Abstract
Many western democracies have witnessed an upsurge of partisan alternative media that explicitly challenge the legitimacy of mainstream media and politics alike and promote populist discourses. Accordingly, alternative media are often discussed in relation to lower levels of political trust and support for populist parties. Yet, only a limited number of studies have investigated these relationships empirically. Drawing on survey data representative of the Danish adult population (n = 1,518) collected in January/February 2023, this study investigates how using alternative media is related to political trust, populist attitudes, and populist vote intentions. Unlike previous studies, it distinguishes between left-wing and right-wing alternative media users and finds notable differences: Users of right-wing alternative media have lower political trust, hold more populist attitudes, and have greater predicted probability of intending to vote for a right-wing populist party than non-users, but similar results were not found for users of left-wing alternative media. The study discusses potential explanations for this and how (different) alternative media might impact democracy.
Keywords
Introduction
Alternative media have been on the rise across western democracies over the past decade. Setting them apart from other forms of partisan news, these web-based news outlets do not only advocate one-sidedly in favor of specific political agendas but also explicitly challenge the legitimacy of mainstream media and politics alike (Barnidge and Peacock 2019). Previous studies have shown that they attack political opponents (e.g., Cushion 2021; Mayerhöffer and Heft 2021), voice explicit criticism of politicians (Brems 2023), and promote populist discourses (e.g., Boberg et al. 2020; Müller and Freudenthaler 2022; von Nordheim et al. 2019). Accordingly, alternative media are often discussed alongside dropping levels of political trust and soaring support for populism, not least because the upsurge of alternative media has coincided with populist election victories sweeping across the United States and many European countries. However, with some notable exceptions (Andersen et al. 2021a; de León et al. 2024; Müller and Bach 2021; Müller and Schulz 2021; Stier et al. 2020), studies empirically investigating how alternative news use is related to political trust, populist attitudes, and populist party choice are limited in number (see also Holt 2020; McDowell-Naylor et al. 2021b). Furthermore, it remains unexplored whether any such relationships are different for use of left-wing and right-wing alternative media as existing studies do not distinguish between left-wing and right-wing alternative news use. Especially in a European context, this distinction is highly relevant given the relatively greater success of right-wing than left-wing populism. Departing from a liberal-democratic perspective, this study explores these matters further with the aim of advancing understanding of how the proliferation of (different) alternative media might impact democracy.
Specifically, the study first investigates how using left-wing and right-wing alternative media is related to political trust and holding populist attitudes. Second, it investigates how using left-wing and right-wing alternative media is related to populist vote intentions, that is, whether alternative media users’ potentially lower levels of political trust and more populist attitudes translate into political action by the ballot box. These questions are investigated using Denmark as a case. The Danish context is interesting because levels of media and political trust are comparatively high (Newman et al. 2023; Stubager 2021) and both left-wing and right-wing populist parties represented in parliament are treated as legitimate political actors. In these regards, the Danish context differs from those of the United States, Germany, and Sweden in which much existing research has been carried out (Andersen et al. 2021a; Müller and Bach 2021; Müller and Schulz 2021; Stier et al. 2020) and may thus contribute new insights. The results show remarkable indications of a left-right divide in how using alternative media is related to the variables of interest: Users of right-wing alternative media have lower political trust, more populist attitudes, and greater predicted probability of intending to vote for a right-wing populist party than non-users, but similar results are not found for users of left-wing alternative media. The article concludes by discussing potential explanations for the observed difference and the democratic implications of the findings.
Political Trust, Populist Attitudes, and Democracy
According to liberal democracy theory, democracy derives its legitimacy from routinely held free and fair elections. They are the process through which the will of the people can be heard, as they give citizens the opportunity to pursue their individual interests by voting for the political candidates they feel best represent them (Baker 2001; Christians et al. 2009; Raeijmaekers and Maeseele 2015; Strömbäck 2005). Moreover, elections make it possible for the people to hold current powerholders accountable by “throwing the rascals out” (Baker 2001; Strömbäck 2005; Stubager et al. 2021). Liberal democracy theory thus places two main tasks on citizens: Choosing who shall govern the people among competing political alternatives and monitoring the conduct of current political powerholders (Moe et al. 2019; Strömbäck 2005).
On the one hand, liberal democracy thus requires that citizens trust the political representatives to whom they must delegate their political influence to act in their best interest (van der Meer and Zmerli 2017) and acknowledge them as legitimate representatives of the people. With this in mind, low political trust and populist attitudes can be seen as challenges to liberal democracy. On the other hand, the obligation of monitoring powerholders laid upon citizens arguably rests on an underlying understanding that politicians might be(come) corrupt. Thus, a certain amount of “suspicion of governmental power” (Rosanvallon 2008: 7) is built into this model of democracy. Accordingly, it is often emphasized that citizens’ trust in politicians should not turn into naiveté (e.g., Schudson 1998: 302; van der Meer 2017; van der Meer and Zmerli 2017). This inherent tension between the need for citizens to trust, yet remain critical of, their political representatives makes political trust and populism of key interest to liberal democracy.
Commonly, the literature distinguishes between diffuse trust in the political system and specific trust in the politicians in the system (Stubager et al. 2021: 189–91). As already indicated, the latter is what is of interest in this study, more specifically alternative news users’ trust in politicians. Political trust can be defined as a relationship between a trustor (citizens) and trustee (political institutions or actors) and expresses the extent to which individuals believe the object of trust (in the present study politicians) will meet certain expectations (e.g., Citrin and Stoker 2018; Geurkink et al. 2020). Trusting politicians thus involves vulnerability as citizens cannot be certain how politicians will act in the future and trusting them means running the risk of being betrayed (Citrin and Stoker 2018; van der Meer and Zmerli 2017; van der Meer 2017).
