Microsoft Word - PDF_Issue_14_2_SeddoneBobbaRoncarolo.docx Italian Political Science, VOLUME 14 ISSUE 2, OCTOBER 2019 © 2019 Italian Political Science. ISSN 2420-8434. Volume 14, Issue 2, 73–92. Contact Author: Antonella Seddone, University of Turin, Dipartimento di Culture, Politica e Società. E-mail address: antonella.seddone@unito.it European versus Domestic Politics: Media Negativity during the 2019 European Elections Campaign in Italy Antonella Seddone UNIVERSITY OF TURIN Giuliano Bobba UNIVERSITY OF TURIN COLLEGIO CARLO ALBERTO, TURIN Franca Roncarolo UNIVERSITY OF TURIN Abstract This article aims at understanding to what extent European issues and actors are covered through a negative tone within the news coverage of the 2019 EP election campaign in Italy. We rely on an original dataset based on content analysis of the 50 days prior to the 2019 EP elections in four mainstream newspapers (n=3,943) and four mainstream TV newscasts (n=1,873). Our findings show that the media negativity is primarily associ- ated with domestic politics, while all in all European politics is covered using a less negative tone. This general picture is the results of two differentiated patterns: newspapers are less negative towards both EU politics and EU and foreign political actors, while TV news are more negative towards EU and foreign political actors. 1. Introduction s is well known, European elections have been defined as second-order elections. Reif and Schmitt argued in 1980 that this type of election is generally character- ized by a high voter turnout, the punishment of incumbents, and greater expressive voting. As regards party campaigns and elections coverage, the main features refer to the more limited budget and the shorter duration of the campaign when com- pared to national elections, the kind of candidate selected (not recruited from among the leaders of parties), and the constant presence within the campaign of national issues, or else European issues framed in a national perspective (Marsh 1998; Cayrol 1991; de Vreese, Lauf, and Peter, 2006a). Yet, it is also the case that this model is evolving (Schmitt 2005; Hobolt and Wittrock 2011; Galpin and Trenz 2018; Roncarolo 2011). Certainly, af- ter the Great Recession, the EU gained considerable salience within domestic debate for imposing austerity measures and (not) managing the migration issue (Hutter and Kriesi 2019). In the same period, several national parties – both Europhile and Eurosceptic – have stressed the relevance of the EU within national political debates (Meijers 2017). In this context, the news media have become increasingly relevant not only as a source of information but also as an active player in the game of (de)legitimation of the EU. A European versus Domestic Politics 74 In this regard, a well-established branch of literature has found that the news cover- age of European Union (EU) politics plays a crucial role in connecting European citizens with institutions and in raising public awareness of EU politics. Banducci and Semetko (2003) found that EU news coverage affects turnout in European elections, while other scholars have shown a correlation between citizens’ support for future European integra- tion and the features of news coverage on the EU (e.g. Maier & Rittberger, 2008; Vliegenthart et al., 2008). EU news coverage also contributes to enhancing the process of Europeanizing national public spheres (e.g. Fossum & Schlesinger, 2007) and strength- ening the legitimacy of the EU institutions (e.g. Trenz, 2008). A second strand of scholars focus more specifically on the negative media coverage of EU-related issues and on their effects on citizens’ attitudes towards the EU. One con- sistent finding, regarding journalism and news values in general, is that bad news is more newsworthy than good news (Soroka, 2014). This also applies to politics at both national and supranational level. In this perspective, the news media can play an active role in am- plifying or restraining EU contestation (Statham & Trenz, 2013). Galpin and Trenz (2018), comparing news media coverage in the UK and Germany, found that the UK news demonstrates a strong negative bias towards EU polity, whereas in Germany EP debates focus more strongly on EU politics and policies and show a positive bias towards EU pol- ity. Other scholars, looking at the effects of this type of coverage, have analysed the tone of EU news and found that exposure to negative evaluations of the EU leads citizens to be more supportive of Eurosceptic parties (De Vreese 2004; van Spanje & de Vreese 2014). This article examines mainstream media coverage of the 2019 EU elections, adopt- ing the case of Italy as an illustration. These elections were the first test for the populist government led by the 5 Star Movement and the League. Both these parties are clearly not Europhile but rather, albeit with some differences, Eurosceptic. The 5 Star Movement has in the past adopted a pretty ambiguous approach by raising criticism against the EU but keeping their distance from the Eurosceptic parties once in Brussels (Franzosi et al. 2015). On the contrary, the League – especially under the leadership of Matteo Salvini – shifted towards hard Euroscepticism, moving the traditional cleavage (and divide) from Rome to Brussels, blaming EU institutions for the bad economic situation (Albertazzi et al., 2018). As a consequence, EU elections have acquired greater importance than in the past. This article aims to understand to what extent European issues and actors are re- ferred to in a negative tone within the news coverage of the 2019 EP election campaign. It allows a thorough understanding of the role of the mainstream media in the (de)legitima- tion of the EU in a country affected by several controversies with the EU (among others, migration issues and public debt containment) and led by two Eurosceptic parties. For a long time, Italy was considered a Europhile country (Huysseune, 2010; Conti & Memoli, 2010, 2013). Since 2014, the situation has radically changed. While Eurosceptic parties did not reach 20% in 2009, they exceeded 50% of the votes in 20141. Italy further 1 In the 2009 EP elections, the Eurosceptic parties were the Northern League (LN) and the Radical Left (SEL). In 2014, besides LN and SEL (renamed the ‘Tsipras List’ for the occasion), a further three parties openly joined the Eurosceptic side. On the right, Berlusconi’s Forza Italia (FI, Go Italy) ran a campaign under the slogan ‘More Italy in Europe, Less Europe in Italy’. In the same area, Fratelli d’Italia (FDI, Brothers of Italy), a Berlusconi ally, campaigned close to the Eurosceptic positions of Marine Le Pen. Fi- nally, the most voted Eurosceptic party in the 2014 elections was the Movimento Cinque Stelle (M5S, Five Star Movement), whose campaign was based on the request for a referendum on the Euro and a refusal of SEDDONE, BOBBA and RONCAROLO 75 evolved toward an EU-critical positioning in 2019 when Lega, M5S and FDI won almost 58% of the votes cast. In this situation, where the two mainstream parties, Forza Italia (FI, Go Italy) and Partito Democratico (PD, Democratic Party), are marginalized, and no pro- European voices are heard in the public debate, it is relevant to investigate the role of the Italian media system in counterposing or fostering the anti-EU narrative. In the light of the above, the article shows how the main Italian newscasts and news- papers used a negative tone to cover domestic and European