Sensationalism and political news in Italian quality newspapers

Sensationalism and Political News in Italian Quality Newspapers

Daniela De Lorenzo

Universiteit van Amsterdam

Master Thesis

Graduate School of Communication

Erasmus Mundus Master’s

Journalism, Media and Globalization

supervisor: Jonas Lefevere

words count: 7623 25th June 2014

2

Acknowledgements

“No formalities on a daily basis” said the Dalai Lama when I met him, in 2012.

I am quite sure he had never dealt with APA style and so forth. From that meeting onwards, I decided to be as less formal as possible in my every-day life. Really impossible in a MA thesis, with its suffocating standards and rules. I felt I had to show a bit of informality, at least in the acknowledgements, to stay truth to my ideas as I always wanted to belong to/ The freak scene/ Or anyone who had set their mind to/ Never follow suit again/ Because they have to1.

This thesis project wouldn’t have been possible without:

The Radio Dept. and their album “Clinging to a Scheme” listened uninterruptedly throughout the whole coding and editing process. Indeed, I was clinging to a scheme.

Me! Four month in the shade. I survived to my thesis. Swear I haven't got a clue. Having had all possible physical stress symptoms, I’d like to thanks my body to not have completely fall apart.

Mamma e papá. Thanks for emigrating to Milan 30 years ago, and for supporting my “emigration”, further north. Thanks for considering education as the most important tool in life. Seemed like a good idea / at one time/ But now it's all wrong. In less than two months distance, you get both your daughters graduated with a MA…bracing the rank of Italian youth unemployment. Congratulations guys, you made it! Thanks to my sister Alessandra, my second coder for the thesis and occasional proof-reader for my exams.

My supervisor, Jonas Lefevere. Please accept this as a token/ of my sincere gratitude, for guidance and most of all, for being patient with my informality, my typos, my delayed emails.

The people who made this M.A. not be just “reading manuals”, and actually proved that the most important lessons were taught out of classrooms:

-the Mundus family. It’s been an honor to share these two years with all of you, and be part of this representative scale of the international society: made of geniuses, dreamers, geeks and social climbers (at the borders). All in all: a tremendous experience. A big thanks to Fanny Chays, for the company in the aseptic computers room of the university - fighting to keep it together, learning new ways to be right on SPSS - and for the lunch breaks on the canals, on sunny days.

- people met in Denmark, especially: Alexandra, Emily, Noémie, Løves’ crew, Hansen family (and with this statement, basically, I am thanking half of Denmark),Shaw family and Professor Georg Sørensen. - the Dutch fellows met at Eurosonic Festival in Groningen: it's good to know we both still care/ And that we have this to share, next festival edition. Thanks Rienk, David, Robert, Wilbert, Auke and Eva.

In Italy, we say carta canta (“paper sings”) when a written text is un-objectionable, incontestable. This acknowledgements page literally sings. Until results will tell how, this thesis is however the clear, incontestable proof, I am done.

1

All italics quotes comes from songs of “Clinging To A Scheme” album, mentioned later on. In order of quotation:

Never Follow Suit, Four Month In the Shade, Memory Loss, Domestic Scene, A Tokens of Gratitude , Heaven’s On Fire, This Time Around, The Video Department, David. The author of this thesis suggests a listen.

3

Content

Abstract

Introduction

Theory & Hypotheses

Production on news in commercialized settings 3

Sensationalism, a definition 4

Sensationalism in the Political News Realm 7

Methodology

Findings

Sensationalism in Il Corriere della Sera 19

Sensationalism in La Repubblica 21

Conclusion

Conclusions 24

Discussion and Limitation 28

References

4

Abstract

In the present study attention is given to the concept of sensationalism and whether its use in reporting political news has increased. In different Western countries, many studies have been looking at the growing usage of sensationalism in television context. The increased of sensationalism have been considered a consequence of commercialization of media systems and their need of profiting. However, no prior study have been focusing on the Italian media system, nor to its newspaper outlets environment. We therefore investigated whether Italian print media’s political coverage became increasingly sensationalistic during the past decade, both at a domestic and at a European political coverage level. A content-analysis of the two most widespread Italian quality newspapers, Il Corriere della Sera (CdS) and La Repubblica (LR), was conducted over the years 2005 – 2013. A codebook to identify sensationalistic textual feature was used to code articles in both domestic and foreign sections (N = 339). Findings from one-way ANOVAs and univariate factorial analysis showed that, domestic political news are reported in a more sensationalistic way, compared to European political news. Furthermore, in the specific Italian context, commercialization and market reasons are not to be considered the only causes of the increase of sensationalism. Sensationalism needs, indeed, to be analyzed in the lights of media outlets’ partisanships.

Keywords: sensationalism, Italian media system, Italian politics, European politics, political

5

Introduction

It has been said that Italian people, disillusioned by their political class, are not interested in politics anymore. However, as Mancini (2000) states, commercialization in Italy has not caused a decrease in the level of attention that mass media devote to political issues, but the nature of this attention has changed substantially (Mancini, 2000, 321).

While on duty, ex-Prime Minister Berlusconi had generally attacked media outlets , from TV channels to newspapers, accusing to mystify reality, especially at his expenses. In 2009, he explicitly said : “Television, is the good part of the media system, as they do not change words. As photographers, do not change images. But newspaper journalists, the mean ones, they can change content by writing it.”2.

From a media tycoon, this is a rather simplistic and explanation on media behaviors, but indeed, journalistic values aside media culture, and economic and legal conditions, is considered to be an important factor in the way news is reported (Esser, 1999, 293).

Nowadays, the main concerns in media is the impact of market-driven journalism on the quality of the news coverage. The primary function of news is, or should be, to provide an objective narrative about the world. However, the increased commercialization makes media outlets choose to increase the production of news able to attract viewers and readership, giving informative value a less determinant role in the production of news.

Aside the increase of news, specific features and tools able to increase readership are in

2

6

place. Newspaper outlets have no possibility to increase the quantity of theirs news, therefore these strategies seems to be the only way they can grasp more attention. Sensationalism is one of the tools able to canalize attention through its specific reporting features.

Whereas, sensationalism is typically utilized in the reporting of sport, crime and showbiz (Grabe et al., 2000), the Italian political scene , and its scandals, aside market-driven causes, may have given reasons to utilize this tool steadily throughout the last decade. Newspapers, therefore, might have sensationalize their political news contents.

In the last decade, as well, the European Union have become a much more defined super-national actor, with a more clear division of powers between its institutions and the super-national ones. As its role became more defined, there might have been an increase of attention and reporting on this actor. However, no set European public sphere is in place (Habermas, 2001; Trenz, 2008), therefore the way national media system portrays national politics determines the discourse around European politics as well. If the Italian media system started utilizing sensationalism as a general tool to cover politics, the chance that the discourse around European politics have become more sensationalized, seems to be higher.

