NEWS FRAMES ON TAX FRAUD AND TAX EVASION by
Diego Ottaviano
Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of
Research Master in Communication Sciences Supervisor Dr. P. H. Sheets University of Amsterdam August 2014
Abstract
This study investigates how news frames are used to present stories about tax fraud and evasion in Italian news media. The investigations propose hypotheses that derive from theoretical approaches on framing theory and political parallelism. Specifically, this study offers insights about how news media of different political orientation use news frames before and after the occurrence of a political scandal, which in this study is Berlusconi’s conviction for tax fraud.
A quantitative content analysis is performed using a codebook, which was developed for this study. News stories published five months before and five months after the conviction were analyzed, corresponding to a time range that covered almost the entire duration of the first Italian coalition government. Large attention is given to how partisan news media used negative frames differently than non-partisan
newspapers, and to how populist news media used political and solution frames, differently than non-populist news media. Moreover, attention is given to how news media of different political orientation use news frames to present political actors.
Overall, findings indicate strong differences between how news media use news frames to present stories about tax fraud and evasion. Specifically, news media’s political orientation impacts how news frames are selected and used to
present stories about fraud before and after the occurrence of a key event.
Introduction
People’s interpretation of important issues can be greatly shaped by how information is presented to them. News media outlets construct information using news frames, which enable the formation of views about issues. Specifically, news frames are tools selected and used to emphasize specific elements of an issue and to construct views that are promoted to people by making them more relevant than other (Entman, 1993). In this sense, news frames are selected and used to promote issues that lead people to think in a particular way (Mazzoleni & Schulz, 1999). In Italy, one area in which news frames can have influence is in presentation of information on tax fraud and evasion.
Tax fraud and evasion represent a large problematic across Italy (Chiarini & Marzano, 2013). In 2013 tax fraud and evasion reached 130 billion euros
(“Saccomanni: Evasione favorisce corruzione”, 2013). This impressive amount
represents a threat to both Italian democracy and European stability because it empowers organized crime, corruption, and illegalities (“Saccomanni: Evasione favorisce corruzione”, 2013). As the last national elections pointed out, Italy is
enduring a clear change in the political asset (Carbone & Newell, 2014). Neither the left coalition nor the right reached the quorum to form a government, and they ultimately opted to group into a right-left wing coalition government. In this context the Italian High Court (Corte di Cassazione) convicted the ex-Prime Minister, Silvio Berlusconi, leader of the right wing coalition, and member of the Chamber of Senate, to four years in prison for tax fraud (Carbone & Newell, 2014).
In this context, Berlusconi’s conviction not only represented an historical moment for both national and international environment but it also reflects an element of concern for the stability of the democracy in Italy but also in Europe (Carugati, 2014). Nevertheless, news frames about scandals like fraud are often biased by news
media political orientation (Cepernicha, 2008). Even though, news media should enable the formation of opinions by providing information that helps people
evaluating issues to interpret, this is not always the case (Cepernicha, 2008). Biased news frames represent an additional threat, as they may mislead how information is presented to the public, and how the public thinks about issues, like fraud
(Cepernicha, 2008).
Theoretical Background
Framing theory is often used in the social sciences, political communication, and humanities. However, despite their wide usage, a universal definition of the properties of the theory is lacking (Entman, 1993; McComb, 1997, Strömbäck & Dimitrova, 2005). However, broad definitions have been proposed by some authors, which see framing theory as the process that yields meaning and helps journalists to present information to the audience (Scheufele, 2000). Framing theory concerns two aspects. First, it relates to the selection of elements of perceived realities or issues based on their salience (Entman, 1993; Weaver, 2007). Second, it relates to how issues and perceived realities are presented to the public (De Vreese, 2005). Framing theory enables agenda setting by selecting specific elements, which promote what audiences should think and worry about (Scheufele, 1999; McComb 2004).
Framing theory also helps to individualize casual elements of an issue (McComb 2004; Strömbäck & Dimitrova, 2005). This aspect is important in three directions. First, casual elements help to individualize which aspects create the issue, and establish what they do, and their costs and benefits. Second, they serve to
construct news media’s views of an issue, which is subsequently promoted to the
public. Finally, they facilitate the public’s causal interpretations (Entman, 1993; Strömbäck & Dimitrova, 2005). Therefore, news frames can be used as features that
define problems, diagnose causes, draw up moral judgments, and propose solutions and remedies (Entman, 1993).
Thus, framing theory constructs news media’s views and helps guide what people should think about, while news frames are subtle tools used to construct and promote news media’s views. Gamson & Modigliani (1989) state that news frames
refer to an interpretative package, which illuminates on what is at issue by providing elements that make sense of it and by excluding those that do not. Nevertheless, news frames are also influenced by public opinion, and by the national and cultural context in which reporters work (Strömbäck & Dimitrova, 2005). Thus, news frames can influence the general consensus, but can in turn be influenced by their setting, creating a ‘self-supporting’ environment. Consequently, news media promote
elements that fit cultural and national frameworks (Clausen, 2003). In this study, news frames are defined as the tool that makes news stories appear natural, by determining what is selected, what is excluded and what is prominent, on the basis of cultural, national and political contexts.
Journalists employ news frames in combination with cultural, national and political contexts, which leads to a further aspect of influence: the relationship between media systems and the democratic structure of a country. In their ‘Media System Model’, Hallin & Mancini (2004) suggest that national setting influences
journalistic professional role, journalistic freedom, media system independency, and the normative, which regulates the media system itself. As identified by Hallin & Mancini (2004), because of the national context in which news media outlets promote and provide information, Italy is a prime example of ‘Polarized Pluralistic Model’ better known as ‘Mediterranean Model’.
One phenomenon that is typical of Mediterranean Model is political
parallelism (Hallin & Mancini, 2004). Political parallelism refers to the nature of links between news media and political actors, and to the extent to which news media promote and reflect political differences. Specifically, political parallelism emphasizes the structural and cultural system that directs the communication
employed by political parties (Mazzoleni, 1987). In Italy political parallelism is strong and conservative (Hallin & Mancini, 2004). In political parallelism, a regular
development of press independency is denied, while an already rooted relationship between political parties and the media environment is reinforced (Hallin & Mancini, 2004). In addition, the political environment largely controls the Italian media system. Thus, journalism offers low quality of professionalism, the media market is controlled by the state, and politics impact news selection and production (Hallin & Mancini, 2004).
An example of political parallelism is provided by the system of allotments, which refers to the distribution of media system control by political parties. In the 90’s Italy offered prime cases of allotments, recognized as lottizzazione, which linked
the high degree of State control and State intervention in the public broadcasting Radiotelevisione Italiana (RAI) (Mazzoleni, 1987). RAI proposed three television blocks (RAI Uno, RAI Due, RAI Tre) controlled by governmental parties and opposition, and further positions of power (e.g. RAI CEO) only shared by
governmental parties (Mancini, 2000). Thus, political parallelism reflects national context and news media political orientation (Hallin & Mancini, 2004; Strömbäck & Dimitrova, 2005).
