Abstract
This paper analyzes the discursive dynamics of mediated truth contestation by examining how assertions of truthfulness and references to falsehood are constructed in Austrian and Czech news coverage of migration. Employing automated content analysis on an extensive dataset of 162,943 news articles (2013–2019), we apply latent semantic scaling to assess differences between tabloid and broadsheet newspapers and between mainstream and alternative news media, investigating how media formats and media types shape the discursive construction of truthfulness and falsehoods in news articles. We further use named entity recognition techniques to examine the role of populist politicians, government representatives, and EU actors in mediated truth contestation, and consider cross-national differences to investigate how political contexts influence these dynamics. By integrating insights from political communication and computational text analysis, this study contributes to debates on post-truth politics, media polarization, and the strategic use of truth and lies in democratic discourse.
Keywords
Introduction
Truths and falsehoods play a central role in mediated public discourse, shaping political decision-making, citizen beliefs, and democratic accountability (Waisbord 2018; Zaller 1992). While exposing lies can foster democratic dialogue and transparency and enable citizens to hold political actors accountable (Graves and Wells 2019; Koliba 2025), (un)truthfulness has increasingly become a strategic discursive resource used to delegitimize opponents, undermine credibility, and construct alternative versions of reality (Egelhofer and Lecheler 2019; Kluknavská and Eisele 2023). Accusations of news media of spreading fake news, challenges to journalistic authority, competing claims to factuality, and relativizing shared assessments of reality reflect a broader shift in which emotional appeals, strategic deception, and alternative epistemic standards can overshadow evidence-based deliberation (Egelhofer et al. 2021; Hameleers and Minihold 2022; Van Aelst et al. 2017). Against this backdrop, what matters is not only whether political information is factually accurate, but how claims about truthfulness and falsehood are discursively constructed, attributed, and contested within mediatized public debate (Waisbord 2018; Kluknavská and Eisele 2023).
This article analyzes the discursive dynamics of mediated truth contestation by examining how assertions of truthfulness and references to falsehood are constructed in Austrian and Czech news coverage of migration. Migration constitutes a highly salient and polarizing issue, and the two countries represent contrasting political and discursive contexts, particularly during and after the 2015/2016 refugee crisis (Eberl et al. 2018). While Austria adopted a more inclusive approach to migration, welcomed a larger number of refugees, and showed a public sentiment characterized by solidarity and support, Czechia pursued stricter migration policies, which resulted in lower refugee intake and a divided public debate (Kluknavská et al. 2021). These differences create distinct opportunity structures (Koopmans and Statham 2010) for contesting truthfulness and falsehood, making migration an ideal case for studying how mediated truth contestation varies across national contexts.
This study employs automated content analysis on an extensive dataset of 162,943 news articles published between 2013 and 2019. Relying on latent semantic scaling (Watanabe 2021), this approach identifies the extent to which the language in news articles aligns with seed words representing assertions of truthfulness or references to falsehood, thus capturing whether the news coverage frames migration more strongly through discourses emphasizing truthfulness or falsehood. We scale media coverage of mainstream national newspapers (two broadsheets and one tabloid) and the two alternative news online media platforms per country. This helps us to explore how media characteristics, particularly political goals, format, ideological orientations, and audience positioning, shape truth contestation in public discourse. In addition, we use named entity recognition techniques (Balluff et al. 2024) to analyze the role of political actors, specifically populist politicians, government representatives, and EU actors.
The study contributes to political communication research. While existing scholarship has examined the delegitimization of news media through the use of fake news labels (Egelhofer and Lecheler 2019), our interest lies in the textual patterns through which truthfulness and falsehoods are represented and interact within a broader discursive field of mediated public debates. Furthermore, it offers new insights into how these dynamics vary across different media types and national contexts. This is particularly relevant given the negative portrayals of migration in alternative news media, often designed to fuel public fear and resentment (Culloty et al. 2022). The comparative approach examining cases in Western and Eastern Europe contributes to a more general understanding of how national political and media environments shape epistemic claims in public debates. Lastly, the study employs computational techniques to analyze large-scale datasets, providing a scalable framework for examining truth contestation across diverse contexts.
Theorizing Dynamics of Truth Contestation
The Construction of (Un)Truthfulness in a Mediated Public Debate
Mediated truth contestation refers to the discursive negotiation of truthfulness and falsehood in the public sphere, that is, how claims of what is presented as “true” or “false” are constructed, attributed, and contested in mediated public debate (Hameleers and Minihold 2022; Kluknavská et al. 2025; Kluknavská and Eisele 2023). This perspective shifts attention away from the verification of factual accuracy toward the discursive construction of truthfulness and falsehood, reflecting broader struggles over the credibility and legitimacy of claims. Public debate unfolds within a hybrid media system, where journalistic norms, media formats, and political strategies intersect (Chadwick 2017). Within this system, news media actively shape how reality is interpreted and provide a space for actors to advance claims, challenge opponents, and contest epistemic authority (Van Aelst et al. 2017; Waisbord 2018).
To analytically capture these dynamics, we conceptualize mediated truth contestation through two interrelated dimensions: references to falsehood and assertions of truthfulness. These dimensions emerge through the reporting, amplification, or contextualization of claims related to truthfulness and falsehoods (Kluknavská et al. 2025). They may occur as direct or reported claims from political actors, journalists, or other public actors, or as broader thematic references embedded within news coverage.