We therefore question, has sensationalism been adopted as a journalistic tool to cover

political news?. Secondly, we question: has sensationalism been utilized to a different extent when portraying European or Domestic politics in Italian Quality Newspapers?. Through a quantitative content analysis of two major quality newspapers we will try to assess whether in the last decade (2005 - 2013) sensationalism became a constant feature in political news and whether it has increased.

7

Theory and Hypothesis

Production on news in commercialized settings

As previously mentioned, one of the main concerns in media is the impact of market-driven journalism on the quality of the news coverage. Infotainment, together with tabloidization is the most direct result of commercialized media (Brants, 1998) and has promoted the increase and emphasis on specific themes (Esser, 1999). However, at the same time, many scholars stressed how these features diminished the importance of public service and quality of their news (Picard, 2004; Bernstein, 1992). As the attempt of attracting readers seems having a purely profiting aim by media outlets, sensationalism and its use, takes usually a negative connotation. Three main concerns have been underlined: it violates notions of social decency; it displaces socially significant stories; and it is seen as a new-sprung drift into excessiveness (Grabe et al.,2001, p. 636).

In media systems theory, symbolic capital is no longer seen in terms of the economics of information, but rather in terms of the economics of attention (Lanham, 2006). In the current time of high commercialization and competition, sensationalism is useful for media platforms as it is designed to attract attention and able to reach higher viewership ratings (Berkowitz, 1993; Scott & Gobetz, 1992). Sensationalism is a theoretical concept that encompasses those features of journalistic products that are capable of attracting the attention of the audience (Hendriks Vettehen, 2007).

8

Studies have showed how, public and private broadcasts have increasingly utilized sensationalism in their news stories to attract readership. Hendriks Vettehen et al. (2008) demonstrated that in a period where competition between Dutch TV news programs increased, a trend toward the use of more sensational production techniques was found. As well, Hjarvard (2000) looked at the case of Denmark broadcast system in the 1990s, finding that sensationalism in news could be enhanced by the introduction of competitors in the media (television) market. An increase of competitors resulted with an increase of news production characterized with sensationalistic features.

Thus, higher commercialization of the media environment have pushed media systems to change their news production. As well, we have shown how sensationalism could be seen as a media news forces, as sensationalist content consumption can bring possible effects on readers. Looking at the latest researches on the topic we will now formulate a conceptualization of sensationalism, to further develop our study.

Sensationalism - a definition

Sensationalism has been studied since the late 1930s, as reported by Tannenbaum & Lynch (1960) and from then on a variety of definitions of sensationalism have been developed. This, especially because of the growth and raise of different media outlets, that introduced new way of “experiencing the news” (visual, audio, multimedia…). Nowadays, sensationalism is being mainly recognized as a popular journalistic practice aimed at ‘crafting the experience of engagement’ (Peters, 2011, 297).

9

Traditionally, sensationalism in the news has been conceived mostly in terms of story content to make reluctant audiences read (Tennenbaum & Lynch , 1960) as its dramatic tone would be able to grab public attention more easily. Researchers have also linked sensationalism to the concept of infotainment (Graber, 1994) and tabloidization (Grabe et al., 2000) as this kind of coverage utilizing dramatic elements increases the level of sensation.

We will focus on the definition of sensationalism by Hendriks Vettehen & al. (2005) who propose a conceptualization dividing sensationalism news features into a three element system consisting of: arousing content , tabloid packaging and vivid storytelling. As a feature, it has generally been utilized in reporting on specific topics such as sports , crime and showbiz (Grabe et al., 2000), however in recent studies, it has been argued that the emotion-eliciting qualities of news are not only bound by topic type, but also by specific linguistic features that cut across news topics (Uribe & Gunter, 2007, 210).

Audiences are sensitive to certain categories of information: arousing content, or “basic needs content” which implies information with survival value (Hendriks Vettehen et al., 2005) from an evolutionary perspective, seen as fundamental in the relation between news production and reception. Examples may refer to crimes and violence (Hendriks Vettehen & Nuijten, 2006).

Secondly, sensationalism cannot only refer to the content, as previously mentioned, but also to the way it presents itself. Aside from a sensationalist choice of topics to be covered, therefore, sensationalism also requires a sensationalizing way of packaging information to make it appear more interesting and relevant, and thus newsworthy. (Molek-Kozakowska, 2013, 177). They refer to this information packaging as tabloidized following Grabe et al. (2001) definition.

10

Finally, vivid storytelling refers to the possibility of further characteristics inside the news able to increase proximity of the story in a spatial, temporal or sensory terms (cf Nisbett & Ross, 1980). Vivid information, such as giving exemplary case histories in a news story, rather than general information, appeared to have positive effects on information processing (Hendriks Vettehen et al.,2006) and therefore be able to attract way easier audience to the news story.

In a recent study Hendriks Vettehen et al. (2008) further developed vivid storytelling in two other features: concreteness , and proximity. The former concept means the act of presenting exemplary information in a news story rather than the general information. The latter, is also a key news value that instead allows newsrooms to develop a closer relationship with their viewers by adding some elements such as the choice of angle and narrative style, able to develop a ’vis-à-vis’ news reporting rather than “top-down” messages (Hjarvard, 2000, 65).

Bourdieu’s (1996) critique of journalism highlights how journalism is losing its autonomy to economic and market principles. A commercialized environment means a competitive system in which the will of having higher readership is also related to other media competitors. When reading a newspaper online or watching television news, sensationalism is used in order to grab attention to that specific channel or media, but as well, to avoid channel zapping.

As we see that commercialization drastically changed the news media landscape, we question: Has sensationalism in political news increased in Italian quality newspapers between 2005 and 2013?

11

reporting throughout the nationals media systems (Slattery et al., 2001; Slattery & Hakanen, 1994; Hendrik Vattehen & Nuijten, 2006; Hendrik Vattehen et al., 2006, Hendrik Vattehen et al., 2005). In the specific case of Italian media system, Mancini (2000) notes that infotainment combines politics and entertainment, however he argues that in the Italian media landscape, despite holding similar attributes, infotainment cannot define the style and form of news. In Italy, politics continues to hold a privileged position in the media, but in a new way that has combined commercialization and partisanship (Mancini, 2000, 322). As mentioned by Hanretty (2007) Italian journalists are, disproportionately likely to emphasize the role of political parties and lack autonomy from them. Mancini (2000) therefore defines this condition as political sensationalism, meaning the dramatization and intensification of political conflict and dramatic and exciting content involving confrontations in order to attract viewers and readers.