Populism falls into the category of essentially contested concepts (Mudde 2017; Wuttke et al. 2020). This study takes an ideational approach and considers populism a “thin” ideology (Mudde 2004). It has the advantage that it considers populism a set of ideas that politicians or parties and citizens can share (Andreadis et al. 2019) and which can be identified among citizens in the form of populist attitudes (ibid.; Fawzi 2019; Mudde 2017). Mudde (2004) defines populism as “[. . .] an ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite’, and which argues that politics should be an expression of the volonté générale (general will) of the people” (Mudde 2004: 543, emphasis in original). Populism thus implies a monolithic understanding of the people-as-one who share the same interests, juxtaposed by the deceitful elite who betray the interests of the people in favor of special interests (Mudde 2017).
Because populism is a thin ideology, it is, in practice, usually combined with a host-ideology, which determines the identity of the homogeneous people and the culprit elite (Golder 2016; Rummens 2017). Right-wing populism understands the identity of the people in ethno-cultural terms and creates homogeneity of the people through exclusion of specific groups, for example, based on culture, religion, or ethnicity. Left-wing populism understands the identity of the people through economic terms such as the working class and creates homogeneity through inclusion of minorities who are seen as part of the people to whom privileges and political rights should be extended (ibid.). Likewise, different groups can be identified as the culprit elite, for example, established political parties, mainstream media, and journalists (Golder 2016; Hansen 2021; Herrmann et al. 2021). This study considers in- or exclusion of minorities and attitudes toward the media/journalist elite attitudes in their own right (i.e., immigration attitudes and trust in media and journalists), separate from the core traits of populism, and conceives of populist attitudes as perceptions of a fundamental conflict between the people and the political elite (for a similar approach, see Stubager 2021).
Both (lack of) political trust and populist attitudes thus tap into opposition to the political elite. Setting the concepts apart, however (lack of) political trust does not capture the antagonistic relation between the pure people and the corrupt elite (Geurkink et al. 2020; Hameleers et al. 2017; Schulz 2019). The concepts are thus theoretically distinct and have also been found to be so empirically (Geurkink et al. 2020; Stubager 2021).
Alternative News Users and Their Attitudes Toward Political Powerholders
Over the past decade, alternative media have made their way into many western democracies and managed to establish considerable user bases (Heft et al. 2020; Newman et al. 2018). The phenomenon is probably most well-known in the United States with examples as Breitbart and Infowars, but alternative media are also on the rise in Europe (ibid.). As a common denominator, these online news sites claim to serve as a counterweight to a perceived media (and political) mainstream (Holt et al. 2019). While they are a heterogeneous group, they commonly feature explicit questioning of the credibility of mainstream media (Brems 2023; Cushion et al. 2021; Figenschou and Ihlebæk 2019) and disseminate partisan news content favoring their own ideological or, in some cases, party-political agendas (e.g., Benkler et al. 2018; Brems 2022; McDowell-Naylor et al. 2021a). Many recent studies focus on right-wing cases of alternative media and show that they often advocate immigration-critical agendas by portraying (Muslim) immigration as a threat or burden to European societies (Klawier et al. 2022; Nygaard 2019; von Nordheim et al. 2019).
Along these lines, a study on German-language alternative media found that the most extreme right-wing alternative media portrayed immigration as an invasion by a homogeneous Muslim culture whose clash with a German homogeneous culture will lead to civil war and portrayed political opponents that did not share right-wing values as threats to German culture. This study also included left-wing cases and found that one left-wing outlet had a similar harsh style and emphasized cultural homogeneity by invoking a working-class identity inclusive of refugees and portrayed political opponents, also moderate and left-wing politicians, as reactionary or racists (Freudenthaler and Wessler 2021). Much in line with this, other studies have shown that alternative media promote populist discourses (e.g., Boberg et al. 2020; Müller and Freudenthaler 2022; von Nordheim et al. 2019) and delegitimize political opponents (e.g., Cushion 2021; Mayerhöffer and Heft 2021; Mayerhöffer and Schwarz 2020; McDowell-Naylor et al. 2021a). This is also the case in Denmark where a study found explicit criticism of politicians in left-wing as well as right-wing cases of alternative media, in some of the right-wing outlets accompanied by radical distrust and construction of divides between the people and the elite (Brems 2023).
Thus, it is not without reason that alternative media are often associated with the rise of populism (e.g., Barnidge and Peacock 2019; Heft et al. 2020; Herrmann et al. 2021; Rae 2021). Indicating an overlap between alternative news users and populist citizens, previous studies have shown that users of alternative media trust mainstream media less (e.g., Andersen et al. 2021a; Schulze 2020) and populist citizens, likewise, evaluate mainstream media negatively (Fawzi 2019; Schulz et al. 2018). However, empirical studies that directly investigate the relationship between alternative news use and holding populist attitudes are relatively sparse (see also Holt 2020; McDowell-Naylor et al. 2021b). The same goes for studies investigating the association with political trust. As notable exceptions, a survey study from Sweden found that more orientation on alternative news is related to lower political trust (Andersen et al. 2021a) while a study combining panel surveys and web-tracking found that lower political trust predicted use of hyperpartisan, alternative, and conspiracy media (considered one group) among German and Swiss participants (de Léon et al. 2024).