References to falsehood include mentions of lies, deception, misinformation, manipulation, untruthfulness, dishonesty, or labels such as fake, false, wrong (Hameleers 2022; Kluknavská et al. 2025). For example, accusations that journalists disseminate fake news or that political opponents spread disinformation are often reported in news articles, thereby embedding references to falsehood into the mediated discourse. Mediated references to falsehood also encompass reporting on claims that politicians, experts, or institutions engage in deception or manipulation, as well as broader discussions of misinformation, rumors, or hoaxes (Egelhofer and Lecheler 2019; Hameleers et al. 2023). While such coverage does not necessarily endorse these claims, it nonetheless situates actors, issues, and events within semantic contexts associated with deception or falsity (Farhall et al. 2019; Hameleers and Minihold 2022).
Assertions of truthfulness, by contrast, involve discursive affirmations of factuality, objectivity, authenticity, or epistemic certainty. Actors invoke facts, expertise, common sense, or moral authority to present their interpretations as credible and legitimate, thereby claiming epistemic authority and delegitimizing alternative perspectives (Waisbord 2018). Such assertions do not necessarily reflect empirical accuracy but function as rhetorical strategies that position particular claims as trustworthy or authoritative. In post-truth political contexts, competing assertions of truthfulness often reinforce political identities rather than resolve factual disagreement (Kluknavská and Eisele 2023), enabling the construction of parallel interpretative frameworks or “alternative realities” that challenge journalistic or scientific authority (Hameleers and Minihold 2022).
Existing research has examined related phenomena using concepts such as truth claims, fake news label, disinformation and misinformation accusations, discourses on untruthfulness, post-truth claims, or accusations of untruthfulness (e.g., Egelhofer and Lecheler 2019; Egelhofer et al. 2022; Galande et al. 2023; Hameleers 2022; Hameleers and Minihold 2022; Kluknavská and Eisele 2023; Sørensen and Krämer 2024). Recent work has, for example, analyzed how populist leaders employ “fake news” accusations (Ross and Rivers 2018) and accusations of untruthfulness (Kluknavská et al. 2025) or how disinformation accusations are used as persuasive tools in an authoritarian context (Alyukov and Zavadskaya 2024) and as a means of deflecting responsibility (Radnitz et al. 2025). While these works have provided important insights into strategic communicative tactics, the concept of mediated truth contestation captures the broader discursive field for the circulation and amplification of (un)truthfulness within news media coverage.
Research on misinformation and disinformation has generated crucial insights into the production, diffusion, and effects of false or misleading information. However, it is primarily oriented toward questions of factual accuracy, intentional deception, and corrective interventions. As recent scholarship suggests, however, contemporary political struggles over truth are increasingly shaped by the performance of truth claims rather than their factual content, with authenticity and identity serving as key sources of epistemic legitimacy (Sørensen and Krämer 2024; Van Zoonen 2012). Radnitz (2025) further argues that the misinformation debate plays out among distinct discursive coalitions contesting authority over misinformation and epistemological competence. Audience-centered research demonstrates that citizens evaluate truth claims based on divergent epistemic logics, negotiating trust in news based on fundamentally different understandings of “truth” ranging from factual verification to values, emotions, and identities (Ryzhova and Toepfl 2025).
Given the growing importance of performative credibility and contestation of authority over epistemic truth (Radnitz 2025; Sørensen and Krämer 2024), the misinformation and disinformation approaches face a challenge in capturing discursive dynamics of mediated truth contestation, where labels such as “misinformation,” “disinformation,” and “fake news” are strategically diffused to delegitimize opponents, journalists, or experts (Egelhofer and Lecheler 2019; Egelhofer et al 2022; Ross and Rivers 2018). The two dimensions of truth contestation, references to falsehood and assertions of truthfulness, become empirically observable because struggles over epistemic authority are performed through recurrent linguistic cues that explicitly evaluate the validity, credibility, or integrity of claims. In mediated public debate, actors and journalists do not contest epistemic legitimacy abstractly but rather do so by invoking terms such as reality or truth to affirm factuality and authority, or by using labels such as lies or manipulation to delegitimize alternative claims. These lexical terms are treated as entry points to a broader semantic field. When such terms appear in news coverage, they situate issues and actors within discursive contexts structured around epistemic legitimacy. As such, rather than treating these terms as indicators of objective truth or falsity, we conceptualize them as discursive resources through which actors perform epistemic legitimation and delegitimation within news coverage. This approach enables us to identify how media coverage becomes semantically structured around competing constructions of epistemic legitimacy, without making claims about the factual accuracy of the underlying statements.
References to Falsehood and Assertions of Truthfulness in Migration News Media Coverage
Migration is a salient and polarizing political issue, which makes it a particularly suitable case for examining mediated truth contestation. Its divisive nature has intensified public emotions and political conflict, producing heated, uncivil, and intolerant public discussions (Kluknavská, et al. 2024). These dynamics were especially pronounced during the European refugee crisis of 2015/2016, which further politicized migration and generated starkly contrasting responses, from hostility and securitization to appeals to solidarity, empathy, and humanitarian responsibility (Gessler and Hunger 2022; Kluknavská et al. 2021).
