The coverage of Golden Dawn by Greek online media
Eralda Agalliu
Student ID: 10696865
Master’s Thesis University of Amsterdam
Graduate School of Communication
Master’s Programme in Communication Science / Political Communication
Supervised by: Dr. Knut de Swert
Abstract
The prosecution of the Greek far-right party, Golden Dawn in 28 September 2013 gave impetus to academics to start investigating the political course of this party. This paper focuses on the Greek media coverage of the party since it entered the Greek Parliament. In other words, it is an effort to investigate whether Greek media coverage can explain the ‘success’ of Golden Dawn. A content analysis of three online Greek newspapers (To Vima, Kathimerini and To Proto Thema) was conducted for investigating whether the use of threat and opportunity frames, the tone of coverage and the political affiliation of media can give a thorough insight of how Golden Dawn is perceived in a national level. The present study reveals that Golden Dawn is framed more as a threat by the Greek media over time and the partisanship of the newspapers determines the use of the relevant frames. Media coverage is a crucial factor for evaluating Golden Dawn’s political performance and the way it is perceived by Greeks.
Keywords: Golden Dawn, far-right, online newspapers, threat frame, opportunity frame, partisan media
The outbreak of the European economic crisis, in 2008, revealed a tremendous change in the European political field, namely, that of the massive rise of far-right parties. A number of popular European far-right parties highly supported by citizens (Eurostat,2014) such as the French Front National (FN), the Flemish Vlaams Blok (VB), the Dutch Partij Voor de Vrijheid (PVV) , The Danish People's Party (DPP) and The Swedish Democrats (SD) started dominating the political field. Among those far-right parties a new far-right party becomes very prominent, Golden Dawn. This far-right party enters the Greek Parliament for the very first time in the country’s history (Ellinas, 2013; Psarras, 2013). The first step towards a historical and political change for both Greece and Europe began right after the Greek Parliamentary elections of June 2012. Despite the fact that economic crises are characterized by academics as ‘a fertile ground’ for the emergence of extreme political reactions in terms of electoral behavior and political options (political parties) (Boomgaarden & Vliegenthart, 2007; Kriesi ,2012; Dalakoglou 2013;), the Greek case appears to be a ‘special case’. In fact, the presence of Golden Dawn has been rather weak in the political stage until 2009, when the consequences of the economic crisis were more than apparent in Greece (Psarras, 2013). Both the socially and economically ‘fragile situation’ in Greece due to the strict austerity measures applied by relevant governments in order to cope with the crisis , not only paved the way for
Golden Dawn’s success but also rendered the party as a unique political construction (Dalakoglou, 2013). As a matter of fact, the party is evaluated by scholars as a Neo-Nazi party rather than simply a far-right one (Banteka, 2013; Dalakoglou, 2013; Ellinas, 2013; Psarras, 2013). Several incidents in which Golden Dawn was involved reveal that the political struggle of the party expands beyond lawful actions. In September 13th of 2013 the party reaches its peak of attention with the involvement in the murder of Pavlos Fyssas which was also the main motive for the Supreme Greek Court to start investigating Golden Dawn. Only two weeks later a number of prosecutions followed, that of Golden Dawn’s leader, Nikos Michaloliakos, as well as of several of Golden Dawn’s MPs. The arrested leader and MPs were accused of directing a criminal organization (Psarras, 2013). Due to those circumstances the Greek media coverage about Golden Dawn increased dramatically. The tabloidization of the political discourse (Mudde, 2012) by the far-right party could also justify both media attention and the impact of the party on the Greek public opinion (Bistis, 2013) . Despite the fact that the party was accused of being involved in 32 cases since the August of 2013, the percentage of Golden Dawn in the European Elections as well as the Greek National Elections of 2014 and 2015 still remained high (Ekloges.ypes.gr, 2014; Ekloges.ypes.gr,2015).Thus arises the question about the reasons due to which Golden Dawn still achieves to attract voters’ attention and support.
A number of previous studies (Bistis, 2013; Dalakoglou, 2013; Ellinas, 2013; Koronaiou & Sakellariou, 2013; Toloudis, 2013) have already dealt with the reasons that made Golden Dawn successful explaining this phenomenon by focusing on traditional aspects like economic, social and anti-immigrants factors. According to Brants and van Praag (2006), we are living in the era of media logic, where media lead the agenda, hence in order to understand the political fermentations of a country it is necessary to understand also how the national media deals with politics. Additionally, media content is considered by previous studies as an important factor for explaining the rise of far right parties (Walgrave & De Swert, 2004; Boomgaarden & Vliegenthart, 2007). The rise of the European far-right parties has also been interpreted by scholars as a result of the agenda setting and issue ownership of those parties in the political discourse (Walgrave & De Swert, 2004). Media seem to build the context within which political fermentations take place by controlling the information flow, selecting to cover certain issues and interpreting them in certain ways (Mazzoleni et al., 2003). In fact, there is an extensive coverage of ‘media-genic’ issues (tabloid-style and emotional issues) and the qualities of political figures which explain how far-right parties
achieve to approach and affect the public opinion (Plasser & Ulram, 2003). The high influence of media on mass audience opinion and electoral behavior (Boomgaarden & Vliegenthart, 2007) reveal that politics and media are strongly intertwined. Therefore taking into consideration all the above mentioned and the fact that the presence, the political course and struggle of Golden Dawn has not yet been interpreted on the basis of media content, media content analysis could give a more thorough insight of the ‘phenomenon’ called Golden Dawn. Nowadays, the traditional press is read by a lower percentage of readers compared to the online newspapers (Yang & Grabe, 2011), thus the present paper will focus on the Greek online newspapers content. More precisely, the aim of this paper is to focus on how Greek online newspapers cover the course of Golden Dawn over time (2012-2015).
Taking into account that far-right parties are perceived in terms of threat and opportunity by scholars (Varga, 2008) it is crucial investigating the following research question: To what extent is Golden Dawn covered as a threat and/or opportunity by the Greek online press over time? Online media content analysis will contribute in understanding how this far-right party is perceived by media, one of the most important factors of the political system. Media portrayal of Golden Dawn could contribute to the interpretation of the party’s course since it entered the Greek Parliament and the respective impact on the Greek public opinion.
Theoretical background
The aim of this study is to evaluate the political course of Golden Dawn through the media coverage as media has always been an important factor for understanding and evaluating politics , especially since the rise of mass media in 1970 (Scheufele & Tewksbury, 2007). The media role, as lapdog, as watchdog or even as a hypodermic needle (McQuail, 2006) has always had an impact on public opinion about political parties. The third age of political communication revealed that media outlets have the potential to affect both the political agenda and the public opinion (Blumler & Kavanagh, 1999). The influence on public opinion is achieved through certain media ‘coverage mechanisms’; namely, the selection of certain issues instead of other known as agenda setting, the intervention in the interpretation process known as framing and by providing a basis upon which public opinion will be based for evaluating the political parties in our case, known as priming (Scheufele & Tewksbury, 2007).
