When challengers become politicians: The protest paradigm
applied to the coverage of new political parties
Aida Lumbreras Higuera
12328618
Master’s Thesis
Graduate School of Communication
Master’s programme: Communication Science
Supervisor: Joost van Spanje
31/01/2020
Abstract
New parties’ success in the polls has been largely studied in the past. While scholars point at different reasons to obtain parliamentary representation, the common
understanding is that the media, amongst others, influences how citizens perceive parties, which ultimately impacts democracy through voting. However, media coverage is inevitably biased since every media holds their own political, economic and societal preferences. In this study we apply the protest paradigm, which measures the biases of the media regarding challenger actors, to the coverage of two new parties with
opposed ideologies. Through content analysis, we examine whether coverage in the same media differs between parties. This study constitutes a novelty since little previous research has applied the protest paradigm, traditionally associated to social movements, to measure biases towards new parties. Our results partly support the existence and application of the protest paradigm to the coverage of new parties that do not match in ideology with the newspaper they are being portrayed in, while further research should be conducted on this issue.
Keywords: Protest paradigm, new parties, Vox, Podemos, media biases, El Mundo, El País
Introduction
The appearance of new political parties within democratic systems is not uncommon. However, their success and entrance into parliament is, indeed, quite uncommon. Many of them never get the minimum support needed from citizens and, as a result, end up dissolving. Nevertheless, some parties are able to gain the support to position themselves as valid voting options for citizens.
Media attention has shown to be critical, even crucial, in bringing success to new parties, given that it is through media that new parties can spread their messages and gain
supporters. In this study, we will look at the way the media, and more specifically the press, gives voice and covers new parties.
However, media coverage regarding political matters is never completely neutral, and every media, whether private or publicly run, holds their own political, economic and societal preferences. These preferences can translate into biases when covering political parties with different ideologies and propositions and, in turn, influence the perception readers have about said political party. As an example, a more ideologically right-wing conservative newspaper could give a more positive coverage to parties which support a higher financing of the Church than those advocating for a strict division between Church and Government. Knowing all this, we wonder if these media’s ideological biases produce a different coverage and treatment of new parties with a more right-wing or left-wing ideology.
Thus, the research question of this study will be the following:
RQ: Are new left and right-wing parties treated differently by the media?
New parties generally act as challenger actors, since they can threaten the position and power of the legacy parties. Some of them are even born as social movements and later gain the structure to become a political party. Therefore, in this study we will examine press coverage making use of the protest paradigm, which has been mostly used to analyse the media coverage social movements and demonstrators but can be applied to any actors such as new parties challenging the status quo.
While many studies have examined new parties, the scientific relevance of this study lies in the fact that very few have examined the coverage of new parties in the media, despite the effect that media has shown to have in creating and reinforcing opinions. Also, as no previous studies have analysed the coverage of new parties through the
protest paradigm, the novelty of using it and the possibility that we find this paradigm does apply to new parties adds up to the scientific relevance of this study.
Similarly, the societal relevance lies in the importance that media has in keeping citizens informed about politics (Schulz, 1994), whilst at the same time, as previously mentioned, in how the media shapes the opinion of its readers. Research has not only shown that the media shapes the opinion of readers (Norris et al., 1999), but also that this could have an effect on party choice and voting (Chiang & Knight, 2008; Hopmann et al., 2010). The different types of coverage and evaluations that media does of different parties could have an impact on how readers perceive new parties and, therefore, who they vote for. This ultimately has an impact on democracy, as it would mean that the media plays a key role in deciding which parties will be successful, and which ones will not obtain parliamentary representation.
Theoretical framework
New parties in democratic systems: success, media coverage and biases
In countries in which democracy is the form of political organization, it is not uncommon that new parties come into scene and try to join the political sphere. Of all these new political groups, most of them never get to succeed and have parliamentary
representation (Bolleyer & Byztek, 2013; Berrington, 1985), but a few do.
Nevertheless, defining new parties has proven to be difficult, with no clear agreement on what constitutes a new party (Šárovec, 2019). Several authors have proposed different conceptualizations, with some stating that a new party is one that is not successor to any previous party, with new faces and a fresh name and structure (Sikk, 2005); others considering new only those parties where “no more than half of its top candidates […] originate from a single former party“ (Barnea & Rahat, 2011, pp.311); while others think that a new party is only a “genuinely new organization that appoints,
for the first time, candidates at a general election to the system‘s representative assembly" (Hug, 2001, pp.14).
However, and for this study, we will take Hug’s broader definition of a new party, which is “one that either results from a split from an existing party or is genuinely new in the sense that it emerges without any help from members of existing parties” (Hug, 2001, pp.79).
Reasons why new parties emerge and try to join the political arena are several. The likeliness of success, diversity within the population or the appearance of new issues in society seem to be the reasons why new parties try to just enter the competition
(Tavits, 2007; Harmel & Robertson, 1985). On the other hand, access requirements such as petitions and electoral deposits could prevent new parties from even trying to compete (Rosenstone et al., 1984).
But to have political success is a different thing than just running for office. Several scholars agree that periods of crisis trigger the success of new political formations. Political dissatisfaction, corruption or unemployment are some of the reasons that can help rise new political parties (Tavits, 2007; Žúborová, 2015; Vidal, 2017). Moreover, a new party does not only need resources to enter the competition, but also to be able to win it. In words of Lucardie (2000, pp.179), a party will need “a minimal number of members, a minimal campaign budget and a minimal amount of publicity” and, without it, it will never gain representation.
Furthermore, and along with the previously mentioned, scholars have started to consider the influence of the media in parties’ success. The mass media, for example, can decide whether to give coverage or not to new parties, but can also prevent people from voting for it by ridiculing it (Lucardie, 2000). Research has already studied the effects that media can have on vote and party preference on established parties. Van Aelst et al. (2008) found that media attention contributes to electoral success, and that
both a media logic and a party logic influence how much media attention is given to different parties and politicians.
Similarly, Eberl, Boomgarden and Wagner studied how media biases in the coverage of political parties can influence party preference. In their study, these researchers found two types of media biases that influence party preferences. Firstly, the agenda bias refers to “the extent to which political actors appear in the public domain in conjunction with the topics they wish to emphasize”, while the “tonality bias measures whether evaluations present in media coverage are systematically more favourable to one political party compared to other parties”. (2017, p.1128)
These biases are reflected in the protest paradigm, which will be used in this study. Most of the elements of this paradigm, which will be discussed in the next section, are based on the neglect of certain groups’ ideas and propositions as well as their negative portrayal in media coverage.