Research on migration coverage shows that issue salience fluctuates over time and is closely linked to shifts in framing, tone, and conflict intensity in news media (Chouliaraki and Zaborowski 2017; Eberl et al. 2018). Periods of heightened salience, such as the 2015/2016 refugee crisis, are associated with more polarized and emotional reporting, stronger elite conflict, and problem-centered framing emphasizing security, moral, or cultural threats (Eberl et al. 2018), often structured around dichotomous representations of refugees as either vulnerable victims or dangerous outsiders (Chouliaraki and Zaborowski 2017; Kluknavská et al. 2021).
Beyond tone and framing, such periods of heightened salience may also act as catalysts for mediated truth contestation. Crises generate uncertainty, contested expertise, and competing interpretations of responsibility, numbers, and policy effectiveness, making disputes over “what is really happening” central to political conflict (Eisele et al. 2022; Gessler and Hunger 2022; Kluknavská and Eisele 2023). In these contexts, journalists increasingly report about credibility and legitimacy related to borders, integration, and state capacity, which are articulated through references to falsehood (e.g., accusations of deception or misinformation) and assertions of truthfulness (e.g., claims to factuality, expertise, or moral authority).
As issue salience rises and falls over time, migration coverage may therefore become more or less embedded in discourses emphasizing truthfulness and falsehood. Against this backdrop, we examine temporal dynamics and issue salience as key contextual drivers of mediated truth contestation and ask:
Beyond these general dynamics, we investigate truth contestation in migration-related news coverage across three dimensions: (1) the most popular national newspapers and alternative news websites, (2) the role of different national and European political actors, and (3) the countries’ political context.
News Outlets’ Characteristics and Dynamics of Mediated Truth Contestation
Characteristics of news outlets shape how mediated truth contestation unfolds, particularly whether coverage leans toward references to falsehood or assertions of truthfulness. News values, professional norms, and editorial styles influence how conflict is framed and which discursive resources are mobilized (Harcup and O’Neill 2017).
Tabloid newspapers are more likely to promote “othering,” reinforce us-them distinctions, and amplify conflictual narratives (Walter and Fazekas 2023). In migration coverage, tabloids often delegitimize the refugees or asylum seekers and provide disproportionate visibility to anti-elite, populist, or protest actors, who thrive on polarization (Akkerman 2011; Eberl et al. 2018; Rooduijn 2014). Their reliance on conflict, simplification, and scandal highlights blame and wrongdoing (Hanusch 2013), making references to falsehood, such as lies, deception, or manipulation, particularly compatible with tabloid news logic. Such references offer clear antagonists, dramatize conflict, and provide emotionally resonant cues that attract audience attention. By contrast, broadsheet newspapers place greater emphasis on contextualization, procedural accountability, and policy substance, and are therefore less structurally reliant on discursive delegitimization through references to falsehood (Kroon et al. 2016; Walter and Fazekas 2023).
While existing research has documented systematic differences between tabloid and broadsheet coverage of migration, particularly in terms of tone, framing, and levels of conflict, it has not directly examined how these differences translate into the balance between references to falsehood and assertions of truthfulness. Building on this literature, we hypothesize:
The digital media environment has facilitated the rise of alternative news websites, which often blend elements of professional journalism with ideological or partisan perspectives to mimic mainstream news formats, thereby enhancing their credibility (Bakir and McStay 2017; Mourão and Robertson 2019). These outlets position themselves in explicit opposition to mainstream news organizations, arguing that dominant journalism systematically excludes certain viewpoints or misrepresents social and political reality (Holt et al. 2019). Central to this oppositional self-positioning is a discursive strategy of boundary drawing, in which alternative news outlets question the epistemic authority of mainstream news media while presenting themselves as corrective actors (Ihlebæk et al. 2022).
To legitimize this role, they frequently claim to expose bias, deception, or deliberate misinformation in dominant news coverage, framing their own reporting as revealing suppressed or distorted truths (Holt et al. 2019). As such, alternative news media can be understood as occupying a symbolic epistemic periphery, positioned outside or at the margins of institutionalized journalistic authority. From this peripheral position, they may challenge the legitimacy of dominant truthfulness claims while simultaneously advancing alternative constructions of epistemic credibility. This positioning creates strong incentives for alternative news media to engage in mediated truth contestation through references to falsehood, which serve to delegitimize mainstream actors and justify the outlet’s claim to epistemic authenticity. Compared to mainstream news, which is more constrained by professional norms of balance and institutional credibility, alternative news media are therefore more likely to rely on references to falsehood as a recurring discursive resource. Accordingly, we hypothesize:
Political Actor Characteristics and Dynamics of Mediated Truth Contestation
The characteristics of political actors contributing to mediated public debates on migration can significantly influence the dynamics of mediated truth contestation. Populist politicians, government actors, and European representatives occupy particularly salient positions. Populist actors attract media attention through conflictual, emotional, and dramatized communication, aligning with media logic (Bos and Brants 2014). Government actors benefit from incumbency, as journalists rely on them as official sources (Wolfsfeld 2011). European actors influence migration debates by shaping cross-border policies and agendas that resonate within national media contexts.