News media political orientation influence political scandals coverage by the American press (Puglisi & Snyder, 2011). Democratic newspapers dedicate
significantly more attention to scandals involving as main character Republican
politicians than members of the Democratic Party (Puglisi & Snyder, 2011). Similarly, Similarly, news media political orientation influence how Italian newspapers use news frames to support political actors (Poletti & Brants, 2010). Right leaning newspapers use more negative frames to present left wing government as the main character of news stories about political scandals, than right wing, and vice versa (Poletti & Brants, 2010). The main character refers to the person or object that receives the highest attention and coverage (Knobloch-Westerwick & Keplinger, 2007).
News media’s political orientation influences how news frames are selected
and used to promote views on political actors (Poletti & Brants, 2010; Scheufele, 2000). For example, news media’s tendency is to support governments that reflect their political orientation. However, during a coalition government news media may be supportive and they may empathize with it even though political actors of opposite orientation are part of it (Kleinnijenhuis & De Ridder, 1998; Strömbäck & Shehata, 2007). Such support is stronger where political parallelism is more effective, and where the political system does not support strong majority governments
(Kleinnijenhuis & De Ridder, 1998; Strömbäck & Dimitrova, 2005; Strömbäck & Shehata, 2007).
However, key events may affect how news frames are used and whether political orientation impacts news frames and how political actors are supported. Key events are stories that receive wider coverage than other news stories of related topic (Brosious & Eps, 1995; Kepplinger & Habermeier, 1995). Higher attention is justified by four main qualities of a key event. First, a key event concerns a particular issue, perceived to be important by the public. Second, a key event has the capacity to
increase the attention of the public and engage it with wider audiences. Third, it impacts people, journalists, politics, and media decision-making changing views on specific issues (Kepplinger & Habermeier, 1995). Fourth, key events relate to situations of exceptional occurrence and low frequency (Kepplinger & Habermeier, 1995).
Political scandals are generally treated by news media as normal events in countries where their occurrence is relatively frequent (Cepernicha, 2008). However, large and uncommon political scandals represent exceptions. Agreeing with Zelizer (1993), a fitting example of a key event as political scandal is the Watergate case. This had an extremely wide coverage, and largely impacted the public and how journalists promoted news stories and gathered information about them (Zelizer, 1993). In Italy, the Bribesville Case (Tangentopoli), a large political scandal of fraud and corruption, had a similar impact. This scandal represents a key event that
influenced media agenda setting, journalist’s reporting tone, public’s thinking, and the overall democratic system (Cepernicha, 2008; Mancini, 2000; Mazzoleni, 2008).
Berlusconi’s conviction can be considered a key event (Carrugati, 2014). First, Berlusconi’s conviction received extremely wider coverage from both national and
international media. Second, it was considered as a highly important event for the stability of national and international political spheres. Third, massive audiences followed the event, which influenced people’s, journalist’s, newsroom’s and political environments’ decision-making (Carrugati, 2014). Therefore, in this study
Berlusconi’s conviction is considered as a key event.
Generally, journalists tend to employ a higher presence of negative news frames to promote scandals (Cepernicha, 2008). However, as argued by Poletti & Brants (2010) news media outlets are less likely to use negative frames, when a
Government reflecting their political orientation is involved in the scandal. Similar views are advanced by Romenti & Valentini (2010) who analyzed news frames about the 2008 Alitalia Crisis Case, which occurred during right wing government. Overall they demonstrated that left leaning newspapers were more likely to use negative news frames, to promote an exaggerated negative view of strategic-solutions proposed by the right wing government (Romenti & Valentini, 2010).
An additional element that may influence how news frames are used is represented by political clientelism. Political clientelism defined as the action of supporting a specific political actor over another for a remunerative return (Hallin & Papathanassopoulos, 2002). Such return can be represented by public subsidies. In this sense, political clientelism becomes a threat for a democracy, as news media may use news frames that promote information that is influenced by the political
environment (Hallin & Papathanassopoulos, 2002). Because of this remunerative relationship, news media do not serve the maintenance of the democratic system and lack independence (Hallin & Papathanassopoulos, 2002). In Italy, such independence is owned and expressed principally by the populist paper Il Fatto Quotidiano, the first that does not receive public subsidies, and that operates as watchdog of the
democracy (Cornia, 2013).
Populist news media cover issues of higher interest for the public (Blumler & Kavanagh, 1999). Moreover, the ways they use news frames, especially in relation to the language, is different from how other newspapers use them (Mazzoleni et al., 2003). For example, populist news media often default to news frames that promote casual language and which reflects the public rather than the elites (Blumler & Kavanagh, 1999). Indeed, their language is generally drawn out against power and political elites and in favor of democracy and political renewal (Statham, 1996). In
this sense, when news media employ populist language they recall a good portion of resonant features of popular culture addressed to a wider audience rather than to opinion leaders or any sort of niche (Statham, 1996).
Populist news media are more common in countries where political parallelism and state control over the media system are stronger (Mazzoleni et al., 2003). In this context populist news media seize public responsibilities by using news frames that promote information on specified issues, and by implementing their coverage to address the public to think more about them (Mazzoleni et al., 2003; McCombs 2004, 2005). As argued by Kowalewski (2012), news media tend to increment coverage on specific issues so people are more likely to mention them as more important than those not discussed. Consequently, populist news media spotlight issues like fraud so citizens may cite them as more important than others (Mazzoleni et al., 2003).
In this sense, populist news media tend to also emphasize news frames similar to newsroom or journalist’s comments, to engage more with people and to influence their thinking (Mazzoleni, 2008). In addition, populist news media use political frames to promote particular stories as politically relavant (Mazzoleni, 2008).
Therefore, populist news media promote information about issues such as corruption and fraud, with news frames that make them more noticeable in the public arena. This action may increase, when the debate among the issue is constructed within a political angle and evaluated within an anti-political class sentiment (Mazzoleni, 2008). Within these actions populist news media tend to employ news frames to promote opinions, ideas, and views that support political renewal and solutions to issues, especially after the occurrance of larger scandals (Lundström, 2013; Statham, 1996).
In these views a deeper investigation of news frames on tax fraud and evasion becomes a priority. Specifically, it is important to point out how news frames are selected and used by news media outlets to influence public’s thinking, interpretation, and evaluation of issues (De Vreese, 2005; Mazzoleni & Schulz, 1999). Thus, this study aims to highlight how news frames on tax fraud and evasion are used by news media with different political orientation to promote news stories on tax fraud and evasion. The following research question is formulated:
RQ: How have Italian online newspapers framed and covered news about tax fraud and evasion before and after Berlusconi’s conviction for tax fraud, of the 1st
August 2013?
Finally, it is expected that before and after Berlusconi’s conviction, left and right leaning newspapers, in this study partisan news media outlets, are less likely to promote coalition government using main character news frames in stories about tax fraud and evasion than non-partisan news media outlets. The following hypotheses are formulated,
H1: Partisan news media outlets present news stories about tax fraud and evasion in which coalition government is the main character, less frequently than non-partisan news media outlets.
