Populists Prefer Social Media Over Talk Shows: An Analysis of Populist Messages and Stylistic Elements Across Six Countries

Abstract

For studying populism in a hybrid and high-choice media environment, the comparison of various media channels is especially instructive. We argue that populism-related communication is a combination of key messages (content) and certain stylistic devices (form), and we compare their utilization by a broad range of political actors on Facebook, Twitter, and televised talk shows across six countries (CH, DE, FR, IT, UK, and US). We conducted a content analysis of social media and talk show statements (N = 2067) from 31 parties during a nonelection period of 3 months in 2015. We place special emphasis on stylistic devices and find that they can be grouped into three dimensions—equivalent to three dimensions used for populist key messages. We further find that political parties are generally more inclined to use populism-related communication on Facebook and Twitter than in political talk shows and that both new challenger parties and extreme parties use higher amounts of populist key messages and style elements.

Keywords

populist communication Facebook Twitter challenger parties extreme parties

Today, it has become almost impossible to read the news without noticing a reference to populism. Scholarly interest in this transnational phenomenon has been growing because an increasing number of politicians and parties are apparently resorting to populist communication repertoires. Several studies analyzed the utilization of populist rhetoric (e.g., Wodak, 2015), populist messages (e.g., Rooduijn, 2014), or populist communication styles (e.g., Jagers & Walgrave, 2007) and demonstrated that investigating populist communication is crucial to fully understand the rise of political populism, as populism is mostly reflected in the oral, written, and visual communication of political actors (Aalberg & de Vreese, 2017). In a hybrid (Chadwick, 2017) or high choice media environment (van Aelst et al., 2017), politicians have gained new options for action because they have a greater number of communication channels—which differ in their affordances—at their disposal. Hence, it has become increasingly difficult to understand the role of a single medium in isolation (Bode & Vraga, 2017). These difficulties call for a comparative analysis that considers different media systems and different channel types. This type of analysis is especially important in the context of populism because there is conclusive empirical evidence that features of specific media channels influence the amount of populist communication (Bos & Brants, 2014; Cranmer, 2011; Ernst, Engesser, Büchel, Blassnig, & Esser, 2017). Our study compares the communicative self-presentation of political actors in three prototypical media channels for populism, Facebook, Twitter, and political talk shows, across six Western democracies. In addition to the particularities of the channels, we investigate whether the characteristics of a political party influence the amount of populist communication. Although populism in Western democracies is often associated with right-wing political actors, several studies with a wider scope have revealed that parties at both edges of the political spectrum adopt populist communication (Bernhard, 2016; Ernst et al., 2017; Rooduijn & Akkerman, 2017). In addition to the left/right divide, we are also examining whether younger parties that adopt a challenger position toward the establishment are more likely to use populist communication.

We define populist communication as the communicative representation of the populist ideology (what is being said) and the use of populism-related stylistic elements (how something is being said) by all sorts of political actors. We, therefore, follow a communication-centered approach (Sorensen, 2017) by defining key characteristics—messages and styles—of populist communication. Following Kriesi’s (2018) suggestion, we will first translate the key concepts of populist ideology into empirically measurable “key messages.” Next, we will examine the “stylistic elements” that politicians use when going popular. Because Kriesi (2018, p. 13) argues that “populist content and populist style tend to go together,” we will finally investigate similarities in politicians’ use of “key messages” and “stylistic elements.”

An important contribution of our study is to identify and systematize stylistic elements that politicians use in a similar way as they utilize populist key messages. In particular, we will determine whether characteristics of the communication channel and properties of a political party influence in similar ways the use of populist key messages and certain style elements. For this purpose, we compare a broad spectrum of political actors on two social media platforms and 12 political talk shows in six Western democracies (CH, DE, UK, US, IT, and FR). We find that (a) a variety of stylistic elements that previous literature has seen as part of populist actors’ communication strategy can be condensed to three dimensions—similarly to three ideological dimensions of populist communication, (b) politicians are generally more inclined to use populist key messages and related style elements on Facebook and Twitter than in TV talk shows, and (c) politicians from both new challenger parties and extreme parties use greater amounts of populism-related communication elements than established mainstream parties.

Populist Communication

Populism is a contested concept with a broad variety of different definitions (e.g., Canovan, 1999). Focusing on political actors’ self-presentation and their communicative approaches, we can identify two main traditions in the literature, as populist communication has either been defined as an ideology (Mudde, 2004) or a communication style (Jagers & Walgrave, 2007). Engesser, Fazwi, & Larsson (2017) argue that these two traditions are not mutually exclusive and only represent different aspects of populism. Their conceptualization of a populist communication logic distinguishes between four main approaches. First, populist ideology conceives populism as a set of ideas and focuses on the content—the what—of populist communication. Second, populism as a style emphasizes populism as a mode of presentation and focuses on the form and how the content is presented. Third, populism as a political strategy conceives populism as a means to an end and is interested in the strategic motives and aims of populist communication. Fourth, research on populism can focus on actors by analyzing the messengers.

In terms of a starting point, we are building upon foundations of the political science literature that conceptualizes populism as a relational concept with a distinct set of political ideas (Hawkins, 2009; Kaltwasser & Taggart, 2016; Rooduijn & Akkerman, 2017). We, therefore, conceive and define populism as a thin ideology. Populism claims that the people have been betrayed by the elites in charge who are abusing their positions of power, and it demands that the sovereignty of the people must be restored. Furthermore, we follow Hawkins (2009) who emphasizes that these basic ideas are expressed with specific discourse patterns, and we follow Wodak (2015, p. 3) who argues that populist political communication always “combines and integrates form and content” by providing “a dynamic mix of substance and style.” Some scholars—most notably Moffitt (2016)—have built ideological elements into stylistic definitions of populism. However, this is a proposal that we expressly do not want to follow, because we want to keep the basic ideological components of populist ideas separate from stylistic elements.

A populist vision of democracy basically separates society into two homogeneous and antagonistic groups, “the good people” versus “the bad elite,” and postulates the unrestricted sovereignty of the people (Abts & Rummens, 2007; Albertazzi & McDonnell, 2008; Mudde, 2004; Wirth et al., 2016).¹ Following this conceptualization, Mény and Surel (2002) have identified three key notions of populism: glorification of the people (people-centrism), condemnation of corrupt elites (anti-elitism), and claims for the restoration of popular sovereignty (popular sovereignty). When communicated in public, these ideological dimensions are broken down by political actors into nine key messages (Table 1). Previous studies have translated these key messages into empirically measurable categories of quantitative content analysis (see Ernst et al., 2017).

Table 1.

Conceptualization and Operationalization of Populist Key Messages.

