State and camp dependencies on framing refugees:
A content analysis of newspaper articles in Germany, Greece and the
United Kingdom
Name: Renz Ringsleben
Student ID: 10975497
Master´s Programme of Communication Science: Master Thesis
Graduate School: Graduate School of Communication
Supervisor: Rachid Azrout
Date: 24.06.2016
1
Abstract
The refugee crisis was one of Europe´s main issues in the year 2015 and still is. This paper offers findings of the framing of refugees, as victims or intruders, in newspaper articles in Greece, Germany, and the UK. Additionally, the paper explains why the particular countries use framing in a certain way. In Germany, newspapers used less intruder and victim framing. The usage of intruder and victim framing was higher in the UK and Greece. Overall intruder framing was used lesser than victim framing in all countries. Newspapers followed their liberal or conservative ideology in Greece and the UK. In Germany no such effect was visible. Tabloid newspapers in Greece and the UK use framing to a greater extent than their broadsheet counterparts.
Introduction
The refugee crisis was one of the main issues in Europe’s news media 2015 and still is. The escape routes of refugees fleeing the war in Syria, authoritarian regimes in North Africa or economic standstill and corruption in the former Yugoslavian countries are leading through the South East of Europe and its so-called transition states, like Greece, Croatia, or Austria. The final destinations are the wealthy European countries in the center and north of the continent, like Germany, Great Britain, Sweden or France. The intention to find peace and a chance for a better life makes refugees travel thousands of miles on their dangerous journey on shabby boats across the Mediterranean with the help of criminal human traffickers (United Nations, 2016). On the other hand, the influx is threatening Europe’s infrastructure massively. Authorities and institutions are pushed beyond their capacities. Citizens,
2
especially from the countryside, are confronted with new foreign neighbors, following different religious views and cultural values, which are not always easy to combine. The budget of the state is more and more needed to build housing and to support the new arrivals. Feelings of anxiety, anger and envy spread over Europe and rightwing parties know how to use these emotions for their benefits. The rise of the "Alternative für Deutschland" party in several German state parliaments, and the establishment of the right-wing parties PVV in the Netherlands or PiS in Poland on a national level reveal the development of a more right-wing political environment in Europe (Aisch et al., 2016).
The point of view shapes the opinion about social issues as the current refugee crisis. In communication science, this theory is defined as framing (Entman, 1993; McQuail, 2000; Gamson & Modigliani, 1987; Tankard et al., 1991). The previous paragraph and its two different inherent narratives are linked to the concept of van Gorp, which defines refugees either as victims or as intruders (van Gorp, 2005). I apply this concept in this paper and elaborate it in the further sections.
This paper will observe in how far the framing of the current refugee crisis is presented in different countries, with different socio-economic background, different media landscapes and different perceptions of the whole idea of Europe. Furthermore, types of newspapers and their traditional ideology will be taken into account. This study will give insights about a transnational discourse beyond country borders and offer an explanation for the framing regarding structural, historical, socio-economic reasons and newspaper characteristics.
Framing refugees as either victims or intruders in newspaper articles is dependent on several circumstances. To describe these circumstances I apply Wimmel’s concept of state and camp dependencies (Wimmel, 2009).
3
State dependencies define that framing is shaped by different perceptions of the European Union of member states as a consequence of national political cultures and public attitudes (Diez, 2001; Waever 1998, 2002: Larsen, 1999). Elite politicians and parties may build collective identities and lead the discourse with their public voice in particular directions (Marcussen, 1999). Geography and wealth also play a leading role in how a social issue is framed in the national discourse. In this paper, I observe these state dependencies on framing the social issue of the refugee crisis as socio-economic aspects. After that, also the media landscape creates attitudes about social issues (Magin, 2015). As camp dependencies of framing, I consider newspaper ideologies, which are shaped by traditional political orientations to frame social issues (Hallin & Mancini, 2004; Eurotopics, 2016). The extent of this ideological effect is state dependent and takes specific media system characteristics into account (Hallin & Mancini, 2004). Newspaper ideology is described as an interaction of camp and state dependencies. Additionally, framing depends on newspaper types. Broadsheet newspapers are opposed to tabloid newspapers and quality to sensational newspapers (McQuail, 2010; Vettehen et al., 2010; van Santen, 2008).
The societal relevance of my thesis is to show if the press is biased in portraying a social issue, which is influencing the whole continent of Europe in the long run. A biased newspaper environment in here means that the press has no balance in reporting and, therefore, a significantly dominant way of portraying refugees as victims or intruders. It is necessary to observe the coverage of issues and to ensure the diversity of media outlet content and raise awareness of a biased press frequently. It is also relevant regarding a European public sphere. Does Europe have a common public sphere, or do opinions about European issues still spread a lot between its members? The academic relevance is to involve nationally-oriented
4
reasons, like specifications of media systems, and newspaper ideology and general reasons, like newspaper types as a possible factor of influence in the framing of articles about refugees in various countries with different cultural background. It is crucial to observe if these specifications go hand in hand or differ in particular cases. Several times before, framing was only observed in a single country (van Gorp, 2005; Nickels, 2007; Roggeband & Vliegenhart, 2007). The findings of the studies were narrowed because it was not possible to measure dependencies between countries and formulate a general conclusion beyond country borders. The current refugee topic is an unprecedented new crisis and, therefore, needs observation. In the outcome, it will be possible to see to what extend framing is influenced by camp and state dependencies.
In this paper, I conduct a content analysis of 703 articles, which focuses on the print coverage of refugee-related articles in the UK, Germany, and Greece. For each country a left-wing/liberal broadsheet, a right-wing/conservative broadsheet, and tabloids will be taken into account.
Theory Framing
A lot of attention in communication science has been given to the way how news is presented or, take the words of communication research, how stories are "framed". Entman (1993, p.52) writes "To frame is to select some aspects of perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, more evaluation and/or treatment recommendation for the item described." This statement implies the perpetual incompleteness of a story. Applied on articles, which are related to the
5
current refugee crisis, this means that some aspects like the suffering on the journey, the untenable conditions in their home country could be specified. On the other hand, the massive economic burden or cultural incompatibility of religions could be stressed.
The selection of aspects might sound like a biased and narrowed way of reporting news, but contrary to this perspective framing also supplies the consumer with a central organizing idea for the content of news (Tankard et al., 1991, p.3). Journalists cannot cover stories completely. They have only the provided space available in a newspaper and therefore the task to be selective. In this process of selection, journalists deal with certain potential factors of influence. The value of news (Galtung & Ruge, 1965) plays a major role in the selection. Of course, the previously mentioned aspects of camp and state dependencies regarding the journalistic ideology and culture take their part. For the implementation process journalists seek opportunities to use tools as helpers to cut the story in shape. Hence, framing provides a nearly unambiguous idea, which establishes the essence of the issue (Gamson & Modigliani, 1987, p.143). The adoption of framing leads to an emphasis on some facts and the omission of other perspectives and details. In the journalistic context, stories are assigned by news values, which enable the reader to connect certain events to similar ones (McQuail, 2010, p.380). Overall, the frame presents the way a story is told.