H1A: After a key event, partisan news media present news about tax fraud and evasion in which coalition government is the main character, less frequently than non-partisan news media outlets.
Furthermore, it is expected that, partisan news media outlets tendency is to use less negative news frames, than non-partisan news media outlets to promote news stories about tax fraud and evasion in which coalition government is involved. The following hypothesis is formulated,
H2: After a key event, partisan news media outlets only use negative frames in news stories of tax fraud and evasion in which Coalition Government is involved, less frequently than non-partisan news media outlets.
In addition, it is expected that populist news media outlets tendency is to use political frames to promote news stories on tax fraud and evasion, more than non-populist news media. The following hypothesis is formulated,
H3: Populist news media outlets use political news frames to promote stories on tax fraud and evasion, more frequently than non-populist news media outlets.
Finally, it is expected that, populist news media outlets will use a higher amount of commentary frames, and solution frames than non-populist news media outlets after the occurring of a political scandal, to promote political renewal and views that can solve the issue. The following hypotheses are formulated,
H4: After a key event, populist news media outlets use commentary news frames like author’s view to promote stories on tax fraud and evasion, more frequently than non-populist news media outlets.
H5: After a key event, populist news media outlets use solution news frames to promote stories on tax fraud and evasion, more frequently than non-populist news media outlets.
Methods Procedure
This study employed a content analysis of news stories on tax fraud and evasion from four Italian online news media outlets. Furthermore, we analyzed news frames use to promote stories on tax fraud and evasion before and after Berlusconi’s conviction. A codebook was developed and coders were employed and trained at the University of Amsterdam. Prior to the study, two training sections were performed for a total of six hours. The first training section was used to explain the general purpose of the codebook, how to use it, and how to locate codes within the units of analysis. The second section was used to practice. Finally, we performed a Pearson chi-square test, and two-way Anova on the data collected.
Sample
A multistage sampling was performed. Initially, sampling concerned the individualization of newspapers in base of their political angle and online circulation. Four newspapers were then selected with the aim of representing a range of political opinions present in the Italian Parliament. Therefore, the selection was based on newspapers that reflect the largest political currents within the Italian Parliament: populism, left wing electorate, right wing electorate, and center/neutral. Based on these criteria, Il Fatto Quotidiano, La Repubblica, Il Giornale, and Il Corriere della
Sera, were selected.
Il Fatto Quotidiano is the first independent newspaper in Italy and does not
independency (Cornia, 2013). For instance, a single shareholder cannot own more than 16 percent of the capital share and journalists can only hold 30 percent. Il Fatto
Quotidiano shows populist connotations acting as watchdog, using public language,
promoting political renewal, and providing political views firmly against Berlusconi and corrupt politicians (Demir, 2012).
La Repubblica is representative of a center-left leaning newspaper, supporting the ex political party Ulivo, which is a stable component of the larger Partito
Democratico (PD) (Triandafyllidou, 2003). Of opposite position, Il Giornale is a right leaning paper representing the angle of Forza Italia Party of which Berlusconi is leader (Demir, 2012; Semino, 2012; Triandafyllidou, 2003). Finally, Il Corriere della
Sera gives support to the industrial environment of the country and maintains a
neutral political position (Poletti & Brants, 2010; Semino, 2012).
The above newspapers were also selected because of their online circulation. According to the Italian Federation of Editors and Journalists (FIEG, 2013) during the two-year period 2010-2012, La Repubblica website had the highest number of views as N= 9,368,000, followed by Il Corriere della Sera webpage, N= 8,300,000.
Different situation concerned the online pages of Il Fatto Quotidiano N= 903,000, and
Il Giornale, N= 871,000 (FIEG, 2013), lower than other six online papers.
Nevertheless, since these six online newspapers did not offer opportunity of political orientation diversity (Cornia, 2013), Il Fatto Quotidiano and Il Giornale were considered.
The period studied ranged from March 1, 2013 to January 31, 2014. After selecting the newspapers a ‘time period’ of interest was defined as five months before Berlusconi’s conviction until five months after. The time range of ten months was
first Italian Coalition Government (Governo Letta – 28th April, 2013 – 14th February, 2014; established after the election of the 25th of February, 2013). Second the study aimed to span a time range that provided an extensive coverage of news stories on tax fraud and evasion published before and after the sentence.
Finally, the units of analysis were news articles. Nexis/Lexis service was used to collect data from Il Corriere della Sera, while news items from La Repubblica, Il
Fatto Quotidiano, and Il Giornale were collected using the online archive and search
engine offered by their online pages. News stories containing at least one keyword, either ‘frode fiscale’ (tax fraud) or ‘evasione fiscale’ (tax evasion) were selected. An initial sample size of N= 2,485 items was collected and then filtered to news articles which were strictly related to tax fraud and/or tax evasion.
To filter each news article two approaches were adopted. First approach, headlines and subtitles were used as filter. Specifically, news items in which the headline or subtitle included one or both keywords and a clear reference to the issue through their wording and content were selected. For example, titles like ‘Come giustifico l’evasione fiscale’ (How I justify tax evasion), or ‘Pressione fiscale al 53% contrasto all’evasione’ (Fiscal pressure at 53%, contrast to the evasion) were sampled.
Alternatively, where headlines and subtitles were not clearly referring to the issue, news articles text were partially read. In this case, attention was given to the main theme of the item. Therefore, if the content referred to the issue as central topic, the item was sampled otherwise it was discarded. A final sample size of N = 588 news articles was obtained. A total number of n = 274 news stories were collected from La
Repubblica, n = 119 from Il Corriere della Sera, n = 104 from Il Fatto Quotidiano,
and n = 91 from Il Giornale. Codebook
The codebook of this study was developed consulting, Eurosphere Media
Codebook I & II (Cushion et al., 2008; Sicakkan & Tønnevold, 2008), and
Neuendorf’s guidebook (2008). In addition, adjustments were made conducting a
pilot study prior to the final coding. Corrections were generally improved on the base of feedback received during the training sessions and disagreement percentage
obtained through the reliability check. Overall, changes were generally about difficult and unclear wording.
Three main sections, corresponding to identification, description, and framing, composed the codebook. The identification section was composed of six variables, which concerned information about coders, news article identification, source name, news title, news subtitle, and news date. The aim of the identification section was to collect data that could be used to label the basic characteristics of each news article. For example, to identify each article in base of their source four codes were provided as following: 1 = Corriere della Sera; 2 = La Repubblica; 3 = Il Fatto Quotidiano; 4
= Il Giornale.
The descriptive section was composed of seven items, used to depict the contextual theme of each news article. Coders were asked to report on the main theme of the article and which actors were mentioned in the story. Actors were identified using a list of 29 categories. Categories included national and international
institutional authorities and elites. Between categories, emphasis was placed on the political environment of the Italian country. Coders were asked to locate whether or not Coalition Government, politicians, and parties were promoted as main character of each article. Main parties (Democratic Party, Party of Freedom, SEL, M5S, etc.), leaders of parties and long serving politicians (Berlusconi, E. Letta, Grillo, Renzi,
etc.) of the Italian Parliament were selected to provide an exhaustive coverage of the political spectrum of the country.
