The Portrayal of Refugees in Public and Private German News Reporting Before and After the Cologne Mass Sexual Assault

Table of Contents

1. Introduction
1.1 The Cologne Mass Sexual Assault
1.2 Research questions

2. Theoretical Framework
2.1 Conceptual definitions relating to the refugee debate
2.2 The German refugee regime and its history
2.3 Refugees in the media and politics
2.4 Migrant, asylum and refugee discourse in Germany
2.5 Agenda-setting and framing as cognitive media effects
2.5.1 Agenda-setting
2.5.2 Framing
2.6 Private and public broadcasting in Germany

3. Research Methods
3.1 Data collection
3.2 Data analysis
3.3 Coding protocol for quantitative framing analysis

4. Results
4.1 Sample summary
4.2 The portrayal of refugees before the incident
4.2.1 Private channels
4.2.2 Sat
4.2.3 RTL
4.2.4 Public channels
4.2.5 ARD
4.2.6 ZDF
4.3 Private – public comparison
4.4 The portrayal of refugees after the incident
4.4.1 Private channels
4.4.2 Sat
4.4.3 RTL
4.4.4 Public channels
4.4.5 ARD
4.4.6 ZDF
4.5 Comparison and summary
4.6 Neutral, negative, and positive articles
4.6.1 Private channels
4.6.2 Public channels
4.7 Content analysis of selected positive and negative articles
4.7.1 Private channels
4.7.2 Public channels
4.8 Agenda-setting
4.5.1 Media agenda
4.5.2 Audience agenda and evaluation
4.8.3 Xenophobic crimes

5. Discussion

6. Conclusion

7. Bibliography

Figures

Figure 1: Number of asylum applications in Germany since 1953

Figure 2: Integrated process model of framing

Figure 3: Market shares in percentage of selected German TV channels 2015

Figure 4: Three step agenda-setting research model 21

Figure 5: Composite frequency of frames for all channels before and after the incident

Figure 6: Number of articles using the political frame, problem/threat frame, and conflict frame per month

Figure 7: Frequency of neutral, negative, and positive articles for private and public channels

Figure 8: Number of articles published per month

Figure 9: Number of articles per month per channel

Figure 10: Number of negative articles per month

Figure 11: Voting forecasts for AfD according to ARD DeutschlandTrend July 2015 – June 2016

Tables

Table 1: Frequency of articles per channel before and after the incident

Table 2: Frequency of frames for all channels before and after the incident

Table 3: Frequency of sub-frames

Table 4: Frequency of frames for private channels before the incident

Table 5: Frequency of frames for public channels before the incident

Table 6: Frequency of frames for private channels after the incident

Table 7: Frequency of frames for public channels after the incident

Table 8: Three most popular frames for private and public channels before and after the incident

Table 9: Composite frequency of frames for private and public channels before and after the incident

Table 10: Frequency of positive, neutral, and negative articles per channel before and after the incident

Table 11: Sum of positive and negative articles per month

Table 12: Media agenda, negative articles, AfD voting forecast, and State Parliaments with AfD participation

With love and dedication for my late father.

The Portrayal of Refugees in Public and Private German News Reporting Before and After the Cologne Mass Sexual Assault

Abstract/Summary

Taking framing and agenda-setting as theoretical approaches, the present study analyzed the portrayal of refugees in German news reporting before and after the Cologne mass sexual assault on New Year’s Eve 2015. Comparing the two largest public (ARD and ZDF) and private (RTL and Sat.1) TV broadcasters, the results indicated that the incident had an immediate impact on the news reporting of all channels, but not in the long term. The problem/threat frame peaked in January 2016 for all channels. Overall, private channels showed a stronger tendency to emotionalize the news and portray refugees negatively. Especially RTL strongly changed its portrayal of refugees after the incident. The political frame was most popular for all channels after the incident, but much stronger so for public channels. The number of articles published peaked for public as well as private channels in the month following the incident. Public opinion data indicated that the popularity of the German right-wing party AfD has increased to an all time high since the beginning of the refugee debate in the media. The Cologne mass sexual assault did however not have an immediate impact. It is argued that the effect is cumulative rather than a short-term immediate reaction to specific events.