actors and themes, by investigating different patterns of Europeanization (Koopmans and Erbe 2004). In the section below, we discuss how literature has addressed this issue and, on this basis, we present our research questions. In section 2, we outline the research design, while in sec- tion 3 we examine the coverage of the 2019 EP election campaign in Italy. Our findings suggest that TV and daily press present different patterns of negativity. Newspapers, in particular, appear more prone to using a negative tone when featuring politics. As con- cerns the dualism between EU and domestic politics, Europe still benefits from a Europhile media coverage if compared with domestic politics, and this applies to both press and TV. Conversely, when the distinction between EU and domestic politics is viewed through the lens of political actors, we observe a radical difference between TV and press. EU political actors are indeed more likely to be covered negatively on TV news- casts than in newspapers. Further differences can be identified when taking into account the dimensions of Europeanization, results suggesting that when the media cover the in- teraction between EU and domestic politics there are higher chances of negative coverage. There is one exception: when Italian political actors approach EU institutions or EU representatives, TV newscasts tend to be less negative 2. EU coverage on mainstream media In the past two decades, several scholars have focused their studies on media coverage of the EU. Analyses of media coverage of European politics have answered questions about the Europeanization of national public spheres (e.g. Fossum & Schlesinger, 2007) and contributed to our understanding of the democratic legitimacy of the EU institutions (e.g. Trenz, 2008, Savini, 2019). Findings show that irregular and limited news coverage could unfavourably influence public citizens' awareness of the EU and especially the relation- ship between the representatives and the represented (Shephard, 1997). Regular coverage, on the contrary, would enhance opportunities for citizens to take note of EU ac- tivities and foster trust in European institutions (Norris, 2000; Nardis, 2015). The development of the European public sphere is challenging journalism cultures and practices; despite country-based differences the EU has indeed become a common issue for journalists (Kunelius 2008). In this respect, Pfetsch et al. (2008), for example, found that the press plays a crucial role in the Europeanization of national public spheres, and in particular some countries seem to share similar approaches when dealing with EU issues as a consequence of a similar degree of involvement in the EU integration project (i.e. the press of France, Germany, Italy, the Netherlands and Spain share a similar point of view on EU issues when compared with the UK or Switzerland). Besides the functional the restrictive policies imposed by the EU (e.g. the inclusion of the fiscal compact and balanced budget principle in the Italian Constitution). European versus Domestic Politics 76 relationship with the EU, press coverage is affected also by the organizational infrastruc- ture and the editorial policies setting the allocation of resources for EU newsgathering, and by the journalistic culture as well. Similarly, Cornia et al. (2008) suggest that a Euro- pean journalistic culture is developing, but the domestic perspective still prevails and national relevance remains the core criterion for news selection. Similar results are found by Preston (2009), underlining that EU issues are still largely addressed and covered through domestic interpretative schemes. News about the EU is not traditionally widespread in mainstream media and some differences can be found between media types. While quality newspapers usually tend to cover European affairs (Trenz, 2004; Machill, Beiler & Fischer, 2006), EU news on tele- vision has been limited, irregular, and mainly focused on domestic issues or actors (e.g. de Vreese, Peter & Semetko, 2001; Peter & de Vreese, 2004; Meyer, 2005; Boomgaarden et al., 2010). Starting from 2008, the economic and Euro crises began to change this ap- proach to covering the EU (e.g. Kleinen-von Königslöw, 2012). At the same time, another series of issues, including humanitarian intervention, common defence, immigration policy and terrorism, are becoming more transnational, potentially contributing to the Europeanization of national public spheres (e.g. Koopmans, 2004; de Vreese & Kandyla, 2009; Kantner, 2011). Scholars have mainly addressed the study of European Union news coverage by tak- ing into account several dimensions, such as the visibility of EU news (e.g. Boomgaarden et al., 2010), the degree of Europeanization (or domestication) (e.g. Koopmans, 2007; McQuail & Bergsma, 1983; Peter & de Vreese 2004), the tone of the news regarding Euro- pean integration, and the framing of EU affairs (e.g. de Vreese et al., 2006). As clearly illustrated in several studies by Boomgaarden, de Vreese, and colleagues (Boomgaarden et al., 2010; Boomgaarden et al., 2013; Schuck & de Vreese, 2011), the vis- ibility of the EU in the news is relevant, since it provides citizens with 'information about the issues at stake and cues them about the importance of European Parliament elections' (Boomgaarden et al., 2013: 611). In the light of the second-order elections thesis (Reif & Schmitt, 1980; Schmitt, 2005), the Europeanness of the news, namely the extent to which EU news is (or is not) domesti- fied (Boomgaarden et al., 2013; Boomgaarden & de Vreese, 2016), has been assessed through different dimensions. Europeanization has been defined as a vertical or horizontal process: vertical Europeanization ‘consists of communicative linkages between the na- tional and the European level’, and horizontal Europeanization ‘consists of communicative linkages between different member states’ (Koopmans & Erbe, 2004: 101). By using both of these dimensions, Schuck & De Vreese (2011: 289) showed that news- papers are still more important as agents of Europeanization than television news; the quality press and public television are more European in focus than their commercial coun- terparts. Mainly focusing on the vertical dimension, Boomgaarden et al. (2013: 623–624) examined the prominence of EU actors in news coverage and found that when there is in- creasing elite conflict, there is less of a European focus in the news. This also applies to negativity among the political elites: the more negative, the less European (or vice versa). In light of the above, we investigate whether and to what extent the coverage of Euro- pean politics and actors differentiates from the coverage of domestic politics and actors. We focus in particular on the negativity dimension, which is a recurrent element in the SEDDONE, BOBBA and RONCAROLO 77 coverage of politics in general (Lengauer et al. 2011). No study has so far focused specifically on the negative media coverage of European politics. Therefore, we do not have clear expec- tations but have structured our analysis around four different research questions. A preliminary question concerns possible differences between newscasts and newspa- pers in the coverage of politics, in general. As mentioned above, we know that some differences emerge in terms of visibility but, to our knowledge, there are no studies that ac- count for systemic differences between the two media regarding negativity. We therefore wondered: RQ1: Does the