The framing section was provided to identify which frames are used to promote stories on tax fraud and tax evasion. Using Entman’s (1993) theory on framing the section was composed of fourteen variables, which helped coders recognize specific six news frames: main character frames, negative frames, main theme frames, commentary frames and solution frames.
Finally, the codebook was composed of a total amount of twenty-eight variables. The codes, ‘coders ID’, ‘article ID’, and ‘date’, were considered as
categorical variables. In addition, coding only pertained to the text body of news items, as reading through the whole news story was required. Photo, video, graph, or additional features linked to each unit of analysis were not pertinent to the coding and analysis of this study.
News Frames
Main character frames were coded to point out which actor was given more
attention in each news story. In doing so, the item ‘power elite category’ of nominal measurement level was used. The item listed a set of twenty-nine characters, each corresponding to a different actor. Coders were asked to locate which of the listed actors received the highest attention in the news story. Coders were first told to highlight with different colors which of the actor listed were nominated in the story, and then to manually count which of the highlighted actors were the most cited. The most cited actor corresponded to the main character of each news story analyzed.
Negative frames were coded to investigate news story overtone, with the item
‘Positive vs. Negative’, of ordinal measurement level, which was reported on a
frame codes were ‘only positive’, ‘more positive than negative’, ‘more negative than positive’, and ‘only negative’. The first frame code, ‘Only positive’ was selected
when a news story only promoted positive views, for example about plans that could resolve tax fraud and evasion. ‘Only negative’ frame code was selected when a news story only promoted negative views about tax fraud and evasion. While the first two codes were quite straightforward, the last two were not, as coders had to rely the number of negative & positive connotations/wording, like negations, negative adjectives, and words that expressed negative connotations like ‘arrested’, ‘jailed’, ‘convicted’ as so on.
Theme frames represented the main topic of each news story. Specifically, the
main topic is intended as the focus of the news story to which was given more attention and coverage. To locate theme frames coders were asked to search for the most covered topic of the story. Themes frames were coded through the item ‘Main Topic’ of nominal measurement level, which listed seven theme frame codes. Each theme frame code was accompanied by a description to explain how to locate them within a text. In this study, particular attention was given to political frames, which are a type of theme frames. Political frames were located searching for elements, like political actors promoting political program against tax fraud and evasion, political debates about tax fraud and evasion, or political speeches about tax fraud and evasion. If a news story promoted more than one theme frame, coders were asked to code the first in order of appearance, starting from the title, subtitle, and text.
Commentary frames were coded within the item ‘Comment’, a nominal
variable of ‘yes/no’ values. To locate commentary frames, coders were asked to search for expressions, phrasing and wording that reflected journalists and or newsroom’ views. For example, journalist/newsroom’ opinions, evaluations,
interpretation, or phrasing like ‘I/we think tax fraud is…’ or ‘tax evasion is a made in Italy story’, were recognized as commentary frames. The aim of commentary frames
was to point out whether news media use commentary frames to influence the public and to engage more with it.
Solution frames were located within the item ‘Solution’, which present five
nominal solution frame codes. Solution frames were located by searching elements in the news story that support plan, strategies, opinions or ideas, that promote political renewal or solution to the issue of tax fraud evasion. Each solution frame was
accompanied by a brief description. When multiple solution frames were located in a news story, coders were asked to code the solution frame that received the highest coverage in the story. The highest coverage was checked by counting how many times a solution frame was mentioned. Solution frames were coded to investigate if news media outlets were likely to promote and support a potential solution for tax fraud and evasion. These were ‘higher control by the authorities’, ‘lower taxation’, ‘increment of penalties’, and ‘other solutions’.
Inter-coder Reliability
For testing the inter-coder reliability of the codebook, a pilot study was conducted using two native Italian-speaking coders. Finally, inter-coder reliability check was conducted on 11% (n= 66) of the sampled news items. Reliability
Calculator - ReCal (Freelon, 2011) was used to check the inter-coder reliability of
each variable. Variables were separately checked using the data provided by both coders.
Krippendorff's Alpha was used to analyze the variable ‘date’, while Scott’s Pi coefficient was used to test the reliability of all the other variables. The variable ‘date’, of interval measurement level, reached a minimum Krippendorff's Alpha of
.80, while the other variables, all of nominal measurement level, reached a minimum Scott’s Pi of .70 or higher. All the variables showed strong and acceptable inter-coder reliability. From the codebook three variables were left out from the reliability check as they corresponded to ‘coders ID’, ‘news Article ID’, and ‘Title’. A final of 25 variables were checked for their reliability (see table 1.0 in Appendix A)
Results
H1 tested whether partisan news media outlets presented coalition government as main character, less frequently than other news media outlets. Partisan news media outlets promoted coalition government as main character, 41.1% (n = 30), while non-partisan news media outlets did 58.9% (n = 43). Person chi-square test indicated significant differences between partisan news media outlets and non-partisan news media outlets, 2
(1, 581) = 15.99, p < .001. Therefore, partisan news media outlets
promoted coalition government as main character significantly less frequently than non-partisan news media outlets, supporting the hypothesis.
H1A tested whether there was a difference/change in the frequency with which partisan news media outlets and non-partisan news media outlets promote coalition government as main character, before and after a related key event occurs. Overall, before Berlusconi’s conviction, partisan news media outlets promoted coalition government as main character, 50%, (n = 16) as well as non-partisan news media outlets, 50%, (n = 16). Pearson chi-square indicated no significant differences,
2(1, 316) = 2.57, p > .05. However, after the conviction, results showed that
partisan newspaper decreased their coverage with coalition government as main character, 34.1% (n = 14), while non-partisan news media outlets increased its coverage as 65.9% (n = 27). Pearson chi-square indicated significant differences between partisan news media outlets and non-partisan news media outlets, 2
(1, 265) = 15.34, p < .001. Therefore, supporting the hypothesis.
Note: Figure 1.0 Errors bars with percentages at 5%. The graph shows frequency differences between
partisan newspapers and other newspapers, in how they presented coalition government as main character of stories of tax fraud and evasion, before and after Berlusconi’s conviction.
H2 tested whether, partisan news media outlets use negative frames to
promote news story in which coalition government was involved, less frequently than non-partisan news media outlets, before and after a related key event occurs. A two-way Anova was performed. Findings showed that partisan news media outlets used a lower amount of negative frames (M = 2.7, SD = .72) than non-partisan news media outlets (M = 3.0, SD = .81). In addition, there was a significant main effect of coalition government, F(1,577) = 10.80, MSE = 5.22, p <.001, as it was of partisan news media outlets F(1,577) = 6.11, MSE = 2.95, p <.05. However, there was not a significant interaction between, partisan news media outlets and coalition government,
F(1,577) = .02, MSE = 0.11, p >.05, indicating no significant differences between
partisan and non-partisan news media outlets, thus not supporting the hypothesis.