Regarding populist attitudes, a survey study from Germany found a positive association between frequent, but not occasional, exposure to alternative media and holding populist attitudes (Müller and Schulz 2021). Similarly, a tracking-based study found positive associations between what was termed “hyperpartisan” news exposure and holding populist attitudes in Germany and the United States, but not in France, Italy, Spain, and the United Kingdom, indicating that the relationship might be dependent on the media and political context (Stier et al. 2020). Existing evidence is thus mixed. Furthermore, these previous studies did not distinguish between using left-wing and right-wing alternative media. Therefore, the present study asks: RQ1: How is left-wing and right-wing alternative news use related to a) political trust and b) holding populist attitudes?
Alternative News Use and Party-Choice
As previously noted, liberal democracy theory puts prime on elections as the act of voting is the key way in which citizens can channel their desires and grievances into the representative political system (Strömbäck 2005; Stubager et al. 2021). Moreover, casting the ballot demarcates an important difference between holding specific attitudes (e.g., toward politicians) and acting on them politically. For these reasons, an additional relevant question is whether alternative news users’ potential dissatisfaction with political elites translates into voting for populist political parties representing these attitudes. The aforementioned German survey study found a positive association between frequent exposure to alternative media and voting for the right-wing populist party Alternative für Deutschland (Müller and Schulz 2021). A tracking-based study similarly found a relationship between exposure to alternative media and greater probability of voting for AfD as well as for the left-wing populist party Die Linke, but results indicated a stronger association with right-wing populist party choice (Müller and Bach 2021). Adding to existing literature, this study distinguishes between using left-wing and right-wing alternative media and asks: RQ2: How is left-wing and right-wing alternative news use related to populist vote intentions?
The Danish Case
Unlike other countries, Denmark has not witnessed a decrease in news media trust over the past decade (Andersen et al. 2021b). A total of 58 percent of the population say they trust news media in general and approximately four in five say they trust the public service broadcasters DR and TV2 (Schrøder et al. 2022). At the same time, there is notable variation across political preferences with indications that right-wing voters are more critical of mainstream media than left-wing voters: 26 percent of citizens identifying as right-wing say mainstream media do not represent their views in a fair way compared to 17 percent of citizens identifying as left-wing (Schrøder et al. 2021) and especially voters of the right-wing populist party New Right have lower trust in mainstream media and journalists than the population average (Andersen et al. 2021b). Trust in politicians dropped from 2007 to 2015 but has since stabilized and is still comparatively high (Stubager and Kjøller 2024).
Danish public debate is described as more inclusive of what could be perceived as extreme political views than neighboring countries as Sweden and Germany, especially with respect to immigration (Heft et al. 2020). Three right-wing populist parties were elected for parliament in the most recent election in 2022 (the Danish People’s Party, New Right, 1 and the Denmark Democrats) and shifting center-right governments have relied on the former as a support-party over the past two decades. 2 Two left-wing populist parties also have representation in parliament (the Red-Green Alliance, and the Alternative) and the former recently acted as support-party for the 2019–22 Social Democratic government. Left-wing and right-wing populist parties are thus included in the parliamentary processes.
In these respects, the Danish media-political context differs from those of the United States, Germany, and Sweden in which many of the existing studies on the association between alternative news use and political trust and populist attitudes and party choice have been carried out (Andersen et al. 2021a; de León et al. 2024; Müller and Bach 2021; Müller and Schulz 2021; Stier et al. 2020). Trust in news media is markedly lower in the United States than in Denmark and also somewhat lower in Sweden and Germany (Newman et al. 2023). As noted, public debate in Germany and Sweden is described as less inclusive of what could be perceived as extreme political views (Heft et al. 2020, 2021) and center-right governments have not collaborated with populist parties Alternative für Deutschland and (until recently) the Sweden Democrats. This makes Denmark an interesting setting for the present study. On the one hand, the characteristics outlined should make the political mainstream relatively robust to challenges to its legitimacy. On the other hand, the electoral success of populist parties give reason to expect that alternative news use might be related to lower political trust, populist attitudes, and populist voting in the Danish case. If this is the case, any such associations can be expected to be more pronounced in dissimilar contexts where levels of trust in news media and politicians are lower and non-centrist points of view and political parties representing them are ostracized.
Data and Measurements
Sample
This study forms part of a larger research project on alternative news use in Denmark (see also Brems 2024) and draws on a subset of data from a survey fielded in early 2023 (17.1.23–7.2.23). Respondents were drawn from YouGov’s online panel, using quotas for sex, age, education, and geography to obtain a sample representative of the Danish population above the legal age (18 years). 3 In total, 3,116 panelists were invited and 2,735 of them completed the questionnaire (87.8 percent). A total of 280 of them were excluded because they finished the survey faster than four minutes (the median time for the rest of the participants was twice as high). The survey included a survey experiment, and this study only includes respondents who were not exposed to the experimental treatment conditions. This resulted in a final sample of n = 1,518. 4 Ethical approval was obtained from the relevant IRB before the data collection was carried out.
Measurements
Political trust was measured using the item “In general, you can trust our political leaders to make the right decisions for the country” (Hansen and Stubager 2020). Responses were measured on a seven-point Likert-scale (1 = “completely disagree,” 7 = “completely agree”) and the item rescaled to 0–1 where high values indicate high levels of trust. “Don’t know” responses were coded as missing and not included in the analysis.
