-wing in their coverage, these sources often appear isolated from right-wing sources.This creates an image of a balance of views across articles, while, in reality, readers of specific articles are more often met by a separation rather than a confrontation of sources.

Discussion and Conclusion
Applying a cross-topical perspective, this paper has studied source diversity and balance in the referencing practices of a diverse sample of alternative news content published by right-wing alternative news media from six countries.We have argued that the hybrid nature of alternative media could result in practices following established journalistic norms more strongly in certain fields of reporting, while other themes more pertinent to these sites' partisan core could be used to gain partisan credibility.
First, our findings show that the media analyzed cover a relatively broad topical spectrum, including the full range of political, economic, societal, and cultural topics.These results support earlier findings on the content-based variety of right-wing alternative news sites (e.g., Kaiser, Rauchfleisch, and Bourassa 2020;Müller and Freudenthaler 2022;Nygaard 2019), underlining that these media are seldom single-issue or exclusively partisan-issue entrepreneurs.However, these media's strong focus on political topics highlighting competition and (potential) anti-elitism regarding societal issues, such as migration, or crime closely mirror their partisan leanings.While this finding holds across the variety of alternative news content included in our study, the results also add to single-country studies in that they underscore country-and area-specific patterns.The differences in the topical diversity of the Scandinavian, English-and German-speaking media point to those sites' embeddedness in a specific political and cultural context which requires further scrutiny in future studies.
Concerning the sources that right-wing alternative media draw on in their reporting, the overall picture reflects diversity in the sense that a wide range of actor types is visible.The biggest share of these sources cannot be characterized by a clear political leaning.However, the share of sources with explicit right-wing political leanings is almost double that of sources with a left-wing stance, clearly indicating the sites' partisan leanings in the overall selection of sources.One could even go further by including the sites' self-references to their prior reporting, which means that readers are confronted with coverage based on more than 40% right-wing sources.We have also shown that particular topics at the core of these sites' partisan agenda, such as the coverage of migration or mass media, are more likely to feature right-wing sources.In addition, the most obvious form of partisan bias occurs either when right-wing partisan sources are explicitly supported in these sites' coverage or when left-wing partisan sources are explicitly criticized.Our data support the notion that the right-wing news sites abandon a neutral journalistic tone and thus sway away from the journalistic ritual of displaying balanced content, particularly in issue areas relevant to their partisan core, by taking sides when reporting dayto-day politics.Thus, by looking at individual referencing practices across these sites' reporting, we have shown relevant differences across topic fields.
Regarding partisan balance at the article level, we have shown that explicitly rightleaning articles are outweighed by articles with more diverse referencing practices.The right-wing alternative media more generally show a degree of normalization in the sense that they adhere to reporting based on a variety of sources.However, the article-level analysis verifies that this practice is indeed topic-driven.In topic fields with core rightwing partisan themes, these sites more often produce articles that rely on right-wing sources as the majority backing for their claims.Moreover, we have shown that, in practice, most articles with a "majority" of right-wing sources have no other sources at all (and vice versa).To readers, such articles will either give the impression of a regular reporting without drawing on right-wing sources at all or will display a strong partisan outlook by exclusively relying on sources from a broader right-wing political and media ecology.
Here, it seems that we find exactly the distribution of labor that undergirds the hybrid nature of right-wing alternative news sites.This hybrid nature might result in an impression that these outlets contribute to external diversity by giving voice to partisan sources deemed excluded elsewhere as well as providing internal balance between rightwing and other sources.The article level analysis, however, reveals that the outlets clearly deviate from journalistic professionalism and its traditional strategic rituals of balance and diversity by employing a greater share of right-wing sources in articles on core partisan topics as well as separating them from other sources.
On one hand, these sites can execute their partisan voice in issue fields at the center of interest for their partisan readers and allies.On the other hand, they can use a variety of issue fields to display-or mimic-legacy media's adherence to journalistic diversity and balance, which may help them gain credibility outside of an already inclined readership.This pattern may express a process of professionalization in the sense that these sites would rather rely on a non-partisan source than on a partisan source in instances where other "opportune witnesses" (Hagen 1993) are available to back their claims.However, this pattern may also simply reflect the fact that these alternative media sites are re-using reports from other media, press releases and official accounts to fill their sections, thus adopting practices of "churnalism" as they can also be found in other types of digital news sites.The fact that alternative news outlets at least partly display a varied and balanced collection of sources must therefore not be unequivocally interpreted as a sign of quality journalism.To the dedicated readership, outright partisan reporting might not even be needed as they know how to decode implicit criticism.
How the right-wing alternative news media in our study evaluate sources-specifically left-wing sources-and use right-wing sources for partisan topics most relevant to themselves and their audiences reveals how they differ from the legacy media they simultaneously imitate and distance themselves from.This shows the complexity of the mainstreaming of right-wing alternative media sites, caught between journalistic legitimacy and partisan credibility.Either way, it contributes to the blurring of boundaries and might further undermine the perception of professional journalism in the dedicated publics.
We have approached this complexity by analyzing a diverse sample of cross-national alternative news content, focusing on source diversity and balance at the level of the entire news sample and each individual article.The in-depth analysis of variance at the topic level, thereby made possible, is inevitably accompanied by a neglect of variability at the country level and at the level of individual news sites.Given the heterogeneity of right-wing alternative news media, the patterns observed are at least partly also affected by the particular style and practices of individual right-wing alternative news media.Future studies should take into account the political and cultural embeddings and the particularities of individual news media in different countries.In addition, future studies could add to our understanding of the hybridity of alternative newsmaking by focusing on the question of how alternative news audiences encounter and engage with professionalized versus partisan content.

Notes
Table A2.Evaluation of external sources per partisanship (in %).Note: Total percentages may not add up to 100 due to rounding.x2 = 291.392,df=4, p < .001. 1 n represents the actual number of external sources in the sample, and the data have been weighted to equalize country-specific differences in distributions.
RQ1a: How topically diverse are articles published by right-wing alternative news media in six countries?RQ1b: What types of sources are referenced in right-wing alternative news content across different topics?RQ2: To what degree do right-wing alternative news media reference partisan and non-partisan sources differently across distinct topic areas?RQ3: To what degree are partisan and non-partisan sources referenced in a balanced way in individual alternative news articles across topics?

Figure 2 .
Figure 2. Partisanship of external sources across main topics and evaluation of left-wing and rightwing sources (in %).Note: Left-wing sources have been split into negative (dark red) and neutral/positive (light red) evaluations.Right-wing sources have been split into neutral/negative (light blue) and positive (dark blue) evaluations.Non-partisan sources are included in this figure, but their evaluations are omitted.n represents the number of external sources in the sample, and the data have been weighted to equalize country-specific differences in distributions.Partisanship across main topics: x2 = 231.555,df = 20, p < .001.

Figure 3 .
Figure 3. Balance between right-wing and other sources on article level (n = 1000).

Figure 5 .
Figure 5. Dominance of right-wing and other sources on article level (n = 903).Note: Excludes articles with no external sources.

Table 1 .
Main topical categories and subcategories across countries at the article level.

Table 2 .
Types of sources across main topics (in %).