0% 10% 20% 30% 40% 50% 60% 70% 80% Before After Fr equ ency
Timeline before and after Berlusconi's conviction
Figure 1.0. Coalition government before and after Berlusconi’s conviction
Partisan Newspapers Other Newspapers
H3 tested whether populist news media outlets use political frames to promote news story with a political angle, more frequently than non-populist news media outlets. Populist news media outlets used political frames 10.6% (n = 11) of news stories with political frames and 89.4%, (n = 93) without political frames. Similarly, non-populist news media outlets used political frames 6.4% (n = 31) of news stories with political frames and 93.6%, (n = 453) without political frames. Pearson chi-square indicated no significant differences between populist news media outlets and non-populist news media outlets, 2
(1, 588) = 2.24,
p > .05. Thus, not supportingthe hypothesis.
H4 tested whether populist news media outlets use more frequently
commentary frames, than non-populist news media outlets, before and after a related key event occurs. Before Berlusconi’s conviction, populist news media outlets promoted 31.1% (n = 19) of news stories with commentary frames and 68.9%, (n = 42) without commentary frames. In contrast, non-populist news media outlets promoted 12.7 % (n =33) of news stories with commentary frames and 87.3% (n = 226) without. Pearson chi-square indicated significant differences between partisan news media outlet and non-populist news media outlets, 2
(1, 320) = 12.29, p <
.001. After the conviction, populist news media outlets promoted 41.9% (n = 18) of news stories with commentary frames and 58.1%, (n = 25) without commentary frames. Meanwhile, non-populist news media outlets promoted 23.1 % (n =52) of news stories with commentary frames and 76.9% (n = 173) without. Pearson chi-square indicated significant differences between partisan news media outlets and non-populist news media outlets, 2
(1, 268) = 6.57, p < .05. Thus, supporting the
Note: Figure 2.0 Errors bars with percentages at 5%; The graph shows frequency differences between
how populist newspapers and other newspapers used commentary frames, before and after Berlusconi’s conviction. In blue news stories that included commentary frames. In red those that do not include commentary frames. The first four columns from the left refer to news stories before the conviction. The other four columns refer to news stories after the conviction.
Finally, H5 tested whether populist news media outlets use solution frames, more frequently than non-populist news media outlets, before and after a related key event. Before Berlusconi’s conviction, populist news media outlets promoted 24.6% (n = 15) of news stories with solution frames and 75.4%, (n = 46) without solution frames. Similarly, non-populist news media outlets promoted 23.9% (n =62) of news stories with commentary frames and 76.1% (n = 197) without solution frames. Pearson chi-square indicated no significant differences between how partisan news media outlets and non-populist news media outlets use solution frames, 2
(1, 320) =
.01, p > .05. However, after the conviction, populist news media outlets promoted
37.2% (n = 16) of news stories with solution frames and 62.8%, (n = 27) without commentary frames. In contrast, non-populist news media outlets promoted 18.2 % (n
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Populist Other Populist Other
Fequ
ency
Before Conviction - After Conviction Figure 2.0. Commentary frames before and after Berlusconi's conviction
Commentary Without
=41) of news stories with solution frames and 81.8% (n = 184) without. Pearson
chi-square indicated significant differences between populist news media outlets and non-populist news media outlets, 2
(1, 268) = 7.77, p < .05. Thus, after the conviction
populist news media outlets increased the amount of solution frames while non-populist news media outlets decreased it, supporting the hypothesis.
Note: Figure 3.0 Errors bars with percentages at 5%; The graph shows frequency differences between
how populist newspapers and other newspapers used solution frames, before and after Berlusconi’s conviction. In blue news stories that included solution frames. In red those that do not include solution frames. The first four columns from the left refer to news stories before the conviction. The other four columns refer to news stories after the conviction.
Discussion
The purpose of this study was to investigate how online newspapers employ news frames to promote news stories about tax fraud and evasion. Specifically, the study addressed two main aspects, first, how news media outlets of different political orientation employed news frames. Second, to highlight whether the use of news
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Populist Other Populist Other
Fr
equ
ency
Before Conviction - After Conviction Figure 3.0. Solution frames before and after conviction.
Solution Without
frames change after the occurrence of a key event, (in this study Berlusconi’s conviction for tax fraud).
Our findings confirmed that news media outlets of different political orientation use news frames differently. Partisan news media covered coalition government with main character frames less than populist and neutral news media. Considering the bipartisan composition of the Italian coalition government, it is likely that partisan newspapers avoid covering news story that can threaten the stability of coalition government, and consequently the relationship between politics and media. Prior research suggests that partisan news media present news on scandals, like fraud, involving a coalition government as the main character, less frequently than non-partisan news media outlets. Our findings confirmed this trend. Specifically, non-partisan newspapers tend to support the coalition government when there was not a strong majority in power (Kleinnijenhuis & De Ridder, 1998; Strömbäck & Shehata, 2007).
In addition, our findings suggest that political parallelism was present in Italian news media, indicating strong influence by the political system. Reduced promotion may also be the result of political clientelism, as partisan newspapers may be supportive of coalition government even when a scandal occurred, to gain some sort of remunerative return, like public subsidies. This further implies that news media’s political orientation may influence how news stories are covered and
promoted to the public (Hallin & Mancini, 2004; Strömbäck & Dimitrova, 2005). Finally, findings may also suggest that Italian journalism and media system are underdeveloped in their role as government watchdog, their professional quality, in information flow, and media market development.
In addition, negative frames were investigated. Specifically, we predicted that partisan news media use negative frames less frequently than other news media
outlets to promote news stories in which coalition Government is involved. Previous studies suggests that partisan news media outlets are less likely to use negative frames than other news media outlets when they promote political scandals involving a supported political actor (Poletti & Brants, 2010; Puglisi & Snyder, 2011). However, our findings only partially support the assumption. On average, partisan newspapers used negative frames less frequently than non-partisan news media. Therefore, no relationship was found between news media political orientation and how negative frames were used specifically in stories in which coalition government is involved.
After Berlusconi’s conviction, two political events may have switched the attention of news media outlets and how negative frames were used when describing stories of tax fraud and evasion and coalition government. The main components of the coalition Government changed its leader. Matteo Renzi became leader of Partito Democratico and then Italian Prime minister of the second coalition government. Second, due to his conviction, Berlusconi was expelled from the Senate Chamber, and the coalition government experienced a period of great pressure. Therefore, these events may have caused a change in the agenda setting and in how negative frames were used to promote coalition government.
Finally, we predicted that populist news media would use more frequently political frames, commentary frames and solution frames than other news media. Populist news media used solution and commentary frames more frequently than non-populist news media. However, findings did not show differences in frequency of political frames used by populist, neutral, and partisan news media outlets. Mazzoleni et al. (2003) suggest that populist news media act as watchdog by using comments and language of the public, political frames that promote political renewal, and
solution frames that promote answers to solve issues. Our study partially supported this trend.