Items tapping populist attitudes were phrased so that each of them refers to and taps the antagonist relationship between the political elite and the people and contains a clear juxtaposition of these two groups. This served the purpose of avoiding compensation between items tapping the notions of the corrupt elite and the homogeneous people separately (for a discussion of compensation see Wuttke et al. 2020). The three items were “There is a big gap between the political elite and the ordinary population in today’s Denmark”; “When politicians get elected to Parliament, they quickly lose touch with the everyday life of ordinary people”; and “It should be required that politicians have at least five years of work experience from the real world in order to get elected to Parliament.” Responses were measured on a seven-point Likert-scale (1 = “completely disagree,” 7 = “completely agree”). The three items were combined to form an index (α = .78) and rescaled to 0–1, where high values indicate more populist attitudes. “Don’t know” responses were coded as missing and only respondents with non-missing values on at least two items were included.
Vote intention was measured by inquiring respondents what party they would vote for if the election for Parliament were to be held today. As response categories, the fourteen parties that ran for Parliament in 2022 (the most recent general election at the time) and the following categories were provided: “Other party/candidate outside the parties,” “would not vote,” “would cast blank vote,” “do not have voting right,” “do not wish to answer,” and “don’t know.” Responses were coded as non-populist (0), left-wing populist (1), and right-wing populist (2) vote intention, and the variable treated as categorical. Based on Hansen (2021, 2024), the Danish People’s Party, New Right, and the Denmark Democrats were considered right-wing populist parties and the Red-Green Alliance, and the Alternative were considered left-wing populist parties. The recently launched party the Free Greens (not elected for parliament in 2022) was also considered left-wing populist. All other parties were considered non-populist. Respondents who chose any other response category than a party were coded as missing and not included in the analysis of vote intention.
The list-frequency technique (Andersen et al. 2016) was employed for tapping the independent variable of main interest, alternative news use. Respondents were asked how frequently (0–7 days) they used twelve Danish alternative news sites over the last seven days. The twelve alternative media were selected based on previous studies (Blach-Ørsten and Mayerhöffer 2021; Brems 2023) and included three left-wing (Solidaritet, Netavisen Pio, Konfront) and nine right-wing (NewSpeek, Kontrast, Indblik, Folkets Avis, Document, ditOverblik, Den Korte Avis, 180Grader, 24NYT) outlets (reflecting greater supply of the latter, ibid.). Respondents were coded as using no alternative media (0), using only left-wing alternative media (1), using only right-wing alternative media (2), or using both left-wing and right-wing alternative media (3). In total, 588 indicated they used alternative media at least one day over the last seven days (LW = 81, RW = 403, LW + RW = 104). 5
Using alternative media is positively related to using (some types of) mainstream media (Andersen et al. 2021a; Brems 2024; Müller and Schulz 2021), and previous studies have found associations between using different types of mainstream media and holding populist attitudes (Hameleers et al. 2017; Schulz 2019). Therefore, variables for use of three different types of mainstream news were included as controls. Specifically, respondents were inquired how often they used Denmark’s two national public service broadcasters (TV2, DR), the three legacy newspapers with the biggest reach (Politiken, Jyllands-Posten, Berlingske), and the two tabloids with the biggest reach (Ekstra Bladet, BT), also employing the list-frequency technique. The reported number of days of exposure to the individual news sources from each type of mainstream news were summed to form indexes reflecting use of public service news, legacy newspapers, and tabloids (with a range of 0–7).
Users of alternative media position themselves further toward the left and right on the ideological spectrum (Brems 2024; Newman et al. 2018). At the same time, ideological orientation is associated with political trust (Stubager 2021) and is one of the most important predictors of party choice (Stubager et al. 2021). Therefore, ideological orientation was included as control, measured as left-right self-placement on an eleven-point scale (0 = left-wing, 10 = right wing). 6 “Don’t know” responses were coded as missing and excluded from the analysis. According to some studies, users of alternative media have a strong political interest (Brems 2024; Schulze 2020). Following the notion that stronger political interest might also be related to holding stronger attitudes toward politicians, a variable for political interest was also included as control, measured on a four-point scale (from “very interested” to “not at all interested”) (Hansen and Stubager 2020). Due to skewed distributions, political interest was coded as an indicator variable. 7 Finally, sex, age, residence, and education were included as sociodemographic controls. Summary statistics can be found in Appendix A. The full questionnaire is supplied in the Supplemental Information file.
Analytical Approach
The first step of the analysis tests how alternative news use is related to political trust and holding populist attitudes. To do so, the indexes for political trust and populist attitudes were regressed on the variable for alternative news use and the controls outlined in two separate Ordinary Least Squares (OLS) regression models. The associations were tested in separate models because political trust and populist attitudes can be expected to be strongly correlated, and there is no theoretical basis for expecting that the one precedes the other in a causal chain and thus should be included in the model before the other. In these regression models, a quadratic term for ideological orientation was included because the outcome variables could be related to more extreme ideological positions in both directions. To account for skewed distributions on some of the variables, the OLS regression models were run with robust standard errors.
The second step of the analysis tests how alternative news use is related to populist vote intentions. This was done using multinomial logistic regression analysis where the variable for populist vote intention was modeled as outcome variable and alternative news use, the controls outlined above, and the variables for political trust and populist attitudes were modeled as predictor variables. Including the latter two variables makes it possible to test whether alternative news use is related to populist vote intentions even when controlling for political trust and populist attitudes, both of which can be expected to influence populist party choice.
