During the last decade, right-wing alternative media have gained public and scholarly attention due to their controversial positions in the media landscape and speculations that they might contribute to polarisation and radicalisation. The term “alternative media” encompasses a broad range of semiprofessional to nonprofessional actors motivated by different ideological ideas, but broadly, it refers to news outlets that openly proclaim they want to be a “corrective” to the perceived mainstream, and that are critical of the established press and cultural elites (Holt et al., 2019). Alternative media coming from a right-wing position share an interest in covering political topics deemed important to the far right, such as immigration, multiculturalism, Islam, or crime. While this kind of media can contribute to diversity by giving voice to sources and topics that are consciously or unconsciously ignored by the established press, they are also considered problematic by some due to their antagonistic criticism of elites, unclear relationship to journalistic ethics, provocative reporting style, and promotion of exclusionary viewpoints (Figenschou & Ihlebæk, 2019; Ihlebæk et al., 2022; Mayerhöffer, 2021; Nygaard, 2021; Pedersen et al., 2024). Furthermore, they may in some instances also contribute to the spread of conspiracy theories and rumours (Frischlich et al., 2021, 2023; Staender et al., 2024).
Existing studies on right-wing alternative media audiences have brought to light the complexity of motivations that guide the use of these media (Rauch, 2007; Schwarzenegger, 2021, 2023), still indicating that distrust in the mainstream media and immigrant-critical attitudes function as strong predictors (Brems, 2024; Schulze, 2020). Despite increasing knowledge about how political attitudes, trust, and right-wing alternative media use are interrelated (Theorin & Strömbäck, 2020; Tsfati et al., 2024), there is still a need for comparative studies that shed light on similarities and differences between various media and political systems (but see Schulze, 2020; Stier et al., 2020). The present article contributes to filling this gap by providing the first fully comparable study of right-wing alternative media users in the three Scandinavian countries. The study furthermore adds to knowledge about how a broad spectrum of political factors and forms of trust are linked with right-wing alternative media use, thus making it possible to assess their relative importance. Such analyses are important for deepening our understanding of what right-wing alternative media use represents as a political force and its democratic implications.
We base our analysis on an original population-representative survey of citizens in Denmark, Norway, and Sweden, including questions measuring political, media, and social trust, party affiliation, and a broad spectrum of far-right beliefs (populism, authoritarianism, anti-immigration, and anti–gender equality attitudes). In line with previous research, we start from the basic assumption that the presence and impact of alternative media – along with users’ understanding of them – should be understood as a reaction to a country’s political information environment (Cushion et al., 2021; Heft et al., 2020; Holt, 2016; Steppat et al., 2023; Stier et al., 2020). Thus, it makes sense to study a specific region and to use similarities and differences between the three countries to underscore the importance of context. Scandinavia lends itself well to such an exploration. On the one hand, the three countries share many institutional traits and are considered to belong to the same “media welfare state” model (Syvertsen et al., 2014). On the other, they have over time varied in the importance and place of right-wing alternative media within the media systems (Ihlebæk & Nygaard, 2021; Nygaard, 2020; Mayerhöffer et al., 2024; Pedersen et al., 2024), and in how political issues, such as immigration, have been covered in the news (Hovden & Mjelde, 2019; Mjelde & Hovden, 2019) and in political discourse (Brochmann & Hagelund, 2012).
We first ask what characterises right-wing alternative media users in terms of political attitudes and political, media, and social trust, and second, which of these factors is most strongly associated with right-wing alternative media use. For both questions, we investigate whether there are distinct patterns in Denmark, Norway, and Sweden. The results show that those who regularly consume news from alternative media sources have lower levels of various forms of trust, they often identify with far-right political parties, and they are likelier to support far-right ideas. Considering all factors, low trust in traditional media and support for far-right ideas are the most important factors. Despite significant cross-country differences in alternative media penetration, these patterns are recognisable in all three countries. However, right-wing alternative media users represent a less distinct group in Denmark, in terms of ideology and trust.
The three Scandinavian countries are stable, small-state democracies with generous welfare systems, high levels of trust and equality among citizens, and relatively low degrees of political conflict (Brochmann & Hagelund, 2012). Even though trust levels overall are high, the Reuters Institute Digital News Reports (Newman et al., 2018, 2019) indicate that generalised media trust is comparably lower in Sweden then in Denmark and Norway. All exhibit strong media policies; proximity between policymakers and the media industry; a strong self-regulating journalistic profession; and high levels of news consumption in the population (Brüggemann et al., 2014; Hallin & Mancini, 2004; Ihlebæk et al., 2024; Syvertsen et al., 2014). Their journalistic cultures are relatively similar in terms of ethics and work practices (Ahva et al., 2017; Hovden et al., 2016).
In the context of this article, some important country-level differences are worth highlighting. A core point is that right-wing alternative media is a more widespread phenomenon in Sweden than in Denmark and Norway in terms of the number of influential sites, their readership, and their roles in online digital networks. Scholars have pointed to factors related to both the political and the media context to explain why alternative media thrive (Brems, 2022, 2024; Cushion et al., 2021; Heft et al., 2020; Ihlebæk & Nygaard, 2021; Mayerhöffer et al., 2024; Newman et al., 2019; Schulze, 2020). Politically, Sweden has historically had the most liberal approach to immigration, and, until recently, Sweden’s political discourse on immigration was more restricted, with immigration-critical voices less visible than in the other two Scandinavian countries (Brochmann & Hagelund, 2012). The populist immigration-critical party, the Sweden Democrats, was also long excluded from coalitions through a political “cordon sanitaire”, a factor that has previously been associated with a stronger gate-keeping role by journalists (Wettstein et al., 2018).
Heft and colleagues (2020) furthermore suggest that the supply of right-wing alternative media must be seen in the context of the wider media landscape, and particularly how the mainstream media covers certain topics important to the far right. They argue that the low number of right-wing alternative media in Denmark is linked to the fact that the Danish press has had a more vocal, critical, and tabloid debate about immigration and Islam, especially in the aftermath of the Muhammad cartoons published in Jyllands-Posten in 2005, limiting the potential for novel immigration-critical media actors to enter this field. Compared with Denmark, the Norwegian and particularly the Swedish news media have historically provided less room for immigration-critical voices (Hovden & Mjelde, 2019; Hovden et al., 2018). This may have impacted the potential for new niche and alternative actors to grow (Holt, 2016). It is important to stress, however, that the comparative results must not be mistaken for a non-critical press in Sweden or Norway. Single-nation studies on the news media and immigration provide a much more complex picture of national discourses on immigration (Alghasi et al., 2020; Strömbäck et al., 2017). Finally, studies further indicate that there has been a harsher relation between right-wing alternative media and the wider journalistic field in Sweden than in the other two countries, which indicates a more tense media environment (Nygaard, 2021; Pedersen et al., 2024).