While this paradigm has normally been applied to social movements, researchers have also applied it to political parties (van Spanje & Azrout, 2019). Some political parties have been created from protests themselves. Well acknowledged examples are
Podemos in Spain, which raised from the 15-M movement (Borge Bravo & Santamaría Sáez, 2016) and the Bolivian Socialist movement, which became a party in the 1960’s (Somuano Ventura, 2007).
Somuano Ventura (2007, p.40) even argues that “social organizations derived from movements and that create their structures can get to resemble a political party”. Therefore, we can argue that social movements and political parties can share certain similarities and that movements can become political parties, which means that it is logical to apply the protest paradigm to political parties.
Being aware of the importance of the media in giving coverage and bringing success to new parties, we wonder whether their own ideological biases produce a different coverage and treatment of new parties with different ideologies. We wonder whether the same media applies these biases to different parties with contrary ideologies, and we will examine this by applying the protest paradigm.
Protest paradigm and its application to this study
The protest paradigm is, to put it in McLeod’s words, a “set of news coverage patterns that typifies mainstream media coverage” (2007, p.1). In this coverage, which results in biases, protesters or challenger actors are described as ineffective or threatening, while calling public opinion to stand against the protesters (McLeod & Hertog, 1998).
As we have previously argued, news biases can have an effect on new parties’ success. The protest paradigm gathers several biases that can affect the image of challenger actors and, therefore, the perception that citizens have of them (McLeod, 2007)
1. News frames: some journalistic frames are more commonly used than others within the protest paradigm. Some examples are framing the challenger group as rioters or criminals (McLeod, 2007).
These frames can be used to “weaken legitimacy, obscure a protest’s social/political concerns, or both” (Weaver & Scacco, 2013, p.64). When news articles are framed in a way that emphasizes the violence of protesters, for example, no emphasis is put on the roots of the conflict but the message sent to readers is that the protesters are violent and have no right to demonstrate. (Weaver & Scacco, 2013). As an example, in an experiment conducted by McLeod and Detenber (1999), participants exposed to news about protests that challenged the status quo gave less support to them than
Therefore, for our first hypothesis, we would expect that:
H1: A newspaper covering two different parties will more often negatively frame the party with a different ideology to the one of the newspaper itself than the party with a similar ideology.
2. Reliance on official sources and official definitions: Official sources give the news more credibility and it is easier to get this information. But, giving voice to the “powerful” minimizes the perspectives and views of the challengers.
For McLeod and Hertog (1998), reasons why journalists rely so much on official
sources are that this adds prestige to a story, increases the efficiency of the process of news production since these sources are normally readily available and, finally, using external sources gives an image of objectivity. The downside of this is, according to McLeod and Detenber (1999), that this only reinforces the status quo when narrating protest stories, since the story is only being told from the side of the power incumbents.
Therefore, we would expect that:
H2: A newspaper covering two different parties will more often employ official sources in the coverage of a party with a different ideology than in the coverage of a party with a similar ideology to the newspaper.
Similarly, we wondered whether the newspapers would change the sources they use depending on their party preference, once again giving preference to external sources than the party being covered. Therefore, our next two hypotheses will be the following:
H3: A newspaper covering two different parties will more often use a journalistic or external party source in the coverage of a party with a different ideology than of a party with a similar ideology.
H4: A newspaper covering two different parties will more often use the party being covered as a source when the party has a similar ideology to the newspaper than when it has a different one.
3. Invocation of public opinion: Journalists try to highlight the differences between the challengers and the rest of society. Several tools can be used to create this separation.
Both McLeod and Hertog (1992) and Dardis (2006) elaborate on these different tools. For example, journalists can use opinion polls or statistics to refute the protesters’ stance and generalize or dramatize the reaction of the public opinion towards the protest. Also, they can give voice to witnesses or bystanders to help create this exaggeration and differentiation between them and the rest of society who follows social norms. We assume that these tools will be greatly used when ideological differences are greater, therefore our next hypothesis will be as follows:
H5: A newspaper covering two different parties will more often employ negative public opinion in the form of negative citizens’ opinions or polls when the party being covered has a different ideology than the newspaper than when it has a similar one.
4. Delegitimization: The media can portray the challengers and their actions as irrational or failures or can fail to explain the context and reasons of the protests, which leads to perceive them as pointless.
Delegitimization can have strong effects in the case of political parties, given that, according to Bos and van der Brug (2010, p.780) voters “probably only support parties that they perceive to be legitimate”. In their study on perceptions legitimacy and
effectiveness, the authors found that both traits had a strong effect on electoral support and party preference. We can therefore argue that if media delegitimize a political group when having opposite opinions, citizens will be more reluctant to cast their vote in favour of this party. Therefore, it is arguable that:
H6: Parties that do not match the newspapers’ ideology will be delegitimized when covered in the news more often than parties that match the newspaper’s ideology.
5. Stigmatization: Media coverage can exaggerate the violence and threats of the protesters, stereotype them and negatively label them, or focus mainly on the negative consequences of the protests such as the monetary costs of the protests.
A stigma appears when, in words of Link and Phelan, “elements of labelling, stereotyping, separation, status loss, and discrimination occur together in a power situation that allows them” (2001, p.377). If a party is negatively portrayed and
stereotyped it will also lose legitimacy, that could lead once again to a lack of support at the polls. We argue, then, that stigmatization will occur more often when there is a mismatch of opinions between the party and the media:
H7: Parties that do not match the newspaper’s ideology will be more often stigmatized when covered in the news than parties that match the newspaper’s ideology.
6. Demonization: In words of van Heerden (2014, p.10), this is ‘‘portraying an actor as the embodiment of absolute evil’’. In their study, conducted in The Netherlands, to demonize a group would be to relate it with Nazism. However, we can argue that, when the context varies, demonization strategies also change.
Demonization can work as a warning or as a strategy. The first one is used to frighten people about the consequences of the possible success of that actor. The latter one is used as a political strategy to obtain an electoral benefit in the polls, while there is no real fear. (van Heerden, 2014) Therefore, we argue that actors disagreeing with the media’s opinions will be more likely demonized, hence being our last hypothesis as follows:
H8: Parties that do not match the newspaper’s ideology will be more often demonized when covered in the news than parties that match the newspaper’s ideology.
To sum up, these eight hypotheses will help us measure to which extent the protest paradigm is applied to new political parties and whether it is more likely applied to the parties with whom the newspaper have greater ideological differences.
The model of analysis: Spanish new parties within the media
In this study, we will compare the two most recently emerged successful parties in Spanish politics: the right-wing party Vox and the left-wing party born from the Indignados movement, Podemos. Spain offers a good political scenario to work with, as new parties with very different ideologies have recently not only started to emerge but also to obtain representation, after being a bipartisan system for nearly forty years. (Oñate, 2006) The main reasons for the upsurge of new parties in Spain were,
according to Rama (2016), the economic recession, the corruption cases in legacy parties, as well as the acceptance of the austerity measures imposed by the European Union.