Findings from the regression analyses will be displayed in figures in the analysis below, but tables with the full results from the regression models are available in Appendix B. The findings from the multinomial logistic regression analysis will be displayed as predicted average probabilities of populist vote intentions for the different groups of alternative news users and non-users. This has the advantage that the results can be interpreted independent of the reference category used and be understood as the predicted average probability each (non-)user group has of intending to vote for a left-wing or right-wing populist party expressed in percent. For a similar approach, see for example, Stubager et al. (2021).
Results
Figure 1 displays the relationships between using alternative media and political trust and populist attitudes, respectively, based on the OLS regression models outlined above. Beginning with RQ1a, the left-hand panel in the figure displays the predicted average levels of political trust for the different user-groups compared to non-users, controlled for the relevant background variables. As the figure displays, the results show that the users who only use right-wing alternative media have significantly lower predicted average levels of political trust compared to non-users. A similar pattern is not found for users who only use left-wing alternative media or users who combine left-wing and right-wing alternative media. Turning to the right-hand panel in Figure 1 and the question of how using alternative media is related to holding populist attitudes (RQ1b), a left-right divide similar to that for political trust can be observed, only more clear-cut. As the figure displays, the results from the OLS regression analysis show that the users who only use right-wing alternative media have significantly higher predicted average levels of populist attitudes compared to non-users. Unlike this, the users who only use left-wing alternative media have lower predicted average levels of populist attitudes compared to non-users. While the latter result falls short of significance by a whisker (p = 0.051), this is likely due to the relatively small size of the user-group. The results show no significant differences between users who combine left-wing and right-wing alternative media and non-users.

Predicted average levels of political trust and populist attitudes by alternative news use.
Political trust: N = 1294, R2 = 0.09. Populist attitudes: N = 1290, R2 = 0.11. The figure displays the predicted average levels of political trust (left panel) and populist attitudes (right panel) for the different groups of alternative news users compared to non-users. Higher values indicate higher levels of trust and more populist attitudes. The dashed horizontal line displays non-users’ predicted average levels of trust/populist attitudes. Controls: sex, age, residence, education, ideological orientation, political interest, and mainstream news use (public service, legacy newspapers, tabloids). CI error bars at 95 percent.
These findings have the important implication that, when looking at the Danish case, the relationships between using alternative media and lower political trust and holding more populist attitudes identified in previous studies (see above) seem only to pertain to use of right-wing alternative media and only when not used in combination with left-wing alternative media. It should, however, be noted that while significant, the differences between users of right-wing alternative media and non-users are not big, indicating only moderate associations. Relatedly, it is notable that the users’ predicted average levels of political trust do not fall far below the midpoint of the scale.
The final section of the analysis turns to RQ2 and the relationship between alternative news use and left-wing and right-wing populist vote intentions. In response to this, Figure 2 displays the predicted average probabilities of intending to vote for a left-wing populist and right-wing populist party for non-users and the different groups of alternative news users, based on the multinomial logistic regression model outlined above. As the right-hand panel in the figure displays, the findings show that users who only use right-wing alternative media have significantly greater predicted average probability of intending to vote for a right-wing populist party compared to non-users. Specifically, the results suggest that, when holding all other variables in the model at their observed values, the predicted average probability of intending to vote for a right-wing populist party is approximately 17 percent for non-users compared to 27 percent for users who only seek out right-wing alternative media, a sizeable difference of 10 percent points.

Predicted average probabilities of left-wing populist and right-wing populist vote intentions by alternative news use.
The findings thus suggest that users of right-wing alternative media do not just have lower political trust and hold more populist attitudes compared to non-users. They also (say they intend to) act on their dissatisfaction with the political elite and channel their dissatisfaction with politicians into the representative political system by voting for right-wing populist parties representing such sentiments. This is not surprising as antielitist attitudes are an important predictor of right-wing populist vote cast (Hansen 2021). It is notable, however, that using right-wing alternative media predicts right-wing populist vote intentions even when controlling for political trust and populist attitudes.
The findings displayed in the right-hand panel of the figure also indicate that users who combine left-wing and right-wing alternative media have higher predicted average probability of right-wing populist vote intention compared to non-users, but not significantly so. Looking at the left-hand panel in Figure 2 and the predicted average probabilities of left-wing populist vote intention, the results displayed show a tendency that users of only left-wing alternative media have slightly greater predicted average probability of left-wing populist vote intention compared to non-users. However, the difference is not statistically significant. The findings thus suggest that there is a significant relationship between using right-wing alternative media and (intentions of) voting for right-wing populist parties, but not when combined with left-wing alternative media, and that there is not a similarly strong relationship between using left-wing alternative media and (intentions of) voting for left-wing populist parties. This is in line with previous studies that have found an association between frequent alternative news use and voting for the right-wing populist party Alternative für Deutschland in Germany (Müller and Schulz 2021) and that alternative news use is more strongly associated with right-wing than left-wing populist party choice in Germany (Müller and Bach 2021).
Discussion and Conclusion
This article set out to explore the relationships between alternative news use and political trust, holding populist attitudes, and populist vote intentions. Altogether, the findings presented indicate a left-right divide where only using right-wing alternative media is significantly related to lower political trust, holding more populist attitudes, and greater probability of voting for (right-wing) populist parties, whereas using left-wing alternative media is not, at least in the Danish case.