Victim and Intruder Framing
Migration has been a popular subject of framing studies (Helbling, 2014; Semetko & Valkenburg, 200; Huysmans, 2006). One commonly used frame, when it comes to migration and refugees, is Helbling’s moral universal frame (2014, p.24). It
6
covers the reference to humanitarian rights and general moral principles like compassion, generosity or sense of duty. Moreover, a positive and negative economic frame needs to be applied. Here, the advantages and threats of high migration numbers towards the national economic and financial consequences are stressed (Semetko & Valkenburg, 2000, p.96). Keeping the idea of migration as a threat in mind, a security frame, which focuses on national security regarding the perceived security and fear of terrorism within a country, is also relevant (Huysmans, 2006, p.28). On the other hand, nationalism, especially in right-wing and conservative coverage is covered in the national frame (Helbling, 2014, p.24). It claims for a homogeneous society to keep the national identity, how it traditionally used to be, alive. It is opposed by the multicultural frame, which underlines cultural openness to other cultures and the ideology of coexistence (Roggenbrand & Vliegenhart, 2007, p.530). The reason why I chose van Gorp’s framing concept is that all these different frames are somehow included in van Gorp’s Codebook. The frames theoretically synthesize.
On this account, Integrated Threat Theory needs to be mentioned as an element of framing concepts regarding the refugee crisis and migration in general. It describes the perceived belonging to a particular group (in-group), which fears potential negative outcomes in intergroup interaction with another group. The group outside the in-group is called out-group (Stephan et al., 1999, p.2223). In newspaper coverage, this used concept might be identified by differentiation of "we" or "us" the good ones, and "them" the bad ones. Since intruders are out-group members, the in-group perceives several kinds of threats towards them. Victims, on the other hand, belong to the in-group and claim for support. A clear explanation of victim and intruder will be given by a set of statements in the operationalization part of the paper.
7
Additionally, Integrated Threat Theory describes realistic and symbolic threats. Realistic threats focuse on the existence of in-group members, their physical and material well being. Hence, realistic threats are included in the security frame and negative economic frame. On the other hand, symbolic threats aim at values and beliefs of and potential conflicts of worldviews (Stephan et al., p.2222). Hence, symbolic threats are part of the nationalistic frame. All of them contribute to the intruder frame. Analyzing van Gorp’s Codebook (van Gorp, 2005, p.491) step by step, we can identify these elements of Integrated Threat Theory. It is important to explain this theory because the victim is considered as member of the in-group and the intruder as member of the out-group regarding the Integrated Threat Theory.
Camp and State Dependencies
Framing depends on journalists’ and editors’ choices. The reasons why they choose to frame stories in particular ways may depend on various factors of influence. Wimmel’s (2009) theoretical framework examines the dominant impact on public discourse of European issues. He gives two options, which might shape the dominant impact: Firstly the impact differs by nation states (state dependencies), and secondly the impact lies in the interpretation of social groups like e.g. conservatives or the unions beyond state borders (camp dependencies) (Wimmel, 2009, p.224).
State dependencies: Socio-Economic Aspects
The concept of the article is to formulate the hypothesis from theories about the dependencies, which might influence the media coverage in the particular countries and the specific media outlet types. I begin with the hypothesis about the first category of state-dependent influences.
8
Explaining the difference only by media systems, economic or geopolitical interests are insufficient (Marcussen et al., 1999). Besides journalists, also politicians have the power to set the media agenda on social issues. This relationship is often described as an intimate Tango, in which the leading role is disputed (van Aelst & Vliegenhart, 2014). Therefore, the elements of political attitudes as dependencies on framing social issues need to be observed. The image of Europe influences the political discourses about social issues (Diez, 2001). Hence, also the framing in the news media is influenced by the image of Europe. It is crucial to elaborate whether and to what extent it is possible to explain state-dependent differences in national discourses by ideas of the European Union (Waever 1998, 2002; Larsen 1999).
Both countries, Germany, and the UK differ immensely in their perception of Europe (Ludlow, 2002; Kassim, 2004). Also, Wimmel used Germany/France and Great Britain as poles. Overall, this supports my selection of countries. Starting with the UK, the so-called "awkwardness thesis" describes the complicated relationship towards continental Europe and the EU. The unique history, distinctive geography, culture and social institutions shall explain these uncertainties and antipathy to any cooperation that may challenge Britain`s unique position in the European Union (Kassim, 2004, p.263). Of course, the level of scepticism has not always been steady. Whereas Tony Blair considered himself as a real European and other prime ministers also saw the benefit of the EU, the public opinion always tended to be more anti-European. This aversion towards the continent is very visible nowadays when the polls frequently showed a majority voting “yes” to leave the EU ("Brexit") from the 25th of May on (Financial Times, 2016). Anglo-American pop music was separated from worse "Euro pop", the uninspiring cars designed in continental Europe were labelled "Euro boxes" and films as "Euro puddings" (Ludlow, p.101). These are
9
indicators that UK’s newspaper coverage of the European issues of the current refugee crisis differs towards the continental coverage.
Marcussen et al. (1999) give an additional explanation for the difference in nation-state identities. Here, scholars apply the concept of "collective identity". These identities are mainly shaped by party elites, which are major actors on the political stage, which are required to justify their actions and which demand confirmation of the public and constituencies (Marcussen et al., 1999, 615). Another interesting point in here is that these national identities do shape not only in-groups but also out-groups (Marcussen et al., 616). A closer look at quotes of British and German authorities concerning the current refugee crisis stresses this effect. UK’s Prime Minister David Cameron expressed that he wants to build higher fences and bring more dogs to Calais (Segalov, 2015). Contrary, with the quote "Wir schaffen das" ("we will make it") (Hildebrand & Ulrich, 2015) from September 2015, Chancellor Angela Merkel took a very clear welcoming view towards refugees. Greek´s Prime Minister Tsipras said that his vision of Europe does not include the building of fences (TO BHMA, 2015). These examples of the statements of national authorities of the UK and Germany express in how far the handling of the issue diverge and might be guided by ideological and historical background. Merkel and Cameron are involved in shaping two collective identities, which also promote separation and out-groups. What is interesting, but difficult to determine is, whether these political attitudes can be seen as origins or followers of certain perspectives.
Nevertheless, the reasons for the different idea of Europe and aversion towards the membership of the EU stay vague and give no clear direction which framing might be dominant in certain countries. Therefore, I take a look at some facts about geography and wealth of the three countries, which are undoubted and might even
10
give a direction whether one of the countries could frame the social issue of the current refugee crisis in a certain way. As a so-called transition-state, Greece is the first contact point for refugees in Europe. The Refugee Agency UNHCR counts over one million refugees in 2015 and 2016 (UNHCR, 2016). Over and above, Greece faces an overwhelming financial crisis (ABC News, 2015). Hence, Greece is dependent on foreign aid to handle the current refugee crisis.