1. Introduction

Since at least mid-2015 it has been virtually impossible to avoid being confronted with the European refugee debate in the media. As tensions in the Arab world and elsewhere rise, more and more people flee to other countries in search for safety. These people are referred to as refugees. According to the 1951 United Nations Convention Relating to the Status of Refugees, refugee status is given to a person who:

“[...] owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it.” (UNHCR, 2007, p.16)

Conflicts such as in Syria, Iraq, Afghanistan, and the rise of ISIS, among others, have led to a considerable increase in the number of refugees worldwide. By mid 2014, the number of refugees in the world reached 13 million, the highest since 1996, with Syrians and Afghans being the largest refugee populations (UNHCR, 2014). While many refugees flee to other relatively poor or neighbouring countries (Castles & Miller, 2009), considerable numbers have also reached the borders of the European Union (EU). Within the EU, Germany is the country that hosts most refugees in terms of absolute numbers (Castles & Miller, 2009) with nearly half a million asylum applications in 2015 (BAMF, 2015).

In spite of its legal definition, the term refugee is open to much public debate and one interpretation varying from the other. Previous studies have repeatedly shown that the media can heavily influence public discussions about refugees, often conflating the terms refugee, asylum-seeker, migrant, or foreigners (Boomgaarden & Vliegenthart, 2009; Bradimore & Bauder, 2011; Erjavec, 2003; Gale, 2004; Innes, 2010; Lacroix, 2004; Leudar et al., 2008; Saxton, 2003). Or as Benson (2013) puts it: “Immigration is reshaping our lives, yet much of what we know about immigration is limited by the information and analyses we receive from the news media. Are the media up to the task?” (p.1) Given Germany´s central role as a refugee receiving country not only within the EU but worldwide, it is important to understand the discursive construction of refugees in German news media reporting as they are among the most important sources of information in this debate. At times, key events like the mass sexual assault in Cologne can have implications for how subsequent stories on similar issues are presented. During the first half of the 1990´s Germany witnessed four violent outbursts against aliens and asylum-seekers that made national news (Brosius & Eps, 1995). These key events then, in turn, influenced subsequent news selection in terms of victims, location, and the type of event (Brosius & Eps, 1995). The next section therefore presents a brief recapture of the Cologne mass sexual assault.

1.1 The Cologne mass sexual assault

The socio-economic, cultural, and political consequences associated with the sudden large scale influx of refugees have of course not gone unnoticed by the media. While the refugee subject matter has been prevalent in the German media landscape since at least mid-2015, one incident intensified the debate even more: The mass sexual assault that occurred in Cologne during New Year´s Eve 2015, where allegedly several hundred men of north African and Arab origin sexually assaulted and mugged women that were passing by on a public square between Cologne main station and the Dome. Some reports put the number as high as up to a group of 1000 offenders. A total of at least 500 reports have been filed at the police, reporting sexual assault, robbery, and libel. The estimated number of victims is around 80. The incident was quickly picked up not only by German but also international media and hotly debated in the context of the refugee subject matter. The incident sparked far-reaching public discussions about limiting the number of refugees, tightening border controls, Merkel´s refugee policies in general, and about refugees being nearly impossible to integrate. Different media might however pursue different agendas in this debate, most notably private versus public channels, which will be further explained in section 2.6.

1.2 Research questions

The present study therefore attempts to answer the following research questions:

RQ1: To what extent, if any, has the portrayal of refugees in German news media reporting changed before and after the Cologne mass sexual assault?
RQ2: What are, if any, the differences in the portrayal of refugees between public and private broadcasters?
RQ 3: How has the media agenda influenced public opinion?