In December 2015, when it first run for general election, Podemos obtained 69 seats. We consider Podemos a new party before the 2015 elections since the party, founded on the 17th January 2013, had never stood for general election before and, therefore, never gotten any parliamentary representation. Podemos has been described in the literature as a left-wing populist party that derived from the social movement 15-M, the Indignados, and that opposed the austerity measures imposed by the EU during the economic recession, the political corruption of PSOE (main labour party in Spain) and PP (main conservative party) and the bipartisan system. (Kioupkiolis, 2016; Zarzalejos, 2016; Casero-Ripollés, Sintes-Olivella & Franch, 2017)
On the other side of the political spectrum, Vox was founded on the 17th December 2013 and obtained parliamentary representation for the first time in April 2019, with 24 deputies. Vox has been described as a far-right populist party by several authors, given their advocacy for more restrictive immigration laws and their nationalist rhetoric. (Rubio-Pueyo, 2019; Hernández Conde & Fernández García, 2019; Climent Sanjuan & Montaner Goetzenberger 2019).
There are several reasons why we will choose these two parties. First, the great ideological differences between them, which will in turn show clearly whether
newspapers treat them differently. Secondly, the fact that both parties were founded and gained representation. And ultimately, the fact that both parties have populist traits, which means we should not expect differences in the type or amount of news coverage for this reason. We believe that these characteristics will make it easier obtain accurate answers to our research question.
Likewise, several reasons justify our model for analysis, which will examine the
coverage of these two parties within one newspaper and not the difference of coverage between newspapers for the same party. Firstly, the focus of this study consists in examining how some biases might be present in a specific media only when the reporting falls on the party with opposed ideas to this media. Therefore, analysing the coverage of one party in two different media would answer an entirely different
question, which is whether different media covers a party differently.
Secondly, while the periods between the party foundation and the reach of
representation which we will analyse are not strictly the same, they overlap for nearly half of the time, about two years. Podemos was only founded one month after Vox, but it reached parliamentary representation earlier and became an established party, and therefore could no longer be analysed as it had lost the condition of new party. We can argue that, the period overlapping for that long means no greater differences in the
newspaper’s ideology should be found when the analysis continues for Vox, and also that, at the point Podemos obtained representation, possible biases about Vox should have already emerged and not substantially change.
Moreover, it would be extremely difficult to examine the different portrayal of two new parties in the same media at the same exact period considering the unlikeliness that two new parties with opposed ideologies are not only founded at the same time, but also gain representation at the same time, which as previously mentioned, is extremely rare.
Also, we are not trying to find a correlation that indicates that Podemos’ positive coverage is correlated with Vox’s negative coverage (or vice versa). We will examine whether media biases comprised by the protest paradigm apply greater to a party which has ideological differences with the media it’s being covered in than one with ideological similarities, but no relationship between the parties is being looked for, which therefore, indicates that having the exact same period for analysis is not strictly necessary.
Methods
For our analysis, we decided to conduct a content analysis. To obtain our sample, we used Nexis Uni to browse the news published about Podemos and Vox in the two media chosen for the analysis, El Mundo and El País. The election of these two newspapers relies on the fact that they are from different ideologies, with El Mundo being more conservative and right-wing and El País more progressive and left-wing, as well as the fact that both of them are the two national largest newspapers, which means they count with the resources and personnel to publish detailed political coverage and not rely on external sources such as news agencies. Choosing newspapers as the media to analyse was decided upon the ready availability of the data online.
Considering the time and resources available for coding, we decided to obtain a sample of around 320 news articles. Since we have two newspapers and two political parties, we would ideally obtain a sample of at least 80 articles for each newspaper and party. To reduce the number of results and make sure all of them would be of interest, we decided to sample news that contained several related words. Therefore, we decided to search for the name of the party (Vox and Podemos), the surname of the general secretary of the party, which is Abascal, for Vox, and Iglesias, for Podemos, and also for the Spanish word for “party” (partido).
Therefore, to gather the news about Vox published by El Mundo, we searched for the words “Vox”, “Abascal” and “partido” while selecting the dates of interest, which
comprises the dates between the 17th December 2013 and the 28th April 2019, the date of the party foundation and the general election day when they first obtained
representation. Then we set Nexis Uni to only show news published by El Mundo and sorted them from oldest to newest. We obtained 440 results, and since we need at least 80, we chose every fifth article, obtaining a final sample of 87 news.
Similarly, to obtain the news about Vox published by El País, we just adapted the previous search to only show articles published by this newspaper. We obtained a similar number of results, which is why we also chose every fifth article and thereby obtained 85 articles.
When looking at Podemos, we searched for the words “Podemos”, “Iglesias”, and “partido” on the dates between the 17th January 2014 and the 20th December 2015, which are the dates of party foundation and the date they first got parliamentary
representation. For the news published by El Mundo, we obtained 1,344 results, so we chose every 15th article, obtaining 89 news for our sample. Finally, for the articles published in El País, we firstly obtained 1,903 results and therefore chose every 25th article. This only reported 78 articles, which is why we decided to include the oldest and the newest article to obtain the 80 news items aimed for. From these 341 articles,
we had to take out four since they either appeared twice or were non-related to the new parties we were coding. Therefore, the final sample was N=337.
While our sampling procedure might not be a random sampling method, it falls closely, since it allows us to randomly select news and to select news all through the sampling period.
The codebook included 25 questions coded dichotomous and two coded on a 5 point-scale to assess positive versus negative evaluations. These questions, that can be found in the appendix, assessed all indicators of the protest paradigm previously mentioned.
Firstly, for the news frames bias, we looked at a negative framing and negative labels. Therefore, we coded whether the party mentioned was evaluated and, if so, how positively or negatively on a 5 point-scale, being 1 the most negative and five the most positive. On this same indicator, we assessed whether the party or politicians were being called performers or fake, whether they were called radicals and were associated and/or compared to other radical figures or if they were called criminals and were associated and/or compared to criminals.
Table 1. Variables measuring news frames bias.
Variable Frequency Percentage Mean S. D.