Dimension	Populist key message	Underlying ideology	Categories
Anti-Elitism	Discrediting the elite	Elites are corrupt.	Elites are accused of being malevolent, criminal, lazy, stupid, extremist, racist, undemocratic, and so on. The elite are called names and denied morality, charisma, credibility, intelligence, competence, consistency, and so on.
	Blaming the elite	Elites are harmful.	Elites are described as a threat/burden, responsible for negative developments/situations, or as having committed mistakes or crimes. Elites are described as not being a source of enrichment or responsible for positive developments/situations.
	Detaching the elite from the people	Elites do not represent the people.	Elites are described as not belonging to the people, not being close to the people, not knowing the people, not speaking for the people, not caring for the people, or not performing everyday actions.
People centrism	Stressing the people’s virtues	The people are virtuous.	The people are bestowed with morality, charisma, credibility, intelligence, competence, consistency, and so on. The people are exempt from being malevolent, criminal, lazy, stupid, extremist, racist, undemocratic, and so on.
	Praising the people’s achievements	The people are beneficial.	The people are described as being enriched or responsible for a positive development/situation. The people are described as not being a threat/burden, not being responsible for negative developments/situations, nor as having committed mistakes or crimes.
	Stating a monolithic people	The people are homogeneous.	People are described as sharing common feelings, desires, or opinions.
	Demonstrating closeness to the people	The populist represents the people.	The speaker describes himself as belonging to the people, being close to the people, knowing the people, speaking for the people, caring for the people, agreeing with the people, or performing everyday actions. The speaker claims to represent or embody the people.
Restoring sovereignty	Demanding popular sovereignty	The people are the ultimate sovereign.	The speaker argues for general institutional reforms to grant the people more power by introducing direct-democratic elements or increasing political participation. The speaker argues in favor of granting more power to the people within the context of a specific issue (e.g., election, immigration, security).
Restoring sovereignty	Denying elite sovereignty	The elites deprive the people of their sovereignty.	The speaker argues in favor of granting less power to elites within the context of a specific issue (e.g., election, immigration, security).

According to Kriesi (2018), the use of these key messages is part of a political strategy that manifests itself in broader communication patterns. In his view, a populist political communication strategy is also expressed by the use of a specific communication style. Kriesi (2018, p. 12) expects this style to be characterized by elements such as “emergency rhetoric,” “emotionalization” as well as “assertive/absolutist” and “colloquial” language, among others. His position is fully compatible with Sorensen’s (2017) stand that “approaching populism from a communications perspective (. . . ) inevitably involves considerations of style as well as ideology” (p. 139). We follow Hofstadter’s (2008) definition of a communication style being the way ideas are believed and advocated by a political actor rather than the truth or falsity of the content (Block & Negrine, 2017).

We are interested in finding systematic parallels in the use of populist key messages and the use of certain stylistic elements. We have searched the research literature to identify the communicative stylistic elements attributed to populist actors. We have finally identified seven of them, which we have summarized in Table 2. It is important to make clear that until we have examined these stylistic elements in more detail, we do not yet claim that they are populist in themselves. Following Kriesi (2018), we only say that they can be considered expressions of the same communication strategy that can also lead to the use of populist key messages.

Table 2.

Conceptualization and Operationalization of Populism-Related Style Elements.

Dimension	Stylistic devices	Underlying style element	Categories
Negativity	Negativism	Paint society and its members (part of the people) “in black” by attributing negative characteristics or condemning actions/situations with negative outcome.	Targets are accused of being malevolent, criminal, lazy, stupid, racist, and so on or are denied being benevolent, likable, intelligent, credible, loyal, consistent, and so on.
Negativity	Crisis rhetoric	Portraying a situation/development as a crisis using exaggerations, emergency rhetoric or declaring a scandal.	Speaker uses rhetorical elements of immorality, exaggeration, scandal, emergency, or war rhetoric.
Emotionality	Emotional tone	Sharing positive and negative emotions or revealing feelings.	Speaker uses emotional language by expressing discrete positive (e.g., happiness, contentment, hope, pride, trust) or negative (e.g., anger, uneasiness, sadness, fear, regret, affection) emotions.
	Absolutism	Using an assertive tone and lacking relativizing words. Tendency to paint world in black and white without any shades of gray.	Speaker uses rhetorical figure of absolutism by presenting something as the only conceivable option or as preposterous or unbearable.
	Patriotism	Emphasizing of the superiority of own country by referencing an idealized and utopic heartland.	Speaker uses rhetorical figure of patriotism by emphasizing superiority of own country or some obscure heartland.
Sociability	Colloquialism	Preference for a simple, dialect, colloquial, or vulgar language and use of nicknames to reach the ordinary people.	Speaker uses vulgar language or slang, employs sarcasm or rhetorical questions and address targets with nicknames.
Sociability	Intimization	Recounting personal and intimate details about personal life.	Targets are described in their predominately personal life by emphasizing their family or love life and making references to personal way of life or leisure activities.

The first style element with an affinity to populism is negativism. It refers to the tendency of populist actors to paint society darkly by attributing negative characteristics to the elites or dangerous others or by condemning situations or actions with a negative outcome (Alvares & Dahlgren, 2016; Block & Negrine, 2017; Bracciale & Martella, 2017; Engesser, Fawzi et al., 2017). Second, populist actors gravitate toward portraying various situations or developments as crises. To employ this crisis rhetoric, populist actors usually adopt rhetorical elements of immorality, exaggeration, scandal, emergency, or war. Third, the emotional tone style comprises populist actors’ tendency to share positive or negative emotions or reveal feelings (Block & Negrine, 2017; Bos, van der Brug, & de Vreese, 2013; Bracciale & Martella, 2017; Canovan, 1999; Engesser, Fawzi et al., 2017; Hameleers, Bos, & de Vreese, 2017). While most authors stress the importance of negative emotions that are raised against others or elites, positive emotions can be directed to the people or the populist leader. Fourth, absolutism describes the affinity of populist actors to paint the society in black and white terms without any shades of gray. The style expresses itself in the use of an assertive tone and a hesitation to use relativizing words in their communication (Bos & Brants, 2014; Engesser, Fawzi et al., 2017; Hawkins, 2009). Fifth, patriotism as a populist communication style portrays the tendency of populists to long for a time when everything was much better and emphasize the superiority of their own country by employing rhetorical elements referencing an idealized and sometimes utopic vision of the country or heartland (Block & Negrine, 2017; Rydgren, 2017; Taggart, 2000). Sixth, populist actors are prone to reduce complexity by employing a colloquial style, which manifests itself in simple, dialect, colloquial, or vulgar language to reach ordinary citizens (Albertazzi & McDonnell, 2008; Bracciale & Martella, 2017; Engesser, Fawzi et al., 2017; Moffitt, 2016). Finally, populist actors do not shy away from using an intimization style in which they recount personal and intimate details about their personal lives to portray themselves as approachable and down-to-earth politicians (Bracciale & Martella, 2017; Stanyer, 2012).

Overall, we take a communication-centered approach and conceive of populist communication as an outcome of a strategy that uses both ideological key messages and certain stylistic elements (see also Bracciale & Martella, 2017; Krämer, 2014; Stockemer & Barisione, 2017; Wirth et al., 2016). How these two components manifest themselves is an open empirical question that we want to clarify. But Krämer (2017) has already pointed out that there is often a “homology between ideologies and styles” (p. 1305) and that scholars should not refrain from style-based reconstructions of populism—however “thin” they may be. Especially because content and form tend to interfere and interact with one another (Stockemer & Barisione, 2017). By combining both perspectives, we wish to explore similarities in the use of populist key messages and stylistic devices on social media and political talk shows.