There is no reason to believe that Greece could care for a proper social and economic integration with this background. Since refugees are arriving in Greece but do not have the intention to stay in the country, the population has no purpose to fear economic or cultural threats in the long run. The countries, which need to facilitate the integration are Germany and the UK. Therefore exhausted refugees in overcrowded boats entering the beaches of the country will be overwhelmingly be portrayed as victims.
H1: The victim framing will be the most dominant in Greece within all countries. Germany, on the other hand, has a stable economy. It counts the highest employment rates since decades and steady economic growth (BMWi, 2016). The UK economy is a little slower but steadily growing. Services are rising robustly (Cadman et al., 2016). These numbers, the good reputation regarding freedom and rights and a migration history lead to the role as destination country of refugees. Instant aid, a long-term integration and, finally, a safe and better life seem not far-fetched. These circumstances could lead to a material loss for British and German society. People could feel threatened, and national identities, which are shaped by elites, could even strengthen these emotions. These feelings are reinforced by Cameron’s claim to build higher fences to keep the intruder out of the British Island, which already prove distinctive scepticism towards continental Europe. On the other hand, Merkel’s “Wir
11
schaffen das” implies trust in society’s ability to handle the problem. It also stresses the confidence in available resources in the country. These expressions can not only lessen the fear of intruders but flip the role of refugees in people, which should be supported. Especially by the countries, which are better off. For this reason I formulate the following hypothesis for the UK and Germany. It is important to mention that the usage of framing is observed independently. A high use of intruder framing does not necessarily imply a low use of victim framing and vice versa. H2: The intruder frame is more dominant in UK newspapers than in German newspapers.
H3: The victim frame is more dominant in German newspapers than in UK newspapers.
Interaction of camp and state dependencies: newspaper ideology
As the next set of characteristics, which might shape the framing of refugee-related newspaper coverage, I elaborate the camp dependencies. They underline the influence of social groups, which define the transnational ideological memberships by similar argumentation on social issues (Wimmel, 2009, p.224). I apply this theory to identify social groups in different transnational newspapers.
Newspapers tend to certain ideological values (Hallin & Mancini, 2004, p.27). In the long term, newspapers commonly share constant ideologies but do not stick one-to-one to party opinions about certain social issues. But what are these ideologies and what do they stand for regarding the framing of social issues? Usually, conservative and liberal ideologies polarize the media landscape, even if their validity is doubted (Alterman, 2003). Some scholars even question the existence of ideologies at all (Jost, 2006), but this does not do any abort to the categorization of newspapers
12
to these values (Eurotopics, 2006). Therefore, it is necessary to explain per definition of liberal and conservative, which framing they are likely to apply. Farewell and Weiner (2000) title their study about popular perceptions of liberals and conservatives “Bleeding hearts and the heartless”. It stresses the common stereotypes of both ideologies. Furthermore, liberalism emphasizes equality and therefore the high standing of the community, which refers to the idea of the in-group. On the other hand conservative values stress the importance of traditional ideas and structures, which do not advocate a change of society (Merriam Webster, 2016). According to the current refugee crisis, this would imply a more frequented use of intruder framing in the conservative press and more victim framing in the liberal press.
It was stated that newspaper ideology matters regarding framing the refugee crisis. Furthermore, the extent of in how far newspapers follow their ideologies in different countries needs to be discussed. Therefore, the theories of media systems (Hallin & Mancini, 2004) are relevant.
The most important factor here is political parallelism towards the media in particular country systems. The term “political journalist” describes someone, who interprets his role in the triangle of media, politics and the public as an influencer concerning certain political ideas (Hallin & Mancini, 2004, p.26). Consequently, a newspaper, which interprets stories always in a certain pattern, has the tendency to drift to particular political ideology, when the reporting goes beyond printing facts and figures, which is usually the case.
The first aspect of political parallelism is in what extent media content reflects different political orientations. Organizational connections between media and political parties, or institutions, which are closely connected to the parties, and the tendency of media workers to be engaged in politics were only relevant in historical
13
components of political parallelism. But, the choice of media institution by political affiliation as an employee stays relevant. The audiences or readers need to be satisfied. Therefore, their partisanship needs to be considered as part of political parallelism as well (Hallin & Mancini, 2004, p.28). In the end, national journalistic role orientations and practices, which influence the role of reporting biased or more neutral, underline the extent of political parallelism. It involves the writing habits, e.g. more writing style, of different nations. (Hallin & Mancini, 2004, p.29)
Accordingly, Hallin and Mancini label two types of pluralism. External pluralism defines pluralism on a level of media systems as a whole. It describes the scenario in which several outlets express the public opinion. It contributes to high political parallelism. On the other hand, internal pluralism names pluralism within media outlets. It underlines the avoidance of institutional boundaries and argumentation convergence. It refers to the balance of reporting and neutrality. It contributes to low political pluralism (Hallin & Mancini, 2004, p.30).
The Mediterranean or Pluralist Model, where Greece belongs to, has an elite-oriented press, which is highly focused on political issues. It involves a high external pluralism and is instrumentalized by the government. Strong ties between media and parties are common. The active state plays a significant role as owner, regulator, and funder of the press (Hallin & Mancini, p.73/74). Since press freedom is not very developed yet and the parties have still strong ties to the media, it is appropriate to expect the ideology effect to be strongest in Greece.
In the North Atlantic or Liberal Model, the UK plays a unique role. It has a highly partisan press, whereas other countries within this system have a more independent press and low political parallelism. It is highly commercialized, and journalistic freedom is commonly more limited through commercial pressures (Esser
14
and Matthes, 2013). In Britain, it is more political instrumentalization, which applies pressure and is underlined in an ongoing commentary tradition (Hallin & Mancini, p.75). Since the UK shares many aspects of Greece concerning political parallelism and external pluralism, but the freedom of the press is higher developed I expect the ideology effect to be still present, but to a lesser extent than in Greece.
The North/Central European or Democratic Corporatist Model, where Germany belongs to, has a history of strong party newspapers, which are more and more vanishing. The focus nowadays lies in information-oriented journalism. Press freedom coexists with strong state dependence. It is characterized by moderate pluralism (Hallin & Mancini, p.74/75). In Germany, Die Sueddeutsche Zeitung is listed as center-left, and Die Welt am Sonntag more to the right. Nevertheless, it does not support the thesis, which social democrats policies are backed by Die Sueddeutsche Zeitung or the Christian Democrats by Die Welt am Sonntag. It does not show a one-to-one similarity of policy opinions, but general political tendencies (Hallin & Mancini, 2004, p.27). Since the government is led by a member of a conservative party, these exceptional circumstances and the moderate pluralism could make newspapers cover the topic more balanced within one newspaper outlet. Therefore, I expect the ideology effect to be the weakest in Germany and the coverage to be more balanced within one outlet.