The media ‘mechanism’ of framing is extremely important when interpreting public opinion on politics as it determines the way people will perceive a certain issue according to the context that has been set by media. Moreover, according to the theory of “cascading activation”, frames are spread more easily by the elites (media organizations in this case) to the lower levels of society, the voters (Entman, 2003). Thus, the frames regarding political parties that are created by media will be easily spread and adopted by citizens. The course of the far-right political parties cannot be regarded separately from the media context and especially framing. Studies investigating the factors that determine the success of far-right populist parties have shown that media framing is a significant factor (Bosman & d’Haenens, 2008). For example in the Dutch case newspapers framed the far-right leader Pim Fortuyn by demonizing him and causing a general feeling of disapprovement to voters (Bosman & d’Haenens, 2008). The political success of political parties can also be seen as a result of favorable media portrayal of the party or the issues related to it and its members (Walgrave & de Swert, 2004). Due to the fact that media relate certain issues to certain parties known as “ownership of issues”, voters tend to believe that the relevant party is capable to provide solutions to those policy questions (Walgrave & de Swert, 2004). As a matter of fact, far-right parties are highly related to anti-immigrant issues; therefore extensive media coverage on immigration issues will have a positive impact in the voters’ attention for the party (Boomgarden & Vliegenthart, 2007; Plasser & Ulram, 2003; Walgrave & de Swert, 2004).
Previous studies confirm that the exclusivity in anti-immigrant issues create political opportunities for far-right populist parties to be prominent in the political arena and consequently affect voters’ attention and preference (Eatwell, 2000; Kriesi, 2012; Varga, 2008). Indeed, in the Greek case scholars have confirmed that Golden Dawns support is attributed to the political actions on immigration issues (Bistis, 2013). Far-right parties are generally evaluated as threat or as opportunity for a country and consequently its voters (Kriesi, 1999; Varga, 2008). In the first case, extreme right parties are perceived as a threat for the political system and subsequently democracy (Ellinas, 2013) when framed by media outlets as a potential danger (Varga, 2008). Although this latter is true, framing an issue as a problem and not the party has the exact opposite influence in public opinion. For instance, a substantial framing of certain issues as problems (frame problem), for example immigration, increases the sense of fear and risk (Altheide, 1997) promoting at the same time the far-right parties which provide policy solutions to those problems. Henceforth, when a far-right party is portrayed by media as a different one, in terms of political solutions than the other rival
political parties, it is evaluated by voters as an alternative option, an opportunity in the political system (Varga, 2008). The difficult political, economic and social circumstances favor the occurrence and prevalence of the far-right populist parties which also try to frame themselves as a political opportunity for changing the respective situation (Rydgren, 2005). Golden Dawn could be perceived as a political opportunity in such a context, too. Several studies portray the rise of Golden Dawn as a threat for Greece, especially by referring to the party in a context of negative events in which the party was involved (Ellinas 2013; Psarras 2013; Toloudis, 2013). Besides this latter, the media coverage of Golden Dawn is expected to be high on negative issues since media preference for conflict and sensational issues will easily attract the mass audience (McManus, 1995; Plasser & Ulram, 2003). Although, the opportunity frame is not examined in the case of Golden Dawn, the party can be perceived as such when taking in consideration previous studies about other similar far-right European parties (Bale, 2003; Rydgren, 2005, 2010).Taking into consideration all the above mentioned, the coverage of Golden Dawn by the Greek media will be evaluated within the framing context of threat and opportunity.
Framing functions through priming in order to shape or alter the way audience interprets and makes choices (Entman, 2007). Namely, media will promote certain issues by emphasizing certain aspects and ideas about news that will encourage the audience to perceive, evaluate and decide in particular ways (Iyengar & Simon, 1993). It is important for this paper to consider also the ‘nature’ of an event. It has been confirmed that the genre of an event will affect the relevant frame used by media for it. More precisely, media will report more negatively on the issue after a negative event than before it took place (Boswell, 2007; Entman, 2003). Respectively, positive frames will be used in the coverage of positive events. The genre of a real-life event is used in priming and can be used as criteria for evaluating those far-right parties (Scheufele & Tewksbury, 2007). Elections period are characterized by a high information flow for all parties and the use of frames is ‘balanced’ for all parties (Dalton, Beck & Huckfeldt, 1998).Under those circumstances elections can be considered as a neutral event. Consequently, it is expected that Golden Dawn’s coverage will be also driven by real life events that will shape media framing according to the genre of those events.
Though, it is crucial to mention that agenda setting, framing and priming lead to media bias. Through the depiction of several aspects of an issue and the promotion of the relevant ideas and aspects (framing) very often news are distorted, not presented equivalently and therefore the final media content is biased (Entman, 2007). The coverage of mere facts and evaluation
provide a distorted picture of the news which is far from stating facts and being objective (Bosman & d’Haenens, 2008). Thus, the degree of framing determines the objectivity of a media outlet (Semetko & Valkeburg, 2000). As revealed by Gan et al. (2005) the attitude of a media outlet towards objectivity will determine which frames will be used more frequently. Scholars underline also the fact that framing offers richness in the media discourse about an issue by covering it in divergent ways (Tankard, 2001). Consequently, the degree of frame use will reveal whether Greek media bias is high in the coverage of Golden Dawn.
The editorial slant (tone of coverage) is an important part of the media content as it reveals the use of the relevant frame (Entman, 2007). Tone is considered to have an impact on framing by emphasizing or minimizing it. The use of visual and text information, components the tone of media coverage, can reveal the way a certain event is framed (Entman, 2007). The editorial slant can be neutral, positive or negative about an issue and thus shape the frames respectively. Therefore, it is expected that both pictures and the use of certain verbal expressions referring to a far-right, in this case, will determine the framing of the party. However, Druckman and Parkin (2005) emphasize that not only the kind of tone used matters but also the quantity of it. In this context, it is expected that both the genre of the picture and the verbal context will both determine and reveal the way Golden Dawn is framed by the Greek online newspapers. The quantity of tone will also determine the strength of the frame used, in particular the use of many verbal expressions and/or pictures will intensify a frame. Τhe editorial tone has the capability to affect public opinion by making them favoring or disapproving the party and/or the politicians (Entman, 2007). Due to this latter, it is necessary that the tone of coverage that the Greek newspapers use in the coverage of Golden Dawn is taken into consideration.
Furthermore, when analyzing the media content on the coverage of far-right parties it is necessary to consider the media system of the country in which a political party is created and acts. The media system of a country allows evaluating the kind of impact media has on public opinion about political parties and thus the political choices (vote). Media and political parties in some cases seem to be related with clientelistic relationship which determines the final media product (Hallin & Papathanassopoulos, 2002). Greece is an illustrative example of the polarized media system, characterized by a strong political parallelism and low professionalism (Hallin & Mancini, 2004). In addition, previous studies have revealed that Greek media are still highly partisan with a coverage emphasis on certain political parties
rather than politicians (Papathanassopoulos, 1999) despite the media deregulation and commercialization. Partisan media report differently on the same topic because of their political predisposition (Niven, 2001).The coverage of political parties in the Greek context will not only reflect the respective political ideology that Greek media espouse, but also shape accordingly the media reportage on other ‘adversary’ political parties. Selective exposure to those partisan media according to the political predispositions of the electorate reveals the significant media impact on citizens’ preferences of certain parties (Stroud, 2007). For example right wing newspapers are more conservative in the coverage of immigration issues compared to left wing ones that are more ‘socially tolerant’ due to their political affiliation (Green-Pedersen & Odmalm, 2008) and will influence their readers accordingly. Therefore, the right wing newspapers will cover more critically on immigration issues than the left-wing ones that will be more willing to initiate a public discourse on the matter. In this case, the coverage of Golden Dawn is expected to differ among the newspapers because of their political ideology and party affiliation. The political ideology and the partisanship to political parties of the Greek media outlets could contribute in understanding any possible difference in the coverage of the political course of the Greek far-right party.