Within a quickly growing research field, scholars have studied the characteristics of alternative media audiences, including rituals and media habits (Rauch, 2007; Schwarzenegger, 2021, 2023); interpretations, motivations, and gratifications (Rauch, 2015, 2020; Noppari et al., 2019; Steppat et al., 2023); and political attitudes and trust (Andersen et al., 2021; Brems, 2024; Müller & Schulz, 2019; Schulze, 2020; Reiter & Matthes, 2021). A key takeaway is that alternative media use is linked to high news media consumption and social media use (Brems, 2024; Leung & Lee, 2014; Müller & Schulz, 2021; Schulze, 2020). Furthermore, alternative media users share a strong interest in politics (Brems, 2024; Leung & Lee, 2014; Reiter & Matthes, 2021; Schulze, 2020).
To interpret and explain why citizens use right-wing alternative media, several theoretical mechanisms have been discussed. The theory of “selective exposure” starts from the assumption that users are rational in their media use, and in a context of complexity will turn to certain media because they support existing political attitudes and beliefs (Skovsgaard et al., 2016; Stroud, 2011) or because they already trust them (Tsfati & Cappella, 2003). Right-wing alternative media use may also be explained by “media effects”, assuming that people are influenced by what they read, which in turn increases or sustains a certain practice, level of trust, and attitude (Shehata et al., 2023). This literature suggests that by being exposed to mainstream journalistic content, readers may increasingly come to trust it, because of trust-building journalistic practices, including formats for presentation and argumentation (Nelson & Kim, 2021; Tsfati et al., 2024). Conversely, right-wing alternative media may have a negative effect on trust in mainstream media, since media criticism is at the core of many right-wing alternative media’s projects (Figenschou & Ihlebæk, 2019, 2025; Heft et al., 2020; Mayerhöffer, 2021). Finally, building on the influential “reinforcing spirals” model (Slater, 2007, 2015), there is the possibility that a reciprocal relationship exists between initial dispositions such as political attitudes and trust. In other words, people may turn to right-wing alternative media because they correspond to their political views and because they trust them (and distrust mainstream media), and the use may reinforce or maintain these attitudes (Slater, 2015; Tsfati et al., 2024).
Given the cross-sectional nature of our study, we are not in a position to test which of these mechanisms are at play when people use right-wing alternative media. But in the formulation of our hypotheses, we still make use of the fact that they all postulate an association between political attitudes, trust, and right-wing alternative media use, either through selectivity or through media effects. These models also have different implications for how to evaluate the democratic consequences of what we find, which we return to in the discussion.
In delimiting far-right ideology, scholars tend to point to defining ideological features, such as authoritarianism or a strong state, nationalism, racism and xenophobia, antidemocracy and populism, and anti-establishment (Carter, 2018). As some of these characteristics overlap, scholars have categorised, combined, and collapsed them in different ways (e.g., Carter, 2018; Mudde, 2007). Indeed, concerning the role of (right-wing) political attitudes in relation to alternative media use, studies have used different measures and reported varying results. Investigating German Internet users, Müller and Schulz (2021) found that frequent use of alternative media is connected to populist attitudes and a sense of personal deprivation or a negative view regarding one’s own opportunities in society. Brems’s (2024) study on Danish alternative media users revealed a significant association between alternative media use (both left and right) and ideological beliefs. Associations have also been found between right-wing alternative media use and anti-immigration attitudes, showing that such use reinforces negative attitudes, in particular towards perceived remote immigration (Theorin, 2019; Theorin & Strömbäck, 2020). In a cross-national survey, Schulze (2020) found that while alternative media use is strongly related to political interest and immigration-critical attitudes, it was unrelated to support for direct democracy or self-reported right-wing–leaning orientation. Finally, in a study of those who used hyper-partisan, alternative, and conspiracy media during the Covid-19 pandemic, de León and colleagues (2024) found that they were placed at both the left- and the right-wing extremes.
In the set of different approaches to studying the political attitudes of right-wing alternative media users, there is a lack of studies exploring the full range of far-right ideas while distinguishing between those who hold moderate and strong support for these. In our study, we apply a broad measure that includes populism, anti-immigration, right-wing radical, and right-wing extremist ideas (see the methods section for further description). Given the anti-elitism and critical stance towards immigration that right-wing alternative media have taken in the Scandinavian countries, it is likely these media will appeal to people holding such views. We seek to capture affiliation to far-right ideology as expressed both in far-right party affiliation and in adherence to a broad set of far-right ideas. As studies have also reported that support for some aspect of far-right ideology is linked to alternative media use, we formulate the following hypotheses:
H1a. Far-right party-affiliation is associated with right-wing alternative media use. H1b. Support for far-right ideas is associated with right-wing alternative media use.
When it comes to expected comparative differences, we formulate the following hypotheses:
H2a. The relationship between far-right party affiliation and right-wing alternative media use is weaker in Denmark than in Norway and Sweden. H2b. The relationship between support for far-right ideas and right-wing alternative media use is weaker in Denmark than in Norway and Sweden.
Since the market potential for right-wing alternative media is more limited in Denmark than in the two other countries, given how immigration has been covered in the mainstream news, we assume that alternative media users in Denmark might, ideologically speaking, be more similar to traditional media users.