Finally, as an interaction of camp dependencies, regarding the influence of newspaper ideology’s on framing and the different extent of this ideology effect between countries I formulate the following hypothesis.
H4: The ideology effect will be present in Greece and the UK and not present in Germany.
15
Newspaper Types
Not only newspaper ideologies will influence the framing of social issues, but also the newspaper type plays a significant role in portraying stories in specific ways. The two contrary types are broadsheets or so-called quality newspapers and tabloids. Tabloid journalism expresses a populist nature. (Örnebring & Jönnsson, 2004, p.292) Since the examples in the introduction underline the high impact of rightwing populism and the populist movement in Europe today, a populist voice claims nothing but right-wing ideologies.
Marvin Kalb describes tabloid news as "a downgrading of hard news and upgrading of sex, scandal and infotainment" (Kalb, 1997). "Infotainment" names a mixing of information with drama and excitement (Esser & Matthes, 2013, p.171). Örnebring and Jönsson state that tabloids combine sensationalism and emotionalism (Örnebring & Jönnsson, 2004, p.287). This argumentation is also supported by numerous scholars (McQuail, 2010; Vettehen et al., 2010; van Santen, 2008) and comes along with the ongoing process of commercialization of the media (Esser & Matthes, 2013, p.182). It is sufficient to state that an upgrade of dramatization and emotionalism in tabloid towards broadsheet newspapers is present. In combination with tabloid’s populist voice, this theoretical background leads to the assumption that the ideology effect of right-wing/conservative outlets is stronger in tabloids than in broadsheets.
H5: The ideology effect will be more dominant in right-wing/conservative tabloid newspapers than in right-wing/conservative broadsheet newspapers.
16
Methodology
Media content from the period of July 1, 2015 until April 24, 2016 was analyzed. The analysis contained three newspaper outlets weekend edition in three different European countries, whereas one was considered to be more left, one more conservative and one tabloid, which was more conservative.
The German newspaper sample consists of Die Sueddeutsche Zeitung (broadsheet, left-liberal), Die Welt am Sonntag (broadsheet, conservative), and BILD am Sonntag (tabloid, conservative). The search term was “Flüchtlinge OR Geflüchtete OR Asyl-Suchende”. The sample in the UK contains The Observer (broadsheet, liberal), The Sunday Telegraph (broadsheet, left-liberal), and the Mail on Sunday (tabloid, conservative). To analyze the true intention of the newspaper, we tried to avoid to code to many event-driven articles. Since weekend editions are providing more background articles in an essayistic style and are less event-driven by covering the daily news beat, this sampling was appropriate.
The search term was “Refugee OR Asylum-Seeker”. In Greece the newspapers were Kathimerini (broadsheet, center-right), Efimerida ton Syntakton (broadsheet, center-left) Proto Thema (tabloid, right). The search term was “πρόσφυγες ή προσφυγική κρίση”. As search tool, we mainly used LexisNexis database. For the BILD am Sonntag and the Greek newspapers, we had to officially purchase the availability of the articles by the publisher. The website Eurotopics offers a categorization of political ideologies of nearly all important newspapers in Europe based on observation of foreign correspondents. (Eurotopics, 2016)
We chose to assign articles randomly with at least one article of each weekend outlet, if this was possible. In The Sunday Telegraph 80 articles were coded out of 117 articles, with at least one paragraph about the current refugee crisis. In The
17
Observer 76 articles were coded in total. We found 153 articles about the relevant topic. The Mail on Sunday and Daily Mail were gathered on LexisNexis. As a sum, we found 898 hits for the search term for the weekly and weekend edition. 132 articles were found to have the current refugee-crisis as a main topic. In the end, 74 articles were coded. Since, the Mail on Sunday did not contain enough articles, we added 8 Daily Mail Saturday editions.
In Die Welt am Sonntag 327 articles had the current refugee crisis as the main topic. 78 of them were coded. In Die Sueddeutsche Zeitung, overall, 122 articles were about the current refugee crisis. We coded 79 articles. BILD am Sonntag was only available by hard copy in the newspaper archive of the “Staatsbibliothek zu Berlin”. I scanned the newspaper manually and found 99 articles about the current refugee crisis. Finally 75 articles were coded. Again, the articles were randomly assigned.
The Greek newspapers were not available in LexisNexis database. After contacting the newspapers’ offices, we were offered free access to the online archive of Kathimerini. Here we found 223 articles about the refugee crisis. Overall 76 articles were coded. Because of time restriction we did not manage to find a similar solution for the other newspapers on time and Efimerida ton Syntakton and Proto Thema were bought from the online platform www.readpoint.com. The terms used for the search of the articles were ‘πρόσφυγες ή προσφυγική κρίση’, which are the equivalent of “refugees OR refugee crisis” in Greek. In Efimerida ton Syntakton we found 146 articles about the current crisis. We coded 77 of them in total. In Proto Thema we found 94 articles about the current refugee crisis and coded 75 articles.
In total, we found 1413 articles about the current refugee crisis and coded 690 articles by applying the codebook, which can be found in the appendix.
18
Operationalization
The first part of the codebook contained general information about the newspaper and article. It started with the Name of the coder, Name of the newspaper, Date of publishing and serial number. After that the publication size was asked. We estimated four categories, which clustered the size of an article in words for articles coded on LexisNexis and on page filling, for the hard copies. The clusters were labelled a quarter page for articles with up to 200 words, a quarter to half a page with up to 500 words, half a page to a full page with up to 1,000 words and more than a page with over 1,000 words. We also distinguished between the types of publications. We estimated pure news (including only headlines), background analysis, interviews and opinion pieces. Afterwards, we asked about the sources in the article. We coded up to three sources and categorized them as local, national, or international politicians, experts, civil society representatives, journalists, citizens and refugees. If the source was a national politician we asked whether he was incumbent, opposition, or not determinable. After that we started with the framing section. Firstly, we asked on a subjectivity scale if the framing in the articles was perceived as victim and/or intruder frame. We used a three point Likert scale with 0 as not present, 1 as present, and 2 as very present. After that we conducted a objectivity scale in terms of van Gorp’s codebook, in which a sample of broad framing theories lead to a set of specific questions, which identifies whether a victim or intruder framing is applied in the article.
We formulated 11 statements. In the beginning, it is important to ask whether the refugees are defined as passive people in need of help (V1). The second question contributing to the victim frame is if people are fleeing to Europe because of violence, persecution or poverty (V2). Moreover, it is asked for the moral duty to help people in
19
affliction calls for compassion (V4). All of these three statements imply Helbling’s moral universal frame and contribute to the first indication of a victim framing. The next statement on the victim scale is that the solution to solve the crisis lies in a humane and flexibly, cautiously and effectively applied asylum policy (V3). Here elements of Roggenbrand & Vliegenhart's multicultural frame and Semetko & Valkenburg´s positive economic frame are taken into account. Agreeing to this statement means to agree to migration and to appreciate an organized implementation policy and potential long-term benefits for the economy. Finally, it is asked for metaphors like shelter, a helpless and anxious victim (moral universal frame) (V5).