In summary, media coverage of Golden Dawn will be evaluated according to a number of media characteristics considering the important role media play in politics. Namely, the use of the threat and the opportunity frames, the editorial slant, the genre of the covered events concerning Golden Dawn and the political ideology of each newspaper, will contribute in assessing the way Greek media cover this far-right party. It is expected that an analysis of the aforementioned factors will offer a more accurate picture about the way Greek media deal with the phenomenon called Golden Dawn.
Hypotheses
Taking into account the above mentioned theories about far-right parties ,especially the fact are perceived as threat or opportunity (Kriesi, 1999; Varga, 2008) and that Golden Dawn is related to several negative events (Ellinas 2013; Psarras 2013; Toloudis, 2013) it is expected that Golden Dawn will be framed more as a threat rather than an opportunity for Greece. Thus the following hypothesis needs to be investigated: H1. Golden Dawn is framed more as a threat than an opportunity for Greece by the Greek press from 2012 up to 2015.
The genre of an event has an impact on the frames that will be used by media when covering a political party (Boswell, 2007; Entman, 2003), therefore it is expected that certain events will have an impact on the frame that will be used by online Greek newspapers when covering Golden Dawn. More precisely, a negative event in which the far-right party is involved is expected to be followed by an extensive use of threat frame and a decrease of the opportunity frame regarding Golden Dawn (Kaparesearch.com, 2014, 2015). Hence, it is interesting to see whether this is valid in the Greek media context. The chosen negative event in this case is the leaked video of a conversation between Golden Dawn’s spokesperson Ilias Kasidiaris and the former cabinet secretary Takis Baltakos (Democratic Party). It is actually a conversation between the two aforementioned MPs which reveals that there were secret communication channels between the far-right party and the former Democratic government. The conversation was videotaped secretly by Golden Dawn’s spokesperson and condemned as an illegal action (Το Vima, 2015). The conversation referred to the prosecution of Golden Dawn’s MPs for directing a criminal organization. In fact it was an attempt of Golden Dawn’s spokesperson to frame the prosecution of his party’s MPs as an effort of the former government to harm the far-right party. This latter was criticized and covered negatively by the Greek press (To Vima, 2015; Kathimerini, 2015; To Ethnos, 2015).
On the other hand positive events are expected to be framed positively by media as mentioned in the theoretical part (Scheufele & Tewksbury, 2007). The lack of positively related event to Golden Dawn within the time frame of the present study (see sample section below) a neutral event was chosen in order to investigate whether framing will differ. Thus, the chosen event is the Greek national elections of 2015. The time period prior to elections is characterized by a high information flow for all parties in order to present their political agendas and proposals and framing is expected to be ‘equilibrated’. Specifically, both negative (threat frame) and positive framing (opportunity frame) are expected to be at the same extent. In this context we are interested in investigating whether there is a change in the intensity of the threat frame and the opportunity prior to elections. It is expected that there will be a change in the use of both frames before elections as the far-right party will promote its policies in order to attract voters but at the same time it will be framed negatively by the rival political parties. The following hypotheses occur: H2a. The ‘nature’ of a real life event can explain the change in the use of the threat frame in the coverage of Golden Dawn by the online Greek newspapers and H2b will investigate whether: The ‘nature’ of a real life event
can explain the change in the use of the opportunity frame in the coverage of Golden Dawn by the online Greek newspapers.
Furthermore, a difference is expected to be found in the use of editorial slant (tone) among the Greek online newspapers. According to scholars partisan media will make a use of frames respective to their political ideology (Papastathopoulos, 1999). Taking in consideration that the tone of coverage determines the use of frames it is expected a difference in the tone, the use of threat frame and opportunity frame in the coverage of Golden Dawn. Media bias is an important factor for interpreting any differences in the use of frames (Entman, 2007). In the case of partisan media framing is expected to reflect the respective political ideologies (Niven, 2001). Thus the following hypotheses (H3a, H3b and H3c) need to be investigated for locating any possible difference in the use of tone and the relevant frames among the Greek online newspapers since they are of different political ideology (Opperman & Viehirg, 2011).
H3a. The tone of the coverage of Golden Dawn is expected to differ among the three online Greek newspapers.
H3b. The use of opportunity frame about Golden Dawn is expected to differ among the three online Greek newspapers.
H3c. The use of the threat frame about Golden Dawn is expected to differ among the three online Greek newspapers.
Methods
In order to investigate the research question and the relevant hypotheses a deductive approach was used. The hypotheses are based on a theoretical background and subjected to empirical scrutiny (Bryman, 2008). A quantitative, non-automated content analysis was employed for testing whether online Greek media cover Golden Dawn as “a threat” or an “opportunity” since the far-right party entered the Parliament in 2012.
Definition of the concepts
This study focuses on two frames, the threat frame and the opportunity frame taking in consideration that they are mainly evaluated in terms of threat and opportunity (Schuck & de Vreese, 2006; Varga, 2008). The independent variables are the three online Greek newspapers
used for this study (To Proto Thema, Kathimerini and To Vima) whereas the dependent variables are the threat frame and the opportunity frame used by those newspapers.
In this context any indication of imminent harm like: pain, injury, damage or any other hostile action, (Highhouse & Paese, 1996; Scholtz, 2000) caused by Golden Dawn’s MPs or supporters and reported in the online Greek press are perceived as a threat frame. Actually, the threat frame is defined by Highhouse & Paese (1996) as a potential danger of loss of benefits and an imminent risk of danger. Scholtz (2007) underlines the emotional effects of threat in the future. All the negative actions or involvement of the party in negative events imply that they have implications for the Greek society by putting in risk democracy, social cohesion, stability and creating a number of problems (e.g. increasing xenophobia, racism, criminality etc.) in Greece (Psarras, 2013). In this study the opportunity frame is defined as any indication of the Greek online press to present Golden Dawn as an alternative party for attaining several goals (economic development, ending corruption, improving Greeks’ quality of life etc.), praising the party for the beneficial actions or justifying the party’s engagement in several incidents. Both terms, threat and opportunity are defined according to “The Oxford English Dictionary” (Simpson & Weiner, 1989). Additionally, the opportunity frame is also based on the paper of Schuck and de Vreese, 2006: Between Risk and Opportunity News Framing and its Effects on Public Support for EU Enlargement.