Populist Actors
Populism, understood as a thin-centered ideology opposing a “pure people” to a “corrupt elite” (Mudde 2004), structures how populist actors engage in public discourse. Populist communication relies on simplified, black-and-white narratives, emotional appeals, and claims framed as common sense or authentic truth (Waisbord 2018). In migration debates, populists frequently depict the situation as a crisis and position themselves as defenders of national sovereignty against allegedly deceptive elites (Krzyżanowski et al. 2018). This often involves accusations of falsehood directed at political opponents or news media, framed as “establishment lies,” while presenting populist claims as revelations of hidden realities (Hameleers 2022; Kluknavská, et al. 2025; Waisbord 2018). Given this reliance on anti-elite rhetoric and epistemic claims, populist actors are expected to be particularly visible in mediated truth contestation. Therefore, we ask:
Government Affiliation
Incumbent politicians have incentives to appear less confrontational and more inclusive, emphasizing competence, responsibility, and policy performance. In contrast, opposition actors are structurally incentivized to challenge the legitimacy and effectiveness of the government (Nai 2020; Vliegenthart and Walgrave 2011). In migration debates, government representatives played a central role in shaping debates about asylum policies during the refugee crisis by shaping policy narratives and justifying decisions (Kluknavská et al. 2021). However, the extent to which government actors engage in truth contestation may vary depending on the intensity of the crisis, electoral competition, and opposition pressure. This leads us to ask:
Populist Governments
The inclusion-moderation thesis posits that populists in government moderate their rhetoric due to institutional constraints, coalition bargaining, and the need to appeal to broader electorates, which helps them stay in power (Akkerman et al. 2016; Krause and Wagner 2021). Governing requires compromise, which is expected to tame confrontational and anti-elite discourse. While some empirical works support this argument (e.g., Bernhard 2020), others challenge it. Schwörer (2022) shows that populist governments often retain core elements of populist rhetoric but redirect attacks toward new targets and selectively redefine which elites are corrupt and positioned against the people’s interests. This ambiguity motivates the following question:
Supranational Level
Migration is a core European policy issue, shaped by EU-level governance, that balances humanitarian obligations with national sovereignty concerns (Chouliaraki and Zaborowski 2017; Brändle et al. 2019). Although national media debates remain primarily confined to the borders of individual states (Lodges and Sarikakis 2013), EU representatives play a significant role in justifying EU-wide policies and financial contributions to migration management (Krotký 2023). At the same time, the EU is a frequent target of blame in migration debates, as exemplified by its central role in the Brexit campaign (Hobolt 2016). EU actors are therefore likely to be embedded in mediated truth contestation, motivating the research question:
The Political Context and Dynamics of Mediated Truth Contestation
Austria and the Czech Republic provide a valuable comparative context for studying truth contestation in migration debates, as differences in real-world events and political and discursive opportunities shape the mediated public debates (Koopmans and Statham 2010). Political environments can create more or less favorable conditions for truth contestation, since journalistic coverage is closely tied to real-world events, political reactions, and newsworthy developments (Esser et al. 2019). The 2015 refugee crisis, in particular, intensified media attention and provided fertile ground for disputes over truthfulness and falsehood in migration coverage.
The two countries experienced markedly different levels of exposure to the refugee crisis. Austria, located on a central refugee route, received 88,160 asylum applications, with many additional refugees passing through the country en route to Germany or other Western countries, making migration a visible and tangible issue. In contrast, the Czech Republic received only 1,515 asylum applications, rendering migration a more distant and abstract concern (Kluknavská et al. 2021).
Beyond real-world circumstances, political rhetoric, policy responses, and public attitudes further shaped mediated debates. In Austria, political elites and a significant portion of the public initially adopted a comparatively welcoming and pro-refugee stance. In contrast, many Czech politicians expressed a more cautious or even restrictive approach, which resonated with public anxieties. Concerns over the arrival of migrants from culturally different backgrounds fueled polarized and, at times, hostile or intolerant debates (Navrátil and Kluknavská 2023). Such differences in political rhetoric and public sentiment could influence the leaning of migration-related news toward truthfulness or falsehoods. Given these contrasting political and discursive contexts, we ask:
Methodology
In this study, we conducted an automated content analysis of media coverage related to migration in Austria and the Czech Republic. The study focuses on identifying systematic and longitudinal patterns in mediated truth contestation across various media types, making a computational approach particularly suitable. To ensure conceptual and empirical validity, the automated analysis was complemented by qualitative validation steps.
Data Collection and Sample
We utilize a unique dataset on migration coverage from two media types—mainstream news media and alternative news media. The sample comprises the print versions of the three most popular national newspapers with varying political leanings (conservative vs. liberal) and formats (tabloid vs. broadsheet), as well as the two most popular alternative news online media platforms per country (see Table 1). This allows for a comparison between mainstream media and outlets that adopt a more ideological stance and cater to a specific audience.
Newspapers and Alternative News Websites Included in the Sample.
Newspaper articles were downloaded using the Application Programming Interface (API) of the Austrian Press Agency (Austria) and the Anopress database (Czechia). The alternative news websites were scraped using R and Python.