On the other hand, the solution might lie in the deportation of refugees, cutting benefits and a discourse towards discouragement (I3). Concluding this statement, a control question, whether refugees are defined as active adventurers or criminals who resort to illicit illegal practices (I1) is applied. In both questions a negative economic frame (Semetko & Valkenburg, 2000, p.96) is included, which condemns illegal refugees’ application for state support. Additionally, Huysmans’ security frame is intrinsic and illustrates the realistic threat of citizens wellbeing and economic disadvantage. Even more specifically, two questions aim at aspects of realistic and symbolic threat in terms of the Integrated Threat Theory. "Refugees are a threat to our culture and achievements and social devices", which contains Helbling’s nationalistic frame (I2a) and "Refugees are a threat to our economy and infrastructure" (I2b), which includes the negative economic frame. Also, a statement about the intention to protect the interest of native people and calls for aversion and distrust (I4) is raised and covers the nationalistic frame and is referring to the in- and out-group element. Finally, it is asked for metaphors like dangerous stranger, human flood, hunting, fortress Europe (security frame) (I5).
20
Overall, the victim framing will consist of six statements, including the subjectivity questions. The Intruder framing will consist of 7 items, because one question about the potential threats is divided in symbolic and realistic threats. The full questionnaire can be found in the appendix. In the end, the coding of the question gives a clear tendency if a newspaper article is framed in a certain way.
For the purpose of analysis, the data will be viewed as dichotomous. The value 1 meaning “not present” both options 2 and 3 were coded as “present”. This operation was done for all seven items regarding content suggesting refugees are an intruder and for 6 items regarding content suggesting refugees as victims. This operation was done for all Variables contributing the victim and intruder frame (Q19-Q21_13). Afterwards, I created a variable to measure the total intruder score. The same was applied for the variable regarding the victim score. Afterwards I create a filter variable to filter newspapers by country, by newspaper ideology and by newspaper type. After that, all hypothesis were tested.
Inter-coder-reliability-calculations
To check the inter-coder reliability a series of tests was conducted. I checked the level of inter-coder agreement (% agreement), the Kappa value and then the Krippendorffs Alpha (KAlpha) for all variables, which were not general information starting with the type of publication. Furthermore, the subjective framing scale and all objective statements were tested. The number of coders was two. The full sample size was 703 and the reliability test size is 64. The measure method for all three variables was nominal. As for the set of statements the tendency of victim or intruder framing is crucial, I merged 2=”a little” and 3=”yes” as 2. Generally the level of inter-coder agreement shall be higher than 80% (Lombard et al, 2010). The probability
21
accounting measures (Kappa, KAlpha) should be higher than 0,67. All results, except of the question about moral duty and compassion, are greater than the limit values. I did not calculate the values for the second and third source, because the way of coding is the same. Even if the question about moral duty was underneath the limits the total of 10 more Objectivity Scale statements makes this finding negligible. All results can be found in table 1 in the appendix.
Descriptive Variables
Overall, 11,9% of all 703 articles coded were pure news with less words than 200. 23.,5% of the articles had up to 500 words. 32,6%, and therefore the highest amount of articles consisted of 500 to 1,000 words. Furthermore, 29,3% of the articles had more than 1,000 words. In total 9,7% of the articles were pure news. The majority with 64,6% were background analyses. 7,1% were interviews and 15,6% were opinion pieces. The top three sources were journalists with 18,2%, national politicians with 14,8% and international politicians with 11,1%. These variables were not relevant for the testing of the hypothesis, but worth to mention, as they give an idea about the samples’ general characteristics.
Analysis
Since a scale variable was observed, a one-way analysis of variance (ANOVA) test for the total score of victim and intruder framing in country groups was appropriate. To answer the hypothesis H1, H2 and H3 a comparison between countries and not only to the overall Means needed to be observed.
There was a statistically significant difference between countries regarding the usage of victim framing in newspaper articles as determined by one-way ANOVA (F(2,700) = 8,836, p =0) (Table 4). Germany showed the lowest score in victim
22
framing in the articles (M = 1,79, SD = 1,599). The UK was presenting the highest victim framing score in the sample of articles (M = 2,47, SD = 2,029). Greece scored in the middle between Germany and the UK by using the victim framing in refugee– related articles (M = 2,27, SD = 1,797) (Table 2). To avoid the accumulation of the alpha error, we applied the Bonferroni correction, when carrying out the post hoc test. (Table 6)
There was an also statistically significant difference between countries regarding the usage of intruder framing in newspaper articles as determined by one-way ANOVA (F(2,700) = 6,673, p = 0,044) (Table 9). The results showed the same order as in the victim framing. Germany showed the lowest score in intruder framing in the articles (M = 0,76 SD = 0,81). The UK was presenting the highest intruder framing score in the sample of articles (M = 1,06, SD = 0,097). Greece scored in the middle between Germany and the UK by using the intruder framing in refugee– related articles (M = 1,03, SD = 1,613) (Table 7). To avoid the accumulation of the alpha error, we applied the Bonferroni correction, when carrying out the post hoc test. (Table 11)
H1: The victim framing is the most dominant in Greece within all countries.
The test results referring Greece were twofold: The positive difference with respect to Germany was significant at a 1,7% level (p=0,014). The negative difference in comparison with the UK was insignificant at all common significance levels (p=0,678). Therefore the hypothesis was partly rejected and partly confirmed. All results of the post hoc test can be found in Table 6.
23
H2: The intruder frame is more dominant in UK newspapers than in German newspapers.
The outcome of the test procedure indicated that there are no significant differences between Germany, the UK and Greece referring the potential picture of refugees as intruders. The F-Test of the ANOVA gives the impression of significant differences between the countries (Table 9). Nevertheless, the approach is not completely applicable in our case due to heteroscedasticity (Levene-Test: p-value = 0,001) (Table 8). Instead of the classic ANOVA we need to apply the rank-based ANOVA, which showed that country-specific differences are not present (p = 0,098) (Table 12, Table 13). Furthermore, all post-hoc-tests do not show any disparities between the countries mentioned, confirming the result of the rank-based ANOVA (Table 11).
H3: The victim frame is more dominant in German newspapers than in UK newspapers.
The test results show exactly the opposite direction. The victim frame is less dominant in Germany than in the UK (Difference: -0,686, p=0) and Greece (Difference: -0,482, p=0,014). (Table 6)
H4: The Ideology effect will be present in Greece and the UK and not present in Germany.
To test H4 the newspaper ideologies per country and its usage of victim and intruder framing were observed. Again, a one-way ANOVA was conducted.