level of negativity vary between TV and newspapers? On the basis of this preliminary RQ, we developed a further set of queries aimed to better clarify the main drivers of negativity within the two media types. The aim is thus to compare the negative tones of TV and the press by focusing on the distinction between domestic and EU politics at different levels. Indeed, we found expectations that domestic politics is more relevant (in terms of visibility and space) than European politics, but there are no clear indications as to the patterns that a negative tone follows when covering the former or the latter. Similar findings concern political actors: domestic actors are more salient within national public debates, but no clear expectations are found regarding media negativity. In this regard, our questions read as follow: RQ2: Does the level of negativity differ when European or domestic politics are cov- ered? RQ3: Does the level of negativity differ when European or domestic political actors are covered? Finally, literature shows that the Europeanization of public debates involves the cov- erage of different dimensions of the European integration process. Our last question, therefore, wonders whether there are different probabilities of negative coverage being associated with vertical and horizontal Europeanization: RQ4: Does the level of negativity differ according to different ways to cover Euro- pean politics? 3. Methods 3.1 Sample This study investigates the Italian media coverage of the 2019 EP elections. Our research questions are addressed by mean of a dataset including all news items focusing on polit- ical issues published or broadcast by the main Italian press and TV media outlets during the seven weeks before the EP elections (7 April–24 May 2019). The unit of analysis was the news story and it was studied through a human content analysis carried out by three supervised coders. The data collection comprehends five newspapers and five TV news- casts that were selected according to two different criteria: the readership/audience and the outlet’s political leaning2. Therefore, among the most widespread Italian 2 The outlets’ political leanings were identified by referring to individual media diet and voting behaviour in the 2018 General Elections by relying on Itanes 2018 survey data (Itanes, 2018; Roncarolo and Mancini, 2018). European versus Domestic Politics 78 newspapers, we considered Corriere della Sera and La Stampa as less-politically-ori- ented/neutral outlets, while Il Giornale and La Repubblica were included as representatives of, respectively, right- and left-wing views. Finally, Il Fatto Quotidiano was covered by our analysis since its political leaning appears closer to the Movimento 5 Stelle. As regards TV, the evening editions of the following newscasts were examined: TG1, TG2, TG3, TG5 and La7 News. Also in this case, TG1 and La7 News were selected as less-politically-oriented/neutral newscasts, while the others account for different polit- ical areas. More precisely, TG2 was considered because it most reflects the government’s political views, while TG3 and TG5 were included in our study for their traditional close- ness to, respectively, the centre-left and centre-right positions. For newspapers, all political news items published on the first page and recalled in internal pages were coded, while news items focusing on EU-related issues were all in- cluded in the data collection, regardless of their position within the newspaper. As regards TV, all political news items broadcast were coded. The final dataset includes 5,816 news items; in detail: 1,873 TV news items and 3,943 press news items. 3.2. Variables3 GENERAL TONE OF THE STORY: NEGATIVE TONE. Our dependent variable (DV) accounts for the negative tone of media coverage. Following Lengauer et al. (2011), indicators of neg- ative tone were found in those news items containing elements referring to (i.e.) crisis, disaster, negative outlook, conflict, defeat, fiasco The DV was derived from a categorical one, which according to the coding procedure aimed to distinguish between news items providing a negative, positive, balanced or neutral tone. Then, for the sake of simplicity, it was recoded to isolate the news items showing negative tones against all the other news items framed by positive, neutral or balanced tones. MEDIA TYPE AND OUTLETS. The first research question (RQ1), exploring the negative bias in press and TV coverage, was addressed by means of a dichotomous variable (IV1) distinguishing between news items reported by press (=1) and news items broadcast by TV newscasts (=0). Furthermore, we also investigated possible differences in the likeli- hood of negative tone among the outlets analysed. Indeed, according to the criteria adopted for selecting the outlets, we could rely on a sample reflecting different political leanings. This allowed us to clarify whether there were ideological elements underpin- ning the outlet patterns of covering political issues. We investigate this dimension through a set of dummy variables identifying each of the ten media outlets analysed (IVs2). EU POLITICS VS DOMESTIC POLITICS. According to RQ2, within a general climate of opin- ion characterized by citizens’ increasing criticism of the European Integration project, we aim to assess if Italian media outlets contributed to this climate of opinion by covering European-related issues and political actors in a negative way, or whether EU politics still benefits from Europhile coverage (Bobba and Seddone, 2018). With this purpose in mind, we classified the news stories according to two main categories: those dealing with domestic politics and those referring to European politics. We intended this latter 3 Coding instructions are reported in Appendix. SEDDONE, BOBBA and RONCAROLO 79 dimension as all the references to EU institutions, EU policies or EU political actors. In addition, when the story was devoted to issues relating to other EU member states it was also identified as European Politics. Conversely, all the stories covering the Italian polit- ical debate or Italian political actors were categorized as Domestic Politics. We derived, then, a dichotomous variable scoring 1 if news stories focused on European Politics and coded as 0 when Domestic Politics was addressed within the news item (IV3). POLITICAL ACTORS. RQ3 examines the impact of political actors’ visibility on negative bias within media coverage. The presence of political actors within the news story was detected by mean of a categorical variable accounting for the main Italian political lead- ers as well as any possible reference to EU or foreign political actor in news items. For each news item, coders were asked to register up to five political actors. These data were then recoded into a set of five different dummy variables (IVs4) indicating, respectively, whether the story presented references to: Berlusconi, Di Maio, Salvini, Zingaretti or EU political actors. EUROPEANIZATION. The fourth research question (RQ4) addresses the dimension of Eu- ropeanization. It is meant as a process of reciprocal connections between the EU political space and the domestic one. The contribution of Koopmans and Erbe (2004) appears fundamental. In particular, their theoretical framework identifies two dynamics of in- teractions between the national and the EU public spheres: Vertical and Horizontal. Vertical Europeanization is intended as the communicative linkages between