Operationalization of the frames
For measuring each concept, dependent variable, namely the threat frame and the opportunity frame several variables (questions) were included in the codebook. More precisely, the threat frame concept was measured through:
a battery of negative incidents which present Golden Dawn as a victimizer (Q7), nominal level (binary)
negative consequences occurring by Golden Dawn’s actions in politics (Q9),nominal variable (binary)
a number of statements about the negative impact of Golden Dawn in the Greek society (Q8), interval level of measurement (5 point Likert scale) and
Whereas, the opportunity frame is measured by the following variables in the codebook:
five variables (positive statements about Golden Dawn) Q10,Q11,Q12,Q13 and Q14, interval level of measurement (5 point Likert scale)
In both cases the variables were recoded into new ones, when it was necessary, and computed in order to have two variables that would measure directly the threat frame and accordingly the opportunity frame. Further information about the variables used for both frames are included in the data analysis section and in the Appendix.
Sample
Due to the lack of Greek official reports regarding the press readability and the fact that the use of technology by Greeks is quite popular (Arampatzis, 2004; Kountzeris & Konstantatos, 2010) the study focuses only on the Greek online press. Online press offers readers an easy access and free of charge (Schoenbach, 2005) which seem to be two of the main reasons justifying readers’ preferences for online newspapers rather than the traditional ones. The target population for this study is all the Greek online newspapers since the research question refers to the Greek online press. Therefore, the ranking website www.alexa.com was used for choosing the sample of this study as it has been used from previous academics (Spyridou & Veglis, 2008) as a trustworthy and official site for website rankings. The three most visited, thus popular online newspapers were included in the sample. The newspapers: To Proto Thema (12th in ranking), Kathimerini (49th in ranking) and To Vima (62nd in ranking) were included in the final sample according to the “site traffic”. The Huffington Post and Naftemporiki newspapers were higher in ranking than Kathimerini and To Vima, but were excluded as they were accordingly new and an economic newspaper. The newspaper To Proto Thema is a weekly, sensationalist “yellow press” newspaper, Kathimerini is a daily conservative newspaper, while To Vima is a daily center-left one (Opperman & Viehirg, 2011). The sample can be considered representative as it includes two prestige press newspapers and a “yellow one”, each online newspaper espouses different political ideology and they are weekly and daily accordingly. For all the aforementioned reasons the sample of this study provides an accurate picture of the Greek online press.
The time frame of this study starts from 2012 when Golden Dawn entered the Greek Parliament for the first time and extends up to ten days before the last Greek national elections of 2015. In order to avoid sample bias the chosen time period for each consequent year (2012-2015) were based on 12 surveys launched by Kappa Research during those four years. Other polling companies are not included in the present study as there is a marginal
difference in the percentages comparing to Kappa Research and this latter is considered to be one of the most trustworthy and popular ones in Greece. The surveys measure the public opinion by phone interviews on the Greek political developments and the relevant elections (see Table1 in Appendix). The time period during which each survey was launched is extended by two days prior to the starting date of each survey. This latter resulted in a sufficient sample size for this study providing an accurate picture of the media reportage on Golden Dawn. The sampling units of this study are the articles that occurred by using the search term “Golden Dawn”, “Chrysi Avgi” in Greek in each newspaper’s database. The most relevant articles to the search term were included in the present study. Furthermore, the relevant articles from each newspaper insert were also included in the final sample.
The searching engines of each newspaper resulted in a sample size (N) of 473 articles in total for all three newspapers. The articles from letters to editor, those included twice in the list and the ones that required subscription were excluded from the sample; therefore the final sample size of the study was N=428 articles. Namely, the final sample consists of N=78 articles for Kathimerini, N=188 for To Vima and N=166 for To Proto Thema.
Coding and Intercoder Reliability
A questionnaire (see Appendix B) was designed in Qualtrics in order to test the aforementioned hypotheses. One extra coder was trained to test whether the questionnaire was reliable and replicable. The coder was a native Greek speaker and had sufficient knowledge of the political fermentations in Greece. Coder training was done so that both coders would code in the same way for each variable. A sample of 59 articles was used for checking intercoder reliability. The coding of this subsample lasted for three days (25th-27th of May). The intercoder reliability was sufficient enough for all variables (Hayes & Krippendorf, 2007) (see Table 2 in the Appendix), apart from one variable, that of protection of Greeks from immigrants (Q14_1). The variable that scored low on Krippendorf’s alpha was not so important for the statistical analysis of the hypotheses, therefore it was not changed. The coding of the total sample (N=428) lasted for eight days.
Data Analysis
In order to answer the aforementioned hypotheses it was necessary to first check whether the variables employed in the codebook for both frames ,threat frame and opportunity
one, (see operationalization part above) are proper indicators of the frames. A factor analysis was run for both frames.
The items: threat to democracy (Q8_1), violence between Greeks and immigrant (Q8_2) and hate speeches of Golden Dawn against other parties’ MPs (Q8_3) were used for measuring the threat frame. A factor analysis was run only for those variables as they are 5-point Likert scale and this statistical analysis requires variables of interval or ratio level of measurement (Field, 2013). The variables negative incidents presenting the party as a victimizer (Q7) and negative consequences of Golden Dawn’s actions (Q9) were not included in the factor analysis but they are considered as reliable measures of the threat frame according to previous studies (Varga, 2008).
The variables (Q8_1, Q8_2 and Q8_3) mentioned above were analyzed with a principal component analysis (PCA). The sample size of the study was (N=428). Samples of a size (N=300) and larger meet the requirements for a factor analysis (Field, 2009; MacCallum, Widaman, Zhang, and Hong, 1999). Two components with an eigenvalue bigger than one were provided by the eigenvalues analysis which according to Kaiser’s criterion are adequate enough. The first component explained 47.1 percent of the variance with an eigenvalue of 1.41, whereas the second component explained 34.8 percent of the variance with an eigenvalue of 1.04. The rest of the components scored lower than 1in the corresponding eigenvalues, hence only the two first factors were retained. Both items (variables Q8_1 and Q8_2) loading in component 1 correlated positively with it, while the item with the strongest association was presentation of Golden Dawn as increasing violence between Greeks and immigrants (Q8_2) (see Table 3 below).The item hate speeches of Golden Dawn against other parties MP’s (Q8_3) correlated positively with component 2.The valid cases for the items included in both components were 428 (no case was excluded). The reliability for both components was, (Cronbach’s A =.38) and by excluding the item hate speeches of Golden Dawn the reliability scale increases (Cronbach’s A= .53). Thus, only the items and Golden Dawn as a threat to democracy (Q8_1) and increasing violence between Greeks and immigrants (Q8_2) were retained as reliable measures of the latent variable (threat frame). In later steps of our analysis those items (variables) are computed in order to measure the threat frame concept for testing the above mentioned hypotheses.
A factor analysis was also run for the opportunity frame. The variables (items) that were included in the codebook for measuring this frame were the following: a number of
statements present Golden Dawn positively (Q10), presentation of GD as an alternative party (Q11), presentation of GD as a beneficial party (Q12), presentation of GD as fighting corruption (Q13), and presentation of the party as protecting Greeks from the immigrants (Q14).