We hypothesize therefore that: sensationalism in political news Italian Quality Newspaper increased between 2005 and 2013.

Although political news has never been considered a traditional sensationalistic topic, we expect that sensationalism increases constantly throughout the period of time between 2005 and 2013 throughout all the three aspect in which sensationalism will be analyzed (basic need content, tabloid packaging and vivid storytelling).

Sensationalism in the Political News Realm

As the Italian media culture described by Mancini (2000) incorporate sensationalism in political news, we could expect that the same trend is present throughout the

12

whole political news environment, and not only in the domestic one. However, news framing sets the parameters in which citizens discuss public events (Tuchman, 1978) , and inside the national media sphere, nationalistic and ethnocentric bias comes into place. Journalism is deeply rooted in culturally specific patterns (Trenz, 2008) when reporting on foreign affairs and European matters, and it seems difficult to believe that, national media would present non-nationalistic news similarly to domestic ones.

Our second hypothesis deals with the possible distinction between European political news and national political news: How do coverage on European politics appear to be in national media in terms of sensationalism? Is the same degree of sensationalism utilized when reporting about domestic and European politics?

The most obvious reason is that EU news is only marginally covered (de Vreese 2001; Peter and de Vreese 2004), while the nation-state remains the primary focus for public political debates. Marginal coverage could reflect detachments of citizens, and therefore, no interest of media in making effort to attract already disinterested readers. As well, as noted by Hanretty (2007) Italian journalists’ emphasis on political parties in their coverage of politics, but also, in their news coverage in general, could influence their approach towards the way they report European political news, where no partisanship is in place.

Another research by de Wilde (2011) shows that the debates around European Union is highly politicized, where politicization is defined as an increase in polarization of opinions, interests or values and the extent to which they are publicly advanced towards policy formulation within the EU (de Wilde, 2011, 559). Political sensationalism as previously defined as

13

intensification of political conflict and dramatic and exciting content involving confrontations (Mancini, 2000, 322) should therefore be utilized in reporting about European Union .

We hypothesize that: Sensationalism is more present in domestic political news than in European political news (H2).

We expect therefore an increase of sensationalism in reporting about European Union, but not as substantial as the one present in domestic news.

Methodology

1. Sample

Articles from two most read quality Italian newspapers La Repubblica (LR) and Il Corriere della Sera (CdS) were content analyzed. The Corriere della Sera has a center-liberal/indepentent alignment3 with a circulation of 472.883 daily copies4. , while La Repubblica has a center-leftist alignment5 with a circulation 455.672 copies. As researches by Macini date back to 2000, we decided to test whether the trend identified in the early decade continued until recent time. Article were therefore selected in the subsequent period of time, between July 2005

3

Il Corriere della Sera has been founded in 1876 . It is the most widespread newspaper in Italy, mainly for its independence from political ownership and non-sided position.no political ownership is in place.

4

Data from November 2013, from ADS (Accertament Diffusione Stampa), official certification company for press coverage assessments. No other newspaper reaches more than 257000 copies. Media Concentration data can be found in Agcom’s Herfindahl Hirschman Index (HHI) of concentration

http://www.agcom.it/provv/D365\_00\_CONS.htm

5

14

and June 2013. Our units of analysis where firstly measured in terms of general requirement - as date of publication , page number and size - newspaper and section presence and topic.

Articles from Il Corriere della Sera were partly collected from LexisNexis (2008- 2013) and partly from the archive on the newspaper website (2005-2007) because LexisNexis does not store earlier years coverage. Articles from La Repubblica were all collected through their website database. Articles in the foreign and domestic political sections were the units of analysis of the present study. These sections were selected as the most relevant and appropriate to investigate the news typology involved, political news. Political news was defined as a story in the foreign or domestic political section which was (1) referring to domestic or European government institutions as main actors, dealing with policies, polity and politics, (2) relating to government policy-making as distinguished from administration or law or (3) relating to the parties and the partisan aspects of politics .

To measure the European news coverage we sampled articles in the foreign politics section of both newspapers. Search words selected were names of European Union institutions (“European Parliament”, “European Commission”) and European Union as an entity itself (“EU”, “European Union”). Domestic news coverage was sampled from the domestic politics section, entering search words referring to political bodies and institutions of Italian politics (“Italian government”, “Italian Parliament”).

As our sample was one article per week, no purpose was seen in utilizing methods such as constructed week sampling, the most appropriate when having to sample units from media outlets. For both topics, one article a month was selected in both newspaper, utilizing a simple

15

random sampling technique in order to have an unbiased yet reliable representative sample. The total number of political news stories in the sample was 339 articles with N= 157 in European political news (LaR = 82, CdS = 75) and N= 182 in domestic political news (LaR= 91, CdS=

91).

To test the reliability of the codebook for the current study, intercoder reliability was calculated using Krippendorff's alpha test. All variables from ten articles for each section and newspaper (N = 40) were coded by two coders. The second coder, familiar with the topic, received training from the first coder, whom developed the codebook. As proposed by Krippendorff (2004) it is customary to require α ≥ .800. Tentative conclusions, however, are still acceptable, and α ≥ .667 is the lowest conceivable limit (Krippendorff, 2004, p. 241). The value α ≥ .667 was imposed as minimum standard coefficient.

2. Measurement

Sensationalist features were derived from the three categories of sensationalism: arousing content, tabloid packaging, and vivid storytelling. Following Hendriks Vettehen et al. (2008) research, the last category, vivid storytelling, was divided and analyzed in terms of two further categories: concreteness and proximity. As the categorization made in previous studies from Hjvard (2000), Grabe et. al. (2001) and Hendriks Vettehen et al. (2008) were based on the analysis of sensationalism in television news, we utilized the frameworks of their categorization and adapted it to serve the study of newspaper articles.

16

We therefore developed a new codebook (Appendix I) following their ones, but removing those features intrinsic to television news (such as sound effect, close ups…) and incorporating features reflecting textual analysis from previous studies from Kurtz (1993) and Tannenbaum & Lynch (1960). As we are seeking to determine presence of sensationalism, all our variables are dichotomous, where the variable takes a value of 1 when there is a sensationalism feature, while the variable value corresponds to 0 when sensationalism is not reported.