the na- tional and European political space. The Bottom-Up variant refers to the case of national actors directly addressing European institutions, European political actors or referring to EU policy issues. The Top-Down variant of Vertical Europeanization follows the in- verse flow, i.e., the case of European actors addressing national actors, or intervening in domestic political debate by emphasizing common European issues and interests. Hori- zontal Europeanization, instead, pertains to the connections between the public spheres of EU members states. The Weak variant of Horizontal Europeanization occurs when na- tional media coverage reports issues related to the national political space of other EU member states. The Strong variant of Horizontal Europeanization concerns instead di- rect communicative linkages between two member states’ political spaces. It could take place, for example, when politicians of a given country directly approach political actors of another EU member state. Empirically, the Vertical Europeanization dimension was covered by two dummy variables detecting whether the news item provides reference, respectively, to Top- Down and Bottom-Up Vertical Europeanization (IVs5). Similarly, two dummy variables identified the news items referring, respectively, to Strong and Weak Horizontal Euro- peanization (IVs6). 3.3. Models The research questions leading our study were addressed by mean of four logit models, testing the likelihood of a negative tone in media coverage (DV). According to RQ1, in order to detect differences between media types, we ran separated models for Press and TV. Model 1 and Model 2 include as independent variables (a) the set of media outlets reporting the news story, (b) the dummy variable testing the reference to EU contents European versus Domestic Politics 80 (institutions, policy issues or political actors), and finally (c) the dichotomous variable accounting for the reference to EU political actors. Models 3 and 4 extend this analysis by introducing the set of dummy variables identifying references to Berlusconi, Di Maio, Salvini, and Zingaretti along with the dichotomous variable referring to the presence of EU political actors. In addition, these two models also include the dimension of Europe- anization, by testing the dummy variables detecting Bottom-Up Vertical Europeanization, Top-Down Vertical Europeanization, Weak Horizontal Europeaniza- tion and Strong Horizontal Europeanization. Again, the two models distinguish between TV and press media. Table 1 shows descriptives of variables included in the two models. Table 1. Descriptives of the variables Variable Obs. Mean Std. Dev. Min Max Type of media (Press vs TV) 5816 0.68 0.47 0 1 Corriere della Sera 5816 0.14 0.34 0 1 La Repubblica 5816 0.12 0.33 0 1 La Stampa 5816 0.13 0.34 0 1 Il Giornale 5816 0.14 0.35 0 1 Il Fatto Quotidiano 5816 0.15 0.36 0 1 TG1 5816 0.07 0.25 0 1 TG2 5816 0.08 0.27 0 1 TG3 5816 0.07 0.26 0 1 TG5 5816 0.05 0.22 0 1 TG La7 news 5816 0.05 0.22 0 1 Reference to EU or Foreign Political Actors 5816 0.20 0.40 0 1 Reference to Berlusconi 5816 0.08 0.27 0 1 Reference to Di Maio 5816 0.23 0.42 0 1 Reference to Salvini 5816 0.37 0.48 0 1 Reference to Zingaretti 5816 0.08 0.28 0 1 EU news stories 5816 0.46 0.50 0 1 Bottom-Up Vertical Europeanization 2103 0.52 0.50 0 1 Top-Down Vertical Europeanization 2103 0.14 0.35 0 1 Weak Horizontal Europeanization 2689 0.31 0.46 0 1 Strong Horizontal Europeanization 2689 0.15 0.36 0 1 4. Results We introduce our findings by providing some descriptive analyses illustrating the pat- terns of negative tone within our sample. Precisely, Table 2 shows the proportion of negative tone among the media outlets included in this study4. First of all, we observe that Italian media coverage of the EP election campaign presents high levels of negative tone, especially if compared with 2014 media coverage of the EP elections (Roncarolo 2019; Bobba and Seddone 2018). In total, 39.4% of the news stories analysed present ele- ments of negative tone. This applies generally to all media outlets analysed, but it should be noted that negativity varies according to media type. Indeed, newspapers present a 4 Table 1a reported in the Appendix provides details about the distribution of negative, positive and neu- tral tone. In general, among Italian media outlets a neutral/balanced tone prevails when addressing politics and related issues. However, there are some significant differences. In particular, as concerns media outlets like Il Giornale and TG5 – both owned by Berlusconi – the portion of neutral/balanced tone is quite low when compared with other outlets, while a negative frame prevails for depicting politics. In addition, it should be noted that a positive tone is actually extremely marginal for all the media outlets included in this study. SEDDONE, BOBBA and RONCAROLO 81 higher quota of news stories using a negative tone (42.8%) while TV outlets seem to be less inclined to cover political issues negatively; nevertheless, the portion of negativity remains pretty high since it concerns more than one third of TV news items. Further differences emerge among outlets composing our sample. As concerns the press, news items characterized by a negative tone range from 34.3% for Corriere della Sera to 56.8% for Il Giornale while the other press outlets register portions of negative news items close to 40%. Regarding TV, instead, TG1 has the lowest quota of news stories framed in a neg- ative tone (24.4%), while TG5 seems to be the outlet that covers politics with greater negativity: almost one out of two news stories is depicted in a negative tone. Table 2. Negative tone in media and single outlets News item (N total) Negative news item (N) Negative news item (%) Daily Press 3,943 1,687 42.8 Corriere della Sera 787 270 34.3 la Repubblica 706 291 41.2 La Stampa 773 300 38.8 Il Giornale 808 459 56.8 Il Fatto Quotidiano 869 367 42.2 Newscast 1,873 609 32.5 TG1 393 96 24.4 TG2 451 157 34.8 TG3 431 124 28.8 TG5 305 145 47.5 La7 News 293 87 29.7 Total 5,816 2,296 39.5 Table 3 details the negativity of media coverage when the news story presents refer- ences to political actors5. We can see from these figures that, in general, when news items include EU or foreign political actors, the portion of negative tone is lower than media coverage focusing on Italian political leaders. Only one news story out of three is negative when referring to EU or foreign politicians. Conversely, negativity increases when the media cover Italian political actors. The percentage of negative news items ranges from 38.6% when Silvio Berlusconi is covered to 51.3% of negative tone when the story presents references to Nicola Zingaretti. Interestingly, the share of negativity referring to Luigi Di Maio and Matteo Salvini is quite similar for the two leading exponents of the coalition government (resp. 45.5% and 42.5%). Table 3. Negative tone and political actors News item (N total) Negative news item (N) Negative news item (%) Berlusconi 453 175 38.6 Di Maio 1,333 606 45.5 Salvini 2,161 918 42.5 Zingaretti 483 248 51.3 EU-Foreign Political Actors 1,141 366 32 5 Table 2a in the Appendix provides details about the general tone for political leaders. Data show that with the sole exception of Nicola Zingaretti, the negative tone prevails for all political leaders – including the EU/foreign political actors. The positive frame is also in this case extremely marginal, ranging from 2.9% for Zingaretti