All variables were analyzed with a principal component analysis (PCA). The sample size of the study was the same as above (N=428). Samples of a size (N=300) and larger meet the requirements for a factor analysis (Field, 2009; MacCallum, Widaman, Zhang, and Hong, 1999), as mentioned above in the threat frame factor analysis. The eigenvalue analysis provided two components with an eigenvalue bigger than (according to Kaiser’s criterion are adequate enough). The first component explained 53.11 percent of the variance with an eigenvalue of 5.31, whereas the second component explained 19.96 percent of the variance with an eigenvalue of 1.99. The rest of the components scored lower than 1in the corresponding eigenvalues, therefore only the two first factors were retained. Furthermore, the scree plot also revealed an infection after the second component. All items correlated positively with the relevant components. The item with the strongest association for component 1 was the presentation of Golden Dawn as a beneficial party (Q12), whereas for component 2 was presentation of Golden Dawn as fighting crime (Q13) (see Table).The number of valid cases for both components were 64 (364 cases were excluded). The reliability for both components was (Cronbach’s A= .89) and with the exclusion of the item protecting Greeks from immigrants (Q14) the reliability coefficient increases (Cronbach’s A= .90). Hence, the items (variables) Q10, Q11, Q12 and Q13 were retained as measures for the latent concept (opportunity frame). The aforementioned variables will be computed and used later on for measuring the opportunity frame concept.
In order to answer the first hypothesis: Golden Dawn is framed more as a threat than an opportunity for Greece by the Greek press from 2012 up to 2015, the variables threat frame and opportunity frame were computed in advance. Namely, the variable threat frame was computed by summing the variables threat to democracy (Q8_1) and increase of violence between Greeks and immigrants (Q8_2) (as mentioned in the factor analysis).The mean score of the sum for both variables measuring the threat frame for all articles (N=428) was M=1.94 SD=1.06. Accordingly, the variable opportunity frame was computed by summing the variables: battery of positive statements (Q10), GD presented as an alternative party (Q11), GD presented as doing beneficial actions (Q12), and GD presented as fighting corruption (Q13). The mean score of the sum of those variables measuring the use of opportunity frame
in all articles included in the sample (N=428) was M=1. 13, SD=.40. The minimum score for the threat frame was Min=1 as well as for the opportunity frame); while the maximum score for the threat frame was Max=5 and for the opportunity frame Max=4. Both frame means were compared on the relevant scores for each consequent year (2012-2015). As revealed by the graph below (Table 5), the threat frame was used more than the opportunity frame over the period 2012-2015. More precisely, the threat frame was used 17.9 percent on 2012 while the opportunity one only 11.9 percent. In 2013, the threat frame was used 44.1percent by the Greek online newspapers whereas the opportunity frame 25.6 percent. In 2014, the frames were used 23.3 percent and 13.9 percent accordingly and in 2015, 13.2 per cent and 9.6 percent. In fact, there is a change in the intensity of use of the threat and opportunity frames, though the threat frame is always used more than the opportunity one. The biggest difference in the use of frames (18.5 percent) is located in 2013 while the lowest difference between the frames (3.6 percent) is in 2015.
Table 5: Use of the threat and opportunity frames in the coverage of GD overtime
Note: N= 428 articles
In order to investigate whether the nature of a certain incident can predict the use of threat frame in the coverage of Golden Dawn a multiple regression analysis was run. The variable date (Q3/ nominal scale) was recoded into a two new variables; one involving all the articles published from the day of the negative incident (2nd of April 2014) until a week later (9th of April 2014) and one variable including the articles published on 8th -15th of January 2015 (ten days prior to the Greek national parliamentary elections of 25th January
17,9% 44,1% 23,3% 13,2% 11,9% 25,6% 13,9% 9,6% 2012 2013 2014 2015
Use of threat frame and opportunity frame
threat frame opportunity frame
2015).Therefore each one of the new created variables (dummies) took the value 1 if the article was published in the relevant dates and 0 if it was not. The number of the articles related to the negative event (Baltakos’ leaked video) was N=72, while for the neutral event (Greek national elections 2015) N=76. The outcome variable threat frame (dependent variable) is of ratio level of measurement. Hence two predictors, the articles published after the negative incident and those in advance of the neutral event, will be used for predicting the use of the threat frame in the coverage of Golden Dawn by the online Greek newspapers.
The regression model with the use of threat frame as a dependent variable and articles concerning the negative and the neutral events as predictors is significant, (F (2,425)=19.36, p<.001). The regression model can be used to predict the use of the threat frame by the online Greek newspapers in the coverage of Golden Dawn, but the strength of the prediction is weak: 8.3% of the variance in the use of threat frame can be explained by the articles referring to the negative and the neutral events (R² = 0.083). The articles published after the negative event, b*=0.13, t= 2.64, p=. 009, 95% CI [0.09, 0.62] and those articles prior to the neutral event, b* =-0.23, t=-4.95, p<.001, 95% CI [-0.91, -0.39] have a significant weak association with the use of threat frame by online Greek newspapers in the coverage of Golden Dawn. If an article is published after an event of negative nature then the threat frame use increases by .354 (meaning that Golden Dawn will be covered by online Greek newspapers more as threat). Per every unit (article published prior to a neutral event) that goes up in the scale the use of opportunity frame concerning Golden Dawn will decrease by .651. Regarding multicollinearity, the VIF (variance inflation factor) in this case is less than 10 (VIF<10). Consequently, the VIF and tolerance values are acceptable. All the other independent variables are assumed to be held constant for the present effects.
A multiple regression was also run for testing hypothesis H2b. The same events (see section above) were used also for checking whether the nature of a certain event can predict the use of opportunity frame in the coverage of Golden Dawn by the Greek online newspapers. The cases for the opportunity frame were the same as for the threat frame, hence N=72 articles for one week since the negative incident (Baltakos’ leaked video) and N=79 articles for the time period prior to the neutral event (Greek national elections 2015). The outcome variable (opportunity frame) is ratio level of measurement and the predictors, negative and neutral events were recoded (see multiple regression for threat frame) from nominal level to dummies.
The regression model with the use of the opportunity frame as a dependent variable and the articles concerning the negative event and the neutral event as predictors is significant, (F (2,425)=6.71, p=.001). The regression model can be used to predict the use of the opportunity frame in the coverage of Golden Dawn by the online Greek newspapers, but the strength of the prediction is very weak: 3.1% of the variance in the use of the opportunity frame can be explained by the articles referring to negative and neutral events (R² = 0.031). The articles published after the negative event, b*=0.15, t=3.12, p=. 002, 95% CI [0.26, 0.60] and those articles prior to the neutral event, b* =0.12, t=2.53, p=.012, 95% CI [0.23, -0.03] have a significant weak association with the use of the opportunity frame by online Greek newspapers. If an article is published after an event of negative nature (like the Baltakos’ leaked video) then the use of the opportunity frame in the coverage of Golden Dawn decreases by 0.162 (meaning that Golden Dawn will be covered by online Greek newspapers less as an opportunity). Per every unit (article published prior to a neutral event) that goes up in the scale the use of the opportunity frame in the coverage of Golden Dawn will decrease by .129. Therefore, the nature of both events will lead to a decrease of the portrayal of Golden Dawn as a threat by the online Greek media. As far as multicollinearity is concerned, the VIF (variance inflation factor) in this case also is less than 10 (VIF<10, thus acceptable tolerance values. All the other independent variables are assumed to be held constant for the present effects.