Populist Communication on Social Media and Political Talk Shows

In a hybrid media system where new and old media are intertwined and their logics complement each other, political actors no longer rely on a single communication channel (Chadwick, 2017). Instead, they choose a variety of different channels to achieve their communicative goals. Bode and Vraga (2017) argue that studying single media platforms in isolation ignores the reality of the contemporary media system. This argument is especially important because research has demonstrated that the specific characteristics of the communication channel influence the amount of populist communication (Bos & Brants, 2014; Cranmer, 2011; Ernst et al., 2017). By investigating populist communication on television and social media, we can gain a more detailed understanding of how the features of a platform affect populist communication. Political talk shows and social media represent two different types of communication channels that suggest differences in the way politicians from different parties and countries use them for populist purposes. What unites these channels is that they offer politicians favorable opportunities for self-presentation with little to no interference from skeptical, hard-nosed political journalists (Esser, Stępińska, & Hopmann, 2017). They are moreover prototypical channels for populist communication and both Bos and Brants (2014) and Cranmer (2011) conclude, that especially political talk shows offer great opportunity structures for populist communication.

Social media plays a major role in the political communication strategies of contemporary parties (Stieglitz, Dang-Xuan, 2013). Especially, Twitter and Facebook have emerged as central media platforms that rival traditional news media in reach and influence (Fisher, Marshall, & McCallum, 2018). The possibility to bypass news journalists and the ability of political actors to communicate directly with their publics increases the chances of successful self-promotion (Lilleker & Koc-Michalska, 2013). This gives us already an idea of why social media networks have transpired as a particularly well-suited channel for populist communication (Ernst et al., 2017).

Four opportunity structures of Facebook and Twitter foster the potential for populist communication: They offer the possibility to establish a close connection to the people, they provide a direct access to the public without journalistic interventions, they can create a feeling of community and recognition among otherwise scattered groups, and they foster the potential for personalization (Ernst et al., 2017).

When comparing the two social media platforms in relation to their potential for populist purposes, Facebook has four advantages over Twitter. First, Facebook offers more reciprocal message exchanges; second, it has higher levels of proximity and the connection between Facebook users is generally more intensive, personal, and intimate; and third, Facebook is not subject to certain character limits, which gives political actors greater opportunity to make their case effectively and elaborately. Finally, due to the different characteristics of users (in terms of education, socioeconomic status, or political interest), Twitter has a stronger professional orientation, and political actors may consider it less suitable for spreading populism (Jacobs & Spierings, 2018). In contrast, Facebook is the platform for ordinary citizens to interact with politicians (Kalsnes, Larsson, & Enli, 2017), which makes it more suitable for populist communication. Schulz (2018) has supported this argument from the audience perspective by showing that populist citizens are more likely to use Facebook as their source of political information, while nonpopulist citizens rather use Twitter for information purposes.

Television talk shows belong to a completely different media category, but can fulfill similar functions for politicians. They are important outlets for the articulation of political ideas (Baym, 2013; Jones, 2010; Kessler & Lachenmaier, 2017) and have a positive effect on viewers’ trust in politicians (Boukes & Boomgaarden, 2016). Like social media, talk shows offer the opportunity to bypass the watchdog journalism more commonly found in hard news programs, offer the possibility of self-presentation, and foster the potential for personalization (Boukes & Boomgaarden, 2016; Kessler & Lachenmaier, 2017). Political talk shows, therefore, represent another ideal stage for populist communication (Cranmer, 2011) and make the comparison with Facebook and Twitter more meaningful.

Research Questions and Hypotheses

The aim of this study is twofold. In the first step, we will examine the seven populism-related style elements for their broader underlying dimensions. To examine the relationship between these seven stylistic elements, we examine the following research question:

RQ1: Do populism-related style elements form distinct dimensions?

In a second step, we will investigate whether the characteristics of communication channels and the properties of parties affect the degree of populist communication on social media and political talk shows. Recent research has demonstrated the importance of Facebook and Twitter for populist communication (e.g., Bracciale & Martella, 2017; Engesser, Ernst, Esser, & Büchel, 2017; Hameleers & Schmuck, 2017; Stier, Posch, Bleier, & Strohmaier, 2017). However, these studies lack a comparison of different media channels. In previous research, political talk shows were found to contain the highest level of populist communication compared with other arenas (Bos & Brants, 2014; Cranmer, 2011). Yet, none of these studies compared the amount of populist communication to social media. We argue that Facebook and Twitter are even more advantageous for spreading populist communication than talk shows because they allow politicians to circumvent traditional gatekeepers completely; they further allow for better message targeting and personalization and for the establishment of reciprocal relationships and a more direct line to their followers (Ernst et al., 2017). Hence, the first hypothesis to be tested is

H1: The proportion of populist key messages and related style elements are higher on Facebook and Twitter than on political talk shows.

In addition, the properties of political parties are of special interest to this study. The so-called challenger parties are a particularly relevant party category because they are often perceived as a threat to the party establishment (Meguid, 2005) and assumed to use populist communication to generate attention (Kriesi, 2014). Throughout the various crisis cycles since the 1980s, new challenger parties from both the left and the right have emerged and gained success in many Western democracies (Hobolt & Tilley, 2016; Müller-Rommel, 1998). Kriesi (2014) argues that these new challenger parties may be perceived as a threat to the establishment because they highlight problems that have been neglected by mainstream parties, mobilize outside of the electoral channels, and resort to creative, innovative forms of protest communication. New right- and left-wing challenger parties can thus be expected to blame the elites and complain about a neglect of the people’s true concerns (Kriesi, 2014). Hence, these parties may rely on populist communication to improve their electoral chances (Betz, 2002) and their media visibility (Mazzoleni, 2008). These assumptions are supported by a longitudinal study from Switzerland that found that new parties—independent from their ideological stance—relied on high levels of populist communication in party advertising and press releases during their initial “challenger phase” (Weber, 2017). We would like to examine this mechanism on a broader empirical basis and propose the following hypothesis:

H2: Challenger parties use a greater proportion of populist key messages and related style elements than established parties.

Party extremism is also expected to influence the amount of populist communication. Populism is often too quickly associated with right-wing parties. However, populism should be considered as a latent or thin ideology (Hawkins, 2009; Stanley, 2008) that, due to its “thinness,” can be combined with a variety of “thick” host ideologies (such as socialism, authoritarianism, nationalism) that add more specific content to it. Although right- and left-leaning parties differ in their ideologies, party programs, and social basis, they have several characteristics in common that are related to populist communication. At least in Western Europe, they emerged in recent decades, often compensate for their small-sized party organization with large-sized communication offensives, tend to remain opposition parties, and share a protest attitude against established parties, politics, and state structures (Müller-Rommel, 1998). Moreover, research has substantiated that parties at the fringes of the political spectrum are more inclined than moderate parties to challenge the current government, attack elites, and glorify the people in their communication within party manifestos (Rooduijn & Akkerman, 2017), press releases (Bernhard, 2016), and on social media (Ernst et al., 2017). Against this background, we propose the following hypothesis:

H3: Extreme parties use a greater proportion of populist key messages and related style elements than moderate parties.