Trust in its various forms is considered vital to the functioning of democracy (Norris, 2017; Uslaner, 2003), and it is often described as a particular resource within the Scandinavian welfare states (Rothstein, 2001). Media trust can be understood as a relationship involving citizens and news media, where citizens act under a condition of uncertainty and expect to gain more than they lose from devoting time to and relying on the media in question (Strömbäck et al., 2020). Media scholars have argued that for people to read and consume news media, they need to trust them (Fletcher et al., 2017; Strömbäck et al., 2020; Tsfati & Ariely, 2014; Tsfati & Cappella, 2003). Several studies have shown that lower trust in mainstream media is related to right-wing alternative media use (Andersen et al., 2021; Strömbäck et al., 2020; Fletcher & Park, 2017; Leung & Lee, 2014; Schulze, 2020). Based on a four-wave panel survey in Sweden, Andersen and colleagues (2021) found that decreasing trust in mainstream media was related to growing alternative news orientation over time and that this relationship was reciprocal, thus creating a reinforcing spiral. A panel study by Tsfati and colleagues (2024) found similar dynamics, showing that right-wing alternative media use and trust grew reciprocally over time, while trust in mainstream media decreased.
So far, the main focus in the literature has been on media trust, but there is an emerging interest for including political and social trust in the study of right-wing alternative media use (Andersen et al., 2021; de León et al., 2024). Political trust is a multilayered concept that conveys citizen’s expectations concerning the competence, integrity, and benevolence of political institutions and their ability to take care of public interest (Easton, 1965; Norris, 2017). Such trust is also conducive to different forms of political outcomes, including vote choice, policy preferences, and compliance (Devine, 2024). In the context of social trust, we focus on “generalised social trust”, which denotes people’s trust in strangers and is deemed important to societal cohesion and collective action (Enjolras et al., 2019; Sønderskov, 2009). It has been argued that generalised social trust is easier to achieve under conditions of ethnic and religious homogeneity (Putnam, 2007), which entails that right-wing alternative media’s focus on the problems connected with immigration and multicultural society could be linked to lower social trust among their readers.
Existing survey-based research indicates that right-wing alternative media use is related not only to media trust, but also to political and social trust (Andersen et al., 2021; de Léon et al., 2024, Brems, 2024). For political trust, this is underpinned by qualitative studies showing that alternative media and their audiences express distrust or scepticism toward the political and cultural elites in society (Brems, 2023; Cushion et al., 2021; Ihlebæk & Holter, 2021; Noppari et al., 2019; Thorbjørnsrud & Figenschou, 2020). We thus formulate the following hypothesis:
H3. Low political, media, and social trust are associated with right-wing alternative media use.
Furthermore, in a bivariate analysis, Andersen and colleagues (2021) found that compared with political and social trust, media trust is more strongly associated with alternative media orientation (left and right) in Sweden. This finding makes sense, since criticisms of mainstream media are central to right-wing alternative media (Figenschou & Ihlebæk, 2019; Heft et al., 2020). We expand on Andersen and colleagues’ work by conducting a multivariate analysis to measure the relative importance of various trust measures, with the following hypothesis:
H4. Among the three types of trust, low media trust has the strongest relationship with right-wing alternative media use.
Concerning comparative country differences, we formulate the following hypotheses:
H5a: The relationship between political, media, and social trust and right-wing alternative media use is weakest in Denmark. H5b: The relationship between political, media, and social trust and right-wing alternative media use is strongest in Sweden.
In Denmark, right-wing alternative media appear less distinct from established media, since the latter are more inclusive of immigration-critical views than what is the case in Norway and Sweden. This could make users of right-wing alternative media less dissatisfied with established media, resulting in a weaker relationship between alternative media use and media trust in Denmark than in the two other countries, particularly Sweden. The relationship between such use and political and social trust is less straightforward, but a core factor may be the extent to which far-right ideas are treated with a cordon sanitaire, as has been the case in Swedish party politics and media. It is likely that right-wing alternative media users in Sweden may experience standing outside of the political and societal community, making them less trustful of political institutions and fellow citizens. Therefore, we expect the strongest relationship between right-wing alternative media use in Sweden to be with lower trust. Conversely, given the lesser distinction between right-wing alternative media and established media in Denmark, notably when it comes to criticising immigration policies and the political establishment (Heft et al., 2020; Hovden & Mjelde, 2019), less distinction in terms of political and social trust among their users seems likely.
The study is based on an online comparative survey carried out in Denmark, Norway, and Sweden in the fall of 2020. The survey was conducted using an online questionnaire administered by Kantar, targeting 1,958 (DK), 2,009 (NO), and 1,991 (SE) adult individuals. Following a probability-based process that did not rely on self-recruitment, samples were drawn from Kantar’s pool of respondents. Response rates are not easily estimated in web surveys relying on pre-established panels, as the surveys “close” when sufficient sample sizes are reached. The survey was distributed to 12,000 panel members in Denmark, 4,909 in Norway, and 9,807 in Sweden, and closed when approximately 2,000 interviews were completed, leaving us with “response rates” (% of distributions) at 17, 43, and 23 per cent, respectively. The national samples were skewed in various ways (see Table 1). Notably, highly educated respondents were overrepresented in the Swedish sample, while older respondents were overrepresented in the Danish sample. To solve this issue, all analyses were conducted either by applying post-estimation weight or by using regression analyses, controlling for age, gender, and education. We lacked information on the respondents’ native background, and based on experiences with population-based surveys, we expect people of immigrant background to be underrepresented in the sample (Theorin & Lundmark, 2024). Based on the immigration-critical profiles of alternative media sources, one might expect that people of immigrant descent are less likely to consume news from these sources; thus, the data may inflate estimates of alternative media consumption.
Based on a list of general categories of media types, with examples of main media outlets belonging to each category in brackets, respondents were asked to indicate how often they read, watched, or listened to news from the following sources:
Alternative media (e.g., Den Korte Avis, Trykkefrihed.dk (DK); Document. no, Resett.no (NO); Nyheter Idag, Samhällsnytt (SE))
Tabloid newspapers (e.g., EkstraBladet, BT (DK); VG, Dagbladet (NO); Expressen, Aftonbladet (SE))
National newspapers (e.g., Politiken, Berlingske Tidende (DK); Dagens næringsliv, Klassekampen (NO); Dagens Nyheter, Svenska Dagbladet (SE))
National Broadcasters (e.g., DR, TV2 (DK); NRK, TV2 (NO); SVT, TV4 (SE))
Local newspapers
Responses were given on an ordinal scale: Daily or almost daily; A few times a week; More rarely; Never; Don’t know. The term “alternative media” may be more open for interpretation amongst users than the other categories applied (for discussions, see Rauch, 2015; Steppat et al., 2023), but the examples of right-wing alternative media outlets given were meant to concretise the question. The examples were chosen based on inputs from colleagues in the three countries, as well as available statistics (Newman et al., 2019), and they share an immigration-critical strategy that is visible either in how they write about themselves and their motivations on their sites or through what they publish. Upon a follow-up question, 64 (DK), 75 (NO), and 55 (SE) per cent of those who consumed news from right-wing alternative media mentioned at least one of these outlets as a “regular” news source. (1) Since right-wing alternative media are often portrayed negatively, there is a risk that their usage may be underreported in the survey. However, there is little reason to believe that this potential bias differs significantly across the three countries.