and 7.8% for EU/foreign political actors. European versus Domestic Politics 82 Table 4 shows instead the patterns of negative tone among news items focusing on the EU. Data suggest that media coverage tends to be more negative when addressing is- sues relating to domestic politics. This seems to confirm the idea of a Europhile attitude among Italian media outlets. The political debate among domestic politics catalyses the negativity of Italian media outlets where almost one out of two news stories presents a negative tone (46.6). Conversely, the percentage of negative news items decreases when the EU is covered (31.24%). However, taking into account the modalities of Europeaniza- tion, data indicate some differences. Indeed, as regards Horizontal Europeanization, the portion of negative news items is pretty similar: about 30% for both strong or weak vari- ants. On the contrary, things appear slightly different when considering Vertical Europeanization. While one third of the stories presenting Bottom-Up elements are framed within a negative tone, as regards the Top-Down variant the tone of the coverage appears far more negative (37%). Table 4. Negative tone and European vs domestic politics News item (N total) Negative news item (N) Negative news item (%) News items with references to domestic politics 3,127 1,456 46.6 News items with references to European politics 2,689 840 31.2 Bottom-Up Vertical Europeanization 1,243 369 29.7 Top-Down Vertical Europeanization 303 113 37.3 Weak Horizontal Europeanization 837 246 29.4 Strong Horizontal Europeanization 398 124 31.1 Total 5,816 2,296 39.5 Although these are preliminary results, some interpretative cues can already be drawn. In the first instance, even within a negative media coverage context, Europe and its political actors seem to benefit from a benevolent media attitude. If domestic politics is covered more frequently with negative tones and frame elements, when dealing with issues relating to the EU the media moderate the negative tones. Furthermore, it is pos- sible to identify a substantial difference between press and TV. The newspapers analysed are all characterized by prominent negativity, assuming much higher proportions than TV outlets. However, when figures are observed in more detail, some dynamics emerge clearly. The European theme is, in fact, treated in different ways by the Italian media. In particular, when the EU institutions directly approach the domestic political space, meaning they intervene in the national political debate by raising European interests, values or rules, the media show less indulgence and the share of negative news items grows more significant. These bivariate analyses have to be integrated by multivariate analyses testing these dimensions and measuring their contribution to negative bias in media coverage during the last EP election campaign in Italy. Taken all together, indeed, these elements of me- dia coverage have a different effect on the probability of negative media coverage. When we look at the models presented in Table 5, these patterns appear clearer. Models 1 and 2, indeed, confirm the evidence that has already emerged from descriptive analyses. As regards the press (Model 1), where the Corriere della Sera is the reference category, all media outlets present positive and significant coefficients, meaning that they are all associated with a higher likelihood of negative media coverage. SEDDONE, BOBBA and RONCAROLO 83 Table 5. Drivers of Negative Tone on Press and TV coverage VARIABLES Model 1 Model 2 Model 3 Model 4 PRESS Negative Tone TV Negative Tone PRESS Negative Tone TV Negative Tone PRESS (ref. cat. Corriere della sera) la Repubblica 1.376*** 1.520** - (0.151) (0.277) La Stampa 1.259** 1.267 - (0.135) (0.230) Il Giornale 2.465*** 2.464*** - (0.260) (0.438) Il Fatto Quotidiano 1.262** 1.481** - (0.131) (0.277) TV (ref. cat. Tg1) TG2 1.692*** - 2.878*** (0.262) (0.914) TG3 1.350* - 1.429 (0.217) (0.480) TG5 2.926*** - 4.758*** (0.486) (1.603) La 7 News 1.353* - 1.367 (0.237) (0.495) EU/foreign Political Actor 0.701*** 2.052*** 0.598*** 1.775** (0.0702) (0.288) (0.0879) (0.420) Berlusconi 0.438*** 2.072*** (0.0881) (0.573) Di Maio 1.528*** 1.088 (0.228) (0.301) Salvini 1.185 1.380 (0.162) (0.338) Zingaretti 0.541*** 5.057*** (0.120) (1.225) EU politics 0.513*** 0.552*** (0.0393) (0.0636) Bottom-Up Vertical Europeanization 0.865 0.468*** (0.103) (0.0976) Top-Down Vertical Europeanization 1.417** 1.830** (0.224) (0.554) Weak Horizontal Europeanization 0.952 2.509*** (0.184) (0.848) Strong Horizontal Europeanization 1.333* 0.780 (0.223) (0.250) Constant 0.769*** 0.349*** 0.369*** 0.103*** (0.0633) (0.0438) (0.0596) (0.0328) Observations 3,943 1,873 1,472 631 Notes: SE in parentheses; *** p<0.01, ** p<0.05, * p<0.1 On the contrary, news items focusing on EU-related issues are less likely to be framed within a negative tone, and the same applies for news stories referring to EU or foreign political actors. When we move to observe these patterns on TV (Model 2) some- thing changes. Here we used TG1 as the reference category. Even in this case, odds rations associated with TV newscasts are all positive and significant, signalling that their news items increase the likelihood of a negative tone in media coverage. In addition, EU- European versus Domestic Politics 84 related issues are, also in this case, less likely to be covered in a negative tone. However, when it comes to assess whether the EU political actors act as drivers of a negative tone in media coverage, the coefficients are positive. Indeed, compared to the press, TV news- casts are more likely to frame negatively stories referring to EU politicians. Models 3 and 4 enlarge this analysis by including two elements: the different vari- ants of Europeanization illustrated above and the differentiated contribution of Italian political leaders. Model 3 focuses on the press. In this case, we observe that media outlets still maintain their effect in increasing the likelihood of negative coverage, with the sole exception of La Stampa. On the other hand, if we consider Model 4, the TV outlets show different patterns. Only TG2 and TG5 are associated with positive and significant odds ratios which are far higher than in Model 2, confirming they are drivers of negativity. In general, it is worth noting that elements of political parallelism could be identified if tak- ing into account coefficients from the two media outlets owned by Berlusconi (Hallin and Mancini, 2004; Mancini, 2008). We found, indeed that they are both characterized by a far higher likelihood of increasing negativity coverage during the election campaign. This is a result of the position taken by these outlets, which basically acted as the opposi- tion to the government (and the former ally, Salvini, in particular) while attacking the left-wing parties as well. Differences between press and TV remain prominent even when considering the dimension of political actors. While within press coverage the presence of EU or foreign political actors reduces the chances of negative coverage, TV shows the opposite result. In this case, news items are more likely to be framed by negative elements. Italian polit- ical actors are also differentiated. As regards the press, Berlusconi, indeed, is associated with significant coefficients proving a lower probability of negativity. The same applies to Zingaretti. Only Di Maio, according to our findings, boosts the likelihood of a negative tone in