2. Theoretical Framework

The theoretical framework is structured as follows: First of all, various key terms relating to the refugee debate will be defined, followed by an outline of the German refugee regime and its history. Afterwards, an introduction into the topic of refugees in the media and politics will be given in section 2.3. The following section moves on to analyze the German asylum discourse. Section 2.5 then introduces the theoretical concepts of agenda-setting and framing. Finally, section 2.6 offers an introduction into Germany’s broadcasting system with an emphasis on the public – private divide.

2.1 Conceptual definitions relating to the refugee debate

“[…] the current persistence of the ‘refugee problem’ in international politics and unsuccessful attempts to respond to it is only partly attributable to politics or questions of resources: ‘conceptual confusion – about the meaning of refugeehood, its causes, and its management - also contributes to the misery of both refugees and host and to the inflammation of international tension´” (Shacknove, 1985; as cited in Haddad, 2008, p.25). A conceptual definition of the various terms surrounding the refugee debate in media as well as politics is of importance to avoid terminological confusion. Furthermore it helps to understand the refugee subject in the broader context of population movements in general.

The media’s contribution to this terminological confusion is rather obvious. Just following the media reports about the refugee debate, one can get easily confused about the related terms refugee (Flüchtling), migrant (Migrant), asylum-seeker (Asylbewerber), and foreigner (A usländer). While the term refugee has been defined in the introduction, this chapter defines the remaining concepts. First of all it should, however, be clarified that these terms are not the only ones surrounding the refugee debate in media and politics. As Haddad (2008) notes terms such as economic migrants, illegal immigrants, displaced persons, political refugees, bogus asylum-seekers, stateless persons, and B-refugees are also present in the refugee debate. Defining each of these terms is however beyond the scope of this study as it is specifically concerned with the portrayal of refugees in mainstream German news reporting. Defining only the most closely related terms to the refugee concept is therefore sufficient.

Haddad (2008) notes that refugees fall in the category of ‘forced migrants.’ Thus while there are many motives for migrants to migrate to another country – e.g. economic, romance, education – refugees have no choice but to leave their country in search for safety. Or as Haddad puts it, while migrants often hope for a better life, refugees are merely trying to rebuild the life they have lost (p.27). Once a refugee has applied for asylum and waits in an asylum-shelter for her case to be decided, she is considered an asylum-seeker (Federal office for Migration and Refugees, 2012). Foreigners are all persons living in Germany not having the German citizenship, irrespective of them being refugees, asylum - seekers, or migrants (cf. Irrgang, 2011).

2.2 The German refugee regime and its history

Legally, Germany´s refugee regime began in 1953 with the adoption of the 1951 Geneva Convention relating to the status of refugees. The formal authority in Germany for handling refugee matters is the Federal Office for Migration and Refugees (Bundesamt für Migration und Flüchtline, hereafter BAMF). Germany is known to have quite liberal asylum policies. Contrary to most other states, Germany grants asylum beyond the requirements specified in the 1951 Convention. Asylum is a basic right for foreigners who are politically persecuted, given that they are persecuted by their respective state (BAMF, 2012). Once a refugee has applied for asylum, she has to go through a process in which it is determined whether she will be granted refugee status (BAMF, 2012). For the duration of this process the alleged refugee is placed in an asylum/refugee shelter until her case is decided.

Merkel´s liberal asylum policies have coined the term Willkommenskultur (welcoming culture). In the first half-year 2015 alone, Germany received 114.000 applications for asylum, only 4000 less than the total number for 2014 (BAMF, 2015) By the end of 2015, Germany had received 476.649 asylum applications (BAMF, 2015). Highlighting the extent of the current influx of refugees, Germany received 176.465 only in the first quarter of 2016 (BAMF, 2016). Already in 2006, Germany was ranked fourth in the list of top host countries for refugees worldwide (Castles & Miller, 2009). As of 2013, 16.5 million people -or 20.5% of the German population- had a migration background, i.e. were either migrants themselves or the descendents of such since World War II (Federal Center for Political Education, 2015). Due to the rising number of conflicts worldwide on the one hand and Germany´s rather liberal asylum politics on the other hand, Germany is recently experiencing an almost unprecedented large scale influx of refugees.