Method

We content-analyzed populist key messages and related style elements used by 110 politicians on political talk shows and social media across six countries in 2015 using a semiautomatic coding program. These channels were chosen because they mostly lack journalistic interference and allow a relatively unfiltered view of politicians’ communication.

Sample

To test our hypotheses, we need to construct a sample of politicians from different political parties who appear on talk shows and social media using populist messages and stylistic devices during our investigation period. To explore populist communication as a transnational phenomenon (Aalberg, Esser, Reinemann, Strömbäck, & de Vreese, 2017; Moffitt, 2016), these politicians need to come from different countries. This requires a three-step procedure: we first identified relevant countries, then sampled the relevant political talk shows and listed all appearing politicians, and finally collected the social media material of these politicians. By applying such an individual matching procedure on the microlevel of politicians, the study ensures the comparability of communication on both channel types and thus avoids ecological fallacies.

In the first step, we selected six countries (CH, DE, IT, FR, UK, and US). By performing all analyses under six different macrosocial conditions, our multinational comparative design serves as a robustness check for the meaningfulness of our findings. In other words, our conclusions on which media types and party types are more susceptible to populist communication thus gain more reach and validity. The country sample provides sufficient variability regarding political systems (parliamentary vs. presidential, representative vs. directional, consensus vs. majoritarian systems), party characteristics (strong vs. weak populist parties), and consumer preferences for political information sources (Aalberg et al., 2017; Newman, Fletcher, Kalogeropoulos, Levy, & Kleis Nielsen, 2017).

In the second step, we identified two influential political talk shows per country (see Table A in online appendix) and recorded four episodes of each show during a 2-month period of routine news without national elections from April through May 2015. The twelve selected talk shows all air weekly and achieve high market shares in their segment. They are all primarily political in focus, follow a roundtable format, have a length of approximately 60 min,² and regularly invite politicians as guests. The number of politicians ranged from one to five per show; we coded only their statements and disregarded those made by the moderator, nonpolitical guests, or members of the audience. This led to a total of 1579 statements by 110 political actors across the 48 taped programs.

In the third step, we collected the social media material (Facebook posts and Tweets) of all politicians³ who appeared on the talk shows during the study period.⁴ We considered only Tweets and Facebook posts that included direct statements of the politician and were longer than eight characters. Simple Retweets and Tweets or Facebook posts including only pictures, links, or videos were excluded from the analysis. We drew a random sample of 50 social media statements per politician (N = 5099).

Out of the initial sample, we kept only those statements that included a veritable statement by a politician and expressed either a political position, an elaboration on a political issue, or an evaluation, or an attribution of a target actor (N = 2130). We further excluded politicians with less than five statements in total or no clear party affiliation. This led to a final sample of N = 969 talk show statements, N = 734 Facebook posts, and N = 364 Tweets by 98 politicians from 31 political parties.

Unit of Analysis

The unit of analysis is a single statement made by a politician. It can contain a key message and/or a stylistic device. In the terminology of our codebook, statements are made by speakers (i.e., politicians) about a target actor or an issue. A target actor is the object of a politician’s characterization or evaluation and may include politicians, members of the elite, or the people. A political issue refers to the thematic context of the statement.

A team of intensively trained student coders reached acceptable levels of reliability. The average Brennan and Prediger’s kappa across all messages and styles is .91 (see Table B in online appendix).

Operationalization

Populism-Related Communication

The nine populist key messages and the seven stylistic devices were gathered based on the instructions given in a codebook (for details on the categories used see Tables 1 and 2). For each category, we recorded whether the variable was present in a statement. A key message or stylistic device was considered present if at least one of the related categories was coded. The dependent variable, populism-related communication, was present if at least one of the nine populist key messages or seven stylistic elements were observed.

Party Categories

The 98 politicians belong to 31 political parties. Table 3 depicts the number of politicians per party (for further details see Table C in online appendix). To determine the degree of party extremism, we assigned each party its respective Chapel Hill expert survey (CHES) score (Table 3). We had to rate American and some Italian parties ourselves because the CHES did not include them (Polk et al., 2017). For each party score, we subtracted the theoretical center of the scale (–5) and took the absolute value to obtain a measure of party extremism. We also recorded the age of each party. Since most challenger parties emerged in recent decades (Hobolt & Tilley, 2016; Müller-Rommel, 1998; Weber, 2017), we coded all parties founded after 1980 as challenger parties. Furthermore, two dummies for Facebook and Twitter (vs. political talk shows) were calculated.

Table 3.

Overview of Political Party Sample.

Country	Party information	Political stance
Country	Party information	Left	Moderate Left	Center	Moderate Right	Right
CH	Political party	GPS	SP	CVP	FDP	SVP
	CHES score	1.88	2.13	5.5	6.88	8.25
	Founding year	1983	1888	1912	2009	1971
	Number of represented politicians	3	4	3	3	7
	Populism-related communication score	0.28	0.30	0.24	0.26	0.42
	Political party			GLP
	CHES score			5.25
	Founding year			2007
	Number of represented politicians			1
	Populism-related communication score			0.32
DE	Political party	Linke	SPD	CDU	FDP	AfD
	CHES score	1.23	3.77	5.92	6.54	8.92
	Founding year	2007	1963	1945	1948	2013
	Number of represented politicians	2	3	4	1	1
	Populism-related communication score	0.50	0.40	0.10	0.10	0.42
	Political party		Gruenen		CSU
	CHES score		3.62		7.23
	Founding year		1993		1945
	Number of represented politicians		3		1
	Populism-related communication score		0.47		0.13
FR	Political party		PS
	CHES score		3.83
	Founding year		1969
	Number of represented politicians		3
	Populism-related communication score		0.25
IT	Political party		PD	SC	FI	LN
	CHES score		3.57	5.43	6.71	8.86
	Founding year		2007	2013	1994	1998
	Number of represented politicians		11	1	1	2
	Populism-related communication score		0.37	0.44	0.33	0.63
	Political party		RI	M5S	Fdl
	CHES score		3	4.67	7.86
	Founding year		2001	2009	2012
	Number of represented politicians		1	1	1
	Populism-related communication score		0.50	0.44	0.38
	Political party			IdV
	CHES score			5
	Founding year			1998
	Number of represented politicians			1
	Populism-related communication score			0.40
UK	Political party	Green	Lab	LibDem	Cons	UKIP
	CHES score	1.86	3.57	4.86	7	9.14
	Founding year	1990	1900	1988	1934	1993
	Number of represented politicians	1	4	4	5	3
	Populism-related communication score	0.50	0.38	0.40	0.45	0.55
	Political party		SNP
	CHES score		3
	Founding year		1934
	Number of represented politicians		3
	Populism-related communication score		0.28
	Political party		Plaid
	CHES score		3.25
	Founding year		1925
	Number of represented politicians		1
	Populism-related communication score		0.79
US	Political party		D		R
	CHES score		3		7
	Founding year		1828		1854
	Number of represented politicians		9		8
	Populism-related communication score		0.27		0.36

Note. Gray background stands for challenger parties. For RI, IdV, D, and R no CHES score exist. We placed these parties ourselves and assigned them an individual extremism score. The populist-related communication score (range = 0-1) is based on both messages and styles across social media and political talk shows.