media coverage. Again, the picture changes with Model 4, focusing on the TV sample. Regardless of whether they are domestic or EU, all political actors are more likely to increase the likelihood of a negative tone. Only Di Maio and Salvini disconfirm this pattern, with no significant odds ratio. These patterns could appear as counterintu- itive. Indeed, given the Eurosceptic claims from both M5S and League, one could have expected that the visibility of the two leaders contributed to increasing the likelihood of a negative tone in media coverage. However, according to some previous research, the Italian media tend to mitigate negativity driven by Eurosceptic political leaders by coun- terbalancing the negative frame relating to their positions with a neutral or even positive frame on the EU that results in a more balanced general tone. Furthermore, these coef- ficients appear quite coherent with bivariate analyses (see Figure 1a in Appendix) showing that on TV both Salvini and Di Maio present a lower degree of negative tone compared with other leaders. Finally, distinguishing between variants of Europeanization, we found interesting results. While in Models 1 and 2, EU-related issues were found less likely to be covered in a negative tone, the introduction of variables relating to the four variants of Europeani- zation determines substantial differences between the two media. Within the press sub- sample, only Top-Down Vertical Europeanization and Strong Horizontal Europeaniza- tion are found to be significant and positively associated with the DV. Conversely, as pertains to TV, a negative tone is more likely when stories refer to Top-Down Vertical SEDDONE, BOBBA and RONCAROLO 85 Europeanization and Strong Horizontal Europeanization. On the other hand, when na- tional political actors approach EU-related issues, the likelihood of a negative tone decreases. 5. Conclusions Our study aimed to investigate the drivers of negativity within Italian media coverage during the 2019 EP Election campaign. Italy offers a privileged angle for addressing this issue. On the one hand, Italian politics is going through a shift in climate of opinion on the EU. The country was traditionally considered Europhile, but the success of populist parties – now in government – criticising the EU integration process and endorsing na- tionalist views, indicate that Eurosceptic views are becoming prominent among Italian parties. On the other hand, the media have usually held a supportive attitude towards the EU. Indeed, while media coverage of politics is in general characterized by a negative tone, where the elements of conflict and negativity prevail, the EU institutions and po- litical actors have traditionally benefited from a less negative tone in the media. In other words, the media have often acted as moderators of criticism towards the EU by temper- ing their usual tendency to negativity in covering political issues. Our study relied on an original dataset derived from a human content analysis car- ried out on news items reported by five newspapers and five TV newscasts during the five weeks preceding the EP elections. Our results contribute to a better understanding of the processes of development of the European public sphere, and we identify some interest- ing patterns concerning the Italian case. However, we are aware of some limitations. First of all, we should underline that our data may present some weaknesses due to the news item selection. Indeed, as concerns the press, the sample includes all news items focusing on the EU regardless of their position within the outlet (front page and internal pages), while news items focusing on domestic politics were included in the sample only when reported on the first page. On the contrary, as regards TV coverage, we took into account all political news items broadcast. On the one hand, this choice may entail a problem of comparability between press and TV; on the other hand, our data may suffer from overrepresentation of EU visibility in the press. For future studies, this issue will be addressed by re-defining the research design and including, for the press, all political news. Nonetheless, given that the purpose of the article was to clarify the dynamics of negativity in media coverage and the relationship with Europeanization patterns, the data provided still offer an interesting report on the dynamics of Europeanization during the 2019 European Election campaign in Italy. Secondly, we acknowledge that some re- sults could be further strengthened if we included in our research design a qualitative dimension addressing the peculiarities of journalistic cultures characterizing the single outlets included in our analysis. This would allow us to differentiate outlets according to their editorial policy and organizational structure as pertains to the EU, since literature found these elements to be relevant to understanding the dynamics of the Europeaniza- tion of public debate. Given the peculiarities of the Italian media system, we organized our research ques- tions by addressing in particular whether press and TV may be characterized by different patterns of negativity within their coverage during the election campaign. In this regard, we observed that newspapers tended more to add a negative tone to political news stories, European versus Domestic Politics 86 while portions of negativity within TV appeared more limited. Furthermore, we also identified diverse attitudes among the outlets included in our study. These differences could be understood by taking into account the parallelism characterizing the Italian me- dia system. Indeed, we found the higher quota of negative news items within the coverage provided by Il Giornale and TG5, which are both owned by the Berlusconi fam- ily. As a result of the crisis within the party, the poor results obtained in the 2018 General Elections and the marginal role played within the centre-right area, the two media out- lets engaged in a very negative and critical coverage, attacking both the left opposition and the coalition in government in an attempt to mobilize Forza Italia supporters. On the contrary, media outlets such as Corriere della Sera and TG1 confirm their more neutral approach to covering politics. The distinction between press and TV was explored also with regard to the other re- search questions leading this study. We investigated whether and to what extent the distinction between EU and domestic politics may affect the negative media bias. Two main considerations emerge from our study. First of all, without much surprise, we observed that EU politics benefits from a lower likelihood of negative coverage compared to domestic politics. In other words, the national Italian political debate catalyses media negativity, and this applies to both press and TV. However, when we move to the dimension of political actors, the picture is more nuanced, showing net differences between newspapers and TV newscasts. In this case, indeed, the press confirms its moderation in the use of a negative tone when covering EU political ac- tors, while on TV the dynamic is reversed, meaning that politicians from EU institutions or from other EU member states are more likely to be covered in a negative tone. Secondly, as concerns Europeanization, we observed that the interplay between EU and Italian politics is more likely to be framed in a negative tone when EU political actors intervene directly within the Italian political debate