This suggests that populist news media tend to engage public responsibilities, especially when political and media system are closely associated with political parallelism and political clientelism, supporting the idea that populist news media use more solution and commentary frames to promote information about issues and to implement their salience (Kowalewski, 2012; Mazzoleni et al., 2003). In addition, through such frames, populist news media direct the public to think about
opportunities of renewal and to solve problems related to tax fraud and evasion. Overall, we found strong differences between how partisan, populist and neutral news media outlets use news frames to promote information on tax fraud and evasion, overall as well as before and after major political scandal. Therefore, this study is particularly important for three main reasons. It broadens the knowledge on framing and political parallelism theory. Specifically, it demonstrates that news media political orientation, as well as the political and media systems, impact how news frames are selected and used. Moreover, it points out the tendency of news media to use specific news frames to promote political scandals, on the basis of their political orientation. This underlines a general lack of professional journalism and press freedom, which may threaten the stability of a democracy and development of a free media market.
Finally, this study points out important deficiencies in how news on tax fraud and evasion are promoted to the public especially after a major political scandal. Tax fraud and evasion can facilitate corruption and illegalities, a clear threat to the stability of the Italian democracy and in Europe (Carugati, 2014; “Saccomanni: Evasione favorisce corruzione”, 2013). Nevertheless, only populist news media fully
keep their role of watchdog for democracy. They do this by illustrating and offering solutions to important issues such as tax fraud and evasion. Conversely, partisan news media do not play a role of watchdog of democracy. Rather, their tend to limit the coverage of news stories on fraud that involve political actors of their ideology. This action highlights negligence, dependency on the political system, and both a low degree of professional journalism and press freedom. Finally, when political parallelism is particularly strong, populist news media promote political renewal, while neutral and partisan news media outlets implicitly or explicitly promote the
status quo of the media and political system.
Results of this study were bound by some technical constraints. Despite, the large sample size used, future study should include additional publications and news articles. For instance, we would suggest comparing at least two news media outlets per political orientation. Moreover future studies would benefit greatly by first identifying the connection between news media outlets political orientation and coalition government and its role in creating the Italian Mediterranean Model identified by Hallin & Mancini (2004). Future studies should also focus specifically on the theory of political clientelism to determine how different news media outlets propose political frames to promote news event about politicians. Specifically, they should compare how news media outlets sustained by governmental subsidies
promote news stories of political scandals relative to media outlets that do not depend on governmental subsidies.
Tax fraud and evasion are major issues in Italy. However, this study demonstrated that news media outlets present information using news frames that support their political orientation rather than information to the public. In doing so, news media weaken the development of the democratic system, showing strong marks
of political parallelism. Moreover, solutions to the issues are only provided by a small part of the media market, as tax fraud and evasion keep growing and threating the stability of the Italian democracy and Europe. More research needs to be conducted, this investigation demonstrated that this topic is rare with opportunity and further research can greatly benefit the overall understanding of the relationship between media and political environments.
Reference
Biorcio, R. (2014). The reasons for the success and transformations of the 5 Star Movement. Contemporary Italian Politics, (ahead-of-print), 1-17. Biorcio, R., & Natale, P. (2013). Politica a 5 stelle: idee, storia e strategie del
movimento di Grillo. Milano: Feltrinelli Editore.
Blumer, J. G., & Kavanagh, D. (1999). The third age of political communication: Influences and features. Political Communication, 16, 209-230.
Brants, K., & van Praag, P. (2006). Signs of media logic. Half a century of political communication in the Netherlands. Javnost-the Public, 13(1), 25-40.
Brosius, H. B., & Eps, P. (1995). Prototyping through key events News selection in the case of violence against aliens and asylum seekers in Germany. European Journal of Communication, 10(3), 391-412.
Cappella, J. N., & Jamieson, K. H. (1997). Framing the news. In J. N. Cappella & K. H. Jamieson (Eds.), Spiral of cynicism. The press and the public good. New York: Oxford University Press.
Carbone, M., & Newell, J. L. (2014). Leading from outside Parliament. Contemporary Italian Politics, 6(1), 1-3.
Carugati, F. (2014). What is the rule of law good for? Democracy, development, and the rule of law in classical Athens. Buffalo Law Review, 42 (1), 119-156.
Cepernicha, C. (2008). Landscapes of immorality: Scandals in the Italian press (1998–2006). Perspectives on European Politics and Society, 9(1), 95-109. Chiarini, B., & Marzano, E. (2013). Evasione fiscale e sommerso economico in Italia:
fatti stilizzati, differenze tra periodi e puzzles. Moneta e Credito, 60(239).
Clausen, L. (2003). Global News Communication Strategies – 9.11.2002 around the World. Nordicom Review, 24(2), 105–16.
Cohen A. (2008). Codebook for content analysis. Foreign TV news study. Appendix B.
Country list. Retrieved from http://www.polcomm.org/wp-content/uploads/Codebook-Appendix-B-Country-List.pdf
Cornia, A. (2013). Will Italian political journalism ever change? In R. Kuhn, & R. K. Nielsen (Eds.), Political journalism in transition. Western Europe in a
comparative perspective (47-72). London: I.B. Tauris.
Cushion, S., Garcia-Blanco, I., Guerra, S., & Wahl-Jorgensen, K. (2008). Media codebook I and coding sheet I for print and broadcast media. Variables V1-V15. Eurosphere Research Notes, 27, 1-19
Demir, M. (2012). The role played by Italy in media history: Its historical problems and its recents innovations in journalism and television world. Cross-Cultural
Communication, 8(3), 8-16.
De Rosa, P. (2012). Tasse e Notizie: Il Caso Agenzia Delle Entrate: L’evoluzione Dell’amministrazione Finanziaria Tra Arene Pubbliche e Nuove Forme di Relazionalità (Taxes and News: A Case History). Rivista di Studi Politici-San Pio V, 2, 165-192.
De Vreese, C. H. (2005). News framing: Theory and typology. Information Design Journal & Document Design, 13(1).
Entman, R. M. (1989). Democracy without citizens: Media and the decay of American politics. Oxford: Oxford University Press.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm.
Journal of Communication, 43(4), 51-58.
Entman, R. M. (2003). Cascading activation: Contesting the White House frame’s after 9/11. Political Communication, 20, 415-432.
Freelon, D. (2011). ReCal2: Reliability for 2 coders. Retrieved from http://dfreelon.org/utils/recalfront/recal2/
Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95,
1–37.
Hallin, D. C., & Mancini, P. (2004). Comparing Media Systems: Three Models of
Media and Politics. New York: Cambridge University Press.
Hallin, D. C., & Papathanassopoulos, S. (2002). Political clientelism and the media: Southern Europe and Latin America in comparative perspective. Media, culture & society, 24(2), 175-195.
Kepplinger, H. M., & Habermeier, J. (1995). The impact of key events on the
presentation of reality. European Journal of Communication, 10(3), 371-390.