Findings

Sample Description

In total, 38% of all statements contain at least one populism-related communication element. Overall, populist key messages and stylistic devices are weakly correlated (r = .173, p < .01), and the style elements (31.3%) are used significantly more, t(2066) = 17.35, p < .001, than populist key messages (13.6%). Anti-elitist key messages (9.6%) are more prominent than people-centrist key messages (3.3%). Key messages about restoring sovereignty are almost absent (0.9%) in politicians’ communication on social media and talk shows. Negativity (19.9%) and emotionality (12.2%) are commonly used style elements, whereas sociability (4.8%) is only present in every 20th statement.

Research Question and Hypotheses

In a first step, we conducted a principal component factor analysis with the seven style elements (varimax rotation, Kaiser normalization). The factor analysis identified three distinct dimensions (51.4% explained variance, Bartlett’s test of sphericity = χ²(21) = 282.3, p < .001) without any substantial cross-loadings (Table 4). Negativity comprises negativism and crisis rhetoric; emotionality includes emotional tone, absolutism, and patriotism; and sociability is composed of colloquialism and intimization. We consider this finding an important empirical contribution to the literature that discusses stylistic devices as part of populist actors’ communication strategy. It provides the basis for our further analyses.

Table 4.

Factor Analysis of Populism-Related Style Elements.

	Factor 1	Factor 2	Factor 3
	Negativity	Emotionality	Sociability
Negativism	0.78	−0.08	0.16
Crisis rhetoric	0.73	0.20	−0.07
Emotional tone	−0.10	0.66	0.09
Absolutism	0.05	0.63	−0.02
Patriotism	0.17	0.57	−0.05
Colloquialism	−0.11	−0.02	0.79
Intimization	0.19	0.05	0.70
Self-value	1.23	1.21	1.16
Variance explained (%)	17.50	17.33	16.60
Total variance (%)	51.42

Note. Rotated factormatrix (principal component factor analysis, varimax rotation), KMO = .56, Bartlett’s test of sphericity χ²(21) = 282.3, p < .001, N = 2065.

To test our three hypotheses, we conducted 12 multilevel models (Table 5 for an overview) with maximum-likelihood estimation (ML). In Models 1 to 4, populism-related communication is the dependent variable, which means the use of at least one populist key message and/or at least one style element. Due to the rather weak correlation of populist key messages and stylistic elements, we calculated eight additional models to test the effects for both components separately. In Models 5 to 8, the dependent variable is the use of any populist key message and in Models 9 to 12, the dependent variable is the use of any style element. Units of analysis are statements made by politicians on Twitter, Facebook, or political talk shows. These statements by politicians are nested in political party properties. Hence, the independent variables for Hypothesis 1 (social media) are located at the first level, while the independent variables for Hypothesis 2 and Hypothesis 3 (challenger parties and extremism) are located at the second level. Before assessing the multilevel models, we determined whether it is useful to let the intercepts vary across parties. If we compare the respective baseline models with Model 1 (χ²(1) = 46.26, p < .001), Model 5 (χ²(1) = 42.06, p < .001), and Model 9 (χ²(1) = 38.23, p < .001) we can conclude that in all three models, the intercepts vary significantly across parties and significantly improved the model fits.

Table 5.

Multilevel Model of Populism-Related Communication.

	Populism-related communication				Populist key messages				Populism-related style elements
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9	Model 10	Model 11	Model 12
	β	β	β	β	β	β	β	β	β	β	β	β
	SE	SE	SE	SE	SE	SE	SE	SE	SE	SE	SE	SE
(Intercept)	−0.013	−0.003	0.018	−0.007	−0.008	−0.005	−0.015	−0.008	−0.003	0.006	−0.007	−0.003
	0.050	0.044	0.042	0.033	0.044	0.042	0.040	0.037	0.050	0.043	0.033	0.030
Twitter		0.076	0.078	0.077		0.016	0.017	0.016		0.079	0.082	0.079
		0.024	0.024	0.024		0.024	0.024	0.024		0.024	0.024	0.024
		***	***	**						***	***	***
Facebook		0.169	0.171	0.172		0.083	0.084	0.024		0.171	0.175	0.174
		0.024	0.024	0.024		0.024	0.024	0.022		0.024	0.024	0.023
		***	***	***		***	***	***		***	***	***
Challenger party			0.113	0.102			0.089	0.081			0.121	0.115
			0.038	0.033			0.039	0.036			0.033	0.030
			**	**			*	*			***	***
Extremism				0.078				0.064				0.058
				0.032				0.035				0.030
				*				†				†
AIC	5824.63	5779.04	5775.39	5770.52	5828.83	5820.68	5817.73	5816.88	5832.66	5786.24	5819.73	5776.24
BIC	5841.53	5807.21	5814.83	5809.95	5845.73	5848.85	5851.54	5856.32	5849.56	5814.41	5859.17	5815.68
Log likelihood	−2909.32	−2884.52	−2880.70	−2878.26	−2911.42	−2905.34	−2902.87	−2901.44	−2913.33	−2888.12	−2902.87	−2881.12
Level 1 N (statements)	2067	2067	2067	2067	2067	2067	2067	2067	2067	2067	2067	2067
Level 2 N (parties)	31	31	31	31	31	31	31	31	31	31	31	31

Note. AIC = Akaike information criterion; BIC = Bayesian information criterion.

***

p < .001. **p < .01. *p < .05. †p < .10.

Hypothesis 1 predicted that populism-related communication by various political actors is higher on social media than political talk shows. To test this hypothesis, we added two fixed effects of Twitter (dummy) and Facebook (dummy) to the model. A model comparison demonstrated that the fit of the models significantly increased for Model 2 (χ²(1) = 49.60, p < .001), Model 6 (χ²(1) = 12.14, p < .01), and Model 10 (χ²(1) = 50.43, p < .001). We find clear support for this hypothesis. Model 2 shows that statements on Facebook (β = .17, t(2035) = 7.09, p < .001) and Twitter (β = .08, t(2035) = 3.17, p < .001) significantly predict populism-related communication and demonstrates that the degree of populism-related communication is higher on both social media platforms than political talk shows. This pattern is identical for style elements (Twitter: β = .08, t(2035) = 3.29, p < .001; Facebook: β = .17, t(2035) = 7.16, p < .001 Model 10) in isolation as well. For populist key messages, we only find a significant effect for Facebook (β = .08, t(2035) = 3.45, p < .001). Key messages on Twitter, however, are not more common than in political talk shows (β = .02, t(2035) = 0.66, ns). If we compare the two social media platforms, we can conclude that Facebook is the stronger predictor and that political actors tend to prefer Facebook for their populism-related communication.