We downloaded all articles on migration using a keyword-based search string designed to capture the core semantic field of migration. The Austrian search string was adopted from a previously validated cross-national project (Lind and Heidenreich 2021), in which native speakers developed and tested a comprehensive set of migration-related keywords. For the Czech dataset, the search string was translated and contextually adapted by native speakers and cross-checked against a validated keyword list used in prior research (Kluknavská et al. 2021). The English-language set of keywords includes: asyl, immigrant, immigration, migrant, migration, refugee, emigrant. The string included all these terms, as well as relevant synonyms in respective languages and grammatical variations, combined using Boolean operators and truncation. We validated the search for both Austrian and Czech samples by manually inspecting random samples to assess the relevance of retrieved articles.
The final sample includes 115,698 newspaper articles and 47,245 alternative news articles (see Table 1) in German and Czech. All text analyses described in the following were performed separately on each corpus to avoid distortions resulting from cross-linguistic semantic differences.
Dependent Variable: Mediated Truth Contestation
Our dependent variable measures the dynamics of truth contestation in migration coverage, operationalized as the extent to which the text of an article leans toward references to falsehood or assertions of truthfulness. Consistent with our theoretical approach, this measure does not assess the factual accuracy of statements but rather the discursive orientation toward (un)truthfulness as a communicative claim.
We operationalized this dimension using the LSX package for R (Watanabe 2021), which estimates the average semantic proximity of words in a text to two opposing conceptual poles. Rather than classifying articles as “true” or “false,” the method locates them on a continuous scale reflecting their relative orientation within a shared semantic space. Poles were defined by corpus-independent seed words representing truthfulness and falsehood (see Table 2). The seed words serve as entry points that approximate a broader semantic field of truth-related evaluation within the corpus. LSX does not detect whether a statement is factually correct, nor can it disambiguate negation, irony, or polysemy at the sentence level. It captures the relative semantic proximity of an article’s vocabulary to language conventionally associated with truthfulness affirmation (e.g., fact and reality) or truthfulness denial (e.g., lie and nonsense). The resulting scale, therefore, reflects a discursive orientation toward the epistemic legitimation or delegitimation within mediated debate.
Seed Words for Truthfulness and Falsehood.
Seed words selection followed a theory-driven and qualitatively validated process. Drawing on research on truth contestation, we selected a small set of highly abstract and semantically stable seed words that explicitly denote truthfulness or falsehood across contexts and languages. An initial exploratory phase tested a different set and a broader range of theoretically related terms. However, close readings of the outcome revealed that many captured general conflict, media criticism, or strategic blame rather than references to epistemic (un)truth. The final selection comprises four truthfulness-related and four falsehood-related words, reflecting core epistemic oppositions such as fact versus nonsense or reality versus manipulation. These terms were chosen to be functionally equivalent in German and Czech and to capture comparable evaluative meanings in both corpora.
The theoretical rationale for the selected seed words follows directly from our conceptualization of mediated truth contestation as the discursive evaluation of epistemic authority. We, therefore, focused on lexical items that directly denote the affirmation or denial of factuality in abstract and generalizable terms. Words such as fact, reality, and truth refer to the assertion of epistemic legitimation, whereas lie, nonsense, or untruth denote its delegitimation. The inclusion of seed words such as objective and manipulate may appear conceptually heterogeneous, but they were retained because keyword-in-context analyses showed that, in both corpora, they functioned as evaluative references to epistemic credibility or distortion.
We excluded more context-specific terms such as fake news, propaganda, or bias. Keyword-in-context analyses revealed that these terms are frequently employed in specific contexts, such as media criticism or strategic blame attribution, which may risk conflating truth contestation with evaluations of journalistic performance or political intent. Our aim was, therefore, to capture an underlying discursive orientation toward truthfulness or falsehood that cuts across actors and contexts rather than accusations directed at specific targets.
To assess the contextual appropriateness of the final seed words, we conducted keyword-in-context analyses in both language corpora, confirming that the selected terms were consistently used as evaluative references to (un)truth. Stopwords, numbers, and URLs were removed before the scaling to reduce potential noise in the analysis.
Validation of the seed words proceeded through a combination of manual coding and qualitative contextual analysis. For each language corpus, a random subsample of five-hundred articles was manually coded to assess whether the overall discursive orientation leaned toward truthfulness or falsehood. In the Czech corpus, three independent coders achieved an intercoder reliability of Krippendorff’s alpha of .60; in Austria, the task was performed by a trained coder. Given the abstract and evaluative nature of the concept and the use of a five-point polarity scale (−2 to +2), this level of agreement is consistent with expectations for complex discursive constructs.
Following earlier studies using the LSX package (Heidenreich et al. 2022; Trubowitz and Watanabe 2021), we triangulated the manual coding with the automated LSX scores at the aggregated level. LSX produces continuous, unbounded scores, with negative values reflecting proximity to falsehood and positive values reflecting proximity to truthfulness. In our data, values obtained from the automated analysis ranged approximately from −2 to +2. To reduce noise and account for temporal variation, we aggregated scores into three-month windows and correlated them with aggregated manual ratings (see Figure 1). The alignment between the two measures indicates that the automated scores capture meaningful patterns of mediated truth contestation.
Correlation plots for validation. (a) Austria and (b) Czech Republic.