Table 6: Regression coefficients for the threat frame and opportunity frame
b Standard Error B b* t p VIF N Threat frame Constant 2.00 0.61 32.96 .000 negative event 3.54 0.13 0.12 2.64 .009 1.05 72 neutral event -0.65 0.13 -0.23 -4.96 .000 1.05 79 Opportunity frame Constant 1.18 0.23 50.28 .000 negative event -0.16 0.52 -0.15 -3.12 .002 1.05 72
neutral event -0.13 0.51 -0.12 -2.53 .012 1.05 79
Notes: R² = 0.083 (threat frame) R² = 0.031 (opportunity frame); alpha level p<.05.
The expected difference in the tone of the articles about Golden Dawn among the three Greek online newspapers (Kathimerini, To Vima and To Proto Thema), hypothesis H3a, was investigated by running a One Way Analysis of Variance (ANOVA). The independent variable newspaper is categorical and has three categories, the three online Greek newspapers mentioned above. The dependent variable tone was computed by summing the variable tone coverage of the article (Q16) and overall opinion of the article (Q20). The variable overall opinion of the article (Q20) was recoded beforehand. Namely, the negative values -3 to -1 were recoded into -1 as negative, the neutral value remained the same (0) and the positive values 1-3 were recoded into 1 as positive. Therefore the computed dependent variable tone of the article, categorical also had three values -1 for negative tone, 0 for neutral tone and 1 for positive tone. A one way between subjects ANOVA was conducted for examining whether there are differences in the use of tone coverage of Golden Dawn among the newspapers Kathimerini, To Vima and To Proto Thema. The assumption of equal variances is accepted in this case, thus the variance of the groups is assumed to be equal (Levene’s F (2,425) =0.30, p=.740). The analysis of variance model (see Table 7), reveals that there is no significant difference in the tone coverage of Golden Dawn among the three online Greek newspapers. (F (2,425) =0.26 p=.772). Due to the fact that the groups do not differ significantly a Post hoc Bonferroni test was not needed. It seems that all three Greek online newspapers (Kathimerini, To Vima and To Proto Thema) use the threat frame when covering the far-right party to the same extent. Table 8 presents the average tone of the coverage of Golden Dawn for each Greek online newspaper.
Table 7: One-way analysis of variance for tone of the coverage of Golden Dawn
Sum of Squares Df Mean of Square F P Between groups 0.15 2 0.08 0.26 .772
Within groups 126.30 425 0.30
Table 8: average tone coverage about Golden Dawn
Newspaper Mean Standard Deviation N Kathimerini -0.45 0.56 74
To Vima -0.46 0.54 188
To Proto Thema -0.42 0.54 166
Hypothesis H3.b, a possible difference in the use of threat frame among the three online Greek newspapers was also tested by running a One Way Analysis of Variance (ANOVA). The independent variable in this case is still the three online Greek newspapers (Q3), Kathimerini, To Vima and To Proto Thema (categorical), whereas the dependent one is the use of threat frame (ratio level of measurement). The assumption of equal variances is accepted in this case, thus the variance of the groups is assumed to be equal (Levene’s F (2,425) =0.18, p=.835). The analysis of variance model (see Table 9), reveals that there is no significant difference in the use of threat frame in the coverage of Golden Dawn by the three online Greek newspapers. (F (2,425) =0.52 p=.593). Due to the fact that the groups do not differ significantly in the use of the threat frame a Post hoc Bonferroni test was not run. Table 10 presents the average use of the threat frame in the coverage of Golden Dawn by each one of the three Greek online newspapers.
Table 9: One way analysis of variance for the use of threat frame for Golden Dawn Sum of Squares Df Mean of Square F P Between groups 1.17 2 0.59 0.52 .593
Within groups 476.98 425 1.12
Total 478.15 427
Table 10: average use of the threat frame in the coverage of Golden Dawn
Newspaper Mean Standard Deviation N Kathimerini 1.88 0.96 74
To Vima 2.00 1.90 188
To Proto Thema 1.94 1.05 166
Hypothesis H3.c, a possible difference in the use of the opportunity frame among the three online Greek newspapers was tested by running a One Way Analysis of Variance (ANOVA), too. The independent variable in this case is still the three online Greek newspapers (Q3), Kathimerini, To Vima and To Proto Thema (categorical), whereas the dependent one is the use of the opportunity frame (ratio level of measurement). in the coverage of Golden Dawn by the relevant above mentioned online newspapers. The assumption of equal variances is not accepted in this case, thus the variance of the groups is assumed to be not equal (Levene’s F (2,425) =14.08, p=.000). The analysis of variance model (see Table 11), reveals that there is a significant difference in the use of the opportunity frame in the coverage of Golden Dawn by the three online Greek newspapers. (F (2,425) =4.21 p=.015). Due to the fact that the groups d differ significantly in the use of the opportunity frame a Post hoc Bonferroni test was run. Table 12 presents the average use of the opportunity frame in the coverage of Golden Dawn by each one of the three online newspapers.
Table 11: One way analysis of variance for the use of opportunity frame for GD Sum of Squares Df Means of Square F P Between groups 1.32 2 0.66 4.21 .015
Within groups 66.43 425 0.16
Total 67.75 427
Table 12: average use of the opportunity frame in the coverage of Golden Dawn Newspaper Mean Standard Deviation N Kathimerini 1.19 0.05 74
To Vima 1.07 0.02 188
To Proto Thema 1.17 0.04 166
The post hoc comparisons (Bonferroni test) revealed that the mean use of the opportunity frame by the newspaper To Vima (M=1.19, SD=0.05) was significantly different than the mean of the mean use of the opportunity frame used by the newspaper To Proto Thema (M=1.17, SD=0.04). For the rest of the groups (newspapers) comparisons the Bonferroni post hoc test did not indicate any significant differences. The results indicate that the newspaper Kathimerini makes a makes a more frequent use of the opportunity frame when covering Golden Dawn.
Discussion
The present study revealed that the use of threat frame in the coverage of Golden Dawn by Greek online newspapers is more frequently used than the opportunity over time (2012-2015). This can be explained by the fact that Golden Dawn was involved in a number of negative events in 2013 and afterwards. Namely, the murder of Pavlos Fyssas and the subsequent prosecution of the party’s MPs attracted media attention as media has a preference for conflict issues (McManus, 1995; Plasser & Ulram, 2003). Therefore the theory of agenda setting can justify that the coverage of Golden Dawn as a threat is more intense than the coverage as an opportunity. Indeed, the use of threat frame is higher compared to the other years (44.1 percent). The genre of the event could also predict the frames that were used by media when covering the far-right party. As a matter of fact, the negative event of the leaked video of Baltakos was followed by a high use of the threat frame and a decrease of the opportunity one. Media apparently covered Golden Dawn in a negative way because the party’s involvement in the incident was evaluated as a threat to democracy (To Vima, 2015).
Thus, the use of threat frame seems to be more intense in the time period that follows a negative event in which the far-right party was involved.