Are they evaluated No 182 54 .46 .49
Yes 155 46
How are they evaluated (1 negative to 5 positive)1 1 91 58.7 2.19 1.64 2 16 10.3 3 9 5.8 4 6 3.9 5 33 21.3
Called performers or fake No 305 90.5 .09 .29
Yes 32 9.5
Called radicals No 313 92.9 .07 .26
Yes 24 7.1
Associated with radicals No 319 93.5 .05 .22
1 The N for this variable is not N=337 but N=155, since only those articles that were coded as
Yes 18 5.3
Compared to radicals No 323 95.8 .04 .20
Yes 14 4.2
Called criminals No 333 98.8 .01 .11
Yes 4 1.2
Associated with criminals No 334 97.9 .01 .09
Yes 3 .9
Compared to criminals No 332 98.5 .01 .12
Yes 5 1.5
Source: own production.
To look at the actors and sources indicators, we coded whether the information that mentioned the new party came from a journalist, another party member or the new party itself. We added this indicator since we deemed it relevant to observe whether the media gives voice to the new parties or only covers them through other sources.
Also, and as previously mentioned, relying on official sources is used to avoid giving voice to challenging groups and discredit them while giving an image of objectivity. We looked for the coverage of official figures such as representatives of organizations as for example the United Nations or Amnesty International, the Government, judges, professors or other experts in fields such as sociology or economics. In order to fit into this category, it was necessary that the position of the source was explicitly mentioned
Table 2. Variables measuring political and journalistic sources.
Variable Frequency Percentage Mean S. D.
Source: journalist No 46 13.6 .86 .34
Yes 291 86.4
Source: new party No 230 69.4 .31 .46
Yes 103 30.6
Source: another party No 264 78.3 .22 .41
Yes 73 21.7
Official source mentioned No 320 92.0 .08 .27
Yes 27 8.0
Source: own production.
In the invocation of public opinion, which tries to give the readers the idea that there is a general dissatisfaction towards the new party, we looked at the negative comments about the new parties coming from citizens or the society at large, whether explicit in
quoting or implied by the journalist. Also, we looked for polls or statistics that portrayed the negative opinion of these citizens.
Table 3. Variables measuring the invocation of public opinion.
Variable Frequency Percentage Mean S. D.
Negative citizens comments
No 330 97.9 .02 .14
Yes 7 2.1
Negative polls or statistics No 329 97.6 .02 .15
Yes 8 2.4
Source: own production.
Additionally, the delegitimization indicators were aimed at the depiction of new parties as unsuitable options for a good and democratic government. Hence, we decided to code whether newspapers portrayed new parties as a threat to democracy and political stability, and if the same politicians were being called irresponsible and populist, which would make them look like an unwise choice for voters. Also, we coded whether the plans and ideas of the parties were mentioned and if these were supported or opposed on a 5 point-scale.
Table 4. Variables measuring delegitimization.
Variable Frequency Percentage Mean S.D.
Threat to democracy No 325 96.4 .04 .19
Yes 12 3.6
Threat to political stability No 325 96.4 .04 .19
Yes 12 3.6
Called irresponsible No 333 98.8 .01 .11
Yes 4 1.2
Called populist No 307 91.1 .09 .28
Yes 30 8.9
Ideas and plans evaluated No 255 75.7 .24 .43
Yes 82 24.3
How are they evaluated (1 negative to 5 positive) 1 32 39.0 2.22 1.05 2 4 4.9 3 44 53.7 4 0 0 5 2 2.4
Source: own production.
Stigmatization traits include the highlighting of characteristics of a group, that would give them a bad reputation or image. As stigmatization indicators, we looked at the
portrayal of the new parties as violent, inexperienced and whether their mental abilities or political capabilities were questioned. In addition, we coded whether the physical appearance of the politicians was object of critique or mockery.
Table 5. Variables measuring stigmatization.
Variable Frequency Percentage Mean S. D.
Violent No 336 99.7 .00 .05
Yes 1 .3
Inexperienced No 321 95.3 .05 .21
Yes 16 4.7
Mental abilities questioned No 332 98.5 .01 .12
Yes 5 1.5 Political competence questioned No 322 95.3 .04 .21 Yes 15 4.7 Physical appearance criticised No 336 100 .00 .00 Yes 0
Source: own production.
Finally, with regards to demonization, we looked for a group that could be portrayed as the summum of evil in a Spanish context. For van Heerden (2014), in her study within the Dutch context, Nazism and everything related to it was the group chosen. However, this study has a different context and, therefore, another group is needed. Since
Nazism did not directly affect Spain, we decided that the most resembling example of an unanimously despised group in Spain appears to be the terrorist group ETA (“Basque Country and Freedom”, from its initials in Basque).
Table 6. Variable measuring demonization.
Variable Frequency Percentage Mean S. D.
Related to ETA or terrorism No 333 98.8 .01 .11
Yes 4 1.2
Source: own production.
This Basque nationalist group, formed in 1959, killed 858 people between the years 1968 and 2010. (Alonso, 2011), in over 2,000 attacks (LaFree et al., 2012). This group created “an intense campaign of violence” (Alonso, 2011, p.696) that included
destructive” (Mahoney, 2018, p.916). Interestingly, groups linked to ETA and Basque nationalism were framed as “totalitarian,” “fascist,” or “Nazi” by medias and politicians (Bourne, 2018, p.328) which, along with the previously mentioned, reinforces our argumentation that ETA is a good example of a group portrayed as evil.
To assess the Inter-Coder Reliability with Krippendorf’s Alpha, a second coder coded 15 percent of the articles. In total, this coder coded N=52 news items. He was asked to code items from each newspaper and political party mentioned following the codebook. Most of our results showed a high reliability between coders.
When looking at whether they were evaluated, Krippendorf’s Alpha showed an .77 agreement, and the ranking of those evaluations on a scale from one to five showed a .72 agreement. New parties being called fake or performers showed a .70 agreement. Concerning radicalism, we obtained good reliability results in the question of whether the parties were called radicals (.78), associated with radicals (.78) or compared to radicals (.85). Whether the parties were called “criminals” also showed a good result (.79), just as being associated with radicals (.73) and compared to radicals (.79) did.
The source of the comments also showed a good inter-coder reliability measure. Agreement on the new party as a source was high (.81), while the agreement on another party as a source was a little lower (.77) and the agreement on the journalist source of the comments was high as well (.79). The variable measuring the use of negative polls or statistics showed a perfect result of 1.0. Negative messages from official sources and from citizens or society also reported good reliability results, with .81 and .78, respectively.
The variables measuring whether the parties were portrayed as a threat to democracy and a threat to political stability both showed good results, with .88 and 1.0
being called irresponsible or populists also showed very good results, with .81 and 1.0, respectively.
While the reliability on the variable measuring whether the plans and ideas of new parties were explained was high (.81), the inter-coder reliability for whether these ideas were supported or opposed was lower but still acceptable (.72).
Inter-coder reliability for the portrayal of new parties as violent was below acceptable (.65). The reason for this is most probably the extremely low number of news items that contained a new party named violent. Therefore, we should be cautious when running tests for this variable.