Hypothesis 2 argued that challenger parties use more populism-related communication than established mainstream parties. To test this hypothesis, we added the challenger party dummy as another fixed effect to the models. Again, model comparisons revealed that the fit of all models significantly improved for Model 3 (χ²(1) = 7.45, p < .01), Model 7 (χ²(1) = 4.95, p < .05), and Model 11 (χ²(1) = 10.35, p < .001). In Model 3, we find support for this hypothesis and show that challenger parties use more populism-related communication (β = .11, t(29) = 2.95, p < .01) than established parties. We identified the same pattern for populist key messages in Model 7 (β = .09, t(29) = 2.27, p < .05) and style elements in Model 11 (β = .12, t(29) = 3.70, p < .001).

To test whether party extremism positively influences the amount of populism-related communication, which is postulated in Hypothesis 3, we added extremism as a fourth additional fixed effect. Model comparisons reveal that for model 4 with the combined use of populist key messages and stylistic elements as the dependent variable, the fit increased significantly (χ2(1) = 5.08, p < .05), meaning that parties with high extremism scores use populism-related communication more often (β = 0.08, t(28) = 2.42, p < .05; model 4) than moderate parties with low extremism scores. For the separate use of populist key messages (χ2(1) = 2.85, p = .09; model 8) or stylistic devices (χ2(1) = 2.64.14, p = .06; model 12), we only find somewhat weaker effects. The influence of party extremism on the separate use of either populist key messages (β = 0.06, t(28) = 1.81, p = .08; model 8) or stylistic devices (β = 0.06, t(28) = 1.97, p = .06; model 12) leans toward significance, but narrowly missed the standard p-value threshold. In sum, we only find partial support for hypothesis 3. Party extremism results in a significantly higher use of the combination of populism-related communication; for messages and style elements separately however, we can only report a trend.

All 12 multilevel models present clear support for the three postulated hypotheses, demonstrating that parties are generally more inclined to use populism-related communication on Facebook and Twitter than in talk shows and that new challenger parties as well as extreme parties use higher amounts of populism-related communication.⁵

Discussion and Conclusion

In a communications approach, populist ideology and style elements are considered inextricably intertwined, but scholars need to keep them analytically distinct and analyze them with separate empirical measures. We are aware that some researchers such as Moffitt (2016) have incorporated ideological elements into the stylistic definition of populism, but we deliberately do not follow this approach. We understand populist communication as the outcome of a political strategy that uses both ideological key messages and certain stylistic elements. We have developed systematic operationalizations for messages (content) and styles (form) and examined them with a content analysis that takes into account different countries, communication channels and party types.

We argued that populism as an ideology consists of three programmatic components that are communicated publicly through nine key messages. In a similar way, we have examined stylistic elements of a “going popular” strategy and found that they can also be grouped into three dimensions: negativity, emotionality, and sociability. This deserves attention because the research of style elements is still in development while the research of ideological key messages is already better established. We have placed particular emphasis on stylistic elements because politicians have to convey their messages through media channels and package them effectively. If the stylistic elements we examined were used equally by all political parties (including mainstream parties), one could not assume a close relationship to populism. But our results indicate a different pattern. We find that politicians who use ideological key messages most frequently (namely, those of challenger parties and extreme parties) also use these stylistic elements most often. This is what we had anticipated, because we have primarily examined those stylistic elements the previous literature had attributed to populist politicians. This important finding leads us to conclude that we are dealing here with populism-related or even populist stylistic elements.

We further theorized that Facebook and Twitter are more susceptible to the use of populism-related communication than talk shows because social media makes it easier for politicians to bypass the traditional media, to tailor their messages to their target groups, and to present themselves as close to the people. With regard to political parties, we argue that it is more likely that young challenger parties as well as extreme left- and right-wing parties communicate in a populist manner.

Our study further demonstrates—on the basis of six countries—that parties are generally more inclined to use populism-related communication on social media than in political talk shows. By comparing the three different media channels, we corroborate that populism-related communication is indeed connected to Facebook and Twitter and that the advantages of social media to bypass gatekeepers and disseminate messages without interference are beneficial to populist communication in general. This is especially true for Facebook and confirms previous empirical evidence that Twitter is less suitable for populist communication compared to Facebook (Jacobs & Spierings, 2018). These results provide the first empirical evidence for the theoretical assumption, that online opportunity structures and populist communication logic interact (Engesser et al., 2017). The channel comparison is significant in all investigated countries except for the United States. The difference between social media and talk shows in populism-related communication by Republicans and Democrats is not significant, and politicians of the Democratic Party such as Bernie Sanders or Claire McCaskill tend to use more populism on television. However, Republicans in general and especially politicians such as John Boehner and Newt Gingrich are fully in line with our results and spread their populism-related communication mainly via social media. However, compared to European talk shows, American shows tend to offer more liberty and less journalistic interference, presenting opportunities for political actors to employ populist communication. Overall, these factors thus point to a communication culture in U.S. talk shows that is as beneficial for populist communication as social media.

At the party level, we find evidence that new challenger parties, and to a somewhat lesser extent also parties at the margins of the political spectrum, use more populism-related communication. It is not only right-wing parties such as Swiss SVP, German AfD, Italian Lega Nord, or UKIP, but also left-wing parties such as the German Linke or British Green party that make ample use of populist communication elements (Tables 3 and 5). For challenger parties, we find a predisposition toward populism in all types of statements (Models 4, 8, and 12 of Table 5), for extreme parties most strongly in statements that combine key messages with style elements (Model 4 of Table 5). We thus confirm earlier studies on the communication repertoire of extreme parties (see also Bernhard, 2016; Ernst et al., 2017; Rooduijn & Akkerman, 2017), but would like to call for future studies to pay more attention to the separate use of key messages and style elements (because we found weaker effects there). We find the strongest use of populism-related communication, incidentally, for those parties to which both the characterization “extreme party” and the characterization “new challenger party” apply at the same time (e.g., German Left, Lega Nord, UKIP). But there are also many counter examples. The Swiss SVP is a good example of a party that has a high extremism score, but has been established in Switzerland for a long time and still relies strongly on populism-related communication. The British Liberal Democrats, German Greens, or the Italian Five Star Movement are excellent examples of challenger parties, but have only a moderate degree of extremism, and yet use high levels of populism in their communication.⁶ Overall, these examples demonstrate that while commonly labeled populist parties are challenger and extreme parties at the same time, political parties holding only one of the properties also employ a high degree of populism on social media and talk shows. This confirms our argument that extreme position and challenger status are two independent properties that are separate explanatory factors for populist communication.