On the other hand, the Greek national elections where preceded by a decrease of the threat frame and an increase of the opportunity one. The theories used by other scholars confirmed that framing is more balanced during elections time (Dalton, Beck & Huckfeldt, 1998), in this case the difference between the two frames was lower prior to the Greek elections 2015 (3.6 percent) than in the other time periods (2012- 2014). It seems that Golden Dawn was framed both as an opportunity when promoting its policy proposals and as a threat by the adversary political parties. The opportunity frame though, seems to be more frequently used as elections approach. Regarding the tone of coverage there was no significant difference among the three online Greek newspapers (To Vima, Kathimerini, To Proto Thema) as expected. A possible explanation could be that all newspapers included in the sample of the study are interested in reporting negative events; hence the negative tone will prevail in all newspapers when covering Golden Dawn. Despite the political affiliation of each newspaper (see sample section), the “yellow newspaper” To Proto Thema was expected to have a more positive tone in the coverage of the Greek far-right party, but it was not confirmed by the presents study. This latter can probably be explained by the fact that several journalists of the aforementioned newspaper are located in the centre of the political spectrum. Thus, the coverage of Golden Dawn is several times objective refraining from the subjective and biased coverage.
Regarding the difference in the use of the threat and opportunity frames a difference is revealed in the second case only. The use of threat frame is similar for all newspapers and this could be attributed to the fact that the murder of Pavlos Fyssas attracted media attention on the social and political discourse about the legitimacy of Golden Dawn being a political party and especially its illegal actions (Psarras, 2013). Once again, media devotes an extensive coverage on negative events (McManus, 1995; Plasser & Ulram, 2003) and uses in this case the threat frame. Though, the use of opportunity frame differs between the centre-left newspaper To Vima and the ‘yellow newspaper’ To Proto Thema. In fact, the latter newspaper the use of the opportunity frame is more intense compared to the former newspaper that makes the least use of all newspapers. The difference in the use of the opportunity frame can be explained by the Greek media system. In this case the strong political affiliation of the Greek newspapers to the respective Greek political parties is extremely high (Hallin & Mancini, 2004). To Vima is a leftist outlet and will be consequently
more critical on Golden Dawn than To Proto Thema which will be more supportive as it espouses extreme right political ideology. The difference in the use of the above mentioned frames are also justified by the support percentage revealed by the national surveys conducted after April 2014 (after Baltakos’ issue) and in January 2015 (prior to the Greek national elections). In April the voting intention was 7.7 percent (Kaparesearch.com, 2014) whereas in January 8.3% (Kaparesearch.com, 2015), thus the coverage of Golden Dawn was more positive. In short, the present content analysis of the Greek media coverage was able to locate some differences in the use of framing regarding Golden Dawn. The Greek far-right party is portrayed mainly as a threat by the Greek online newspapers. Furthermore, the system and the political affiliation (partisanship) of the Greek newspapers were able to reveal differences in the use of both frames and explain the respective frame preference.
Those findings are important as there are no previous studies focusing on the media content as an explanatory factor for the political course and the political achievements (voters’ support and elections results) of Golden Dawn. This paper is could be the basis for any further future research in the coverage of Golden Dawn and might be used for investigating the extent of the Greek media impact on Greek voters’ preferences and behavior.
Limitations
The number of the newspapers included in the present study could be considered as a limitation, although the three included newspapers offer a picture of how Greek online newspapers function. It would be interesting and necessary for any future research to include other media outlets, too. Including more online newspapers or even news blogs as they are read by a severe percentage of Greek readers would provide a more accurate picture of the Greek media landscape. It is also important that for the threat frame to be redefined and measured by more indicators so it can be identified more easily in the media content. Moreover, an extensive focus on the visual parts of the media coverage, such as videos, could potentially contribute more to understand the intensity of frame use by paying attention to the tone of coverage. In this context the tone of coverage was found similar in all online newspapers but further and more detailed research could probably offer a more accurate picture as according to previous studies it affects the frame use.
Acknowledgements
The author would like to thank Dr. Knut de Swert for his helpful guidance and advice, and Sophia Gyftakou for her cooperation in the intercoder reliability part.
References
Altheide, D. (1997). "The news media, the problem frame, and the production of fear ". The Sociological Quarterly, 38(4), 647-668. doi:10.1111/j.1533-8525.1997.tb00758.x.
Bale, T. (2003).Cinderella and her ugly sisters: the mainstream right in Europe’s bipolarizing party systems. West European Politics, 26(3), 67-69.
Banteka, N. The Resurgence of Right-Wing Extremism in Greece: A Not So Golden Dawn. SSRN Journal. doi:10.2139/ssrn.2261417.
Bistis, G. (2013). Golden Dawn or Democratic Sunset: The Rise of the Far Right in Greece. Mediterranean Quarterly, 24(3), 35-55. doi:10.1215/10474552-2339453.
Blumler, J., & Kavanagh, D. (1999). The Third Age of Political Communication: Influences and Features. Political Communication, 16(3), 209-230. doi:10.1080/105846099198596.
Boomgaarden, H., & Vliegenthart, R. (2007). Explaining the rise of anti-immigrant parties: The role of news media content. Electoral Studies, 26 (2), 404-417.
Bosman, J., & d'Haenens, L. (2008). News reporting on Pim Fortuyn: framing in two Dutch newspapers. Media, Culture & Society, 30(5), 735-748. doi: 10.1177/0163443708094018.
Boswell, C. (2007). Migration Control in Europe After 9/11: Explaining the Absence of Securitization. JCMS: J Common Market Studies, 45(3), 589-610.
Dalton, R., Beck, P., & Huckfeldt, R. (1998). Partisan Cues and the Media: Information Flows in the 1992 Presidential Election. The American Political Science Review, 92(1), 111.
Druckman, J., & Parkin, M. (2005). The Impact of Media Bias: How Editorial Slant Affects Voters. The Journal Of Politics, 67(04).
Eatwell, R. (2000). The rebirth of the 'extreme right' in Western Europe?. Parliamentary Affairs, 53(3), 407-425.
Ekloges.ypes.gr, (2014). Greek Ministry of Interior. Retrieved 23 June 2015, from http://ekloges.ypes.gr/current/v/public/# {"cls":"main","params":{}}
Ekloges.ypes.gr, (2015). Greek Ministry of Interior. Retrieved 23 June 2015, from http://ekloges.ypes.gr/current/v/public/# {"cls":"main","params":{}}
Ellinas, A. (2013). The Rise of Golden Dawn: The New Face of the Far Right in Greece. South European Society And Politics, 18(4), 543-565.
Entman, R. (2007). Framing Bias: Media in the Distribution of Power. Journal Of Communication, 57(1), 163-173.
Entman, R. (2003). Cascading Activation: Contesting the White House's Frame After 9/11. Political Communication, 20(4), 415-432.
Ethnos, (2014).ethnos.gr, Έκθετος ο Μπαλτάκος για τη Χρυσή Αυγή. Διάλογος - σοκ με Κασιδιάρη. Retrieved 24 June 2015, from ethnos.gr/sok/Kasidiaris.
Green-Pedersen, C., & Odmalm, P. (2008). Going different ways? Right-wing parties and the immigrant issue in Denmark and Sweden. Journal Of European Public Policy, 15(3), 367-381.
Hallin, D., & Mancini, P. (2004). Comparing media systems, 89-92. Cambridge: Cambridge University Press.
Hallin, D., & Papathanassopoulos, S. (2002). Political clientelism and the media: southern Europe and Latin America in comparative perspective. Media, Culture & Society, 24(2), 175-195.