The inter-coder reliability for the variable measuring the questioning of the mental abilities of new parties’ members was high (.85), while it was slightly lower for the one questioning political competences (.77), but perfect for the portrayal of new parties as amateurs or inexperienced (1.0). Finally, no variance was found while testing the inter-coder reliability for the variable that measured the critics of the physical appearance of new parties’ members, and therefore no Krippendorf’s Alpha could be calculated. Given this, this variable will not be used in the analysis or exposed in the results.
Results
To test our hypotheses, we conducted several Chi-square tests per newspaper, since, as mentioned beforehand, we aim to analyse the differences between the coverage of the two parties within one newspaper.
News frames
Our first hypothesis considered that a new party would be more negatively framed when covered by a newspaper with a different ideology than said party, compared to the coverage of a party with a similar ideology. Therefore, we would expect El País to
negatively frame Vox more often than Podemos, and El Mundo to negatively frame Podemos more often than Vox.
Table 7. Chi-square results of the news frames indicators.
Variable El País El Mundo
Are they evaluated X2 (1, N=164) =2.64, p=.104 X2 (1, N=173) =7.92, p=.005
How are they evaluated X2 (1, N=164) =6.45, p=.168 X2 (1, N=173) =5.77, p=.217 Called performers or fake X2 (1, N=164) =.023, p=.881 X2 (1, N=173) =5.80, p=.016 Called radicals X2 (1, N=164) =5.71, p=.017 X2 (1, N=173) =6.14, p=.013 Associated with radicals X2 (1, N=164) =.329, p=.566 X2 (1, N=173) =8.29, p=.004 Compared to radicals X2 (1, N=164) =8.01, p=.005 X2 (1, N=173) =6.14, p=.013
Called criminals N.A. X2 (1, N=164) =8.01, p=.005
Compared to criminals X2 (1, N=164) =.00, p=.961 X2 (1, N=173) =.99, p=.319
Associated with criminals
X2 (1, N=164) =.96, p=.328 X2 (1, N=173) =2.00, p=.157
Source: own production.
Almost half of our variables are significant, with considerably more significant results when examining El Mundo, as can be seen in the table. Firstly, Podemos was
evaluated significantly in more news items than Vox when analysing El Mundo, which was not the case for El País. Moreover, and as can be observed in the table, none of the newspapers showed a difference in the evaluations of both parties.
Podemos was also called fake by El Mundo and accused of performing significantly more times, and the same applies to being called radicals, associated with radicals, and compared to radicals. For El País, a statistically significant difference between parties was found in more likely calling Vox radical and comparing it to radicals. On the other hand, Vox was not more likely called fake, neither was it associated with radicals.
Regarding the topic of criminality, in El Mundo, Podemos and its members were more likely called criminals, but not associated with criminals or compared to them. For El País, none of these three variables showed a difference in the coverage between parties.
Based on these results, we can partly confirm our hypothesis that a party will be
framed more negatively in a newspaper when ideologies between media and said party differ. However, we cannot either refute our hypothesis since we found eight significant results in the direction our hypothesis suggested.
Reliance on official sources
Our second hypothesis examined the appearance of official sources, which are used to give a sense of neutrality and objectivity, but also neglects giving voice to the
challenger parties. From eight analyses conducted regarding this matter, only two resulted significant, as can be observed in the following table.
Table 8. Chi-square results of indicators measuring reliance on official sources.
Variable El País El Mundo
Use of official sources X2 (1, N=164) =.329, p=.566 X2 (1, N=173) =7.75, p=.001
Source: journalist X2 (1, N=164) =006, p=.937 X2 (1, N=173) =11.63, p=.001
Source: new party X2 (1, N=164) =2.42, p=.120 X2 (1, N=173) =.61, p=.443
Source: other party X2 (1, N=164) =1.58, p=.209 X2 (1, N=173) =.91, p=.339
Source: own production.
At El Mundo, official sources were much more used when talking about Podemos than Vox. However, the use of official sources did not differ either between parties when analysing El País. Hence, this hypothesis cannot be completely refuted neither accepted.
Our third and fourth hypotheses focused on the source of the information concerning new parties. We argued that Vox and Podemos respectively would not be the source when covered by a newspaper with a different ideology, and similarly, that legacy parties or journalists would most likely be the source when covering a party with ideological differences.
In El Mundo, the source of the information did not statistically differ between parties when it came from a new party or member source, neither did it in El País. Therefore,
our third hypothesis can be rejected. The use of other parties or journalists as a source did not differ between parties in any of our newspapers, and only in the case of El Mundo, the source was more likely to be a journalist when the coverage was done about Podemos. Considering only one of our four variables measuring our fourth hypothesis is significant, we can refute it.
Invocation of public opinion
For our fifth hypothesis, we expected that negative public opinion in the form of
negative citizens’ opinions or polls would be more often used in the coverage of parties with a different ideology than the newspaper.
Table 9. Chi-square test results of variables measuring invocation of public opinion.
Variable El País El Mundo
Negative opinion of citizens X2 (1, N=164) =.00, p=.972 X2 (1, N=173) =1.82, p=.178
Negative polls or statistics X2 (1, N=164) =.26, p=.610 X2 (1, N=173) =3.89, p=.079
Source: own production.
However, none of our two newspapers examined showed a difference between parties regarding the use of a more negative opinion of citizens or the use of more polls and statistics depending on the political party covered. In conclusion, we can state that our fifth hypothesis is refuted.
Delegitimization
For our next and sixth hypothesis, we asserted that parties which do not match the newspapers’ ideology would be delegitimized when covered in the news more often than parties that match the newspaper’s ideology. Just four out of our twelve analyses showed significant results.
Table 10. Chi-square tests results of variables measuring delegitimization.
Variable El País El Mundo
Threat to democracy X2 (1, N=164) =7.43, p=.006 X2 (1, N=173) =1.02, p=.313
Threat to political stability X2 (1, N=164) =2.57, p=.109 X2 (1, N=173) =6.14, p=.013
Called populist X2 (1, N=164) =1.83, p=.176 X2 (1, N=173) =3.29, p=.070
Plans and ideas mentioned X2 (1, N=164) =9.47, p=.0022 X2 (1, N=173) =.81, p=.368
Plans and ideas supported or opposed
X2 (4, N=36) =10.48, p=.033 X2(3, N=49) =2.23, p=.526 Source: own production.
To test this, we looked at whether there was a difference in the portrayal of new parties as a threat to democracy, which was significant for El País, with Vox portrayed as a greater threat than Podemos, but not for El Mundo. On the contrary, Podemos was, according to El Mundo, a greater threat than Vox to political stability, while no difference was found in El País regarding this matter.