To place this in a broader historical context, it is worthwhile mentioning that Germany has a history as a destination country for refugees and migrants. Historically, the most significant event in this context was Germany’s defeat in World War II. It led to the separation of Germany into two separate sovereign states. The victorious allies of USA, Great Britain, and France gained control over the West of Germany, establishing the Federal Republic of Germany (FRG) with democracy as its political ideal. At the same time the Soviet Union gained control over the East of Germany, leading to the establishment of the German Democratic Republic (GDR) with communism as its political system.

While the Soviet Union led the GDR into political isolation with closed borders, the FRG soon opened its borders for foreign workers to help re-building the country. To do so, Germany introduced its guest worker (Gastarbeiter) policy, beginning in the mid 1950’s (Castles & Miller, 2009). From this period onwards until the 1970’s millions of foreign workers came to Germany. By 1974 11 million guest workers resided in Germany, mainly originating from Greece, Portugal, Turkey, and Yugoslavia (Rattansi, 2011). While several million guest workers left Germany when the post-war economic boom merged to a deep recession, 3 million Turks stayed and were eventually able –with the support of international and German courts– to bring their families (Rattansi, 2011). Besides guest workers, starting with the ending of World War II Germany experienced a massive wave of immigration of ethnic Germans: “Between 1945 and 1949, nearly 12 million German refugees and expellees flocked to the territory of today's Germany. They were either German nationals who had lived in areas intermittently under German jurisdiction prior to 1945, or ethnic Germans from other parts of Poland, Czechoslovakia, Hungary, and Yugoslavia” (Migration Policy Institute, 2004).

While (West) Germany thus has a rich history of hosting migrants and refugees, the current influx of refugees has still attracted enormous media attention. While this can partly be explained by the rapid advancement of mass media compared to the post World War II situation, it is also indicative of the impact that the current large scale influx of refugees has on the political and media agenda. The only time Germany experienced a similar influx of refugees was shortly after its re-unification in the early 1990’s, with the largest refugee groups coming from central and Eastern Europe, which was increased by the civil war in Yugoslavia (ARD, 2015). The impact that this large scale influx of refugees had on the media reporting at the time will be explored in greater detail in the next section. Figure 1 gives an overview of the number of asylum applications in Germany since the 1950’s.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1: Number of asylum applications in Germany since 1953 (retrieved from: http://www.tagesschau.de/inland/asyl-chronologie-101~magnifier_pos-0.html; original source: BAMF, 2015) *January - October 2015; by the end of 2015 the number reached 476.649 (BAMF, 2015)

Finally, it should be emphasized that the immigration debate in Germany takes place in a political and social context different from countries such as the US, Canada, or Australia. While more than 20% of the German population does have a migration background, Germany’s migration history is rather short. As Rattansi (2011) notes, even during the period of large scale influx of foreign workers, Germany’s citizenship policies remained one of ius sanguinis – descent from blood, meaning that only ‘ethnic Germans’ could gain access to citizenship. Countries such as the US, Canada, and Australia have for example been countries of immigration since their establishment with diverse ethical populations. Refugees immigrating to these countries might therefore be confronted with a less mono-cultural history than in European nation-states, including Germany.

2.3 Refugees in the media and politics

Due to the recency of the current large scale influx of refugees to the EU, little systematic research has been conducted so far in the area of media representations of this wave of refugees, especially with regards to the current state of affairs in Germany. As refugees in general are however not a recent phenomenon, the representation and discursive construction of refugees in the media has been the topic of various international studies over the past 25 years. These studies have investigated the subject from different angles, including –but not limited to– communication and media studies, discourse analysis, political science, migration studies, human rights, and anthropology. Especially for studies about media representations of refugees it holds true that the results of these studies are inherently bound to the particular socio-cultural, historical, and political context in the receiving as well as sending countries at the time the study has been conducted.