Some limitations must be considered. First, due to the sampling procedure, the study only includes politicians who appeared in political talk shows. Even though most of these politicians have active Twitter and/or Facebook accounts, the sample does not include politicians who are not regular talk show guests and therefore rely heavily on their social media communication. This sample strategy excludes social media affine politicians that willingly circumvent either tradition elite media (e.g., Trump) or political outsiders that are not able to find a stage in mainstream media. A sample including such politicians might find even stronger support for the argument that social media “beats” political talk shows. Another limitation is the rather low sample size in terms of countries and parties; the findings represent a specific sample and any generalizations must be drawn carefully. Since the party sample includes many more moderate center parties than extreme ones, it would be desirable to further investigate far left- and right-wing parties. The sample for France is especially problematic because it is only represented by the French Socialist Party. In the eight episodes of Le Grand Jury and Le Grand Journal, politicians from other parties were not invited.⁷ Including further key French parties such as “Front National,” “Left Front,” “The Republicans,” or “La République En Marche!” could improve the findings for the French context. A final limitation concerns the lack of any national elections during the routine time period. Populist communication could be different during elections. In particular, the debates and discussions on political talk shows might be enriched and loaded with more populist communication elements. Future studies should strive to sample both routine periods and election campaigns and include a broader sample of countries and parties to compare populist communication across these different modes of operation.

In conclusion, the study at hand contributes to populist communication scholarship in several ways. It integrates the two diverging perspectives on populist ideology and populist style and identifies three populist style dimensions that are of heuristic value for further research on populist styles. It further establishes that social media is more useful fordisseminating populism-related communication than talk shows—contesting previous studies that identified the importance of talk shows. Finally, the study finds support across countries and media platforms that parties still fighting for their place and parties positioning themselves at the polar ends of the political spectrum show the highest tendencies to use populist communication. Future research should follow a communication-centered approach and investigate the broad political spectrum. Furthermore, it is crucial to differentiate between various media channels. A next logical step would thus be to investigate and compare the populist communication by parties on social media and political talk shows with their representation in the traditional news media (print and TV) or online news.

Supplemental Material

Online_Appendix_Review_SMS-18-0070.R2 – Supplemental material for Populists Prefer Social Media Over Talk Shows: An Analysis of Populist Messages and Stylistic Elements Across Six Countries

Supplemental material, Online_Appendix_Review_SMS-18-0070.R2 for Populists Prefer Social Media Over Talk Shows: An Analysis of Populist Messages and Stylistic Elements Across Six Countries by Nicole Ernst, Sina Blassnig, Sven Engesser, Florin Büchel and Frank Esser in Social Media + Society

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Center of Competence in Research on ‘Challenges to Democracy in the 21st Century’ (NCCR Democracy), funded by the Swiss National Science Foundation (SNSF).

Notes

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Nicole Ernst (M.A., University of Zurich) is a research and teaching assistant at the Department of International & Comparative Media Research at the University of Zurich. Her research interests include populism, digital political communication and social media research.

Sina Blassnig (M.A., University of Zurich) is a research and teaching assistant at the Department of International & Comparative Media Research at the University of Zurich. Her research interests include populism, political online communication and its effects on audience reactions.

Sven Engesser (PhD, LMU Munich) is chair of Communication at Technische Universität Dresden, Germany. His fields of interest include populism in the media, political polariziation, and digital political communication.

Florin Büchel (Ph.D. from University of Zürich) is a TV researcher at Mediapulse AG. Among other topics, his research interests include political communication, media systems, media usage, and methodology.

Frank Esser (PhD, Johannes Gutenberg-University Mainz) is professor of International & Comparative Media Research at the University of Zurich and holds an adjunct professorship at the University of Oslo. His research focuses on cross-national studies of news journalism and political communication.

References

Aalberg

de Vreese

(2017). Introduction: Comprehending populist political communication. In Aalberg

Esser

Reinemann

Strömbäck

J. C.

Vreese

(Eds.), Populist political communication in Europe (pp. 3–11). New York, NY: Routledge.

Aalberg

Esser

Reinemann

Strömbäck

de Vreese

(2017). Populist political communication in Europe. New York, NY: Routledge.

Abts

Rummens

(2007). Populism versus democracy. Political Studies, 55, 405–424.

Albertazzi

McDonnell

(2008). Twenty-first century populism: The spectre of Western European democracy. Basingstoke, UK: Palgrave Macmillan.

Alvares

Dahlgren

(2016). Populism, extremism and media: Mapping an uncertain terrain. European Journal of Communication, 31, 46–57.

Baym

(2013). Political media as discursive modes: A comparative analysis of interviews with Ron Paul from Meet the Press, Tonight, The Daily Show, and Hannity. International Journal of Communication, 7, 489–507.

Bernhard

(2016, July). Left or right?: Populist communication of political parties in recent Western European elections (Working paper no. 92). Zürich, Switzerland: National Center of Competence in Research Democracy.

Betz

H.-G.

(2002). Conditions favouring the success and failure of radical right-wing populist parties in contemporary democracies. In Mény

Surel

(Eds.), Democracies and the populist challenge (pp. 197–213). Basingstoke, UK: Palgrave Macmillan.

Block

Negrine

(2017). The populist communication style: Toward a critical framework. International Journal of Communication, 11, 178–197.

10.

Bode

Vraga

E. K.

(2017). Studying politics across media. Political Communication, 9(2), 1–7.

11.

Bos

Brants

(2014). Populist rhetoric in politics and media: A longitudinal study of the Netherlands. European Journal of Communication, 29, 703–719.

12.

Bos

van der Brug

de Vreese

(2013). An experimental test of the impact of style and rhetoric on the perception of right-wing populist and mainstream party leaders. Acta Politica, 48, 192–208.

13.

Boukes

Boomgaarden

H. G.

(2016). Politician seeking voter: How interviews on entertainment talk shows affect trust in politicians. International Journal of Communication, 10, 1145–1166.

14.

Bracciale

Martella

(2017). Define the populist political communication style: The case of Italian political leaders on Twitter. Information, Communication & Society, 20, 1310–1329.

15.

Canovan

(1999). Trust the people! Populism and the two faces of democracy. Political Studies, 47(1), 2–16.

16.

Chadwick

(2017). The hybrid media system: Politics and power. Second Edition. Oxford, UK: Oxford University Press.

17.

Cranmer

(2011). Populist communication and publicity: An empirical study of contextual differences in Switzerland. Swiss Political Science Review, 17, 286–307.

18.

Engesser

Ernst

Esser

Büchel

(2017). Populism and social media: How politicians spread a fragmented ideology. Information, Communication & Society, 20, 1109–1126.

19.

Engesser

Fawzi

Larsson

A. O.

(2017). Populist online communication: Introduction to the special issue. Information, Communication & Society, 20, 1279–1292.

20.

Ernst

Engesser

Büchel

Blassnig

Esser

(2017). Extreme parties and populism: An analysis of Facebook and Twitter across six countries. Information, Communication & Society, 20, 1347–1364.

21.