In Germany, There was no statistically significant difference between ideologies regarding the usage of victim framing in newspaper articles as determined by
one-
24
way ANOVA (F(2,230) = 1,475, p = 0,231) (Table 16). The post hoc Bonferroni test confirmed the ANOVA (Table 18).
German liberal left broadsheets showed the highest score of victim framing in the articles (M = 2,04, SD = 1,516). The conservative broadsheet was presenting the medium victim framing score in the sample of articles (M = 1,66, SD = 1,635). German conservative tabloids scored a lowest by using the victim framing in refugee– related articles (M = 1,65, SD = 1,634) (Table 14).
There is no ideology-effect at all in German newspaper outlets with respect to the usage of victim framing. The post hoc Bonferroni test confirmed the results of the ANOVA.
There was again no statistically significant difference between ideologies regarding the usage of intruder framing in newspaper articles as determined by one-way ANOVA (F(2,230) = 2,761, p = 0,065) (Table 21). The post hoc Bonferroni test confirmed the ANOVA (Table 23).
German liberal left broadsheets scored lowest in using intruder framing in the articles (M = 0,54, SD = 1,041). The conservative broadsheet was presenting the highest average score in intruder framing in the sample of articles (M = 1,433, SD = 0,163). German conservative tabloids scored medium by using the victim framing in refugee–related articles (M = 0,73, SD = 1,181). (Table 19)
There is no ideology-effect at all in German media landscape, with respect to the usage of intruder framing. The post hoc Bonferroni test confirmed the results of the ANOVA.
Since there is no ideology effect for both perspectives, it is to state that the reporting about refugees in Germany is very neutral. It confirms one part of the hypothesis.
25
In the UK, there was statistical significance in the difference of the extent of victim framing in UK newspaper articles (F(2,239) = 46,211, p = 0) (Table 26). The post hoc Bonferroni test confirmed the ANOVA (Table 28).
The UK conservative right broadsheets showed the lowest score of victim framing in the articles (M = 1,74, SD = 1,839). The liberal left was presenting by far highest usage of victim framing in the sample of articles (M = 4,07, SD = 1,803). The UK conservative tabloids scored medium by using the victim framing in refugee– related articles (M = 1,77, SD = 1,512). (Table 24)
There was again a statistically significant difference between ideologies regarding the usage of intruder framing in newspaper articles as determined by one-way ANOVA (F(2,239) = 12,741, p = 0) (Table 31). The post hoc Bonferroni test confirmed the ANOVA (Table 33).
The UK conservative right broadsheets scored medium in using intruder framing in the articles (M = 0,87, SD = 1,432). The liberal left broadsheet was presenting the lowest average score in intruder framing in the sample of articles (M = 0,61, SD = 1,251). UK conservative tabloids scored highest by using the victim framing in refugee–related articles (M = 1,73, SD = 1,603) (Table 29).
Overall, the ideology effect is clearly to see in both directions. In the UK the liberal left press frames refugees as victims and the conservative right press frames refugees as intruders.
In Greece, there was a statistically significant difference between ideologies regarding the usage of victim framing in newspaper articles as determined by one-way ANOVA (F(2,225) = 55,992, p = 0) (Table 36). The post hoc Bonferroni test confirmed the ANOVA (Table 38).
26
Greek conservative right broadsheets showed the medium score of victim framing in the articles (M = 2,76, SD = 1,704). The liberal left broadsheet was presenting the highest victim framing score in the sample of articles (M = 3,19, SD = 1,531). The Greek conservative tabloids scored lowest in using the victim framing in refugee–related articles (M = 0,81, SD = 1,123). (Table 34)
Like in the UK, there is an ideology-effect appearing in all Greek newspapers, with respect to the usage of victim framing.
In Greece, there was again a statistically significant difference between ideologies regarding the usage of intruder framing in newspaper articles as determined by one-way ANOVA (F(2,225) = 46,865, p = 0) (Table 41). The post hoc Bonferroni test confirmed the ANOVA (Table 43).
The Greek conservative right broadsheets scored medium in using intruder framing in the articles (M = 0,57, SD = 1,011). The liberal left broadsheet was presenting the lowest average score in intruder framing in the sample of articles (M = 0,29, SD = 0,646). Greek conservative tabloids scored highest by using the victim framing in refugee–related articles (M = 2,27, SD = 2,042).
Overall, the ideology effect was clearly to see in both directions. In the Greek, the liberal left press frames refugees as victims and the conservative right press frames refugees as intruders.
H5: The ideology effect will be more dominant in rightwing/conservatives than in rightwing/conservative broadsheets.
To answer H5 we had to go back to recap the post hoc tests of victim and intruder framing of the UK and Greece. Firstly, we observed the UK victim framing. Both, the UK conservative right broadsheet and tabloid speak with one voice
27
(Difference in Means: -0,022) (Table 28). On the other hand, the difference in intruder framing was significant for both newspaper types (Difference in Means: -0,861) (Table 33). Therefore, the tabloid newspapers in the UK fill much the role of the intruder frame as the broadsheet newspapers.
Secondly, Greek conservative right broadsheet and tabloids were observed. For the victim framing the difference in here is significant (Difference in Means: 1,95) (Table 38). Also, the conservative right tabloid uses significantly more intruder framing than the broadsheet (Difference in Means: 1,701) (Table 43). It is to state that, again, the intruder framing is more commonly used in Greek tabloids.
Generally, it was possible to confirm H5. The only exception, in which the ideology-effect disappeared was regarding the victim in the UK.
Discussion
In this paper, a set of hypotheses is observed. In the beginning, the central question is, which framing, victim or intruder, is the most dominant in Greece, the UK, and Germany. Secondly, the dependence of newspaper ideology on framing usage in each country is conducted. Thirdly, the dependence of newspaper types on framing is elaborated. To take the set of five hypotheses to a test a content analysis of newspaper articles in all three countries is applied. In each country, one left-wing/liberal broadsheet, one wing/conservative broadsheet, and one right-wing/conservative tabloid are included.
By testing, if the victim framing is the most dominant in Greece within all countries, the result is twofold. Whereas the victim framing is more present in Greece than in Germany, the difference in using the victim framing in its newspaper articles about the refugee crisis in the UK is unremarkable. This hypothesis bases on the
28
assumption that Greece is a transition state, which faces immediate issues, but has not to bear the integration of millions of refugees in the next decade. It seems that this effect is only valid for the difference to Germany, which, as destination country with an open border policy has to manage the integration. As mentioned in the example of Calais, the UK, as an island, is keeping the refugees at the tunnel entry on the French side of the Channel. Taking in more refugees regarding the crisis is not committed yet. Hence, no additional burden of integration is predicted. In that manner, the results mirror the real life in policy decisions about commitments of integration efforts.