The main dependent variable is a dummy distinguishing between those who consume news from right-wing alternative media sources daily or a few times a week (1), and those who more rarely or never do (0). To compare right-wing alternative media users with other media users, we constructed similar variables measuring daily or weekly news consumption through traditional newspapers (e.g., tabloids and national and local papers) and national broadcasters (e.g., public and commercial TV and radio). No respondents consume news only from alternative sources; thus, alternative media users are also traditional media users. (2)
We analysed variations in right-wing alternative media use related to political attitudes (party affiliation and support for far-right ideas) and political, media, and social trust.
Respondents were asked: “Which party best represents your viewpoints?” The response list included all parties represented in parliament, in addition to all far-right-parties (and a few other smaller parties) that had participated in the most recent elections. To reduce complexity, we combined single parties into five party groups (far left, left, centre, right, and far right) based on a combination of “party family” and left–right score (“rile”) reported in the Manifesto Project prior to the fielding of our survey (Volkens et al., 2020). (3) 93–99 per cent (SE and DK, respectively) of the respondents in the far-right category preferred one of the parties represented in the national parliaments (DK: New Right [Nye Borgerlige] and Danish People’s Party [Dansk Folkeparti], NO: Progress Party [Fremskrittspartiet], and SE: Sweden Democrats [Sverigedemokraterna]). The most popular far-right party outside parliament was Alternative for Sweden [Alternativ för Sverige], which was preferred among 16 respondents (6% of those in the far-right category).
The survey included 14 items that measured support for different dimensions of far-right ideas, including immigration scepticism, authoritarianism, and anti–gender equality. Most of the items were gathered from existing surveys, such as the Norwegian National Election Study/Comparative Study of Electoral Systems (questions on immigration, strong leader, satisfaction with democracy). In addition, to measure support for more extreme viewpoints, we also constructed additional items on violence and race. We measured all items using a 7-point scale, from Strongly disagree (1) to Strongly agree (7). A principal factor analysis suggested that 12 of these clustered on one factor, with factor scores higher than 0.4 and a total eigenvalue of 3.8 (no other factors reached a score above 1). Next, we constructed an index measuring support for far-right ideas based on unrotated saved factor scores (see Table A2 in the Appendix).
We generated an index of political trust based on mean values on single survey items of trust in government, parliament, and courts, measured on a 7-point scale, from Very little trust (1) to Very high trust (7), in response to the question, “How great or little trust do you have in …” The items correlated strongly in each country (Cronbach’s alpha: 0.83 [DK], 0.88 [NO], and 0.89 [SE]).
We measured media trust using an index constituting mean levels of trust in national broadcasters and national broadsheet newspapers. (4) All single items were measured using a 7-point scale, from Very little trust (1) to Very high trust (7), in response to the question, “How great or little trust do you have in the following news sources?” Scale reliability varied between 0.76 (DK) and 0.84 (NO).
We measured social trust via the standard generalised trust measure, which used a 7-point scale, from You cannot be careful enough (1) to Most people can be trusted (7), in response to the question, “In general, would you say that most people can be trusted, or that one cannot be careful enough when dealing with other people?”
For ease of interpretation, all continuous variables were standardised to 0–1 scales. The three trust measures correlate at moderate to strong levels, with the lowest single correlation coefficient being 0.34 between media and social trust, and the highest being 0.60 between political and media trust. This indicates that while the measures capture three distinct dimensions of trust, they also reflect a shared underlying trust construct to some extent. Consistent with previous studies, we include them as separate trust measures (Hanitzsch et al., 2018). Table 1 summarises the descriptive statistics for all the variables used in the analyses.
Descriptive statistics
| Denmark (n = 1,958) | Norway (n = 2,009) | Sweden (n = 1,991) | |||||
|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | ||
| Daily/weekly right-wing alternative media use | 0.05 | - | 0.08 | - | 0.14 | - | |
| Party affiliation (party groups) | Far left | 0.15 | - | 0.13 | - | 0.12 | - |
| Left | 0.28 | - | 0.16 | - | 0.26 | - | |
| Centre | 0.07 | - | 0.13 | - | 0.14 | - | |
| Right | 0.20 | - | 0.17 | - | 0.18 | - | |
| Far right | 0.11 | - | 0.10 | - | 0.13 | - | |
| No party/NA | 0.19 | - | 0.20 | - | 0.17 | - | |
| Continuous variables | Support far-right ideas | 0.38 | 0.18 | 0.34 | 0.18 | 0.35 | 0.20 |
| Political trust | 0.70 | 0.21 | 0.71 | 0.21 | 0.63 | 0.24 | |
| Media trust | 0.69 | 0.19 | 0.72 | 0.19 | 0.68 | 0.23 | |
| Social trust | 0.69 | 0.26 | 0.67 | 0.26 | 0.62 | 0.26 | |
| Control variables | Gender (women = 1) | 0.50 | - | 0.49 | - | 0.47 | - |
| < 30 years | 0.10 | - | 0.10 | - | 0.14 | - | |
| 30–44 years | 0.14 | - | 0.21 | - | 0.28 | - | |
| 45–59 years | 0.27 | - | 0.33 | - | 0.28 | - | |
| 60+ years | 0.49 | - | 0.35 | - | 0.30 | - | |
| Education (university/college = 1) | 0.33 | - | 0.41 | - | 0.56 | - | |
Comments: All variables range from 0 (min) to 1 (max).
Party categories (Manifesto Project):
Far left: Enhedslisten, Socialistisk Folkeparti, Alternativet, Veganerpartiet, Frie Grønne (DK); Rødt, SV (NO); Vänsterpartiet, Feministiskt initiativ (SE).