A central limitation to the study is that the causal direction of the associations identified cannot be decided based on the cross-sectional data, making it difficult to determine what explains the observed difference. Based on existing evidence, at least three causal paths are relevant to consider. First, while a recent study showed that both left-wing and right-wing Danish alternative media voice explicit criticism of politicians it also found that three of the right-wing alternative media included in this study stood out from the rest by voicing radical distrust in the political system (Brems 2023). A supply-side/media-effects explanation could thus be that users of (some) right-wing alternative media are exposed to content more extremely critical of politicians than users of left-wing alternative media, causing them to lose trust in politicians, adopt more populist attitudes, and become more inclined to vote for populist parties. Second, as noted above, there are indications that Danish right-wing voters are more critical of mainstream media and journalists than left-wing voters (Andersen et al. 2021b; Schrøder et al. 2021). Departing from the notion of differential susceptibility to media effects (Valkenburg and Peter 2013), another explanation could thus be that right-wing voters’ lower trust in mainstream media makes them more prone to influence from alternative media. Third, it could be a demand-side/selective exposure (Stroud 2011) explanation: Because (some) right-wing alternative media are more critical of the political elite, they appeal more to the part of the population that already has lower trust in politicians, hold more populist attitudes, and are more inclined to vote for populist parties.
To clarify whether the suggested causal paths hold up, we need more studies that directly compare left-wing and right-wing alternative news content focusing on possible differences in severity of their criticism of politicians/the political system and use of populist rhetoric. Furthermore, future studies should rely on panel data or experiments to test (a) the effects of exposure to alternative content with different degrees of extremity in terms of political criticism and populist rhetoric; (b) whether effects of exposure to alternative news content are stronger for right-wing than left-wing voters; and (c) whether citizens with lower political trust and more populist attitudes are more prone to selectively expose themselves to content critical of politicians.
Other than the traits of Danish alternative media and voters discussed above, the observed difference between users of left-wing and right-wing alternative media is possibly influenced by the Danish context where right-wing alternative media and populist parties have a stronger foothold than their left-wing counterparts: Recent findings indicate greater supply of (Blach-Ørsten and Mayerhöffer 2021) and demand for (Brems 2024) right-wing than left-wing alternative media in Denmark, and right-wing populist parties receive higher vote-shares than left-wing populist parties (14.4 vs. 9.4 percent at the latest general election). It also aligns with prior findings that the political attitudes of Danish voters of right-wing populist parties are more distinct from the rest of the population’s (Hansen 2021).
Interestingly, many of the characteristics discussed that might explain the observed difference between users of left-wing and right-wing alternative media can also be found in other countries and across different media systems. In recent years, right-wing populist parties have had greater success than left-wing populist parties in many European countries (Pew Research 2022), and greater demand for right-wing than left-wing alternative media has also been observed in the United States (Benkler et al. 2018; Lorenz et al. 2023). Likewise, the tendency that right-wing voters have lower trust in mainstream news can be observed in other countries, for example, in Germany and the United States (Newman et al. 2021). While few studies compare left-wing and right-wing alternative media content, the above-cited study on German-language alternative media found extremity and populist rhetoric in several right-wing outlets but in only one left-wing outlet (Freudenthaler and Wessler 2021). Also, evidence from Flemish alternative media showed that right-wing alternative media gave more visibility to right-wing populist actors, the equivalent of which was not found in left-wing outlets (Buyens and van Aelst 2022). This indicates that differences in extremity of content are also present in other cases. Additionally, especially right-wing populist parties in many countries attack mainstream media, for example, AfD in Germany, the Sweden Democrats in Sweden (Sehl et al. 2022), and Trump-led Republican Party in the United States (Benkler et al. 2018). Across countries, right-wing populist parties and right-wing alternative media thus fish for voters/readers in the same pool of citizens dissatisfied with mainstream media. Altogether, this could give reason to expect that the difference between users of left-wing and right-wing alternative media observed in this study might also be found in other cases. Conversely, the picture might look different in dissimilar contexts, for example, where the populist left-wing is stronger. Future studies should investigate empirically whether a left-right divide can also be found outside the borders of Denmark.
Regarding the relationships between using right-wing alternative media and political trust and populist attitudes, the findings showed that the differences between right-wing alternative news users and non-users, while significant, were not big, and that the users’ levels of trust did not fall far below the midpoint of the scale. This, too, should be understood in relation to the Danish case. When what could be considered extreme right-wing views and populist parties are included in public debate (Heft et al. 2020) and parliamentary work, the part of the population that alternative media appeal to might be less critical of the political elite and thus less different from the rest of the population than in other countries. If this holds true, the associations between right-wing alternative news use and lower political trust, more populist attitudes, and greater probability of populist vote intention identified in the present study can be expected to be stronger in contexts where extreme views and actors are ostracized like in Sweden and Germany. The small differences could also be explained by the timing of the data collection, which coincided with heated public debate on the Danish government’s decision to cancel a public holiday, a policy that the majority of the population did not support. 8 The salience of the issue might has lowered the general levels of political trust and heightened levels of populist attitudes in the population at that time, which could make the differences between users and non-users appear smaller.