Kleinnijenhuis, J., & De Ridder, J. (1998). Issue news and electoral
volatility. European Journal of Political Research, 33(3), 413-437.
Knobloch-Westerwick, S., & Keplinger, C. (2007). Thrilling news: Factors generating suspense during news exposure. Media Psychology, 9(1), 193-210.
Kowalewski, J. (2012) Does Humor Matter? An Analysis of How Hard News versus Comedy News Impact the Agenda-Setting Effects. The Southwestern Mass Communication Journal, 28(1), 1-29
Mancini, P. (2000). How to combine media commercialization and party affiliation: The Italian experience. Political Communication, 17, 319-324.
Mazzoleni, G. (1987). Media logic and party logic in campaign coverage: The Italian general election of 1983. European Journal of Communication, 2, 81-103.
Mazzoleni, G. (2008). Populism and the Media. Twenty-First Century Populism. The Spectre of Western European Democracy, 49-64.
Mazzoleni, G., & Schulz, W. (1999). " Mediatization" of politics: A challenge for democracy? Political Communication, 16(3), 247-261.
Mazzoleni, P., Stewart, J., & Horsfield, B. (2003). Approaches to Media Roles. Informational role and ‘Media Populism. In P. Mazzoleni, J. Stewart, & B.
Horsfield (Eds.), The media and neo-populism: A contemporary comparative
analysis, (7-15) Westport: Prager Publisher.
McCombs, M., Llamas, J. P., Lopez-Escobar, E., & Rey, F. (1997). Candidate images in Spanish elections: Second-level agenda-setting effects. Journalism & Mass Communication Quarterly, 74(4), 703-717.
McCombs, M. (2004). Setting the agenda: The mass media and public opinion. Cambridge, England: Polity Press.
McCombs, M. (2005). A look at agenda-setting: Past, present and future. Journalism
Studies, 6(4), 543-557.
McManus, J. (1995). A market-based model of news production. Communication Therory, 5(4), 301-338.
Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks, CA: Sage Pubblication.
Poletti, M., & Brants, K. (2010). Between partisanship and cynicism: Italian journalism in a state of flux. Journalism, 11(3), 329-346.
Puglisi, R. (2006). Being the New York Times: The political behavior of a newspaper. (Working Paper No. D72, D78, L82). Social Sciences Research Network. Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=573801 Puglisi, R., & Snyder, J. M. (2011). Newspaper coverage of political scandals. The
Journal of Politics, 73(3), 931-950
Romenti, S., & Valentini, C. (2010). Alitalia’s crisis in the media – a situational
analysis. Corporate Communications: An International Journal, 15(4), 380-396.
Saccomanni: Evasione favorisce corruzione e crimininalità. Befera: miliardi sottratti al fisco, incompatibile con democrazia. (2013, December 10). Il Sole 24 Ore. Retrieved from http://www.ilsole24ore.com
Santoro, M. (2014, May 10). Interview by P Gomez [Personal Interview].
Informazione, web, politica. parla santoro. 1- announo, servizio pubblico e il fatto. Il Fatto Quotidiano. Retrieved from
http://www.ilfattoquotidiano.it/2014/05/10/informazione-web-politica-parla-santoro-1-announo-servizio-pubblico-e-il-fatto/980362/
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of communication, 49(1), 103-122.
Scheufele, D. A. (2000). Agenda-Setting, priming, and framing revisited: Another look at cognitive effects of political communication. Mass Communication &
Society, 3(2&3), 297-316
Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of
Semino, E. (2002). A sturdy baby or a derailing train? Metaphorical representations of the euro in British and Italian newspapers. Text, 22(1), 107-139.
Sicakkan, H., G., & Tønnevold, C. M. (2008). Media codebook II and coding sheet II for print and broadcast media. Variables V16-V22. Eurosphere Research
Notes, 28, 1-8
Statham, P. (1996). Berlusconi, the media, and the new right in Italy. The Harvard International Journal of Press/Politics, 1(1), 87-105.
Strömbäck, J. (2005). In search of a standard: four models of democracy and their normative implications for journalism. Journalism Studies, 6(3), 331-345. Strömbäck, J., & Dimitrova, D. V. (2005). Mission accomplished? Framing of the
Iraq War in the elite newspapers in Sweden and United States. Gazette: The
International Journal of Communications Studies, 67(5), 339-417. Strömbäck, J., & Shehata, A. (2007). Structural biases in British and Swedish
election news coverage. Journalism Studies, 8(5), 798-812.
Triandafyllidou, A. (2003). Research note: The launch of the Euro in the Italian media. Representations of political and economic integration. European
Journal of Communication, 18(2), 255-263.
Valeriani, A. (2013). Micro media event. Costruttori, promotori e narratori per il palcoscenico di Twitter. Comunicazione politica, 13(2), 199-216.
Weaver, D. H. (2007) Thoughts on agenda setting, framing, and priming. Journal of
Communication, 57, 142-147.
Zelizer, B. (1993). Journalists as interpretive communities. Critical Studies in Media Communication, 10(3), 219-237.
Appendix A - Inter-Coder Reliability
Table 1.0
Codebook Inter-coder reliability results
Variable Name N coders Percent Agreement
Scott's Pi Krippendorff's Alpha
n Agreements n Disagreements N Cases N Decisions
Date 2 100 1 1 66 0 66 132 Occhiello 2 95.45 0.89 0.89 63 3 66 132 Source 2 100 1 1 66 0 66 132 Talking About 2 100 1 1 66 0 66 132 Power Elite 2 98.48 0.93 0.93 65 1 66 132 Power Elite N° 2 78.79 0.74 0.75 52 14 66 132
Power Elite Name 1 2 95.45 0.95 0.95 63 3 66 132
Power Elite Name 2 2 89.39 0.88 0.88 59 7 66 132
Headline Content 2 89.39 0.82 0.82 59 7 66 132 Headline Overtone 2 86.36 0.81 0.81 57 9 66 132 Main Topic N° 2 87.88 0.80 0.80 58 8 66 132 Main Topic 2 100 1 1 66 0 66 132 Positive vs Negative 2 90.91 0.84 0.84 60 6 66 132 Illegal vs Legal 2 100 1 1 66 0 66 132 Area 2 80.30 0.72 0.72 53 13 66 132 Cause 2 87.88 0.84 0.84 58 8 66 132 Solution 2 93.94 0.88 0.88 62 4 66 132 Decreas. Vs Increas. 2 95.45 0.80 0.80 63 3 66 132 Issue 2 87.88 0.83 0.83 58 8 66 132 Reporter-Newsroom 2 98.48 0.98 0.98 65 1 66 132 Comment 2 93.94 0.86 0.86 62 4 66 132 Magnitude 2 92.42 0.83 0.84 61 5 66 132 Relevance 2 93.94 0.88 0.88 62 4 66 132 Domestic News 2 100 1 1 66 0 66 132 Foreign News 2 98.48 0.90 0.90 65 1 66 132
36 Appendix B
CODEBOOK
Framing News Stories on Tax Fraud/Evasion
Please, follow the instruction provided with the question and record the code in the corresponding column of the code sheet. Please, consult the appendixes when required.