Left: Socialdemokratiet, (DK); Ap (NO); Socialdemokraterna, Miljöpartiet de gröna (SE).
Centre: Radikale Venstre, Kristendemokraterne (DK); Senterpartiet, Venstre, Krf, MDG (NO), Centerpartiet, Liberalerna, Kristdemokraterna (SE).
Right: Venstre, Det Konservative Folkeparti, Borgerlisten, Liberal Alliance (DK); De kristne, Høyre (NO), Moderaterna (SE).
Far right: Nye Borgerlige, Dansk Folkeparti, Stram Kurs (DK); Fremskrittspartiet, Demokratene, Alliansen, Selvstendighetspartiet (NO); Sverigedemokraterna, Alternativ för Sverige, Nordiska motståndsrörelsen (SE).
Other parties/no party omitted.
To take account of biased samples across the three countries (see above), we modelled the predicted probabilities of right-wing alternative media use by means of binominal logistic regression models, controlling for gender, age, and education. The variables of interest (sociodemographics, party affiliation, attitudes, trust) were interacted with country. To take account of differences between those who more rarely use right-wing alternative media and those who never do, we also estimated multinomial models, which are reported in the Appendix. The analytical strategy was to first describe characteristics of right-wing alternative media users (their sociodemographic, political, and trust profiles), and then to evaluate which factors were most important. This was done by first modelling the association between right-wing alternative media use and sociodemographic characteristics and each political and trust variable separately, before modelling all variables simultaneously. We estimated both country-specific- and cross-country (pooled) models, where country was interacted with the independent variables. Multicollinearity with party affiliation, attitudes, and trust variables included simultaneously was not a problem (the single highest VIF estimates were party affiliation with far-right parties [2.84] and far-right attitudes [2.75] in Sweden).
The results are presented in three steps. First, we describe the demographic profiles of right-wing alternative media users in Denmark, Norway, and Sweden. Second, we compare the rates (predicted probabilities) of daily or weekly use of right-wing alternative media with regard to political attitudes (party affiliation and support for far-right ideas) and trust (political, media, and social). Third, we estimate the relative strength of association between the different independent variables and right-wing alternative media use.
To start with the broad demographic picture, we estimated the predicted probabilities of frequent (daily or weekly) consumption of news from right-wing alternative media sources by means of binary logistic regression models, with gender, age, and education included simultaneously, and interacted with country. Table 2 shows the predicted probabilities, with corresponding confidence intervals, of daily/weekly right-wing alternative media use in total, and across demographic groups. The numbers suggest that Swedes (.14) were almost three times as likely as Danes (.05) to consume news from alternative media sources daily or weekly, with Norwegians (.079) in between (but closer to the Danes).
Daily/weekly use of right-wing alternative media by gender, age, and education
| Denmark | Norway | Sweden | ||
|---|---|---|---|---|
| Total | 0.050 [0.040–0.060] | 0.079 [0.067–0.090] | 0.144 [0.127–0.160] | |
| Gender | Men | 0.062 [0.047–0.078] | 0.106 [0.087–0.125] | 0.190 [0.166–0.215] |
| Women | 0.038 [0.025–0.050] | 0.049 [0.035–0.062] | 0.093 [0.074–0.113] | |
| Age | < 30 years | 0.044 [0.014–0.073] | 0.066 [0.030–0.102] | 0.097 [0.062–0.132] |
| 30–44yrs | 0.051 [0.025–0.077] | 0.080 [0.054–0.106] | 0.133 [0.105–0.161] | |
| 45–59 yrs | 0.053 [0.034–0.072] | 0.077 [0.057–0.097] | 0.145 [0.116–0.174] | |
| 60+ yrs | 0.050 [0.036–0.064] | 0.083 [0.062–0.103] | 0.161 [0.132–0.191] | |
| Highest level of education | Basic/High school | 0.053 [0.040–0.065] | 0.084 [0.068–0.100] | 0.145 [0.122–0.168] |
| University/college | 0.047 [0.031–0.064] | 0.071 [0.052–0.090] | 0.142 [0.121–0.162] | |
Comments: Predicted probabilities and [confidence intervals] from binary logistic regression models. The reference category consists of respondents who more rarely or never consume news from right-wing alternative media.
As shown in Table A3 in the Appendix, a larger share of the Swedes also reported that they more rarely consume news from right-wing alternative media sources, underlining that the share of non-users was significantly higher in Denmark and Norway. Across all three countries, men were more likely than women to use right-wing alternative media sources for news consumption (statistically significant difference in Norway and Sweden, not in Denmark). There were virtually no differences between age groups in Denmark and Norway, while in Sweden, daily/weekly usage increased with age (significant difference between the youngest and oldest age cohorts). Negligible differences existed related to daily/weekly consumption and educational levels.
As can be seen in the Appendix (Table A1), daily/weekly use of traditional newspapers and national broadcasters are generally much higher than daily/weekly use of right-wing alternative media, and by including those who more rarely use these news sources, almost all respondents in the three countries are covered. The sociodemographic profiles of right-wing alternative media users, compared with users of traditional newspapers and national broadcasters, are more skewed in terms of gender (male dominated), but less skewed in terms of age (see Table A2).
Next, we measure political attitudes with party affiliation (party group feeling closest to) and support for far-right ideas. Figure 1 shows the predicted probabilities of daily/weekly news consumption from right-wing alternative media based on the respondents’ affiliation with different groups of parties (left panel) and support for far-right ideas (right panel), controlling for gender, age, and education (at their observed levels). Across the three countries, those affiliating with far-right parties or who support far-right ideas were significantly more likely than other respondents to consume news from alternative media sources. The association between alternative media use and affiliation with far-right parties is especially strong in Sweden, with a predicted daily/weekly reach of 37 per cent [CI: 32–43], compared to 5–13 per cent among other voters. A predicted share of 24 [CI: 19–30] and 15 per cent [CI: 10–20] of far-right voters in Norway and Denmark, respectively, consume news from alternative media sources, compared to less than 10 per cent among other voters.