The identified differences in how left-wing and right-wing alternative news use is related to political trust, populist attitudes, and populist vote intentions suggest that, in these regards, right-wing alternative media might have a greater democratic impact than their left-wing counterparts. Importantly, though, it should be emphasized that it is not necessarily a negative democratic impact. Rather, this is a matter of interpretation. On the one hand, low(er) levels of political trust can be considered concerning from a liberal-democratic perspective, since this representative model of democracy requires that citizens trust the politicians to whom they must delegate their political influence (e.g., van der Meer 2017; van der Meer and Zmerli 2017). Moreover, it can be argued that populism and populist attitudes constitute a fundamental challenge to liberal democracy because the notion of an inherent conflict between the people and the corrupt elite questions the legitimacy of politicians as representatives of the people (e.g., Norris 2020). From this perspective, right-wing alternative media constitute or reflect a greater challenge to democracy than left-wing alternative media.
On the other hand, the liberal model of democracy requires that citizens critically monitor politicians (e.g., Rosanvallon 2008; Schudson 2008) and right-wing alternative news users’ lower political trust could thus be interpreted as necessary skepticism. Moreover, it has been argued that populist attitudes (and lower political trust, one might add) should not necessarily be understood as a democratic challenge in and of itself but could be seen as reflections of underlying democratic problems (Aalberg and de Vreese 2016), in the present case that users of right-wing alternative media do not think politicians do a proper job of representing their views and taking care of the public’s interests. Following this, it could be considered a sign of democracy working when users of right-wing alternative media hold political powerholders accountable by voting for parties representing their grievances—and thus let their disenchantment with the political elite find its outlet within the confinements of the representative political system. Which interpretation is more correct is not easily settled empirically, as it is a question of how liberal democracy theory is interpreted. This study takes the position that both perspectives have theoretical merit. To empirically investigate normative implications of alternative news use, future studies could explore whether users of alternative media entirely turn their backs on or reject the representative political system, manifesting in for example, abstaining from voting or refusing to recognize political decisions as legitimate. Such findings would be concerning from both perspectives outlined.
Furthermore, to understand the democratic implications of alternative news users’ populist vote cast, it is also relevant to investigate what policies populist parties implement once elected. Do they serve as correctives of established parties or do they compromise fundamental liberal-democratic values such as pluralism, inclusive citizenship, protection of minority rights, independence of courts, and freedom of the press (Galston 2018)? Such assessments would call for investigation of the policies of populist parties rather than surveys of alternative news users or populist voters’ attitudes.
As noted, the study has the central limitation that causality cannot be determined based on the cross-sectional data. Additionally, the study relied on self-reported news use, which could lead to over-reporting (Prior 2009) as well as under-reporting (Müller and Schulz 2021) of alternative news use. Therefore, it cannot be ruled out that the measurement has captured respondents who in fact do not use alternative media or has not captured all users. Such issues could potentially influence the results by making users appear less different from non-users. Even with these limitations, the study has provided valuable new empirical insights into the associations between use of left-wing and right-wing alternative media and political trust, populist attitudes, and populist vote intentions on which future studies can build and demonstrated the relevance of distinguishing between different types of alternative news use.
Supplemental Material
sj-do-2-hij-10.1177_19401612241301647 – Supplemental material for A left-right divide? Alternative news use and political trust, populist attitudes, and populist vote intentions in the case of Denmark
Supplemental material, sj-do-2-hij-10.1177_19401612241301647 for A left-right divide? Alternative news use and political trust, populist attitudes, and populist vote intentions in the case of Denmark by Miriam Kroman Brems in The International Journal of Press/Politics
Supplemental Material
sj-docx-1-hij-10.1177_19401612241301647 – Supplemental material for A left-right divide? Alternative news use and political trust, populist attitudes, and populist vote intentions in the case of Denmark
Supplemental material, sj-docx-1-hij-10.1177_19401612241301647 for A left-right divide? Alternative news use and political trust, populist attitudes, and populist vote intentions in the case of Denmark by Miriam Kroman Brems in The International Journal of Press/Politics
Footnotes
Appendix A: Summary Statistics
Variables in the Analysis.
| Variables | M | SD | Min | Max | Percent | N |
|---|---|---|---|---|---|---|
| Ideological orientation | 5.43 | 2.60 | 0 | 10 | 1,320 | |
| Public service news usage | 4.61 | 2.25 | 0 | 7 | 1,518 | |
| Legacy newspaper usage | 1.06 | 1.48 | 0 | 7 | 1,518 | |
| Tabloid news usage | 2.35 | 2.47 | 0 | 7 | 1,518 | |
| Political trust | 0.46 | 0.29 | 0 | 1 | 1,447 | |
| Populist attitudes | 0.75 | 0.23 | 0 | 1 | 1,449 | |
| Gender | ||||||
| Female | 50.1 | 760 | ||||
| Male | 49.9 | 758 | ||||
| Age | ||||||
| 18–29 | 12.1 | 183 | ||||
| 30–39 | 13.8 | 209 | ||||
| 40–49 | 14.7 | 223 | ||||
| 50–59 | 20.7 | 314 | ||||
| 60+ | 38.8 | 589 | ||||
| Level of education | ||||||
| Lower Secondary | 12.8 | 194 | ||||
| General Upper Secondary | 9.8 | 149 | ||||
| Vocational | 44.5 | 675 | ||||
| Short-Cycle Tertiary | 5.8 | 88 | ||||
| Medium-Cycle Tertiary | 18.0 | 273 | ||||
| Long-Cycle Tertiary or above | 9.2 | 139 | ||||
| Residence | ||||||
| Outer municipalities | 7.2 | 110 | ||||
| Rural municipalities | 25.9 | 393 | ||||
| Middle-range municipalities | 17.1 | 259 | ||||
| City municipalities | 49.8 | 756 | ||||
| Political interest | ||||||
| Not at all interested | 6.2 | 94 | ||||
| Only a little interested | 26.7 | 405 | ||||
| Somewhat interested | 43.3 | 657 | ||||
| Very interested | 22.7 | 345 | ||||
| Do not wish to answer | 1.1 | 17 | ||||
| Alternative media use | ||||||
| Use no alternative media | 61.3 | 930 | ||||
| Use left-wing alternative media | 5.3 | 81 | ||||
| Use right-wing alternative media | 26.5 | 403 | ||||
| Use left-wing and right-wing alternative media | 6.9 | 104 | ||||
| Vote intention | ||||||
| Not populist party | 69.0 | 814 | ||||
| Left-wing populist party | 9.2 | 109 | ||||
| Right-wing populist party | 21.8 | 257 | ||||
| Total N | 1518 | |||||
Note. The education category “vocational” includes the technically and commercially oriented upper secondary educations (in Danish HTX and HHX). Residence categories are based on the classification of Danish municipalities in Kristensen et al. (2007). In the survey, the small islands Fanø, Læsø, Langeland, Samsø, and Ærø were collapsed into one category. Respondents belonging to this category were coded as outer municipalities, as four of the five islands belong to this category in Kristensen et al. (2007).