Systematic studies about the representation of refugees in the media date back to the 1990’s and have since been investigated in countries as diverse as Australia (Saxton, 2003), Germany (Becker, 2015; Boomgaarden & Vliegenthart, 2009), Canada (Bradimore & Bauder, 2011; Lacroix, 2004), Costa Rica (Larson, 1995), Slovenia (Erjavec, 2003), South Korea (Yi & Jung, 2015), UK (Gabrielatos & Baker, 2008; Innes, 2010; Philips & Hardy, 1997), and US (Steimel, 2010). With the exception of Steimel’s study investigating the portrayal of refugees in human interest stories in the US print media, the overwhelming tenor of these studies is that the mainstream media portray refugees in negative, stereotyping and partly discriminating ways. In her study about the portrayal of Nicaraguan refugees in Costa Rican media, Larson noted already in 1995 “[…] that we all must be aware of the power of the press, and that in this age of correct speech, we all must bear some of the responsibility to change the errors of the past” (p.25). Despite this early warning call, even the most recent studies have almost exclusively found a continuing pattern of negative portrayals of refugees in mainstream media.

Investigating the causal link between various negative claims made by the media about refugees and their influence on public opinion, Esses et al., (2013) found that negative images circulated by the media in Canada and other Western nations can possibly lead to a dehumanization of refugees. According to their study, the media take advantage of the uncertainty that surrounds the refugee subject to perpetuate stereotypes that (1) immigrants spread infectious diseases (2) refugee claimants are often bogus (3) terrorists may gain entry disguised as refugees.

Similar results have been obtained by Leudar et al., (2008) investigating the refugee discourse in the UK. They found that hostility themes in the media deny refugees of certain aspects of humanity and can eventually lead to exclusion from local communities. Leudar et al., have, among others, analyzed media contents that contained statements made by politicians regarding a new refugee bill to be passed in the British Parliament, pointing towards the interplay between media agendas and politics in creating a negative climate surrounding refugees. In a similar vein, Saxton (2003) investigated the discursive construction of refugees in Australia after an incident where refugees had allegedly thrown their children overboard in order to manipulate the Australian Navy into picking them up. Responding to the incident, the Australian Prime Minister proclaimed: “I certainly don’t want people like that here.” Saxton’s analysis has shown that media reports deliberately controlled the identity of refugees to support government politics of exclusion.

Finally, analyzing an incident in which a boat with 76 Tamil refugees arrived off the coast of British Columbia, Canada, Bradimore and Bauder (2011) as well found a continuous pattern of negative media reports, speaking about mystery ships, infectious diseases, and potential terrorists, spreading a climate of anxiety through public discourse. In an interplay with the media, Canadian politics responded by treating the refugees as a risk that needs to be managed. After the arrival of another boat with Tamil refugees, the Canadian Safety Minister explained via the National Post: “[A]buses of Canada’s immigration system cannot and will not be tolerated. As we deal with this current situation under Canadian law, Canadian officials will look at all available options to strengthen our laws in order to address this unacceptable abuse of international law and Canadian generosity” (as cited in Bradimore & Bauder, 2011, p. 657).

Addressing this negativity in the portrayal of refugees in Western news reporting Wright proclaims that (2014) “[…] it is essential to remain alert to the negative accounts of Western media reporting and maintain pressure on media institutions to be responsible in their coverage of migration issues. Media reports, which do not always reflect the reality of the situation, have a strong effect on public opinions, but people’s perceptions can determine the reality” (p.470).

Summarizing the vast majority of studies conducted in the area of media representations of refugees found a continuous pattern of negative portrayals of refugees in mainstream media. These results have been obtained in different historical and geographical contexts. Some of these studies have further highlighted the influence of politics in the process of controlling the refugees’ identity. The following section moves on to analyze the refugee discourse in Germany embedded in its historical and political context.