Cameron’s statements about the aggressive defense of the UK borders were indicators for the second hypothesis. It states that the intruder framing is more dominant in the UK than in Germany, where Chancellor Merkel uses a soft tone towards refugees entering the country. Viewing the results, this was not confirmed. Overall, the intruder framing appears very rarely in newspaper coverage in all three countries, except the Greek right-wing tabloid. Since van Gorp (2005, p. 498) also reveals only two of eight newspapers in Belgium use more intruder than victim framing, there is reasonable ground to assume a common public sphere of an unpopular intruder frame in Europe. Newspapers with a great extent of intruder framing might just do lead to high sales numbers, and therefore newspaper organization bound to shareholder interest do not use it (McManus, 1995). Further research could be applied by observing more countries regarding its newspaper coverage or comparing more results of articles about the topic.
On the other hand, Germany, which should have the most compassionate and welcoming attitude towards refugees regarding the statements of the government, which can shape collective identities (Marcussen, 1999) reveals the exact opposite. Germany uses less victim framing than the UK. The nation-state theory seems to be
29
rejected. Cameron is not able to manipulate the voice of the press. Also, the comparable economic wealth does not appear to unify the victim framing usage in both countries. Since there is also a higher use of intruder framing in Germany, one could argue that the whole discourse is more moderate in Germany, and the narrative stays more neutral than in the other countries. The balance of state regulation and press freedom is confirmed (Hallin &Mancini, 2004). There could be a different overall style of reporting in Greece and the UK as it is in Germany. Here, the greater extent of framing could be pushed by high economic pressure in the private newspaper sector (Esser and Matthes, 2013). The paper offers additional insight about a cross-country comparison in the usage of victim and intruder framing towards the studies, which only observe one country (van Gorp, 2005; Nickels, 2007; Roggeband & Vliegenhart, 2007). Also, multiple potential explanations by geography and wealth, the European idea and the attitudes of the governmental force are taken into account.
Moreover, the ideology effect in the countries was analyzed. A hypothesis that the ideology is present in the UK and Greece and not present in Germany is defined. Beyond all articles about the framing habits of newspapers of refugees in several countries (Helbling, 2014; Semetko & Valkenburg, 2000; Huysmans, 2006; van Gorp, 2005; Nickels, 2007; Roggeband & Vliegenhart, 2007) this paper is the first, which gives an additional explanation in how far the ideology effect is mediated by the media system characteristics. As we see in the analysis, this effect is dependent on the external or moderate pluralism of the media. In the Mediterranean (Greece) and the UK as an exception in the Liberal Model, the press is characterized by external pluralism and high party parallelism (Hallin & Mancini, 2004). We observe that the framing regarding the assumed direction of the newspaper ideology is much more extensive in the UK and Greece than in Germany. It seems that the culture of
30
information-oriented journalism in Germany overrules the ideological values of the newspaper outlets with its moderate pluralism. On the other hand, in the UK and Greece this effect is still appearing for both ways of portraying refugees as victims and intruders.
Finally, we check the hypothesis if the ideology effect will be more dominant in right-wing/conservative tabloid newspapers than in right-wing/conservative broadsheet newspapers. The hypothesis is applied by the assumption, that tabloid news express a populist nature (Örnebring & Jönnsson, 2004; McQuail, 2010; Vettehen et al., 2010; van Santen, 2008), which underline conservative values and, therefore, a high usage of intruder framing. Furthermore, the downgrading of hard news and upgrading of scandal is taken into account. Also, the term of “infotainment” as an indicator for a higher usage of framing to provide a more thrilling experience than reading neutral news fits in the hypothesis (Esser & Matthes, 2013, p.171). Since we show that there is no ideology effect in Germany, we only take a look at Greece and the UK. The findings reveal that the role of refugees as intruders is more fulfilled by the tabloids than by broadsheet newspapers in the UK, whereas the victim framing is used to the same extent. In Greece, the victim and intruder framing is higher in tabloids than in the broadsheet versions. Three of four possibilities are confirmed. Especially the intruder style seems to fit the entertaining criteria of the tabloid. More drama, whether it is suffering victims or aggressive intruders can be confirmed.
Limitations
The content analysis was limited in the time span. The period from July 1, 2015 until April 24, 2016 might be too short to cover a topic as complex as the refugee crisis, whose overall perception is changing all the time. Also, the number of
31
articles N=703 seems minuscule. As we had to choose articles randomly to cover the whole period in some outlets, important information might be missing. There were times of refugees filling a lot of newspaper space, and there were times when refugees were not covered very often. A longer period should be observed in the future to make the findings more reliable. Especially the time after the peak of a vast number of refugee-related news could be attractive to the applied framing in articles. How do newspapers frame when the refugees entering Europe become normality?
To involve three countries of different media systems and different socio-economic background in the research was not always easy to apply. It is hard to make reliable assumptions why the framing was used in a certain way and to what extent. Multiple potential reasons can be offered to give explanations. On the other hand, for testing the ideology effect, the country sample was appropriate. For further research, it is recommended to test H1, H2, and H3 independently and only observe one state at a time.
The newspaper sample missed a left-wing/liberal tabloid. This type of newspaper is not available in Greece and Germany. Nevertheless, an addition of a liberal left tabloid in UK could have underlined the newspaper type hypothesis.
References
ABC News. (2015). Fact file: The Greek debt crisis explained. Retrieved June 15, 2016, from http://www.abc.net.au/news/2015-07-20/fact-file-the-greek-debt-crisis/6621376
Aisch, G., Pearce, A., & Rousseau, B. (2016). How Far Is Europe Swinging to the Right? Retrieved June 16, 2016, from
http://www.nytimes.com/interactive/2016/05/22/world/europe/europe-right-wing-austria-hungary.html?_r=0
32
Alterman, E. (2003): What Liberal Media? The Truth About Bias and the News. Basic Books.
BMWi- Bundesministerium für Wirtschaft und Industrie. (2016). Die wirtschaftliche Lage in Deutschland im Mai 2016. Retrieved June 15, 2016, from
http://www.bmwi.de/DE/Presse/pressemitteilungen,did=766634.html
Cadman, E., Bernard, S., & Pearson, T. (2016). 44 charts that explain the UK economy. Retrieved June 15, 2016, from
https://ig.ft.com/sites/numbers/economies/uk
Diez, T. (2001). Europe as a Discursive Battleground: European Integration Studies and Discourse Analysis. Cooperation and Conflict 36 (1), 5–38.
Entman, R. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication 43 (4), 51–58.
Eurotopics. (2016). Medien | eurotopics.net Retrieved June 15, 2016, from http://www.eurotopics.net/de/142186/medien
Farewell, L. & Weiner, B. (2000). Bleeding hearts and the heartless: Popular perceptions of liberal and conservative ideologies. Personality and Social Psychology Bulletin.
Financial Times (2016). Brexit poll tracker. Retrieved June 15, 2016, from https://ig.ft.com/sites/brexit-polling/
Galtung, J. & Ruge, M. (1965) The structure of foreign news: the presentation of the Congo, Cuba and Cyprus crises in four Norwegian newspapers. Journal of International Peace Research 1, pp. 64–91.