2.1. Arousing content

Three aspects of arousing content were distinguished.

a) Sensational topicalization Political news can present several different topics referring to election, international relation, public affair and so on. We look specifically on whether political news was told in terms of most popular sensationalism topics. The categories defined as sensational by Grabe et al. (2001) analysis were stories content with survival value such as stories about crimes, violence, as well as justice, war, malpractices and scandals. More than one category could be selected. The score on this variable refers to the presence of at least one sentence (presence = 1, non-present = 0) of one or more of the formulated sensationalist categories in the political news article. If no category of the aforementioned was present, the article did not tell the story using a sensationalist topic.

b) Verbalized Emotions implies the presence of wording referring to emotions. It has been argued that explicit references to basic emotions by anchors, interviewees, or bystanders appeal to basic human needs (Hendriks Vettehen, 2008). More recently, a study of Lecheler et al. (2013)

17

proves that feeling such as anger and enthusiasm are able to cause a shift in opinion and political support, while contentment and fear do not. We included variables for all the four feeling mentioned above, as they have been included in previous studies. Furthermore we included a fifth variable, Others (presence = 1, non-present = 0), referring to the presence of language referring to others specific emotional responses.

Following early categorization of sensationalism by Tannenbaum & Lynch (1960) we further include another category.

c) Activity Factor , whether the vocabulary utilized in the news article is active or passive. Trenz (2008) noted that when media provide unbalanced information about politics and not of opinion, media perform a passive role. We defined vocabulary as passive (value = 0) when the verbs and adjectives mirror the political system and amplify the rational discourse (e.g. adjectives as: understandably, wisely…) while we defined it active (value = 1) when its aim is to critique or prompt action (e.g. verbs as could, should, adjectives as moot, thoughtless, inaccurate...).

2.2. Tabloid packaging

Tabloid is used to describe a type of writing, which tends to be casual, concise and sensationalist in both tone and language (Conboy, 2006), while quality newspaper focus on informing the public, using serious language and tones (McLachlan and Golding, 2000). To measure tabloid packaging, aspects were distinguished in terms of content-style, layout, tone and

18

frame. Variables were based on a study by Kurtz (1993) on tabloidization of quality newspaper and from a study by Tannenbaum & Lynch (1960) on sensationalism measurements in news.

a) Content-Style. Following Kurtz’s (1993) study, we looked at whether the political news was presented focusing on its soft news features (value = 1), rather than its hard news, informing capability (value = 0). Furthermore, we looked at the presence of metaphors ( presence =1 , non-presence = 0), personification of institutions ( non-presence = 1 , non-non-presence = 0) , dramatizing verbs ( presence = 1 , non-presence 0) and adjectives ( presence = 1 , non-presence = 0).

b) Sensationalist Title. We focused on whether the articles titles was utilizing an arousing, and sensationalistic wording ( presence = 1 , non-presence = 0) similarly to the previous theorization made in regards of the content of the article. By being the leading sentence able to grab attention of the reader as, the title might tell whether the whole content can tend to be packaged in a tabloidization way. No further layout features (paragraphing, presence of images) could be analyzed as articles in archives are not presented in the same layout as in the newspaper.

c) Tone. We looked at the tone of journalists in reporting. To determine a tabloidization tone, we used categories from Kurtz (1993) and variables from the evaluative factor of Tennenbaum & Lynch’s (1960) study. Our dichotomous variables therefore tested whether articles were written in a scandalized, and emotional way. Scandalization ( presence = 1, non-present = 0) was determined on the basis of whether the journalists non-present the issue portraying it as immoral or improper, utilizing wording in order to shock and horrify the readers. (e.g. shameful, dishonorable, unbelievable…). Emotionalization ( presence = 1, non-present = 0) was

19

determined on the basis of whether the journalist presents the issue wanting to appeal the readers emotionality with dramatic narrations.

2.3. Vivid storytelling

Two aspects of vivid storytelling were distinguished: concreteness and proximity . Following Hendriks Vettehen et al. (2008) and Hjarvard (2000) explanation and codification of these main categories we developed a own coding of storytelling.

a) Concreteness . Concreteness can be described as the exemplification of a news story. Example can make a story more concrete and direct and therefore vivid, instead of using general descriptions to exemplify the public opinion concerning the specific issue (Hendriks Vettehen et.al, 2008). We therefore looked at it creating two indicators : personalization, and laypersons speaking. As story featuring individuals in their everyday environment or telling about their personal situation (Hendriks Vettehen et al., 2008) increase concreteness, we coded personalization as whenever an individual was portrayed while carrying a specific task which was explained or told. As concreteness refers to exemplification and not to the subject of the examples, we will consider as individuals, no distinction between politicians and common people will be made. In television news laypersons speaking is considered to be commentary from outside and non-introduced sources such as passing tourists and common people (Hvitfelt,1994) . We considered layperson speaking variable (presence = 1, non-present = 0) when quotes and comments from civilians appeared in the text.

20

b) Proximity. Proximity is one of the classic criteria of newsworthiness . Following researches by Hjarvard (2000) it was studied in terms of both content and narration. Not only visual, but also textual features, can actually induce feelings of sensory closeness amongst readers. Presence of proximity was categorized into two main variables. The first one tested whether the presence of a subject matter within viewers' frame of reference was present. It was coded as the presence of social reproduction of policies and therefore whether examples and explanations of effects of policies on citizens’ life was in place ( presence = 1, non-presence = 0). Hjarvard (2000) explained that subject matter related to the recipient of the news, is closer than traditional political news and therefore enhances proximity. The second variable coded was familiarity ( presence = 1, non-presence = 0) of the narration and therefore whether a closer relation between reporters and readers was in place (e. g. “why should we pay”…)

3. Procedure

The two hypothesis formulated in the theoretical part of the current study were tested using analysis of variance (both simple and factorial ANOVA), to test variance of sensationalism as well as its sub-components, over time. From our variables, we created a new sum-scale variable, namely “Sensationalism” (S), from the sum of the three subcomponents: Arousing Content (AC), Tabloid Packaging (TP) and Vivid Storytelling (VS).

We grouped variables belonging to each subcomponents and divided them for the number of their variables. To determine the extent to which an article had Arousing Content (AC) value we used the following formula:

21

AC = Sensational Topicalization [(crime + violence + justice + war + malpractice +

scandals)/6] + Verbalized Emotions [(Anger + Excitement + Contentment + Fear + Oth

er)/5] +Activity Factor (Activity /1)

Therefore , Arousing Content takes a value between 0 and 3, with higher values corresponding with higher degrees of sensationalism.

The same procedures have been carried out for the other two sub components (see Appendix II).

of sensationalism, to finally reach the total sensationalism value

S = AC + TP + VS

that could range from a minimum value of 0 to a maximum value of 8.This results in an overall value of sensationalism for four sets of articles: Sensationalism in European Political News in Il Corriere della Sera (CdSEU) Sensationalism in Domestic Political News in Il Corriere della Sera (CdSDom), Sensationalism in European Political News in La Repubblica (LREU), and Sensationalism in Domestic Political News in La Repubblica (LRDom).