2.4 Migrant, asylum and refugee discourses in Germany

To understand the current refugee debate in Germany and the portrayal of refugees in German media, it is important to place the current debate in its historical and political context. Germany has a democratic, multi-party system. Federal elections take place every four years. Since 2005 Angela Merkel has been the Chancellor of Germany. She is part of the conservative Christian Democratic Union (CDU). The CDU is in coalition with the Christian Social Union (CSU) and the Social Democratic Party (SPD). The Green Party and the Left are in the opposition.^* Furthermore each of Germany’s 16 Federal States has its own State Parliament which can take independent decisions, but is ultimately subordinated to the Federal Government.

In October 2010 Angela Merkel made a precarious statement regarding the living together with migrants in Germany. One week after the Bavarian Premier Horst Seehofer (CSU), part of Merkel’s governing coalition, called for a halt of Turkish and Arabic immigrants, Merkel announced that “the multicultural approach has failed, utterly failed!” (“Der Ansatz für Multikulti ist gescheitert, absolut gescheitert!”) (Spiegel Online, 2010). As Rattansi (2011) however notes, this statement was quite ironic, given that multiculturalist policies had hardly ever been tried in Germany.

Two months earlier, former central bank official Thilo Sarrazin had published his highly controversial book Germany abolishes itself (Deutschland schafft sich ab), in which he vigorously attacked Muslims and migrants in general, and Turkish migrants in particular. In his book, Sarrazin claimed among others that Turkish migrants decrease the country’s prosperity and lower the average intelligence through higher birth rates than native Germans and lower levels of education. While most of his claims were backed up by questionable statistics and he intentionally left out aspects that portrayed migrants in a positive light, his book had struck a nerve. It was followed by heated public and private debates at all levels of society as well as high levels of media attention. It even led to an extensive empirical investigation led by Naika Foroutan of Humboldt University Berlin to discredit Sarrazin’s claims (Foroutan, 2011).

In an article entitled “Beyond Sarrazin?” Irrgang (2011) analyzed the portrayal of migration issues in the German daily BILD (tabloid press) and news magazine DER SPIEGEL for articles published in August and October 2010, thus during politically turbulent times regarding the migration debate similar to the current debates. Using topoi, framing, and agenda-setting as theoretical frameworks, Irrgang found that especially the tabloid magazine BILD portrayed migration as a threat to German society, highlighting crimes, terrorism and unwillingness to integrate on the side of migrants, exclusively focusing on Turkish Muslim migrants while ignoring other migrant groups. News magazine SPIEGEL, to the contrary, portrayed the migration debate in a more balanced way. In doing so, they used for example economic frames to highlight the potential benefits of migration. Quantitatively, Irrgang found that in both media the publishing of Sarrazin’s book led to an increase in the number of articles published about migration, labeling him as an “agenda-setter” (p.29). If the Cologne mass sexual assault had a similar agenda-setting function remains to be seen during the analysis as well as the extent to which private and public news channels differ in their portrayals of refugees.

Recent analyses of the general asylum discourse in Germany are relatively scarce. The exception is Becker’s (2015) excellent in-depth, extensive study of the German asylum discourse for the period July 2013 – June 2014. Using media discourse and corpus analysis, Becker mainly analyzed print media content of various newspapers but also online forums and political talk shows broadcast on the public channels ARD and ZDF. In descriptive terms, Becker found a correlation between the number of asylum applications and the number of articles published about the asylum subject by the print media, “[...] signifying the intense solicitousness of the media landscape and, at the same time, relevance and timeliness of the asylum topic” (p.129). Becker further analyzed dimensions and perspectives on responsibility, fear, and xenophobia as central sub-discourses within the broader asylum discourse.

Regarding the construction of responsibility in the asylum discourse, Becker found that the print media mainly use the term in its prospective form while talk shows and online forums rather tend to ascribe responsibility in retrospect. In its prospective form it refers to a future-oriented ascription of responsibility or a duty to fulfill a certain role. In its retrospective form, the term refers to the ascription of responsibility for past events (Becker, 2015). Another interesting finding concerns the attribution of responsibility in the asylum discourse. While the print media use the term in its collective form attributing responsibility for the well-being of refugees to public authorities and government officials, online forums and partly talk shows attribute responsibility to the refugees themselves or respectively to the government of the sending countries.