Appendix B: Tables With Regression Results
Multinomial Logistic Regression Model Predicting Probability of Populist Vote Intentions.
| Predictor variables | LW-pop |
RW-pop |
|---|---|---|
| Alternative news use (**) | ||
| Use left-wing alt. med. | 0.66 (0.45) |
−0.21 (0.57) |
| Use right-wing alt. med. | 0.19 (0.36) |
0.84***
(0.21) |
| Use left-wing and right-wing alt med. | 0.03 (0.51) |
0.54 (0.39) |
| Gender | ||
| Male | 0.29 (0.29) |
0.10 (0.20) |
| Age (*) | ||
| 30–39 | 0.14 (0.46) |
0.34 (0.44) |
| 40–49 | −0.43 (0.52) |
0.78 (0.42) |
| 50–59 | −0.40 (0.49) |
1.07**
(0.40) |
| 60+ | −0.99*
(0.47) |
0.76 (0.39) |
| Education (**) | ||
| General Upper Secondary | 1.43 (0.76) |
−0.21 (0.40) |
| Vocational | 0.91 (0.69) |
−0.63*
(0.27) |
| Short-Cycle Tertiary | 0.15 (0.95) |
−0.66 (0.45) |
| Medium-Cycle Tertiary | 1.10 (0.72) |
−1.41***
(0.36) |
| Long-Cycle Tertiary or above | 1.54*
(0.76) |
−1.11*
(0.44) |
| Residence (*) | ||
| Rural municipalities | −0.58 (0.76) |
−0.83*
(0.34) |
| Middle-range municipalities | 0.17 (0.77) |
−0.49 (0.36) |
| City municipalities | 0.43 (0.70) |
−0.82*
(0.32) |
| Political interest | ||
| Only a little interested | 0.32 (1.23) |
−1.10*
(0.55) |
| Somewhat interested | 0.48 (1.22) |
−1.25*
(0.54) |
| Very interested | 0.77 (1.24) |
−1.65**
(0.58) |
| Do not wish to answer | −13.48 (1569.74) |
−15.41 (1011.42) |
| Ideological orientation | −0.64***
(0.07) |
0.31***
(0.04) |
| Public service news usage | −0.06 (0.07) |
−0.02 (0.05) |
| Legacy newspaper usage | 0.19*
(0.09) |
−0.09 (0.07) |
| Tabloid news usage | −0.21**
(0.07) |
−0.00 (0.04) |
| Populist attitudes | −0.27 (0.69) |
0.67 (0.51) |
| Political trust | −1.51**
(0.56) |
−2.93***
(0.39) |
| Intercept | 0.23 (1.64) |
−0.67 (0.90) |
p < 0.001, **p < 0.01, *p < 0.05. N = 1,056. McFadden’s pseudo-R2 = 0.32. Nagelkerke’s pseudo-R2 = 0.50. Chi2(52) = 543.76. Reference categories: Party choice (Not populist party), Gender (Female), Age (18–29), Education (Lower secondary), Residence (Outer municipalities), Political interest (Not at all interested). Standard deviations in parentheses. SD for the category in the political interest variable “do not wish to answer” are high due to few respondents in the category. For entire categorical variables, levels of significance are based on joint Wald tests for all categories simultaneously.
Acknowledgements
The author would like to thank Rune Stubager (Aarhus University) for valuable feedback on earlier versions of the manuscript.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The data collection was supported by Aarhuus Stiftstidendes Fond (021738-0010), Dagspressens Fond (J.nr. 2/21), and Familien Hede Nielsens Fond (no grant number) as well as the Graduate School, Arts at Aarhus University. Grant givers have had no influence on the research whatsoever.
Ethical Considerations
Ethical approval was obtained from the Institutional Review Board at Aarhus University prior to data collection.
Consent to Participate
All participants gave written consent to participate before answering the survey questionnaire.
Consent for Publication
Not applicable.
Gender Balance
Approximately a third of the references have a female first author.
Data Availability
Data is available upon request.
Supplemental Material
Supplemental material for this article is available online.
Notes
Author Biography
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