A one-way ANOVA was conducted to compare sensationalism level within years 2005 – 2013 for each of the four sensationalism values. Furthermore, we conducted one-way ANOVA for each of the subcomponents. As we utilized a parametrical statistical method, with equally distributed data, we ran a parametric Levene’s Test to test equality of variance. For all the four sensationalism value, Levene’s test verified the equality of variance in the samples (CdSEU p =

22

0.862; CdSDom p = 0.747; LREU p = 0.473, LRDom p = 0.126) (p > .05) (Martin & Bridgmond, 2012).

Findings

In the second part, two hypotheses were formulated. The first hypothesis concerned whether sensationalism increased throughout time. The expectation was that more sensationalism would be present nowadays than in the early 2000 in both Il Corriere della Sera and La Repubblica. A one-way analysis of variance calculated sensationalism in articles on European politics in the foreign affair section of Il Corriere della Sera and La Repubblica , as well as on domestic politics in the domestic section of Il Corriere della Sera and La Repubblica.

In Figure 1, we first look at the use of sensationalism in the period under investigation, for both the domestic and European political coverage in the two newspapers. For part of the section

23

analyzed it is possible to see a clear increase of sensationalism. The two sections of La Repubblica scored a higher level of sensationalism and a general upwards trend. Il Corriere della Sera’s sections have had a decrease of sensationalism reporting, in both sections, during the years 2009 – 2012, to then increase again in the later year. For both newspapers, however, higher sensationalism appears in their domestic sections, rather than in the foreign one, reporting on European politics. The results of the four sections are explained in detail in the following part. 1. Sensationalism in Il Corriere della Sera

Sensationalism in European Political News

The foreign section of Il Corriere della Sera presents unexpected results as, instead of an increase of sensationalism, a decrease is clearly visible. A one-way analysis of variance predicted sensationalism in articles on European politics in the foreign affair section of Il Corriere della Sera, throughout the time period 2005 - 2013 (IV). Specifically in the years 2010 – 2012, the level of sensationalism was lower than in 2005.

A possible explanation could be the fact that the financial crisis of 2008, attributed to the European Union by media, slowly lost its centrality as a topic to relate to Europe. As well, a replacement of the government Berlusconi with the Monti Cabinet and its credibility at the European level. The new government center-liberal oriented as Il Corriere della Sera, has been seen as a beneficial considering ex-Prime Minister Monti previous experience as commissioner at the European Commission. In the year 2013, the level of sensationalism increases again, although similarly to the levels of the early 200. The reasons could be the shift of government and period of instability, as well as, the approaching of European Election in 2014.

24

Though the graph suggest a decrease trend, the analysis was not significant, F(8,66) = 1,631 , p = .133. No significance was found as well for the analysis of the same section, excluding data from 2013: F(8, 66) =1.554 , p. .166, therefore we reject H1. Further support to our rejections comes by analyzing individually the three sub-components. No significance was found for Arousing Content F(8,66) = 2.061, p = .052 , Tabloid Packaging F(8,66) = 1.080 , p = .388 nor Vivid Storytelling F(8,66) = 1.209 , p = .307, although Arousing Content values were almost generalizable (F critical = 2.082, p =.05) .

Sensationalism in Domestic Political New

A similar trend is visible in the domestic political section of Il Corriere della Sera for which a oneway ANOVA predicted sensationalism in articles throughout the time period 2005 -2013 (IV).

During the years of the Monti Cabinet, also called “the earnest government” by media outlets6, the level of sensationalism is lower, to then increase again in between 2012 – 2013 , when yet another government crisis lead months of uncertainties and towards a left wing government in April 2013.

Although the graphs shows that throughout time sensationalism has generally increased, the analysis was not significant, F(8,82) = 1,088 , p = .380 and therefore we rejects our first hypothesis. However, analyzing individually the three sub-components, significance was found for Tabloid Packaging F(8,82) = 2,239 , p = .033, while, no significance was found for neither

6

The definition comes from Don Ciotti statement to Il Fatto Quotidiano on : “Let’s hope this government will be ethical, transparent and earnest ” on 21st November 2011 http://www.ilfattoquotidiano.it/2011/11/21/ciotti-speriamo-governo-monti-etico-trasparente-sobrio/172135/

25

Arousing Content F(8,66) = 0.722, p = .671 and Vivid Storytelling F(8,82) = 1.541 , p = 0.165. This suggest that we can support our H1 only for the Tabloid Packaging subcomponent.

2. Sensationalism in La Repubblica

Sensationalism in European Political News

A constant increase in the presence of sensationalism seems visible in the articles on European politics in the foreign affair section of La Repubblica. The one-way analysis of variance throughout the time period 2005 -2013 (IV), however, was not significant, F(8,73) = 1,209 , p = .306 and , therefore we reject H1. Reasons might be the fact that level of confidence for the yearly mean was overlapping (especially in two waves 2006-2008 and 2009- 2013). Analyzing individually the three sub-components, significance was not found for Arousing Content F(8,73) = 0.589, p = .784 , and Vivid Storytelling F(8,73) = 0704 , p = .687. Significance was found only in Tabloid Packaging F(8,73) = 2.466 , p = 0.02 , for which H1 can supported.

Sensationalism in Domestic Political News

Significance was found only for sensationalism in articles on Domestic politics in the foreign affair section of La Repubblica, throughout the time frame 2005 – 2013 (IV). A substantial increase can be noted in the years of right wing government and during the government crisis. The one-way analysis of variance was significant, F(8,82)= 5.865, p = .00, which support H1. Analyzing individually the three sub-components, significance was found for

26

Arousing Content F(8,82) = 5.897, p = .00 and Tabloid Packaging F(8,66) = 3.3007 , p = .003, but not for Vivid Storytelling F(8,82) = 0.929 , p = .497, (F critical = 2.053, p =.05) .

The results of the analysis support H1 significantly only for La Repubblica’s domestic section, but not for the other three sections, therefore we reject H1. Although no significance was found for the other three categories, a general increase in sensationalism can be found in La Repubblica’s foreign affair section and Il Corriere della Sera’s domestic section, for which tabloid packaging was significant for both sections. In regards of Il Corriere della Sera’s foreign section, no substantial increase has been proven by the one-way ANOVA test. As well, in any of the four tests, the variable vivid storytelling was found to be not significant.

This suggest that, throughout time, newspapers have generally increased the use of specific content features belonging to the sensationalism categorization, and that the three components of sensationalism can appear independently from one another.

In the second hypothesis the expectation was that Domestic political news would have had a higher level of sensationalism than European political news. We firstly have to note that, in general, European political coverage was lower than domestic political coverage, in both newspaper. Secondly, by looking at the estimated mean of sensationalism (Figure 2) in the four newspapers' sections, we indeed support the hypotheses (H2) for both newspapers.