In addition, we conducted a qualitative contextual examination of a purposive subsample of one-hundred articles (fifty per country), selected from the highest and lowest ends of the automated scale. These articles were closely read by the authors of the study to assess how references to falsehood and assertions of truthfulness were articulated in context. This step served both to validate the automated measure and to illustrate the substantive variation in truth contestation across countries and media types. This mixed-method analytical strategy helps mitigate the risk that aggregate quantitative patterns obscure substantively different communicative practices.
Despite these validation steps, the measure cannot fully resolve issues of context, negation, or polysemy at the micro-level and we acknowledge that individual occurrences may carry context-specific meanings. Nevertheless, the LSX approach mitigates this limitation by estimating semantic proximity across the full distribution of words in an article and captures aggregate discursive tendencies, showing patterns of semantic orientation.
Independent Variables: Migration Topic, News Outlets’ Characteristics, and Country
As a proxy for how extensively the migration topic was covered in the article, we counted the number of occurrences of the keywords that were used to retrieve articles. We then calculated the share of keywords in all words in the article as a measurement of the prominence of the migration topic in the article. Media characteristics were operationalized by distinguishing between tabloid and broadsheet newspapers and between mainstream and alternative news media. Country differences were modeled using country dummies. All metric variables were standardized (M = 0; SD = 1).
Independent Variable: Actors
We created three binary variables to differentiate between the following actors: (1) populist vs. non-populist political actors (politicians or political parties), (2) government versus opposition (politicians or political parties affiliated with incumbents or opposition during a corresponding time period), and (3) EU political actors (including institutions and bodies such as the European Commission, European Parliament, or other EU institutions, and individuals representing them, such as MEPs or Commissioners).
Instead of creating an extensive dictionary for each actor group, we used Named Entity Recognition to inductively identify persons and organizations from the text. For Austria, we relied on the spacyR package in R (Benoit et al. 2018) and included “persons” and “organizations.” Since spacyR does not include a language model for Czech, we relied on the Czech Named Entity Corpus 2.0 (Ševčíková et al. 2007). We included “person names,” “institutions,” and “media names” as the categories most comparable with those of the spacyR package. The extracted list of actors was then labeled regarding actors’ classification (1) as populist based on the established categorization of the PopuList project (Rooduijn et al. 2019), (2) belonging to the government, or (3) representing (parts of) the EU’s institutional system. The latter two were defined based on the information provided on the webpages of the EU as well as of the Austrian and Czech governments.
Analysis
A Bayesian regression model was estimated using Stan (Carpenter et al. 2017) and brms (Bürkner 2017), with weakly informative priors. Given the roughly normal distribution of the response variable (mediated truth contestation), a Normal link function was used. R-hat values for all posterior parameters did not exceed 1, indicating good MCMC convergence (Gelman and Rubin 1992). To account for yearly variability, a random intercept model (with errors clustered by year) was estimated. The posterior estimates are reported in Figure 3 and Table 3 below.
Posterior Distributions of the Regression Estimates.
Results
Addressing RQ1a (see Figure 2a and b), the dynamics of mediated truth contestation in the media landscapes of the Czech Republic and Austria exhibit distinctive temporal patterns between 2013 and 2019. Czech media is characterized by pronounced spikes and dips in the dynamics of mediated truth contestation over time. By contrast, Austrian media display less variability, albeit showing a general tendency toward assertions of truthfulness toward the end of the timeframe.
Dynamics of mediated truth contestation in Austria and Czechia. (a) Across all news media. (b) Across alternative news media and mainstream news media.
Disaggregating the results by news media type (Figure 2b) reveals systematic differences between alternative and mainstream news media. Across both countries, alternative news media exhibit markedly higher volatility in truth contestation dynamics, reflecting stronger and more fluctuating semantic proximity either to references to falsehood or to assertions of truthfulness. Mainstream media, by contrast, tend to cluster around the average value, suggesting a more stable pattern of limited truth contestation over time.
In Austria, this pattern is accompanied by a clear upward trajectory toward assertions of truthfulness, particularly pronounced in alternative news media, where the dynamics exhibit a steeper slope. Even mainstream outlets show a gradual shift toward truthfulness, albeit with a less pronounced slope. Qualitative evidence suggests this trend is driven by a growing emphasis on factual certainty, numerical evidence, and transparency claims. Alternative outlets prominently invoke “true numbers” of asylum applicants and accuse governments of “covering up the real figures.” Broadsheet newspapers often frame truthfulness in more reflexive terms, discussing “moments of truth,” the limits of objectivity, or the responsibility attached to factual claims in migration debates.
In Czechia, the disaggregated results reveal a contrasting pattern. Alternative news media show highly volatile dynamics, oscillating between strong references to falsehood and strong assertions of truthfulness. High-polarity articles invoke reality, facts, and certainty (e.g., “it is clear that it will only get worse,” or “but what will really happen?”), whereas low-polarity coverage is saturated with references to manipulation, disinformation, and propaganda, including claims that media create “a deceptive impression of calm.” Czech mainstream news media, by contrast, consistently lean toward references to falsehood, with only a slight shift toward assertions of truthfulness at the beginning of the research period. Across both tabloid and broadsheet outlets, low-polarity articles commonly reference false assumptions and misleading claims, for example, rejecting statements as “mere rumors,” describing public beliefs as “mistaken,” or labeling political positions as “unrealistic,” with tabloids sometimes explicitly framing such practices as deliberate simplification or manipulation by populist actors.