Being called irresponsible did not show significant results for any of the newspapers. Likewise, no difference was found in the amount of times newspapers called Vox and Podemos populist, neither for El Mundo, nor El País.
Regarding whether the plans and ideas of the parties were mentioned, we found no significant difference between parties in El Mundo. In El País, however, we found that Vox’s ideas and party plans were much more mentioned than Podemos’ ideas, which goes in the opposite direction that we were expecting. We expected Vox’s ideas to be silenced by El País and not amplified. Looking at whether these plans were more supported or opposed depending on the party covered, we found a non-significant result in El Mundo, while El País did show that Vox’s ideas were less supported than those of Podemos, which lies in the direction we were expecting.
Looking at the results, we can state say that our sixth hypothesis is partly supported.
Stigmatization
Considering stigmatization, we argued that a new political party would be more stigmatized when covered by a newspaper from a different ideology. Firstly, some of
2 While this variables shows significance, it is in the opposite direction that we suggested, and
our variables could not be measured. This was the case for the portrayal or parties as violent in El Mundo, where no such items were found. In El País, we found only one news item that portrayed Vox as violent and no news that portrayed Podemos as such, showing a non-significant difference. In any of the news items did we find comments that criticised the physical appearance of party members, and therefore no analyses could be conducted.
Table 11. Chi-square tests results of variables measuring stigmatization.
Variable El País El Mundo
Called violent X2 (1, N=164) =.958, p=.328 3
Physical appearance criticised
Mental abilities questioned X2 (1, N=164) =.958, p=.328 X2 (1, N=173) =1.00, p=.317
Called inexperienced X2 (1, N=164) =.329, p=.566 X2 (1, N=173) =.00, p=.989
Political competence questioned
X2 (1, N=164) =.595, p=.441 X2 (1, N=173) =9.38, p=.002
Source: own production.
Regarding the politicians’ capabilities, we observed that the differences between parties regarding their mental abilities being questioned were non-significant for both newspapers. The difference between parties being called inexperienced or amateurs was also non-significant both for both El Mundo and El País. Finally, El País did not question the political competence of Vox more than it did of Podemos, while in turn, El Mundo questioned the political competence of Podemos more than the one of Vox.
Since most of our variables gave insignificant results for this hypothesis, and also considering some of our variables could not be calculated, we have to refute the hypothesis concerning stigmatization.
Demonization
Finally, for our last hypothesis, we argued that demonization would be greatly applied to parties when covered by a newspaper with a different ideology.
Table 12. Chi-square tests results of measuring demonization.
Variable El País El Mundo
Called terrorists or related to ETA
X2 (1, N=164) =.958, p=.328 X2 (1, N=173) =3.02, p=.082
Source: own production.
El Mundo referred to Podemos as terrorist or related it to ETA in three of the articles and Vox was never called so, but this arrives at a non-significant result. In turn, Vox was related to terrorism only once by El País, and therefore no difference was found between the coverage of both parties. Therefore, we can reject this hypothesis.
Discussion and limitations
Overall, our results suggest that most of the biases we tested did not significantly differ between parties in the news items. The hypothesis of the use of polls or citizen’s opinions to give a negative portrait of a party did not show any significant results, just like the one on party demonization, so they were openly rejected. Following, the different use of sources between the party analysed, other parties and journalists only showed significance in one of our six analyses, just like the stigmatization indicators, and therefore cannot be accepted either. The use of official sources did show
significance for El Mundo, but not for El País, and therefore our hypothesis that more official sources will be used in covering parties with a different ideology than the newspaper is partly accepted.
Concerning delegitimization, only three of our ten analyses showed significance, two in El País, and one in El Mundo, and while this hypothesis cannot be completely rejected, it should be carefully considered. The analyses that considered framing did offer better results. From our 20 analyses, nine of them turned significant, will allows us to at least partly confirm that the framing of parties differ depending of their ideology. Altogether, eleven variables showed significance in El Mundo’s coverage, while four were
Having said that we cannot, in general terms, consider that the biases portrayed by the protest paradigm are more likely present in the parties which politically disagree with the newspaper they are being covered in, and therefore we cannot either confirm our research question that the same media treats new left and right-wing parties differently.
This study also has several limitations that should be highlighted. Firstly, more than half of our sample did not include evaluations of the party we wanted to analyse. A way to solve this would be not to withdraw a larger sample per se, but only proceed to further coding if the party is evaluated.
Secondly, some of our variables only appeared in very small numbers, and some did not appear at all such as the critique to physical appearance, which means that our significant results should also be accepted with caution.
A way to try and maximize the evaluations could be to choose only news items in which the party we want to analyse is the main subject, for example looking for news in which the name of the party or one of its members appears in the headline or sub-header, or in which the name of the party or one of its members should appear several times and not just one. Within our sample, we realized that, even if we tried to reduce our search to only relevant results for us to code, in many of the news items the party we were analysing would only briefly be mentioned.
Thirdly, operationalization of measurements for coding proved to be difficult. While we relied on previous studies, the protest paradigm’s literature did not clearly establish how to correctly measure and operationalize the different biases. Also, the fact that some of our measurements did not appear in the news means that clearer and more defined measures of the protest paradigm should be adopted.
Finally, the time period of analysis for both parties was not the same, even if it
an impact in our results, it is true that the most desirable analysis would include the same dates for both parties. However, the requirement of a party to not have political representation in order to be considered new, and the unlikeliness that within a country two parties with opposed ideologies are established at the same time and become elected at the same time makes impossible to achieve this.
Scientifically, this study manifests the need for further research regarding the operationalization of the protest paradigm and the portrayal of new parties in the media. Our partly supported hypothesis should be considered for further examination, maybe on other political and media scenarios, or maybe applying the advice we have given for better sampling. In addition, content analysis cannot make inferences about the effect that this coverage might have on readers, and therefore different approaches to this topic should be taken that might provide with some clarity about the impact that biases could have.
Societally, the partial confirmation that some biases exist on the framing of different new parties is of relevance since readers should be mindful of the existence of these biases and the already known fact that media influence the success of new parties, and it can change or reinforce party preference. However, this study cannot assess how these biases may have this influence, and more studies should be conducted to test it.
Future research should, therefore and firstly, attempt to optimize the coding and analysis of the protest paradigm’s variables, which could be achieved by conducting a meta-analysis comprising previous research in which coding of different challenger groups or demonstrators has been performed. This could provide new variables that could be used for further studies regarding new parties as well as other challengers. Besides, this same study could be replicated and expanded too by comparing how the same party is covered by different newspapers or even by different types of media such as radio or television.