By comparing values from the two different newspaper we further notice, that the level of sensationalism in Il Corriere della Sera is lower than in La Repubblica. As well, Domestic Political News in Il Corriere della Sera (M = 1.99, SD = .810). are written utilizing less

27

sensationalism than in the European political news of La Repubblica political news (M = 2.27 , SD = 1.012).

A general trend in both newspaper has been to utilize more sensationalism in the titles of articles referring to European political news than in the article content, while the opposite case was visible in the Domestic political news.

To sum up, from both hypothesis, it seems that differences at the media level may explain differences in the extent of sensationalism, despite anyway supporting the hypothesis that political sensationalism is presentmore present in domestic political news compared to European political news.

To provide a more rigorous test of H2 we used a factorial univariate analysis in which we statistically tested whether variance of sensationalism differed systematically between media outlet and sections. Our independent variables were therefore the sections of the newspapers and year as a control. The factorial ANOVA resulted in significant for both time (F = 2.085 ; p =

Figure 2

Level of Sensationalism

Newspaper Outlet European Political News M SD

Domestic Political New M SD La Repubblica 2.270 1.012 2.621 1.039

Il Corriere della Sera

1.365 .973

1.991 .810

28

.037; M = 1.7) and news section (F = 21.798, p = .000, M = 17.772). Further, our analysis found significance for the interaction between both time and news section (F = 2.387 ; p = .000 ; M = 1.946 .) meaning that sensationalism was different between sections, and it increased throughout time.

Using a Scheffe post-hoc test we analyzed the relations between sections. We found that a significant differences ware present between all the sections (p = .000) except for the compared La Repubblica’s foreign section and La Repubblica’s domestic section (p = 0.098) and La Repubblica’s foreign and Il Corriere della Sera’s domestic section (p = .238). For the former case, no significance might be present due to the fact that, by being in the same newspaper, a common practice of sensationalism is performed in all sections of the newspapers, despite at different levels. For the latter case no variance might be present due to the similar levels of sensationalism in both sections. However, it has to be addressed that consistent differences exist not only between sections (H2), but as well, between the two newspapers.

Conclusions

Conclusion

“The brain was not meant to read.” said Michael Gazzaniga, one of the leading researchers in cognitive neuroscience. The alphabetic system is only one of the conventions that people use to interact with, and understand what happens around them. The human mind knows and recognizes the world mainly through its eyes. Nowadays, the newspaper industry is facing a deep decline as television and multimedia content online appeal to audiences as their visual

29

components render information more graspable than textual ones. In an era of competition however, television newscasts raise the stakes by using sensationalistic features to attract a higher viewership (Hendriks Vettehen et al. 2008; Hendriks Vettehen et al. 2006; Grabe et al., 2001). Does the newspapers industry run for cover and follow this pattern to avoid its decline?

In the present study, we investigated whether Italian print media’s political coverage became increasingly sensationalistic. Through the decades, sensationalism has been conceived as a specific discourse strategy aimed at channeling audience’s attention (Molek-Kozakowska, 2013, 175), that can enhance “unwholesome emotional responses” (Mott, 1962, 442), shocks and thrills our moral sensibility (Tannenbaum & Lynch, 1960). In this sense traditional non-sensationalistic topics such as political news can be covered in a non-sensationalistic way using attributes leading to an audience emotional response.

The two most widespread Italian quality newspapers Il Corriere della Sera and La Repubblica, were chosen as cases of our research.

Although a general increase of sensationalism can be seen between 2005 and 2013, sensationalism was found to be significant throughout time only in the Domestic section of La Repubblica. Significance for Tabloid Packaging was found in both sections of La Repubblica and in the domestic section of Il Corriere della Sera. No significance difference was found for the foreign section of Il Corriere della Sera.

It can be concluded that tabloid packaging is indeed recurrently utilized by these newspapers to write political news stories and, therefore, politics in news is featured with

30

sensationalistic characteristics. As aforementioned, significance throughout time, was found only for some of the newspaper’s sections. This, together with our second analysis suggests that sensationalism might not only be determined solely by market reasons, but it is indeed bounded to topics and events in the domestic and European political arena as well as at a media level.

Organizational influences on news content (Shoemaker and Reese, 1996) should indeed be seen in the light of macro-level factors related to the political and national media systems (Hallin and Mancini, 2004; Van Dalen et al., 2012). The particularity of the Italian media system made it possible to develop a specific kind of sensationalism, political sensationalism, that can enhance emotional responses depending on which partisan media outlet is visualized (Mancini, 2000). According to Hanretty (2007) parties remain central to the Italian polity to a greater degree than in many other advanced democracies. Furthermore links between journalists and the political elite brought deleterious consequences for journalistic independence (Hanretty, 2007).

Newspapers have a tradition of partisanship, for which readers buy specific newspapers as they follow a specific political ideology. Once bought, readers, do not have the possibility of changing news provider (as it might happen when changing newspaper online websites or through zapping TV channels). This might protect newspapers journalists from being bounded to the rules of the competitions market as much as other outlets' journalists, to stick, however, to partisanship rules .

Through a factorial analysis and mean comparison of sensationalism level, we could see a clear distinction between La Repubblica’s and Il Corriere della Sera’s newspaper's reporting. In

31

sum, it appears that differences at the newspaper level, in a certain extent, influence the extent of sensationalism.

Regarding our second hypothesis, an higher presence of sensationalism in Domestic political news than in European political news, the mean scores of sensationalism supports the expected differences in the extent of sensationalism used, in both newspapers. Newspapers outlets however shows two opposite outcomes for the European coverage. In the foreign section of La Repubblica is clearly visible a continuous increase of the use of sensationalism, which might support Trenz's (2008) thesis of a higher politicization and presence of European news, and , subsequently, to a higher sensationalism.

Sensationalism in the foreign section of Il Corriere della Sera, mentioning European Union, however did not follow the same trend, but most likely as identified by Bomgaardeen et al., (2010) depends primarily on polity-related and exogenous events reporting. In general, European news is treated in a less sensationalistic way, yet sensationalism is still present. Italian journalists emphasize the role of political parties in their coverage of politics (Hanretty, 2007) and this do not seems possible at a European level, and might explain lower sensationalism.

Reporting in the two political sections was found indeed different. Concerning Tabloid Packaging more sensationalism was found in the Content of Domestic political news, and therefore the full content of the articles, while European political news scored higher in Sensationalized Title. This suggests that different ways of reporting and degrees of sensationalism are in place.