These findings suggest that while Austrian media gradually shift toward framing migration debates in terms of truthfulness, Czech media, especially mainstream outlets, remain more strongly oriented toward references to falsehood, with alternative news media amplifying volatility in both directions.
Table 3 and Figure 3 present population-level effect estimates from the regression model. Starting with RQ1b, the greater salience of the migration topic is accompanied by a trend toward more references to falsehood, as indicated by a negative posterior median coefficient (β = −.04 [−.05; −.039]). This suggests that more intensive coverage of migration is associated with an increased discursive emphasis on falsehood, deception, or contested claims.
Posterior distributions of the model estimates.
Turning to media characteristics, tabloid media exhibit a stronger trend toward references to falsehood compared to broadsheet media (H1a), as indicated by a negative posterior median coefficient (β = −.32 [−.33; −.31]). This pattern is consistent with tabloid coverage that highlights deception, as evidenced by references to “lying press” or claims that public audiences are being manipulated. By contrast, no systematic difference emerges between mainstream and alternative media overall (H1b; β = .0 [−.02; .01]).
Regarding political actors, populist actors are more likely than non-populist actors to appear in news articles that lean toward assertions of truthfulness (RQ2a), as indicated by a positive coefficient (β = .014 [.01; .02]). Government actors, in contrast, are more strongly associated with references to falsehood than non-government actors (RQ2b; β = −.017 [−.02; −.01]. The interaction between populist and government actors (RQ2c) yields a small but positive coefficient (β = .01 [.005; .013]), indicating that for populist actors who are also in government, the predicted level of mediated truth contestation increases more than what would be expected based on the sum of their individual effects alone. This suggests that populist government actors are more likely to be covered in the context of assertions of truthfulness than when they are not in government (see Figure 4).
Estimated marginal effect of populism on government and non-government actors.
EU-level political actors are more likely to be discussed in the news articles with references to falsehood than state-level actors (RQ2d), as indicated in a negative coefficient of β = −.014 [−.02; −.01], suggesting that European actors are more often embedded in mediated discourse highlighting falsehoods, misinformation, manipulation, or contested claims. Finally, Czechia is associated with a trend toward references to falsehood compared to Austria (RQ3), as supported by a negative coefficient of β = −.07 [−.09; −.06], which reinforces the descriptive findings on cross-national differences in mediated truth contestation.
Discussion and Conclusion
Truthfulness, factuality, and objectivity have become increasingly contested in contemporary political communication, extending well beyond populist discourse (Egelhofer et al. 2021; Kluknavská and Eisele 2023). This study examined how such contestation unfolds in mediated public discourse on migration in Austria and Czechia. By combining large-scale automated content analysis with qualitative validation, we analyzed the distribution of assertions of truthfulness and references to falsehood across media systems, political contexts, and actor types. Rather than treating truth contestation as a static feature of post-truth politics, our findings highlight its temporal volatility and contextual embeddedness across countries and media environments.
The over-time analysis reveals pronounced differences between Austria and Czechia. Czech news media exhibit marked oscillations between assertions of truthfulness and references to falsehood, whereas Austrian news media show comparatively less variability and a gradual shift toward assertions of truthfulness. These differences point to distinct national repertoires of mediated truth contestation. In Czechia, heightened volatility may reflect a more polarized epistemic environment in which journalistic authority and factual boundaries are repeatedly renegotiated, consistent with prior research documenting higher media distrust and exposure to disinformation narratives (Štětka et al. 2021). Austrian coverage, by contrast, appears more constrained by institutionalized journalistic norms that stabilize epistemic framing even in politicized debates. Qualitative examples at the extremes of the scale reinforce this contrast. Czech articles more frequently contrast “facts” and “reality” with explicit references to propaganda, hoaxes, or deception, while Austrian articles more often emphasize clarification, logical reasoning, and numerical evidence. The gradual increase in truthfulness assertions after 2015 may also indicate a form of journalistic self-reflection in response to intensified public concerns about misinformation and fake news (Egelhofer et al. 2020).
Across both countries, alternative news media, often operating from a symbolic epistemic periphery relative to mainstream journalism, exhibit substantially higher volatility than mainstream outlets, particularly in Czechia. This suggests that alternative news media function as dynamic arenas of epistemic struggle (Schwarzenegger 2023), where assertions of truthfulness and references to falsehood are mobilized more flexibly. Qualitative examples illustrate that alternative outlets often challenge dominant interpretations, question elites, or emphasize hidden realities, while referencing disinformation or manipulation. Mainstream news media reveal more stability, consistent with routinized professional practices oriented toward verification, balance, and institutional credibility (Hanitzsch and Vos 2018). Austrian upward trajectory toward truthfulness assertions and Czech consistent leaning toward falsehoods underscore how national contexts intersect with media type.