References:
Aelst, P. van, Maddens, B., Noppe, J. & Fiers, S. (2008). Politicians in the news: Media or party logic? Media attention and electoral success in the Belgian election campaign of 2003. European Journal of Communication, 23(2), 193-210. DOI:
10.1177/0267323108089222
Alonso, R. (2011). Why do terrorists stop? Analyzing why ETA members abandon or continue with terrorism. Studies in Conflict & Terrorism, 34(9), 696-716. DOI:
10.1080/1057610X.2011.594944
Barnea, S. & Rahat, G. (2011). Out with the old, in with the “new”: What constitutesa new party? Party Politics, 17(3), 303–320. DOI: 10.1177/1354068810369148
Berrington, H. (1985). New parties in Britain: Why some live and most die. International Political Science Review, 6(4), 441–461. DOI: 10.1177/01925121850060040
Bolleyer, N. & Bytzek, E. (2013). Origins of party formation and new party success in advanced democracies. European Journal of Political Research, 52(6), 773–796. DOI:10.1111/1475-6765.1201
Borge, R. & Sáez, E. (2016). From protest to political parties: Online deliberation in new parties in Spain. Medijske studije, 7(14), 104-122. DOI:10.20901/ms.7.14.8
Bos, L. & van Der Brug, W. (2010). Public images of leaders of anti-immigration parties: Perceptions of legitimacy and effectiveness. Party Politics, 16(6), 777–799. DOI:10.1177/1354068809346004
Bourne, A. K. (2018). Securitization and the proscription of terrorist organizations in Spain. Terrorism and political violence, 30(2), 318–335, DOI:
Casero-Ripollés, A., Sintes-Olivella, M. & Franch, P. (2017). The populist political communication style in action: Podemos’s issues and functions on Twitter during the 2016 Spanish general election. American Behavioral Scientist, 61(9), 986–1001. DOI: 10.1177/0002764217707624
Chiang, C. F. & Knight, B. (2011). Media bias and influence: Evidence from newspaper endorsements. Review of Economic Studies, Oxford University Press, 78(3), 795–820. DOI:10.1093/restud/rdq037
Climent Sanjuan, V. & Montaner Goetzenberger, M. (2019). Los partidos populistas de extrema derecha en España: Un análisis sociológico comparado. Revista Izquierdas, 49, 910–931.
Dardis, F. E. (2006). Marginalization devices in U.S. press coverage of Iraq war protest: A content analysis. Mass Communication & Society, 9(2), 117–135.
DOI:
10.1207/s15327825mcs0902_1
Eberl, J. M., Boomgaarden, H. G. & Wagner, M. (2017). One bias fits all? Three types of media bias and their effects on party preferences. Communication Research, 44(8), 1125–1148. DOI: 10.1177/0093650215614364
Harmel, R. & Robertson, J. D. (1985). Formation and success of new Parties: A cross-national analysis. Intercross-national Political Science Review, 6(4), 501–523.
DOI:10.1177/019251218500600408
Hernández Conde, M., & Fernández García, M. (2019). Partidos emergentes de la ultraderecha: ¿fake news, fake outsiders? Vox y la web Caso Aislado en las elecciones andaluzas de 2018. Teknokultura. Revista De Cultura Digital Y Movimientos
Hopmann, D. N., Vliegenthart, R., De Vreese, C. & Albæk, E. (2010). Effects of Election News Coverage: How Visibility and Tone Influence Party Choice. Political Communication, 27(4), 389–405. DOI: 10.1080/10584609.2010.516798
Hug, S. (2001). Altering party systems: Strategic behavior and the emergence of new political parties in Western democracies. University of Michigan Press. DOI:
10.3998/mpub.16263
Kioupkiolis, A. (2016). Podemos: the ambiguous promises of left-wing populism in contemporary Spain, Journal of Political Ideologies, 21(2), 99–
120. DOI: 10.1080/13569317.2016.1150136
LaFree, G., Dugan, L., Xie, M. & Singh, P. (2012). Spatial and temporal patterns of terrorist attacks by ETA 1970 to 2007. Journal of Quantitative Criminology 28(1), 7–29. DOI: 10.1007/s10940-011-9133-y
Link, B. & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27(1), 363–385. DOI: 10.1146/annurev.soc.27.1.363
Lucardie, P. (2000). Prophets, purifiers and prolocutors. Party Politics, 6(2), 175–185. DOI: 10.1177/1354068800006002003
Mahoney, C. W. (2018). End of the cycle: assessing ETA’s strategies ofTerrorism. Small Wars & Insurgencies, 29(5-6), 916–940. DOI: 10.1080/09592318.2018.1519300
McLeod, D. M. & Hertog, J. K. (1998). Social control and the mass media’s role in the regulation of protest groups: The communicative acts perspective. In D. Demers & K. Viswanath (Eds.), Mass media, social control and social change. 305–330. Ames, IA: Iowa State University Press.
McLeod, D. M. & Hertog, J. (1992). The manufacture of “public opinion” by reporters: Informal cues for public perceptions of protest groups. Discourse & Society, 3, 259– 275. DOI: 10.1177/0957926592003003001.
McLeod, D. M., Detenber, B. H. (1999). Framing effects of television news coverage of social protest. Journal of Communication 49(39), 3–23. DOI:
10.1111/j.1460-2466.1999.tb02802.x
McLeod, D. M. (2007) News Coverage and Social Protest: How the Media's Protect Paradigm Exacerbates Social Conflict. Journal of Dispute Resolution 2007(1), 185-193
Murua, I. (2017). Ending ETA’s armed campaign. How and why the Basque armed group abandoned violence. Milton Park, England: Routledge.
Norris, P., Curtice, J., Sanders, D., Scammell, M. & Semetko, H. (1999). On message- Communicating the campaign. London: Sage.
Oñate, P. (2006). Elecciones, partidos y sistemas de partidos en la España democrática. Published in Murillo, F., García de la Serrana, J. L. et al. (Eds).
Transformaciones políticas y sociales en la España democrática, 399-431. Valencia: Tirant lo Blanch.
Rama, J. (2016). Crisis económica y sistema de partidos. Síntomas de cambio político en España. Working papers 344. Barcelona: Institut de Ciències Politiques i Socials
Rosenstone, S., Behr, R., & Lazarus, E. (1984). Third Parties in America: Citizen Response to Major Party Failure - Updated and Expanded Second Edition. Princeton, New Jersey: Princeton University Press.
Rubio-Pueyo, V. (2019). Vox: ¿Una nueva extrema derecha en España? Study published by the Rosa Luxemburg Stiftung- New York Office.