The association is even stronger when considering support for far-right ideas, especially in Norway and Sweden, where a predicted level of about 70 per cent among those with the highest support score consume news from alternative media, compared with “only” about 20 per cent among those with a moderate support score (0.5). In Denmark, a predicted share of about 30 per cent of those most supportive of far-right ideas consume news from alternative media daily or weekly, compared with less than 10 per cent among those with a moderate score. It is, however, important to emphasise that only a minority of the respondents strongly support far-right ideas. The dashed vertical lines in the figure indicate the interquartile range of observations, meaning that there are about 1,500 observations within each area. Thus, about one fourth of the respondents had values higher than the average level on the index (.5). Only 5 per cent had values higher than .7 (not shown in figure). Thus, while there is a strong relationship between support for far-right ideas and daily/weekly usage of right-wing alternative media, this relationship is driven by a small proportion of respondents.
Predicted probability of daily/weekly use of alternative media, by party affiliation and support for far-right ideas
Comments: Predicted probabilities from binary logistic regression models, controlling for gender, age, education. Dashed vertical lines mark the interquartile range of observations.
Alternative media (examples given): Den Korte Avis, Trykkefrihed.dk, Document.no, Resett.no, Nyheter Idag, Samhällsnytt. Control variables included (at their observed value): gender, age, education.
Party categories (Manifesto Project):
Far left: Enhedslisten, Socialistisk Folkeparti, Alternativet, Veganerpartiet, Frie Grønne (DK); Rødt, SV (NO); Vänsterpartiet, Feministiskt initiative (SE).
Left: Socialdemokratiet, (DK); Ap (NO); Socialdemokraterna, Miljöpartiet de gröna (SE).
Centre: Radikale Venstre, Kristendemokraterne (DK); Senterpartiet, Venstre, Krf, MDG (NO), Centerpartiet, Liberalerna, Kristdemokraterna (SE);
Right: Venstre, Det Konservative Folkeparti, Borgerlisten, Liberal Alliance (DK); De kristne, Høyre (NO), Moderaterna (SE).
Far right: Nye Borgerlige, Dansk Folkeparti, Stram Kurs (DK); Fremskrittspartiet, Demokratene, Alliansen, Selvstendighetspartiet (NO); Sverigedemokraterna, Alternativ för Sverige, Nordiska motståndsrörelsen (SE).
Other parties/no party omitted.
Far-right ideas: composite index, 12 items (see Table A3 in the Appendix).
Detailed analyses of the various dimensions constituting the far-right index suggest that the associations are strongest when considering far-right extremism, anti–gender equality, and populism (see Figure A1 in the Appendix). The association is weakest – but still highly significant – when considering immigration scepticism, which is also the most mainstream dimension. This suggests that alternative media are particularly attractive among those supporting the most radical far-right ideas. In all three countries, very few of those who most strongly reject far-right ideas consume news from alternative sources.
As a robustness check, we also modelled alternative media use with multinomial regression, distinguishing between non-users, infrequent (more rarely) users, and frequent (daily/weekly) users. These models suggest that the main distinction is between non-users and frequent users, and that the relationships between infrequent use of right-wing alternative media and political attitudes are weaker (see Figures A2 and A3 in the Appendix). In Norway and Sweden, infrequent use is most common among those with a medium-high score on the far-right attitudes index (0.6), and in Sweden, there is also correlation between support for traditional right-wing parties and infrequent use. Finally, the “uniqueness” of the association between alternative media use and political attitudes is corroborated by the fact that there is a non-linear or even reversed relationship between daily/weekly consumption of news from traditional newspapers and national broadcasters and political attitudes. Party affiliation is only weakly related to consumption of news from traditional newspapers and national broadcasters, while support for far-right ideas is associated with less consumption of news from these sources (see Figure A4 in the Appendix)
In summary, these analyses of political attitudes provide support for H1a and H1b, stating that far-right party affiliation and support for far-right ideas are associated with alternative media use. The comparative analyses show that H2a is partially supported. Supporters of far-right parties are indeed less likely to consume news from right-wing alternative news sources in Denmark than in Norway and Sweden. Nevertheless, it is Sweden that is most clearly differentiated, with a high share of alternative news users among those who identify with such parties. Regarding far-right attitudes and alternative media, however, the relationship is weaker in Denmark than in the two other countries, fully supporting H2b. While these country-level differences can be partly attributed to generally lower consumption rates of alternative media in Denmark, another interpretation is that traditional media have incorporated far-right segments of the population to a larger extent than in the two other countries, leaving less room (or market shares) for alternative media sources (Nygaard, 2021).
Next, we consider the association between alternative media use and political, media, and social trust. Figure 2 shows the predicted probabilities of daily/weekly news consumption from right-wing alternative media sources among respondents with different levels of trust.
Predicted probability of daily/weekly use of alternative media by political, media, and social trust
Comments: Predicted probabilities from binary logistic regression models, controlling for gender, age, education. Dashed vertical lines mark the interquartile range of observations.
Alternative media (examples given): Den Korte Avis, Trykkefrihed.dk, Document.no, Resett.no, Nyheter Idag, Samhällsnytt.
Political trust: Trust in government, parliament, courts. Control variables included (at their observed value): gender, age, education.
Overall, the figure shows significant negative relationships between right-wing alternative media use and trust, the strongest being media trust (centre panel), followed by political trust (left panel), and social trust (right panel). The associations are strongest in Norway and Sweden. In these countries, an estimated share of about 70 per cent among those with the lowest level of media trust consumed news from alternative media sources daily or weekly. While the relationship is also present in Denmark, it is much weaker. Among those with the lowest scores on political trust, estimated shares of about 50 (SE), 40 (NO), and 15 (DK) per cent consume news from alternative media sources, while the corresponding estimates for social trust are about 25 per cent in Sweden, about 20 in Norway, and less than 10 in Denmark. In Denmark, the association between alternative media use and social trust is not statistically significant. Importantly, however, in all three countries, only a negligible share of those with high levels of political, media, or social trust consume news from alternative media sources.
Again, it is important to emphasise that the relationships are boosted by relatively few respondents. As illustrated by the dark grey vertical lines in Figure 2, about one fourth of the respondents’ express trust levels below the average score on the trust indices.
The importance of (a lack of) trust for alternative media use is corroborated by a reversed relationship between trust and traditional media use, in which the probability of consuming news from traditional newspapers and national broadcasters increases with higher levels of trust (see Figure A5 in the Appendix).