Within mainstream media, tabloids are more likely to highlight deception and manipulation, aligning with research on sensationalism and conflict-driven news values (Esser 1999; Otto et al. 2017). Broadsheets appear less inclined to emphasize falsehood references. These patterns indicate that mediated truth contestation is not merely actor-driven but is shaped by journalistic norms, editorial priorities, market logic, and commercial imperatives. From a market-oriented perspective, such framing may also resonate with audience expectations and enhance attention in highly contentious issue areas such as migration (Boukes and Vliegenthart 2020).
Turning to political actors, our findings show that populist actors are more frequently associated with assertions of truthfulness than non-populist actors. This finding challenges the prevailing tendency to conceptualize populists primarily as producers of accusations (Engesser et al. 2017; Kluknavská et al. 2025; Nai et al. 2021) by showing that mediated discourse also positions them as claimants to epistemic authority. This aligns with theoretical accounts of epistemic populism, which emphasize populists’ claims to exclusive access to the “real truth” of the people (Saurette and Gunster 2011; Kluknavská and Eisele 2023). Importantly, this pattern reflects journalistic framing rather than actors’ intentions. In the context of post-truth politics, where appeals to authenticity and emotional resonance often outweigh factual accuracy (Egelhofer and Lecheler 2019; Kluknavská and Eisele 2023), such portrayals carry significant normative implications. They may weaken journalism’s watchdog function by legitimizing populist truth claims, while simultaneously reflecting democratic responsiveness to publics who feel unrepresented by mainstream elites.
The interaction between populism and government affiliation further nuances this picture. Populist actors are more likely to be covered in articles that lean toward the truthfulness when in government than when in opposition, suggesting a partial taming or inclusion effect. Incumbency appears to redirect populist epistemic claims toward legitimacy and credibility, rather than abandoning these claims altogether. This finding aligns with research indicating that populists adjust their communication strategies once in office, prioritizing self-legitimation over confrontation (Bernhard 2020).
Government actors more broadly, regardless of populism, are covered with a tendency toward references to falsehood. This complicates prevailing assumptions that incumbents primarily rely on positive, achievement-oriented communication (Walter and Nai 2015; Nai 2020). This suggests that incumbency structurally increases exposure to epistemic scrutiny, situating government actors within discourses of contestation irrespective of their communicative intent. Incumbency may thus heighten exposure to accusations, defensive communication, or reactive framing, reflecting intensified expectations of accountability, credibility, and epistemic scrutiny, particularly in crisis-laden policy areas such as migration (Eisele et al. 2022).
EU-level political actors exhibit a similar pattern, being more often covered in contexts characterized by references to falsehood. This likely reflects the structural characteristics of EU politics, which are often covered through the lens of complexity, distance, and crisis (Eisele and Heidenreich 2025). In migration debates, EU institutions are frequent targets of criticism by national actors. They are simultaneously compelled to defend their legitimacy and policy effectiveness, resulting in mediated contexts saturated with epistemic contestation.
Our findings advance a relational understanding of mediated truth contestation. Rather than a uniform symptom of post-truth politics or a communication style confined to populist actors, truth contestation appears as a context-dependent feature of mediated political communication. It varies across national media systems, suggesting that epistemic volatility is embedded in broader political environments. It is also conditioned by media formats and journalistic norms, indicating that truth contestation is influenced by media logics rather than driven solely by political actors (Boukes and Vliegenthart 2020). However, it intersects with actor characteristics, particularly populism and incumbency, pointing to the mediated construction of epistemic populism (Hameleers 2024). By demonstrating these patterned variations, the study shifts the focus from individual actors’ claims to the broader communicative structures that shape how epistemic legitimacy is constructed and challenged in public discourse.
Regarding limitations, automated approaches necessarily abstract from rhetorical nuances and cannot fully capture communicative intent, irony, or the distinction between active accusations and passive attributions of (un)truthfulness within journalistic narratives. Because this approach relies on lexical indicators, it captures aggregate discursive orientations toward epistemic evaluation rather than the contextual meaning of individual statements. Accordingly, this study focuses on the presence and distribution of epistemic references rather than strategic use by actors. Future research could complement these findings through qualitative discourse analysis or actor-centered approaches to examine how truthfulness and falsehood are rhetorically constructed in concrete interactions.
Moreover, our study focuses on two countries with distinct languages and media systems. Future research should extend the analysis to additional contexts and explore how national repertoires of truth contestation interact with language-specific semantic conventions, as concepts such as “truth,” “reality,” or “fact” may carry distinct rhetorical connotations across languages. Comparative designs within the same language (e.g., German-language media in Austria and Germany) could help disentangle country effects from linguistic ones. Finally, while migration provides a particularly salient case in mediated truth contestation, other policy areas, such as climate change, public health, or security, may demonstrate similar or differing dynamics.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work of Alena Kluknavská on data collection and analysis was supported by the NPO “Systemic Risk Institute” no. LX22NPO5101, funded by European Union—Next Generation EU (Ministry of Education, Youth and Sports, NPO: EXCELES). The work of Alena Kluknavská on the draft and revisions was supported by the project “Research on Peripheries to Strengthen the Resilience of Czech Society,” Reg. No. CZ.02.01.01/00/23_025/0008727, co-funded by European Union (Ministry of Education, Youth and Sports of the Czech Republic, OP JAK). The work of Olga Eisele has been supported by the Austrian Press Agency (APA) via the APA-UNIVIE Data Project.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.