Esser

Stępińska

Hopmann

D. N.

(2017). Populism and the media: Cross-national findings and perspectives. In Aalberg

Esser

Reinemann

Strömbäck

Vreese

(Eds.), Populist political communication in Europe (pp. 365–380). New York, NY: Routledge.

22.

Fisher

Marshall

McCallum

(2018). Bypassing the press gallery: From Howard to Hanson. Media International Australia, 167, 57–70.

23.

Hameleers

Bos

de Vreese

(2017). “They did it”: The effects of emotionalized blame attribution in populist communication. Communication Research, 44, 870–900.

24.

Hameleers

Schmuck

(2017). It’s us against them: A comparative experiment on the effects of populist messages communicated via social media. Information, Communication & Society, 20, 1425–1444.

25.

Hawkins

K. A.

(2009). Is Chávez populist?: Measuring populist discourse in comparative perspective. Comparative Political Studies, 42, 1040–1067.

26.

Hobolt

S. B.

Tilley

(2016). Fleeing the centre: The rise of challenger parties in the aftermath of the euro crisis. West European Politics, 39, 971–991.

27.

Hofstadter

(2008). The paranoid style in American politics. New York, NY: Vintage Books.

28.

Jacobs

Spierings

(2018). A populist paradise?: Examining populists’ Twitter adoption and use. Information, Communication & Society. 1–16. Advance online publication. doi:10.1080/1369118X.2018.1449883

29.

Jagers

Walgrave

(2007). Populism as political communication style: An empirical study of political parties’ discourse in Belgium. European Journal of Political Research, 46, 319–345.

30.

Jones

J. P.

(2010). Communication, Media, and Politics: Entertaining politics: Satiric television and political engagement. Lanham, MD: Rowman & Littlefield.

31.

Kalsnes

Larsson

A. O.

Enli

G. S.

(2017). The social media logic of political interaction: Exploring citizens’ and politicians’ relationship on Facebook and Twitter. First Monday, 22(2).

32.

Kaltwasser

C. R.

Taggart

(2016). Dealing with populists in government: A framework for analysis. Democratization, 23, 201–220.

33.

Kessler

S. H.

Lachenmaier

(2017). Ohne Belege in den Talkshow-Olymp: Belegmuster und Akteure in Polit-Talkshows zur Griechenlandkrise. Without evidence in the talk show Olympus: References and actors in political talk shows on the Greek crisis M&K Medien & Kommunikationswissenschaft, 65, 64–82.

34.

Krämer

(2014). Media populism: A conceptual clarification and some theses on its effects. Communication Theory, 24, 42–60.

35.

Krämer

(2017). Populist online practices: The function of the Internet in right-wing populism. Information, Communication & Society, 20, 1293–1309.

36.

Kriesi

(2014). The populist challenge. West European Politics, 37, 361–378.

37.

Kriesi

(2018). Revisiting the populist challenge. Politologický Časopis/Czech Journal of Political Science, 25, 5–27.

38.

Lilleker

D. G.

Koc-Michalska

(2013). Online political communication strategies: MEPs, e-representation, and self-representation. Journal of Information Technology & Politics, 10, 190–207.

39.

Mazzoleni

(2008). Populism and the media. In Albertazzi

McDonnell

(Eds.), Twenty-first century populism: The spectre of Western European democracy (pp. 49–64). Basingstoke, UK: Palgrave Macmillan.

40.

Meguid

B. M.

(2005). Competition between unequals: The role of mainstream party strategy in niche party success. American Political Science Review, 99, 347–359.

41.

Mény

Surel

(2002). The constitutive ambiguity of populism. In Mény

Surel

(Eds.), Democracies and the populist challenge (pp. 1–21). Basingstoke, UK: Palgrave Macmillan.

42.

Moffitt

(2016). The global rise of populism: Performance, political style, and representation. Standford, CA: Standford University Press.

43.

Mudde

(2004). The populist zeitgeist. Government and Opposition, 39, 542–563.

44.

Müller-Rommel

(1998). The new challengers: Greens and right-wing populist parties in Western Europe. European Review, 6, 191–202.

45.

Newman

Fletcher

Kalogeropoulos

Levy

Kleis Nielsen

(2017). Reuters institute digital news report 2017. Oxford, UK: Reuters Institute for the Study of Journalism.

46.

Polk

Rovny

Bakker

Edwards

Hooghe

Jolly

. . . Zilovic

(2017). Explaining the salience of anti-elitism and reducing political corruption for political parties in Europe with the 2014 Chapel Hill Expert Survey data. Research & Politics, 4(1), 1–9.

47.

Rooduijn

(2014). The mesmerising message: The diffusion of populism in public debates in Western European media. Political Studies, 62, 726–744.

48.

Rooduijn

Akkerman

(2017). Flank attacks: Populism and left-right radicalism in Western Europe. Party Politics, 23, 93–204.

49.

Rydgren

(2017). Radical right-wing parties in Europe. Journal of Language and Politics, 16, 485–496.

50.

Schulz

(2018). Where populist citizens get the news: An investigation of news audience polarization along populist attitudes in 11 countries. Communication Monographs, 1–16.

51.

Sorensen

L. N.

(2017). Populism in communications perspective: Concepts, issues, evidence. In Heinisch

Holtz-Bacha

Mazzoleni

(Eds.), Handbook on political populism (pp. 137–151). Baden-Baden, Germany: Nomos.

52.

Stanley

(2008). The thin ideology of populism. Journal of Political Ideologies, 13, 95–110.

53.

Stanyer

(2012). Intimate politics: Publicity, privacy and the personal lives of politicians in media saturated democracies. Cambridge, UK: Polity Press.

54.

Stieglitz

Dang-Xuan

(2013). Social media and political communication: A social media analytics framework. Social Network Analysis and Mining, 3, 1277–1291.

55.

Stier

Posch

Bleier

Strohmaier

(2017). When populists become popular: Comparing Facebook use by the right-wing movement Pegida and German political parties. Information, Communication & Society, 20, 1365–1388.

56.

Stockemer

Barisione

(2017). The “new” discourse of the Front National under Marine Le Pen: A slight change with a big impact. European Journal of Communication, 32, 100–115.

57.

Taggart

(2000). Populism. Buckingham, UK: Open University Press.

58.

van Aelst

Strömbäck

Aalberg

Esser

de Vreese

Matthes

Stanyer

. (2017). Political communication in a high-choice media environment: A challenge for democracy? Annals of the International Communication Association, 41(1), 3–27.

59.

Weber

(2017, January). Populist parties in Switzerland and their integration into the establishment. Conference paper presented at NCCR Democracy Final Workshop.

60.

Wirth

Esser

Wettstein

Engesser

Wirz

Schulz

. . . Schemer

(2016, May). The appeal of populist ideas, strategies and styles: A theoretical model and research design for analyzing populist political communication (Working paper no. 88). Zürich, Switzerland: National Center of Competence in Research Democracy.

61.

Wodak

(2015). The politics of fear: What right-wing populist discourses mean. London, England: SAGE.

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.

0.00 MB

0.02 MB