In sum, with the exception of social trust in Denmark, the analyses provide support for H3, which states that low trust is associated with alternative media use. The analyses also support H5a, which suggests that the association between the various forms of trust and alternative media use is weakest in Denmark. On the other hand, H5b, which suggests that the associations would be strongest in Sweden, must be rejected, since the associations are not significantly stronger than in Norway. In other words, regarding this question, it is Denmark that stands out in Scandinavia. Again, the comparative differences may be attributed to traditional media having incorporated far-right ideological currents in a way that makes reading alternative media less of an issue of distrusting media or politics.
The analyses presented so far has described the “profiles” of frequent users of right-wing alternative media. Since the various political and trust variables are interrelated, we estimate models with all variables included simultaneously to evaluate the relative strength of the association between right-wing alternative media use and each independent variable. Figure 3 plots the predicted probability of consuming news from right-wing alternative news sources daily or weekly when all variables are included simultaneously (based on model 9 in Table A4 in the Appendix; separate models for each country are reported in models 6–8). Overall, the results from the full regression model suggests that supporting far-right ideas and having low levels of trust in traditional media are by far the most important variables correlating with alternative media use. Supporting far-right parties loses significance when taking account of support for far-right ideas (it remains significant with trust variables included), suggesting that supporting such parties is captured by attitudes.
Predicted probability of daily/weekly use of alternative media
Comments: Predicted probabilities from binary logistic regression models, all variables included simultaneously. Dashed vertical lines mark the interquartile range of observations.
In sum, the full regression models weaken support for H1a (far-right party affiliation), further confirm H1b (support for far-right ideas), moderately support H3 (political trust only significant in Norway; social trust insignificant in all three countries), and strengthen support for H4 (low media trust most important trust dimension).
Considering the comparative hypotheses, the combined analysis weakens support for H2a, since country differences related to far-right party affiliation are reduced when taking account of support for far-right ideas and trust in traditional media. The cross-country differences related to support for far-right ideas (H2b) are also reduced in the full regression model. Due to a low n and large confidence intervals, it becomes insignificant among those with the highest scores; however, the country differences remain significant among those with medium to high levels of support for far-right ideas (0.4–0.7). (5)
As for H5a and H5b, Figure 3 shows that having low levels of trust in traditional media is significantly more strongly associated with right-wing alternative news consumption in Sweden and Norway than in Denmark. The explained variance (pseudo r2) is also substantially lower in the Danish sample (model 6 vs model 7/8), suggesting that the variables examined generally have a weaker relationship with right-wing alternative media use in this country. This confirms the support for H5a found in analysis above (the relationship between trust and right-wing alternative media use is weakest in Denmark), while still rejecting H5b.
Finally, and as a side note, the full models in Table A4 also suggest that education is positively related to right-wing alternative media use in Norway, which was not visible in the descriptive statistics (see Table 2). Detailed inspection suggests that this effect appears when introducing far-right attitudes to the equation. By modelling the interaction effect between education and support for far-right ideas in Norway, the reason seems to be that among those who both have higher education and support far-right ideas, alternative media use is higher than among those without a college/university degree (models not shown). Supporting far-right ideas is strongly and negatively correlated with having a university or college degree, and by “taking out” the effect of far-right ideology, education becomes significant.
In this article, we have studied users of right-wing alternative media sources in Scandinavia in terms of their political attitudes and trust levels. Comparing three relatively similar countries with different experiences with alternative media penetration provides an avenue to gauge how various individual factors are related to alternative media use and whether such outlets attract audiences distinct from traditional media users.
The analysis confirms findings from previous research showing that right-wing alternative media consumption is considerably more widespread in Sweden than in Denmark, with Norway in between (Newman et al., 2019). Regarding political distinctiveness, consuming news from right-wing alternative media is more prevalent among those who affiliate with far-right parties and those who support far-right ideas. Notably, far-right parties in Scandinavia differ quite strongly in their ideological foundations and in their relationship to mainstream politics (Jungar & Jupskås, 2014), implying that the ideological worldviews of far-right voters who regularly use alternative media might also differ. Nevertheless, right-wing alternative media use is much more common among those who strongly support far-right ideas than among those who moderately support such ideas. For our study, we included a range of attitudes related to populism, anti–gender equality, immigration scepticism, authoritarianism, and extreme-right views, including the use of violence to protect national culture. Thus, irrespective of party ideology, right-wing alternative news media are the most popular among the most radical supporters of far-right ideas.
Political, media, and social trust are all negatively related to right-wing alternative media use. However, media trust stands out as the single most important factor, while social trust is the least important. This corresponds to what Andersen and colleagues (2021) found in their study of alternative media orientation, but our study further substantiates the role of media trust by measuring the relative importance of media trust vis-à-vis political and social trust. The results indicate that distrust in mainstream media is at the core among right-wing alternative media users.
Given that characteristics of right-wing alternative media users are shared across the region, one higher-level conclusion might be that universal individual predictors are valid across national contexts. Such a view is supported by previous single-country studies finding associations between low media trust, some aspects of far-right attitudes, and right-wing alternative media use (Brems, 2024; Müller & Schulz, 2021; Theorin, 2019). On the other hand, one of the very few comparative studies of right-wing alternative media audiences show national variations in the association between populist attitudes and media use, the association being stronger where right-wing alternative media are widespread (Stier et al., 2020). Likewise, the study at hand finds that Danish right-wing alternative media users are less distinct, in terms of support for far-right ideas and trust, compared with their counterparts in Norway and Sweden. One way to explain this situation is that both the media and the public debate in Denmark have gradually included stronger expressions of scepticism toward immigration, including viewpoints that would otherwise only be found in right-wing alternative media (Heft et al., 2020). As a consequence, it seems sensible that users and non-users of right-wing alternative media should be less distinct when it comes to trust and political attitudes than in the two other countries.
The findings are consistent with theories of selective exposure and media effects, in the sense that we find a correspondence between ideological attitudes and lesser trust in the mainstream media, which could stem either from people holding those attitudes seeking out right-wing alternative media, and from them being influenced by what they read over time. Specifically for explaining country differences, however, two empirical findings could be interpreted in favour of the selective exposure argument: cross-country variations in overall right-wing alternative media consumption combined with cross-country variations in the association between alternative media and trust in traditional media. If traditional media do not offer content that meets the demands of citizens who hold far-right attitudes, these citizens lose trust in these media and start looking for alternative news sources. Conversely, if the demands of these citizens to a larger extent are fulfilled via traditional media outlets, they are less likely to lose trust and seek out alternative news sources. Sweden represents the first situation, Denmark the latter. In Sweden, a lack of trust in traditional media goes hand in hand with an (in relative terms) extensive use of right-wing alternative media.