V1. Coder. Each coder has a unique coder ID number assigned. The ID number will be reported in the first column of the code sheet under the variable name Coder_ID. Coders are assigned the following ID code numbers:
V2. Article ID. An ID number identifies each story. The ID number will be recorded in the code sheet under the variable name Article_ID
Please, record as shown by the example: E.g. 001
002 003 ===
V3. Date. News article date can be found at the end or beginning of each news article. Please, do NOT report the date you see on the top left corner of each item, as it
corresponds to the downloading date. You will code the date in the code sheet under the variable name ‘Date’. Date will be coded in the following format: MM/DD/YY [E.g. 6 marzo 2013 => 03/06/13]
V4. Headline (Title). The title of each news article is to be recorded in its entirety. Copy and paste the title in the code sheet under the variable ‘Title’.
V5. Occhiello (Subtitle). Is the headline accompanied by a subtitle? You will code ‘Yes’ if a subtitle is present, otherwise you code ‘No’. Report the code in the code sheet under the variable named ‘Occhiello’.
0. No 1. Yes
V6. Source. From which news media outlet is the news story? An ID number identifies each source. The ID number will be reported in the code sheet under the variable name Source_ID.
1 = Corriere della Sera
Coder Name Coder Number
Diego 1 Francesca 2 Roberto Other 3 4
37 2 = La Repubblica
3 = Il Fatto Quotidiano 4 = Il Giornale
V7. Talking About. Is tax fraud/evasion discussed in the news story? In this variable you will pay particular attention to the overall content of the news article you are coding. You will point out if tax fraud/evasion is the focus of the news article, or if the story is about something else and tax fraud/evasion is briefly mentioned. You will code ‘No’ when the article does not mention or discuss tax fraud or tax evasion.
You will code ‘Yes’ when the article concerns an event in which tax fraud or tax evasion is mentioned as the central point of discussion of the article.
You will code ‘Yes when the article concerns an event in which tax fraud/evasion is just mentioned.
You will code ‘Yes’, if the story is about a scientific report, or a political speech, or an opinion, or other, and tax evasion/fraud is just mentioned without being central point of the story.
You will code ‘Yes’ when the article concerns a political strategy in which tax fraud and tax evasion are discussed together with other topics*.
You will code, ‘Yes’when the article concerns a specified case** of tax fraud or tax evasion. Round and record the code in the code sheet under the variable named ‘Talking About’.
(*Other topics, education, economy, illegalities, politics, viability, globalization, and
so on)
(**Specified case of tax fraud or tax evasion means a legal action in which someone
is suited for tax fraud/evasion; a legal action in which someone is charged with tax fraud or tax evasion; a story event in which someone is accused of tax fraud or tax evasion)
0. No 1. Yes
If the coding to the previous question is NO, please stop coding the article and skip to the next news story event.
V8. Power Elite. Is the main character of the news story a powerful individual, a private or public organization, or a public or private institution? In this variable you will locate the main character of the news story. To locate the main character you will look at whether or not the article gives particular attention to one or more characters that the reader would give particular attention to.
You will code ‘No’ when the article does not discuss an event in which one or many powerful individuals are given particular attention.
You will code ‘Yes’ when the article discusses an event in which one or many powerful individuals are given particular attention.
38 individual and the news article content gives a large space to that powerful individual. You will code ‘Yes’ when a news story explains an event in which more powerful individuals are investigated for tax fraud/evasion, and the news article gives large space to such powerful individuals.
You will consider powerful individuals, politician(s), representatives of private or public organizations, private or public organization, representatives of public, private institutions, and private or public institutions.
You will consider a public organization any organized group, which is not part of the Public Administration of a country, and that still provides a public service. [E.g. Red Cross, Public Broadcasting.]
You will consider a public institution any organized group, which is part of the Public Administration of a country. [E.g., Government, Parliament, Chambers of Senate, Ministers]. Politicians and Political parties not mentioned in the list (See V8b) are to be considered public institutions. City Hall, assessor, consiglieri and members are to be coded as public institution.
You will code ‘Yes’ when the main actor of the article is the organized crime or members of the organized crime.
You will code ‘Yes’ when the news story has more than one main actor from different categories.
You will code ‘Yes’ when the main actor is a politician or a political party.
Please, record the code in the code sheet under the variable name Power_Elite.
0. No 1. Yes
Please, only code the following variables if you coded ‘1’ in the previous variable ‘Power Elite‘, V8. Otherwise, please skip to V9.
V8a. Please, code the number of Power Elite mentioned in the news story from 0 to infinite, under the variable ‘Power Elite N°’ of the code sheet.
To know each category and category code need, consult Appendix - Power Elite
Category List
You will count each category as ‘one’ power elite/main character. When a news article propose many actors grouped in the same category countries you will count them under the same category –
For example: Consider a news story that involves more than two countries as main character. Countries are part of the category ‘Foreign Institution’. Even though, each country is an actual power elite you will still code them as part of ‘Foreign
39 V8b. Power Elite Name. Which of the following power elite is present in the article? Please, record the code in the code sheet under the variable name
Power_Elite_Name.
The code is the number you see in column ‘code’ of the table below. Code is on the left of the Power Elite Name column
Please, only code the two most important ‘Power Elite Names’, or one if less than two.
You will establish the two most important ‘Power Elite Names’ following the criteria of V8. Therefore, to locate the two most important ‘Power Elite Names’ you will look at whether or not the article gives particular attention to one or more characters that the reader would give particular attention to.)
Power_Elite_Name_1 = First most important and most covered main character; Power_Elite_Name_2 = Second most important and most covered main character; Please Appendix to code Power Elite Categories Table 1.0. Power Elite Categories
In the next section you will code the headline content of each article
============================================================= V9. Headline. Is the title of the news story explicitly about tax fraud and or tax evasion?
You will code ‘No’ when the title does not include explicit wording linked to ‘tax evasion’, or ‘tax fraud’.
You will code ‘Yes’ when the title includes explicit wording linked to ‘tax evasion’, or ‘tax fraud’.
You will code ‘Yes’ when to recall tax evasion or tax fraud the title uses a metaphor, or a pun*. For example, if the headline uses a form of speech, in which a word is applied to one or more objects/actions to which the form of speech is not literally applicable, you will consider this a metaphor, and you will code ‘Yes’. [E.g.
metaphors about tax evasion: Pierluigi Bersani – ex Italian PM -, described tax evasion as ‘the act of not seeing a cow in a corridor. The cow is there but you cannot see it’ - "è come non vedere una mucca nel corridoio. C'è la mucca e non la vedi"]. *Pun = Gioco di parole
You will consider a pun a joke exploiting the different possible meanings of a word. [E.g. pun example ‘Tacks evasion!’]
Record the code in the code sheet under the variable name ‘Headline Content’.
0. No 1. Yes 9. Can’t tell