Gamson, W. A., and Modigliani, A. (1987). The Changing Culture of Affirmative Action. In Research in Political Sociology 3, ed. R. D. Braungart. Greenwich, CT: JAI Press: 137– 77.
Hallin, D. C., and Mancini, P. (2004). Comparing media systems: Three models of media and politics. Cambridge: Cambridge University Press
Helbling, M. (2014). Framing immigration in western Europe. Journal of ethnic and migration studies, 40(1), 21-41.
Hildebrandt, T., & Ulrich, B. (2015). Im Auge des Orkans. Retrieved June 15, 2016, from
33
Huysmans, J. (2006). The politics of insecurity: fear, migration and asylum in the EU. London: Routledge.
Jost, J. T. (2006). The end of the end of ideology. American Psychologist, Vol 61(7), Oct 2006, 651-670.
Kalb, M. (1997) Editorial: The Clock is Ticking. Intellectual Capital. Com 2(1) at http://www.intellectualcapital.com/issues/97/0102/index
Kassim, H. (2004). The United Kingdom and the Future of Europe: Winning the Battle,Losing the War. Comparative European Politics 3 (2), 261–281. Larsen, H. (1999). British and Danish European Politics in the 1990s: A Discourse
Approach. European Journal of International Relations 5 (4), 451–83.
Lombard, M. (2010, October 1). Inter-coder reliability. Retrieved January 30, 2016, from http://matthewlombard.com/reliability/
Ludlow, P. (2002). Us or Them? The Meaning of Europe in British Political Discourse. In: Mikael af Malmborg and Bo Stråth (eds.). The Meaning of Europe. Variety and Contention within and among Nations. Oxford: Berg Publishers, 101–124.
Magin, M. (2015). Shades of Mediatization Components of Media Logic in German and Austrian Elite Newspapers (1949–2009). The International Journal of Press/Politics, 1940161215595944.
Marcussen, M., Risse, T., Engelmann-Martin, D., Knopf, H.J., and Roscher, K. (1999). Constructing Europe? The Evolution of French, British and German Nation State Identities. Journal of European Public Policy 6 (4), 614–633. Matthes, J. and Esser, F. (2013). Mediatization effects on political news, political
actors, political decisions, and political audiences. Palgrave.
McManus, J. (1995). A Market-Based Model of News Production. Communication T heory, p.301-338.
McQuail, D. (2010). McQuail's Mass Communication Theory. 6th Edition. London: Sage Publications.
Merriam Webster. (2016). Def. conservatism. Retrieved June 21, 2016, from http://www.merriam-webster.com/dictionary/conservatism
Merriam Webster. (2016). Def. liberalsim. Retrieved June 21, 2016, from http://www.merriam-webster.com/dictionary/lieberalism
34
Örnebring, H., & Jönnsson, A. M. (2004). Tabloid journalism and the public sphere: A historical perspective on tabloid journalism. Journalism Studies, 5(3), 283-295.
Roggeband, C., & Vliegenthart, R. (2007). Divergent framing: The public debate on migration in the Dutch parliament and media, 1995–2004. West European Politics, 30(3), 524 548.
Segalov, M. (2015, July 31). David Cameron to use 'more dogs and fences' to tackle 'unacceptable' Calais migrant crisis – and warns it will last all summer. Retrieved June 15, 2016, from http://www.independent.co.uk/news/uk/home- news/calais-migrant-crisis-live-david-cameron-warns-crisis-will-last-all-summer-10430199.html
Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A content analysis of press and television news. Journal of communication, 50(2), 93-109.
Stephan, W. G., Ybarra, O., & Bachman, G. (1999). Prejudice toward immigrants. Journal of Applied Social Psychology, 29(11), 2221–2237.
Tankard, J., Hendrickson, L., Silberman, J., Bliss, K., & Ghanem, S. (1991). Media frames: Approaches to conceptualization and measurement. Paper presented to the Association for Education in Journalism and Mass Communication, Boston.
TO BHMA. (2015). PM Tsipras. Retrieved June 21, 2016, from http://www.tovima.gr/en/article/?aid=751546
UNHCR Refugees/Migrants Emergency Response - Mediterranean. (2016, June 12). Retrieved June 15, 2016, from
http://data.unhcr.org/mediterranean/country.php?id=83
United Nations. (2016). Global Issues at the United Nations. Retrieved June 16, 2016, from
http://www.un.org/en/globalissues/briefingpapers/refugees/overviewo fforceddisplacement.html
Van Aelst, P. & Vliegenhart, R. (2014). Studying the Tango, Journalism Studies, 15:4, 392-410.
Van Gorp, B. (2005). Where is the frame? Victims and intruders in the Belgian press coverage of the asylum issue. European Journal of Communication, 20(4), 484-507.
35
Van Santen, R. (2008) Politics and Entertainment: A happy Marriage? Javnost - The Public, 2008, Vol.15(3), p.89-100.
Vettehen, P.H., Nuijten, K. and Beentjes, J.(2010) Research Note: Sensationalism in Dutch Current Affairs Programmes 1992–2001. Department of
Communication at Radboud University, Nijmegen.
Wæver, O. (1998). Explaining Europe by Decoding Discourses. In: Anders Wivel (ed.). Explaining European Integration. Copenhagen: Copenhagen Political Studies Press, 100–146.
Wæver, O. 2002. Identity, Communities and Foreign Policy. Discourse Analysis as Foreign Policy Theory. In: Lena Hansen and Ole Wæver ver (eds.). European Integration and National Identity. The Challenge of the Nordic States.
London/ New York: Routledge, 20–49.
Wimmel, A. (2009). Beyond the Bosphorus? Comparing public discourses on Turkey’s EI application in the German, French and British quality press. Journal of Language and Politics, 8(2), 223-243.
I
AppendixCoding Manual
1) Name of the coder 1= Mina
2= Renz
2) Newspaper
01= Süddeutsche Zeitung am Wochenende
04= Sunday Telegraph, The 07= Kathimerini tis Kyriakis
02= Die Welt am Sonntag 05= Observer, The 08= I Efimerida ton Syntakton 03= BILD am Sonntag 06= Mail on Sunday, The 09= Proto Thema
3) Date
Date of publication of the article (DD/MM/YYYY)
4) Serial Number
A number that will help us to go back to an article if needed. This number will consist of seven digits. The first two are the code number of the newspaper (VAR02), the third and fourth the day of the month of publication, the fifth and sixth refer to the month the publication was made, the next two the page number the article was on. For the articles retrieved from LexisNexis, the coding of the serial number might differ according to whether or not page number is available. In cases where page number is not available, the last two digits will be coded as 00.
When there are more than one articles on the same date, in order to maintain the uniqueness of the serial number, the last two digits will represent in which order the article was coded, starting from 00 and rising to the total number of articles on the date -1.
5) Title of the article
Copy paste the title of the article
6) Is the refugee crisis the main topic of the article? 1= yes