by raising European interests (i.e. threatening sanctions in the event of disregarding EU treaties or agreements). More in- terestingly, when the Bottom-Up Vertical Europeanization variant is addressed within media coverage, TV media are less likely to adopt a negative tone. In other words, when Italian politics addresses EU politics, TV tends to change its approach. This is not found within the press sample of news stories. Also as regards Horizontal Europeanization, press and TV are differentiated, the strong variant being significant only within the press and the weak one significant only within TV. All things considered, it seems that while the press have adopted a negative attitude towards politics, expressing criticism towards domestic politics rather than EU political space, TV shows some signs of adapting to a new climate of opinion. TV media outlets moderate their negativity against domestic politics when compared to the press. But within their coverage they tend more to frame EU political actors negatively in particu- lar, while they temper their criticism when domestic actors approach EU institutions. References Albertazzi, D., Giovannini, A., & Seddone, A. (2018). ‘No regionalism please, we are Leghisti!’The transformation of the Italian Lega Nord under the leadership of Matteo Salvini. Regional & Federal Studies, 28(5), pp. 645-671. SEDDONE, BOBBA and RONCAROLO 87 Banducci, S. A. & Semetko, H. A. (2003). Media and mobilization in the 1999 European parlia- mentary election, in: M. Bond (Ed) Europe, Parliament and the Media, pp. 189–204 (London: Federal Trust for Education and Research). Bobba G., & Seddone, A. (2018). How do Eurosceptic parties and economic crisis affect news coverage of the European Union? Evidence from the 2014 European Elections in Italy. European Politics and Society, 19(2), pp. 147-165. Boomgaarden, H. G. & de Vreese C. H. (2016). Do European elections create a European public sphere? in: W. van der Brug & C. H. de Vreese (Eds) (Un)intended Consequences of Euro- pean Parliamentary Elections, pp. 19–35 (Oxford: Oxford University Press). Boomgaarden, H. G., de Vreese, C. H., Schuck, A. R. T., Azrout, R., Elenbaas, M., Van Spanje, J. H. P. & Vliegenthart, R. (2013). Across time and space: Explaining variation in news coverage of the European Union. European Journal of Political Research, 52(5), pp. 608– 629. Boomgaarden, H. G., Vliegenthart, R., de Vreese, C. H. & Schuck, A. R. (2010). News on the move: Exogenous events and news coverage of the European Union. Journal of European Public Policy, 17(4), pp. 506–526. Cayrol, R. (1991). European Elections and the Pre-Electoral Period. Media Use and Campaign Evaluations. European Journal of Political Research, 19(1), pp. 17-29. Conti, N. & Memoli, V. (2010). Italian parties and Europe: Problems of identity, representa- tion and scope of governance in the Euromanifestos (1989–2004). Perspectives on European Politics and Society, 11(2), pp. 167–182. Conti, N. & Memoli, V. (2013). Italy, in: N. Conti (Ed) Party Attitudes towards the EU in the Member States: Parties for Europe, Parties against Europe, pp. 79–98 (London: Routledge). Cornia, A., Lönnendonker, J., & Nitz, P. (2008). European Councils as first encounters of trans-European media communication?. Journalism, 9(4), pp. 493-515. de Vreese, C. H. & Boomgaarden, H. G. (2006). Media effects on public opinion about the en- largement of the European Union. Journal of Common Market Studies, 44(2), pp. 419– 436. de Vreese, C. H. & Kandyla, A. (2009). News framing and public support for a common foreign and security policy. Journal of Common Market Studies, 47(3), pp. 453–481. de Vreese, C. H., Peter, J. & Semetko, H. A. (2001). Framing politics at the launch of the euro: A cross-national comparative study of frames in the news. Political Communication, 18(2), pp. 107–122. de Vreese, C.H. (2004). The effects of strategic news on political cynicism, issue evaluations and policy support: a two-wave experiment. Mass Communication and Society, 7(2), pp. 191–215. de Vreese, C.H., Lauf, E. & Peter, J. (2007). The media and European Parliament elections: Second-rate coverage of a second-order event? in: W. van der Brug & C. van der Eijk (Eds) European Elections and Domestic Politics. Lessons from the Past and Scenarios for the Fu- ture, pp. 116–130 (Notre Dame: University of Notre Dame Press). Fossum, J. E. & Schlesinger, P. (Eds) (2007). The European Union and the Public Sphere. A Communicative Space in the Making? (London: Routledge). Franzosi, P., Marone, F., & Salvati, E. (2015). Populism and Euroscepticism in the Italian Five Star Movement. The International Spectator, 50(2), pp. 109-124. European versus Domestic Politics 88 Galpin, C. and Trenz, H.-G. (2018). Rethinking First- and Second-Order Elections Media Neg- ativity and Polity Contestation during the 2014 European Parliament Elections in Germany and the UK. ARENA Working Paper 3/2018. Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and pol- itics. (Cambridge: Cambridge University Press). Hobolt, S. B. and Wittrock, J. (2011). The second-order election model revisited: An experi- mental test of vote choices in European Parliament elections. Electoral Studies, 30(1), pp. 29-40. Hutter, S. & Kriesi, H. (2019). Politicizing Europe in times of crisis. Journal of European Public Policy, 26(7), pp. 996-1017. Huysseune, M. (2010). A Eurosceptic vision in a Europhile country: The case of the Lega Nord, Modern Italy, 15(1), pp. 63–75. Itanes. (2018). Vox Populi. Il voto ad alta voce del 2018 (Bologna: il Mulino). Kantner, C. (2011). Debating humanitarian military interventions in the European public sphere, RECON Online Working Paper, 2011/30 (Oslo: ARENA). Kleinen-von Königslöw, K. (2012). Europe in crisis? Testing the stability and explanatory fac- tors of Europeanization of national public spheres, International Communication Gazette, 74(5), pp. 443–463. Koopmans, R. & Erbe, J. (2004). Towards a European public sphere? Vertical and horizontal dimensions of Europeanized political communication, Innovation: The European Jour- nal of Social Science Research, 17(2), pp. 97–118. Koopmans, R. (2004). Integrated Report: Cross-National, Cross-Issue, Cross-Time (Berlin: Eu- ropub.com). Koopmans, R. (2007). Who inhabits the European public sphere? Winners and losers, sup- porters and opponents in Europeanised political debates, European Journal of Political Research, 46(2), pp. 183–210. Kunelius, R. (2008). Journalism and the EU: A relationship in contexts. Journalism, 9(4), pp. 371-376. Lengauer, G., Esser, F. & Berganza, R. (2012). Negativity in political news: A review of con- cepts, operationalizations and key findings. Journalism, 13(2), pp. 179–202. Machill, M., Beiler, M. & Fischer, C. (2006). Europe-topics in Europe's media: The debate about the European public sphere: A meta-analysis of media content analyses. European Journal of Communication, 21(1), pp. 57–88. Maier, J. & Rittberger, B. (2008). Shifting Europe’s boundaries: Mass media, public opinion and the enlargement of the EU. European Union Politics, 9(2), pp. 243–267. Mancini, P. (2008). The Berlusconi case: Mass media and politics in Italy. In I. Bondebjerg & P. Madsen (Eds.) Media, Democracy and European Culture, pp. 107-119 (Bristol-Chicago: Intellect Books). Marsh, M. (1998). Testing the second-order election model after four European elections. British Journal of Political Science, 28(4), pp. 591–607. McQuail, D. & Bergsma, F. (1983). The European dimensions