Šárovec, D. (2019). Is the “new” always “new”? Theoretical framework problems of new political parties’ research: The Czech Republic experience. Politics in Central Europe, 15(1), 55–80. DOI:10.2478/pce-2019-0003
Schulz, W. (1994). Wird die Wahl im Fernsehen entschieden? [Is the election result decided on television?]. Media Perspektiven, 7, 318–327
Sikk, A. (2005). How unstable? Volatility and the genuinely new parties in Eastern Europe. European Journal of Political Research, 44(3), 391–412. DOI: 10.1111/j.1475-6765.2005.00232.x
Somuano Ventura, M. F. (2007). Movimientos sociales y partidos políticos en América Latina: una relación cambiante y compleja. Política y cultura, 27, 31–53.
Tavits, M. (2008). Party systems in the making: The emergence and success of new parties in new democracies. British Journal of Political Science, 38(1), 113–133. DOI: 10.1017/S0007123408000069
van Heerden, S. C. (2014). What did you just call me? A study on the demonization of political parties in the Netherlands between 1995 and 2011 (PhD dissertation,
Amsterdam Institute for Social Science Research, Amsterdam, The Netherlands)
van Spanje, J., Azrout, R. (2019). Tainted love: How stigmatization of a political party in news media reduces its electoral support. International Journal of Public Opinion Research, 31(2), 283–308. DOI: 10.1093/ijpor/edy009
Vidal, G. (2018). Challenging business as usual? The rise of new parties in Spain in times of crisis. West European Politics, 41(2), 261–286. DOI:
Weaver, D. A. & Scacco, J. M. (2013). Revisiting the protest paradigm: The Tea Party as filtered through prime-time cable news. The International Journal of Press/Politics, 18(1), 61–84. DOI: 10.1177/1940161212462872
Zarzalejos, J. (2016). Populism in Spain: an analysis of Podemos. European view, 15(?), 183–191. DOI: 10.1007/s12290-016-0408-4
Žúborová, V. (2015). Newcomers in politics? The success of new political parties in the Slovak and Czech Republic after 2010? Baltic Journal of Law & Politics, 8(2) 91–111. DOI: 10.1515/bjlp-2015-0020
Appendix I. Codebook
News classification:
a) Number assigned to the news article:
b) Title of the news piece:
c) Party mentioned in the article:
1. Podemos
2. Vox
d) Newspaper the article appears in:
1. El País
2.El Mundo
e) Date of publication: dd/mm/yyyy
Protest paradigm indicators:
A. Frames:
1. Are the party or its members evaluated in the text? If answer is no, move to question number 2.
0. No
1. Yes
1.1. On the following scale, how are they evaluated?
1. Only negatively
3. Balanced
4. Mixed but mostly positively
5. Only positively
2. Are the party or politicians explicitly being called “performers” in a spectacle or “fake”?
0. No
1. Yes
3. Are the party or politicians being called radicals?
0. No
1.Yes
3.1. Are the party or politicians being associated with other groups called “radicals” or “rioters”?
0. No
1.Yes
3.2. Are the party or politicians being compared to other parties or politicians that are described as “radicals” or “rioters”?
0. No
1.Yes
4. Are the party or politicians being called “criminals” or “non-democratic”?
0. No
4.1. Are the party or politicians being associated with criminals, terrorists, or non-democratic groups?
0. No
1.Yes
4.2. Are the party or politicians being compared to criminals, terrorists, or other non-democratic groups?
0. No
1.Yes
B. Actors and sources:
For the information that explicitly refers to Podemos or Vox:
5. Is the party or one of its members the source of the information?
0. No 1. Yes
6. Is another party or group the source of the information?
0. No
1. Yes
7. Is a journalist the source of the information?
0. No
1. Yes
8. Does the text include messages from official sources? Official sources are: Parties forming Government, judges, police officers and representatives of associations such as Amnesty International, WHO, etc.
0. No
1. Yes
D. Invocation of public opinion:
9. Are negative opinions of citizens, or the society in general, about the political party portrayed in the news?
0. No
1. Yes
10. Are negative polls or statistics about this political party mentioned? 0. No
1. Yes
E. Delegitimization:
11. Is the party portrayed as a threat to democracy?
0. No
1. Yes
12. Is the party portrayed as a threat to political stability?
0. No
1.Yes
13. Is the party, politicians or its ideas called irresponsible?
1. Yes
14. Is the party called populist or portrayed as having no real ideas behind their messages?
1.Yes
0. No
15. Are the parties’ plans and ideas mentioned? If answer is no, move to question 16.
1. Yes
0. No
15.1. Are the parties’ plans and ideas supported or opposed?
1.Only supported 2.Mostly supported
3.Balanced, mentioned pros and cons
4.Mostly opposed
5.Only opposed F. Stigmatization:
16. Is the political group or the politicians portrayed as violent?
1. Yes
0. No
17. Is the physical appearance of the politicians criticised? For example, implying that having dreadlocks or not wearing formal clothing is indecent.
1. Yes
18. Are the mental abilities of politicians questioned?
1. Yes
0. No
19. Is the political competence of the politicians questioned?
1. Yes
0. No
20. Are the politicians or their ideas explicitly associated with political inexperience or amateurism?
1. Yes
0. No G. Demonization:
21. Is the political party described with words related to terrorism and the terrorist group ETA? E.g. “terrorist”, “etarra”, “pro-etarra”.
1. Yes
0. No
Appendix II. Inter-Coder Reliability results.
Variable Krippendorff’s Alpha
Is the new party evaluated? .77
How? (On a scale 1 to 5) .72
Is the new party portrayed as “fake” or as a “performer”? .70 Is the new party called radical? .78 Is the new party associated with radicals? .78 Is the new party compared to radicals? .85 Is the new party called criminal? .79
Is the new party associated with criminals? .73 Is the new party compared with criminals? .79 Is the party the source of the information? .81 Is another party the source of the information? .77 Is the journalist the source of the information? .79 Does the text include messages from official sources? .81 Are negative messages from citizens about the new party
portrayed?
.78
Are negative polls or statistics about the new party mentioned?
1.0
Is the new party portrayed as a threat to democracy? .88 Is the new party portrayed as a threat to political stability? 1.0 Is the party called irresponsible? .81 Is the party called populist? 1.0 Are the party plans and ideas mentioned? .81 Are these ideas supported or opposed? (On a scale 1 to 5) .72 Is the party portrayed as violent? .65 Is the physical appearance of the members criticised? N.A. Are the mental abilities of the party members questioned? .85
Are the political competences of the party members questioned?
.77
Are the party members portrayed as inexperienced or amateurs?
1.0
Is the party related to terrorism or ETA? .85 Source: own production.