Highhouse, S., & Paese, P. (1996). Problem Domain and Prospect Frame: Choice under Opportunity versus Threat. Personality And Social Psychology Bulletin, 22(2), 124-132.
Iyengar, S., & Simon, A. (1993). News Coverage of the Gulf Crisis and Public Opinion: A Study of Agenda-Setting, Priming, and Framing. Communication Research, 20(3), 365-383. doi: 10.1177/009365093020003002
Kaparesearch.com, (2014). Pan-Hellenic National Survey April 2014. Retrieved 24 June 2015, from http://kaparesearch.com/index.php?option=com_k2&view=item&id=86:jan-15-2015-nationwide-poll&Itemid=137&lang=el
Kaparesearch.com, (2015). Pan-Hellenic National Survey 15 Jan2015. Retrieved 24 June 2015, from
http://kaparesearch.com/index.php?option=com_k2&view=item&id=86:jan-15-2015-nationwide-poll&Itemid=137&lang=el
Kathimerini.gr, (2014). Μπαλτάκος: Δεν γνώριζα ότι ο Κασιδιάρης βιντεοσκοπούσε | Kathimerini. Retrieved 24 June 2015, from /article/epikairothta/politiki/baltakos
Kriesi, H. (2012). The Political Consequences of the Financial and Economic Crisis in Europe: Electoral Punishment and Popular Protest. Swiss Political Science Review, 18(4), 518-522.
McQuail, D. (2006). Mass communication. London: SAGE Publications. 26 (3), 45-48.
Mudde, C. (2011). Who's Afraid of the European Radical Right?. Dissent, 58(4), 7-11.
Niven, D. (2001). Bias in the News: Partisanship and Negativity in Media Coverage of Presidents George Bush and Bill Clinton. The Harvard International Journal Of Press/Politics, 6(3), 31-46.
Oppermann, K., & Viehrig, H. (2011). Issue salience in international politics. London: Routledge.
Papathanassopoulos, S. (1999). The effects of media commercialization on journalism and politics in Greece. The Communication Review, 3(4), 379-402.
Rydgren, J. (2005). Is extreme right-wing populism contagious? Explaining the emergence of a new party family. Eur J Political Res, 44(3), 413-437. Saull, R. (2013). Capitalist Development and the Rise and 'Fall' of the Far-Right. Critical Sociology.
Rydgren, J. (2010). Radical Right-wing Populism in Denmark and Sweden: Explaining Party System Change and Stability. SAIS Review, 30(1), 57-71.
Saull, R. (2013). Capitalist Development and the Rise and 'Fall' of the Far-Right. Critical Sociology.
Scheufele, D., & Tewksbury, D. (2007). Framing, Agenda Setting, and Priming: The Evolution of Three Media Effects Models. Journal Of Communication, 57(1), 9-20.
Scholtz, S. (2000). Threat: Concept Analysis. Nursing Forum, 35(4), 23-29.
Stroud, N. (2007). Media Use and Political Predispositions: Revisiting the Concept of Selective Exposure. Polit Behav, 30(3), 341-366.
To Vima, (2015). tovima.gr. - Υπόγειες συνομιλίες-σοκ Π. Μπαλτάκου - Ηλ. Κασιδιάρη (video). Retrieved 23 June 2015, from http://www.tovima.gr/politics/article/?aid=582948
Toloudis, N. (2013). The Golden Dawn: The Financial Crisis and Greek Fascism's New Day. New Labor Forum, 23(1), 38-43.
Varga, M. (2008). How Political Opportunities Strengthen the Far Right: Understanding the Rise in Far-Right Militancy in Russia. Europe-Asia Studies, 60(4), 561-579.
Walgrave, S., & De Swert, K. (2004). The Making of the (Issues of the) Vlaams Blok. Political Communication, 21(4), 479-500.
Yang, J., & Grabe, M. (2011). Knowledge acquisition gaps: A comparison of print versus online news sources. New Media & Society, 13(8), 1211-1227.
Appendix Α
Table 1: Surveys of Kappa Research upon which the survey timeframe was based
Survey
Time period +2 days
N S.E. Purpose of survey
Articles for all newspapers July ’12 10-12/07/’12 1003 3,09% Pol.developments 19 Dec.‘12 18-21/12/’12 1028 3,06% Pol.developments 35 Jan.‘13 20-24/01/’13 1019 3,07% Pol.developments 61 March ‘13 18-21/03/’13 1002 3,10% Pol.developments 27 June ‘13 10-13/06/’13 1003 3,09% Pol.developments 20 July ‘13 15-18/07/’13 1002 3,09% Pol.developments 26 Nov.‘13 24-28/11/’13 1009 3,08% Pol.developments 56 April ‘14 01-03/04/’14 1003 3,09% EU elections 77
May ‘14 04-08/05/’14 1149 2,89% Electoral trends 45 Oct. ‘14 11-15/10/’14 1028 3,06% Pol.developments 21 Jan.‘15 05-08/01/’15 1010 3,10% Pol.developments 22 Jan. ‘15 Total nr. of articles 11-15/01/’15 1037 3,00% Pol.developments 64 473 Table 2: Krippendorf’s alpha Variables Krippendorf’s alpha Q4 Topic of the article .96
Negative words .85
Beneficial actions .85
Q7_2 other negative words .84
Q5 negative words 1
Q6_5 Fyssas murder 1
Q6_6 Prosecution of GD’s MPs 1
Q6_7 Attack against GD’S office 1
Q6_8 Baltakos leaked video .79
Q8_1 threat to democracy . 97
Q8_2 increasing violence between Greeks and immigrants .77
Q8_3 hate speeches of GD .87
Q10_1 fighting crime .78
Q10_2 helping Greeks .79
Q10_3 fighting against gov. austerity measures .90
Q10_4 opportunity for change in the pol. system .76
Q10_5 solution to crisis 1
Q11 alternative party .77
Q12 beneficial actions .91
Q13 fighting corruption .72
Q14 protecting Greeks from immigrants* .64
Q15 positive consequences . 81
Q16 tone of coverage .70
Q17 picture .96
Q18_1 portrayal of GD in the picture .82
Q20_1 general opinion of the article .78
* This variable scored low on Krippendorff’s alpha as it was coded even when protection was not mentioned in the article, but also implied.
Table 3: Factor analysis for the threat frame
Item Factor loading
Component 1
GD increasing violence between Greeks and immigrants 0.85 GD presented as a threat to democracy 0.82
Component 2
Table 4: Factor analysis for opportunity frame
Item Factor loading
Component 1
GD is a beneficial party 0.91 GD is helping the Greeks 0.89 GD is an alternative party 0.87 GD is an opportunity for a change in the pol. system 0.82 GD offers solution for the economic crisis 0.80 GD fights against austerity measures 0.76 Portrayal of the party in the articles’ picture 0.60
Component 2
GD is fighting crime 0.89 GD is fighting corruption 0.84
GD protects Greeks from immigrants 0.72
Appendix B
Codebook Q1. Coder:
-Eralda -other
Q2. In which newspaper was the article published?
-Kathimerini -To Vima -To Proto Thema
Q3. When as the article published? (DD/MM/YYYY)