32 Discussion and Limitations

Concerning the conclusions of the present study, some critical remarks and limitations have to be drawn. First, a lot of variance in the sensationalism level can be seen between the two newspapers. Most Italian newspapers are partisan press, following party lines. Both Il Corriere della Sera and La Repubblica are not related to any Italian political party, though La Repubblica is known to be affiliated with left-wing ideology.

Although we did not take into consideration the newspaper affiliation in our analysis – as we were mainly interested in measuring sensationalism level – at a glance it might seem that, sensationalism increases due to newspaper’s political orientation and due to changes in the political scenario (e. g. government crisis, election campaign, change of government…) as Figure 3 shows.

33

Though we wanted to test presence of sensationalism over time, isolating political partisanship and political events into variables might have gave us a better chance to define how much market reasons cause the use of sensationalism. This could be tested introducing a right wing newspaper in the analysis, though newspapers indeed are affiliated with parties, or directly belonging to politicians (e.g. Il Giornale and Libero, directly belongs to ex. Prime Minister Berlusconi).

Secondly: we can only account our results as partial. Our sample was drawn only from specific sections of newspapers, and not all the ones involving political news (e.g. first page). An extended study throughout all the politically relevant section might prove different results. The last critical remarks are related to the codebook and measurement instrument of the research. We included variables from different studies, whose sample were not articles, but television news. We therefore attempted to adapt newscast measurements for newspapers articles. In their study, Hendriks Vettehen et al. (2005) suggest that personalization has become one of the main sensationalistic features utilized in newscasts. However, this does not appear to be the case for print media. The reason is mainly structural, as newspapers are not a visual medium. As the core of sensationalism remains its ability to arouse attention and emotions, it can appear, and be studied as well in textual news. Furthermore, the production of newspaper articles is different from the production of online and television news, as it is not immediate, dynamic and it does not follow the same competition setting. We decided not to include the analysis of images , but it might have gave more insight and clearer results for vivid storytelling.

34

As well, regarding the sub-component verbalized emotions some consideration have to be made. To the four variables (Anger, Excitement, Contentment, Fear), a further one was added (Other), in order to cover the possibility that different emotions from the ones previously stated, were mentioned. It might seems controversial as only one variable was supposed to cover all the emotions not included, however, we did not find multiple emotions not quoted in the units of our sample. This problem, however, can be solved by adding an open variable in the new codebook, where the coder would state which emotion/s is/are present.

That said, this study has shown that while sensationalism increased in Italian print media’s political coverage between 2005 and 2013, large differences exist between newspapers and the type of political coverage. As stated above, the Italian media scene is deeply rooted in partisanship. This particular setting tend to make difficult to isolate the power of market forces. As well, this setting presents itself only at a national level. Journalists seem not to apply the same extent of sensationalism when partisanship do not occur or present itself. Furthermore, as according to Eurobarometer data (2013) the media whose are most utilized by Italian citizens are television and online newspapers,. To have a broader vision of sensationalism in Italian media further empirical study for the current issue are required and a cross media extensive analysis at both quantitative and qualitative level might give a clearer representation of sensationalism in the Italian media system.

35

References

Berkowitz, D. (1993). Work roles and news selection in local TV: Examining the business journalism dialectic. Journal of Broadcasting & Electronic Media, 37(1), 67-81

Bernstein, C. (1992). The idiot culture: Reflections of post-Watergate journalism. The New

Republic, 22

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), 506-526.

Brants, K. (1998) Who’s Afraid of Infotainment? European Journal of Communication, Vol. 13, No. 3, 315-355.

Bourdieu, P. (1998). On Television and Journalism, P. Parkhurst, Trans.

Conboy, M. (2006). Tabloid Britain: constructing a community through language. Oxon: Routledge.

De Vreese, C. H. (2001). Europe'in the News A Cross-National Comparative Study of the News Coverage of Key EU Events. European Union Politics, 2(3), 283-307.

De Wilde, P. (2011). No polity for old politics? A framework for analyzing the politicization of European integration. Journal of European Integration, 33(5), 559-575.

36

Esser, F. (1999) Tabloidization of news: A comparative analysis of Anglo-American and German press journalism. European Journal of Communication 14(3): 291–325.

Eurobarometer 54/European Commission (2012) Eurobarometer: Public Opinion in the European Union (Rep. No. 54). Brussels: Directorate-General X.

Grabe, M. E., Zhou, S., & Barnett, B. (2001). Explicating sensationalism in television news: Content and the bells and whistles of form. Journal of Broadcasting & Electronic Media,

45(4), 635-655.

Grabe M. E. , Zhou S., Lang A. and Bolls P.D. (2000) Packaging television news: The effects of tabloid on information processing and evaluative responses. Journal of Broadcasting and

Electronic Media, 44, 581–598.

Graber, D. A. (1994). The infotainment quotient in routine television news: a director’s perspective. Discourse and society, 5, 483-508.

Habermas, J. (2001) Why Does Europe Need a Constitution? New Left Review 11

Hallin, D. C. and Mancini P (2004) Comparing Media Systems: Three Models of Media and

Politics. Cambridge: Cambridge University Press.

Hanretty, C. (2007). The gospel truths of Italian media bias. Comunicazione Politica, 31-48.

Hendriks Vettehen, P.G.J. (2007). Sensationalism. In W. Donsbach (Ed.). The International Encyclopedia of Communication. Oxford / Washington: Blackwell / ICA

37

Hendriks Vettehen, P. H., Nuijten, K., & Peeters, A. (2008). Explaining effects of sensationalism on liking of television news stories the role of emotional arousal. Communication

Research, 35(3), 319-338.

Hendriks Vettehen, P. H., & Nuijten, K. (2006). In Need of an Audience: Sensationalism in Dutch Public Service News and Current Affairs Programs in the 1990’s. In 3rd bi-annual

RIPE conference, Amsterdam, November.

Hendriks Vettehen, P.G.J., Nuijten, C.M. & Peeters, A. (2006). Sensationalism in Dutch current affairs programs 1992-2001. European journal of communication, 21(2), 227-237.

Hendriks Vettehen, P., Nuijten, K., & Beentjes, J. (2005). News in an age of competition: The case of sensationalism in Dutch television news, 1995–2001.Journal of Broadcasting &

Electronic Media, 49(3), 282-295.

Hjarvard, S.(2000) Proximity: The Name of the Ratings Game, Nordicom Review,(2), 63-81.

Hvitfelt, H. (1994). The commercialization of the evening news: Changes in narrative technique in Swedish TV news, Nordicom Review, 15(1), 33–41.

Kurtz, M. L. (1993). Crime of the century: the Kennedy assassination from a historian's perspective. Univ. of Tennessee Press.

Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Thousand Oaks, California: Sage.

38

Lanham, R. A. (2006). The economics of attention: Style and substance in the age of

information. University of Chicago Press.

Lecheler, S., Schuck, A. R., & de Vreese, C. H. (2013). Dealing with feelings: Positive and

negative discrete emotions as mediators of news framing effects.

Mancini, P. (2000). How to combine media commercialization and party affiliation: The Italian experience. Political Communication, 17(4), 319-324

Martin, W. E., & Bridgmon, K. D. (2012). Quantitative and Statistical Research Methods: From Hypothesis to Results. Somerset, NJ: Wiley.

McLachlan, S., and Golding, P. 2000. ‘Tabloidization in the British Press: A quantitative investigation into changes in British Newspapers, 1952 – 1997’. In Sparks, C. and Tulloch, J. (eds.). Tabloid Tales:global debates over media standards. Maryland: Rowman & Littlefield Publishers, Inc. pp 75‐90.

Molek-Kozakowska, K. (2013). Towards a pragma-linguistic framework for the study of sensationalism in news headlines. Discourse & Communication, 7(2), 173-197.

Mott, F. L. (1962). American journalism. New York: Macmillan.

Nisbett, R., & Ross, L. (1980).Human inference: Strategies and shortcomings of social

39

Peter, J., & De Vreese, C. H. (2004). In Search of Europe A Cross-National Comparative Study of the European Union in National Television News. The Harvard International Journal

of Press/Politics, 9(4), 3-24.

Picard, R. G. (2004). Commercialism and Newspaper Quality. Newspaper Research Journal,

25(1), 54-65.

Scott, D. K., & Gobetz, R. H. (1992). Hard news/soft news content of the national broadcast networks, 1972-1 987. Journalism Quarterly, 69(2), 406-441.

Shoemaker P. J., & Reese S. D. (1996) Mediating the Message. New York: Longman.

Slattery, K.L., Doremus, M., & Marcus, L., (2001). Shifts in public affairs reporting on the

network evening news: A move toward the sensational. Journal of Broadcasting &

Electronic Media, 45(2), 290-302.

Slattery KL , & Hakanen E (1994) Sensationalism versus public affairs content of local TV news: Pennsylvania revisited. Journal of Broadcasting and Electronic Media 38(2): 205– 216.

Tannenbaum, P. H., & Lynch, M. D. (1960). Sensationalism: The concept and its measurement.

Journalism Quarterly, 37(2), 381 -392.

Trenz, H. J. (2008). Understanding media impact on European integration: Enhancing or restricting the scope of legitimacy of the EU?. European Integration, 30(2), 291-309.

40

Tuchman, G. (1976). Telling stories. Journal of Communication, 26(4), 93-97.

Uribe R and Gunter B (2007) Are ‘sensational’ news stories more likely to trigger viewers’ emotions than non-sensational news stories? A content analysis of British TV news.

41

Appendix

Appendix I - Codebook Coder

Newspaper La Repubblica (1) / Il Corriere della Sera (0) Section European Politics (1) Domestic Politics (0) Date Size Page Number 1. Arousing Content 1.1 Sensational topicalization

Is the article referring to..?

crime Yes (1) / No (0) violence Yes (1) / No (0) justice Yes (1) / No (0) war Yes (1) / No (0) malpractice Yes (1) / No (0) scandals Yes (1) / No (0) 1.2 Verbalized Emotions

By reading the article it is possible to find words referring to or directing to..

Anger

(“the angry crowd standing outside the parliament…”)

Yes (1) / No (0)

Excitement

(“Firstly trepidation, then joy when the EXPO Commission awarded Italian city of Milan as future centre of…”)

Yes (1) / No (0)

Contentment

(“President expresses its satisfaction with the police forces for…”)

Yes (1) / No (0)

Fear

(“Panic swept between the benches..”) Yes (1) / No (0) Other

(“Commissioners are worried by.., ”)

Yes (1) / No (0) 1.3 Activity Factor By reading the article it is possible to

define the text as ... active rather than passive

( “in the next few days EU will try to quell...”..., “

42 2. Tabloid Packaging

2.1 Content-Style Is the content of the text presenting the news..

in terms of soft news rather than hard news?

Yes (1) / No (0) using metaphors?

(Europe sinks under the burden of...)

Yes (1) / No (0) using personification of institutions?

(Mother Russia,...)

Yes (1) / No (0) using dramatizing verbs?

(collapse,...)

Yes (1) / No (0) using dramatizing adjectives?

(dramatic, shocking...)

Yes (1) / No (0) 2.2 Layout Is the title of the article...

using metaphors? Yes (1) / No (0)

using personification of institutions? Yes (1) / No (0) using dramatizing verbs? Yes (1) / No (0) using dramatizing adjectives? Yes (1) / No (0) 2.3 Tone Is the tone and the reporting of the

topic... scandalized?

(the journalist utilizes language that is evocative of a lack of morality and tends to use exaggerating adjectives to underline horror and shock)

Yes (1) / No (0)

emotional rather than rational? (journalist refers on how citizens should feel in relation to the issue)

Yes (1) / No (0)

3. Vivid Storytelling 3.1 Concreteness

personalization Is the article featuring individuals (civilians/politicians/..) in their environment, giving example on what they are dealing with? (ex. protesters distributes flyers and explains people..., M5S MP stands outside the parliament boycotting the vote... )

Yes (1) / No (0)

laypersons speaking Is the article featuring comments from common people/non experts ?

43

(ex. “We are human people” said X, refugee in the reception center of Lampedusa..)

3.2 Proximity Subject Matter

Subject Matter Is the subject matter portraying a social reproduction of policies on citizens? (“..in practice, the new house tax demands citizen owning a house to pay 450 euros more each year...”)

Yes (1) / No (0)

familiarity Is the narration creating a familiar and closer relation between reporter and readers?

(ex. “we, Italians...”, “our European identity...”, “we might feel..” )

44

Appendix II - Formulas to calculate sub components of Sensationalism. Tabloid Packaging:

TP = Content [(soft news + metaphor + personification + dramatic verb + dramatic

adjective) / 5] + Layout [(metaphor + personification + dramatic verb + dramatic

adjective ) / 4] + Tone [(Scandalized + Emotionalized)/2]

Therefore , Tabloid Packaging takes a value between 0 and 3, with higher values corresponding with higher degrees of sensationalism.

Vivid Storytelling:

VS = Concreteness [ ( Personalization + Layperson speaking) /2] + Proximity [(Subject Matter + Familiarity )/2]

Therefore , Vivid Storytelling, consisting of two sub-variables, takes a value between 0 and 2, with higher values corresponding with higher degrees of sensationalism.