What can this study tell us about the political and democratic significance of alternative media use? On the one hand, the results confirm that consuming right-wing alternative media is linked to support for far-right ideas – and the more radical, the stronger the association. Read through this lens, alternative media could be seen as sounding boards for extreme ideas, thus instigating processes of political radicalisation, in line with the theory of reinforcing spirals (Theorin, 2019; Tsfati et al., 2024). On the other hand, our combined analysis indicates that the distrust of alternative media users centres much more on media trust than on political and social trust. Trust captures citizens’ sense of how society and democracy work and guides what they can expect from their surroundings, including the expectation that institutions and people have the necessary competence to fulfil their roles in good ways and that partners in interactions will honour their obligations and responsibilities (Barber, 1984. From this perspective, trust reflects the efficiency of society’s normativity and morality (Enjolras et al., 2019). Given the inherent criticism of mainstream institutions and politics, the rise of far-right ideology as a political movement marks a moment in which trust is put at stake. Based on our analysis, individuals who express the lowest level of trust combined with the strongest support for far-right ideas only constitute a small minority, and right-wing alternative media also attract users with more moderate levels of trust and support in far-right ideas. Thus, while some users of right-wing alternative media can be described as positioning themselves outside the democratic community, many of them can better be described as ideologically distinct but still bound to political institutions and part of the democratic community (Sniderman et al., 2014). These reflections underscore the usefulness of applying broad measures of far-right attitudes that can capture different degrees of radicalism and extremism, which is a contribution of our study.
There are several limitations of our study, however, that should motivate further research on news consumption of alternative news sources. First, “right-wing alternative media” is a heterogeneous category consisting of multiple outlets with varying ideologies. Alternative media may mean different things to specific audiences (Rauch, 2015; Steppat et al., 2023). Ideally, we would have studied characteristics among users of individual outlets, but because each had limited outreach, we relied on broad categories. Thus, future studies must use larger datasets and make comparisons between alternative media outlets with different ideological profiles. Second, our reliance on cross-sectional data rendered us unable to draw causal inferences. Even though we make the theoretical argument for selective exposure being a key mechanism, we do not know whether respondents chose media outlets based on their political attitudes and trust levels, or whether their media consumption affected their attitudes and trust levels. To clarify such questions, panel survey data over long time periods are needed. Third, our study covered three fairly similar countries. While this allows us to discuss contextual specificities in detail, the possibility of generalising findings beyond Scandinavia is limited. Finally, the survey was carried out during the first autumn of the Covid-19 pandemic, which formed a particular period in terms of political debates, issues, and agendas, and in which right-wing alternative media needed to adapt their roles and message. It is impossible to fully gauge the implications for our analyses here. Nevertheless, the Scandinavian countries were characterised by high levels of political and social trust throughout the pandemic, broad support for social distancing and public health guidelines, limited vaccine scepticism, and generally less polarisation than what was found in other countries (Johansson et al., 2023). Even though the alternative media scene in general is susceptible to changes, right-wing alternative media remains part of the media landscape in all three countries. This indicates an upheld demand for right-wing alternative outlets in Scandinavia.
It was also possible to mention other right-wing alternative media sources, and when including these, the share of regular users increased to 64 (DK), 81 (NO), and 69 (SE) per cent. The full list of mentioned outlets: Den Korte Avis, Trykkefrihed.dk (DK); Document.no, Human Rights Service, Resett.no (NO); Nyheter Idag, Samhällsnytt, Ledersidorna, Samtiden, Nya Tider, Det goda samhället (SE).
The dependent variable identifies regular users of right-wing alternative media, and contrasts them with all others, which is relevant for examining the distinctive characteristics of this group. As a robustness check, and to compare regular users with more infrequent ones, we also ran all models with a dependent variable that distinguishes between those who consume such news 1) daily or a few times a week, 2) more rarely, and 3) never. These analyses as well as analyses of traditional media users are briefly mentioned in the Results section and reported in the Appendix.
The Danish party “Venstre” was coded as right instead of centre (which was suggested in the Manifesto Project), as this has been the main party in centre-right government alliances in the 2000s. Party categories (Manifesto Project):
Far left: Enhedslisten, Socialistisk Folkeparti, Alternativet, Frie Grønne, Veganerpartiet (DK); Rødt, SV (NO); Vänsterpartiet, Feministiskt initiative (SE).
Left: Socialdemokratiet (DK); Ap (NO); Socialdemokraterna, Miljöpartiet de gröna (SE).
Centre: Radikale Venstre Kristendemokraterne (DK); Senterpartiet, Venstre, Krf, MDG (NO), Centerpartiet, Liberalerna, Kristdemokraterna (SE).
Right:Venstre, Det Konservative Folkeparti, Borgerlisten, Liberal Alliance (DK); De kristne, Høyre (NO), Moderaterna (SE).
Far right: Nye Borgerlige, Dansk Folkeparti, Stram Kurs (DK); Fremskrittspartiet, Demokratene, Alliansen, Selvstendighetspartiet (NO); Sverigedemokraterna, Alternativ för Sverige, Nordiska motståndsrörelsen (SE).
Other parties/no party omitted.
Trust in tabloid newspapers was omitted due to low scale reliability, especially in Denmark (alpha = .63, when including tabloid newspapers). Separate analyses of daily/weekly use of tabloid newspapers suggest that the relationships are rather similar to traditional newspapers when it comes to party affiliation, while the relationships with far-right attitudes are generally a bit weaker. Regarding trust, there are marked country differences, with strong positive relationships between political and media trust and frequent use in Norway, weak positive relationships in Sweden, and no relationship in Denmark (results can be retrieved upon request).
If we recode the far-right index into a categorical variable with four values based on the interquartile range, the country differences between Norway and Sweden on the one hand, and Denmark on the other, are large and statistically significant among those in the fourth quartile (predicted right-wing media use: DK: .074 [.051–.097], NO: .142 [.110–.173], SE: .183 [.140–.226].