of the campaign, in: J. G. Blum- ler (Ed) Communicating to Voters: Television in the First European Parliamentary Elections, pp. 258–283 (London: Sage). Meijers, M. M. (2017). Contagious Euroscepticism: The impact of Eurosceptic support on mainstream party positions on European integration. Party Politics, 23(4), pp. 413-423. SEDDONE, BOBBA and RONCAROLO 89 Meyer, C. O. (2005). The Europeanization of media discourse: A study of quality press cover- age of economic policy co-ordination since Amsterdam, Journal of Common Market Studies, 43(1), pp. 121–148. Nardis, Y. (2015). News, trust in the European Parliament, and EP election voting: Moderated- mediation model investigating voting in established and new member states, The Inter- national Journal of Press/Politics, 20(1), pp. 45–66. Norris, P. (2000). A Virtuous Circle? Political Communications in Post-industrial Democracies (Cambridge: Cambridge University Press). Peter, J. & de Vreese, C. H. (2004). In search of Europe: A cross-national comparative study of the European Union in national television news, Harvard International Journal of Press/Politics, 9(4), pp. 3–24. Pfetsch, B., Adam, S., & Eschner, B. (2008). The contribution of the press to Europeanization of public debates: A comparative study of issue salience and conflict lines of European in- tegration. Journalism, 9(4), pp. 465-492. Preston, P. (2009). An elusive trans-national public sphere? Journalism and news cultures in the EU setting. Journalism studies, 10(1), pp. 114-129. Reif, K. & Schmitt, H. (1980). Nine second-order national elections–A conceptual framework for the analysis of European election results, European Journal of Political Research, 8(1), pp. 3–44. Roncarolo, F. (2011). Mediatizing Europe While Overlooking the European Election: The Par- adox of the Italian Case, in: Maier, M. & Strömbäck, J. (eds.) Political Communication in European Parliamentary Elections, pp. 129-144 (London: Routledge). Roncarolo, F. (2019). L’argine fragile all’antieuropeismo: l’informazione sull’Europa nel si- stema italiano dei media. Atlante Elettorale, la Repubblica, May 6. Roncarolo, F., & Mancini, P. (2018). The traditional media, political parallelism and public opinion on contentious issues in the 2018 Italian election campaign. Contemporary Ital- ian Politics, 10(3), pp. 243-266. Savini, P. (2019). Fostering European identity through communication. The missed oppor- tunity of the «Treaty establishing a Constitution for Europe». Problemi dell'informazione, 44(1), pp. 59-86. Schmitt, H. (2005). The European Parliament elections of June 2004: Still second-order?. West European Politics, 28(3), pp. 650–679. Schuck, A.R. & de Vreese, C. H. (2011). Finding Europe: Mapping and explaining antecedents of ‘Europeanness’ in news about the 2009 European Parliamentary elections. Studies in Communication/Media, 1(2), pp. 265–294. Shephard, M. (1997). The European Parliament: Laying the foundations for awareness and support. Parliamentary Affairs, 50(3), pp. 438–452. Soroka, S. (2014). Negativity in Democratic Politics: Causes and Consequences (New York: Cambridge University Press). Statham, P. & Trenz, H.J. (2013). The Politicization of Europe: Contesting the Constitution in the Mass Media (London: Routledge). Trenz, H. J. (2004). Media coverage on European governance: Exploring the European public sphere in national quality newspapers. European Journal of Communication, 19(3), pp. 291–319. European versus Domestic Politics 90 Trenz, H. J. (2008). Understanding media impact on European integration: Enhancing or re- stricting the scope of legitimacy of the EU?. Journal of European Integration, 30(2), pp. 291–309. van Spanje, J. & de Vreese, C. (2014). Europhile media and Eurosceptic voting: Effects of news media coverage on Eurosceptic voting in the 2009 European parliamentary elections. Po- litical Communication, 31(2), pp. 325–354. Vliegenthart, R., Schuck, A. R., Boomgaarden, H. G. & de Vreese, C. H. (2008). News coverage and support for European integration, 1990–2006. International Journal of Public Opin- ion Research, 20(4), pp. 415–439. SEDDONE, BOBBA and RONCAROLO 91 Appendix: codebook General tone of the story What is the overall tone of the news item? Does the report convey primarily a positive, negative, balanced or neutral impression of politics, political records, conditions or views? 1 = predominantly negative tone 2 = predominantly positive tone 3 = balanced/ambivalent 0 = neutral/not applicable Instructions: indications of a negative tone are when the story is framed as political failure, fiasco, dis- aster, crisis, frustration, collapse, flop, denial, rejection, neglect, default, deterioration, resignation, scepticism, threats, cynicism, defeatism or disappointment. Indications of a positive tone are depic- tions of political success, problem solutions, achievement, improvement, advance, prosperity, accomplishment, enthusiasm, hope, benefit, gain, sustainability, gratification or accomplishment. If a report does not reflect indications of a negative or positive tone, then it should be coded as ‘neutral’. EU and domestic politics Does the news item contain relevant references to Europe (EU or any other EU coun- try/political actor apart from Italy)? 1= Yes, reference to EU (in general or single institution) 2= Yes, reference to other EU country/political actor 0= no reference to Europe Instructions: Europe is meant as a single institution, as EU issues, as EU policies as well as other EU countries. Please note that code 2 should be applied only in residual cases, namely only other EU country references and not EU institutions. Any reference to Brexit should be coded as 1. Any reference to EU parliamentary elections should be coded as 1. Vertical Europeanization: Bottom Up Does the news item include references to national actors addressing EU actors or making claims on EU issues/policies (general or specific)? 0= no 1 = yes 999= NA Instructions: EU issues and policies should be addressed from a European perspective. For example, references to the EU election campaign or alliances at domestic level are not to be considered as an EU issue. On the contrary, when the discussion focuses on alliances within the EU Parliament it should be considered as an EU issue. Euro-crisis, Schengen, Lisbon Treaty are to be considered as EU issues. Vertical Europeanization: Top Down Does the news item include references to EU actors intervening in the domestic debate of any EU country in the name of common European/EU interests? European versus Domestic Politics 92 0= no 1 = yes 999= NA Instructions: for example, code 1 to be applied for news items where, for example, Draghi/Junker/Ta- jani (or EU commissioners) intervene on Greek debt crisis. Horizontal Europeanization: Weak Does the news item cover the political debate in another country? 0= no 1 = yes 999= NA Instructions: code 1 to be applied for news items focusing on other EU member states and their issues (i.e. Brexit, Gilets Jaunes…) Horizontal Europeanization: Strong Does the news item cover one political actor (from any EU country) speaking about is- sues or with actors related to another EU country? 0= no 1 = yes 999= NA Instruction: code 1 to be applied when an actor from a member state addresses or refers to actors or is- sues related to another member state or even participating in public debates in another country (i.e. Merkel commenting on Gilets Jaunes; Di Maio speaking about Brexit, Orban commenting on Italian government)