Becker further found the concept of fear to be very important and part of many argumentative structures within the asylum discourse, labeling it as a “leading vocabulary” (p.110) in the asylum discourse. Becker’s analysis further showed that the fear is for all investigated media only portrayed from the viewpoint of the Germans while almost completely ignoring the fears of the asylum applicants. According to Becker, the fear discourse often showed conflicting conceptualizations of the emotion as either being legit in response to the situation or as being irrational.

Finally, Becker’s analyses revealed that the concept of xenophobia is closely related to the concept of fear, either by connecting the two concepts or by explicitly setting them apart. Fear about refugees can be described as a sign of xenophobia, or as an independent emotion that should not necessarily be interpreted as xenophobic tendencies. To what extent the three main concepts of Becker’s analyses of the asylum discourse in Germany 2013-14 are present in the portrayal of refugees in the current debates, is another question that remains to be answered during the analysis. Both Irrgang and Becker started their analysis based on the assumption that the mass media are important organs of knowledge acquisition and public opinion formation. The present study follows this line of thinking.

Most systematic reviews of the portrayal of refugees in German news reporting date back to the 1990’s – the decade in which Germany was re-united and experienced various outbursts of violence and hatred against immigrants and refugees, most notably in East Germany which had previously closed borders under the communist influence of the Soviet Union. As Ohlemacher (1994, 1998) recaptures, during the early 1990’s Germany experienced four infamous incidents of arson and violence against asylum-seekers in Hoyerswerda, Rostock, Mölln, and Solingen (except for Solingen all of these cities are located in the former East Germany). Each of these events was followed by a volatile increase in media coverage (Ohlemacher, 1998). According to Ohlemacher, the first two events significantly influenced subsequent xenophobic crimes and he ascribes a key role in this process to the mass media. Brosius & Eps, (1995) found for example that the media often acted reflectively, using such key events as a standard for future coverage of similar events meaning that previous events function as prototypes for coverage of similar events in the future. This can lead to a distorted coverage as events that are more similar to previous events are covered more frequently (Brosius & Eps, 1995).

Analyzing the media discourse and public opinion formation for the years 1991- 97, Ohlemacher (1998) even found that the aforementioned BILD published articles expressing understanding for the crimes and their motivations. Furthermore Ohlemacher noticed a tendency of some newspapers to clearly separate the actual crime from its xenophobic motivation, for example by intentionally leaving out the xenophobic motivation of the offenders to commit crimes against foreigners. According to Ohlemacher, this has found resemblance in public opinion as less people considered themselves “little or not at all political” on the used five-point scale.

Covering a larger time span, Boomgarden and Vliegenthart (2009) investigated the influence of news content on anti-immigration attitudes in Germany 1993-2005. As Boomgarden and Vliegenthart analyze: “[...] both the frequency and the tone of coverage of immigrant actors in the news significantly influence dynamics in anti-immigration attitudes. The strength of the effect of the news, however, depends on contextual variation in immigration levels and the number of asylum seekers” (p.516). As both immigration levels and number of asylum seekers are at a high point in Germany, it can be assumed that the strength of the effect of the news is accordingly strong.

Summarizing the studies conducted in the last two decades, Irrgang (2011) found a strong tendency for negativism in German news coverage, manifested in certain discursive strategies that highlight exceptions and exaggerations portraying a distorted reality. This highlights the need for the present study, taking into account the Cologne mass sexual assault as a contextual factor. As the theoretical framework has further shown, key events like this can significantly impact subsequent news reporting which, in turn, can influence xenophobic attitudes and crimes.

Indeed a large body of research has emerged over the past 40 years or so indicating that the media and their agendas can influence the audience’s subsequent cognitive processing of a subject. As recent as 2015, Bleich et al., urged scholars of migration and minorities in the media to engage in framing and agenda-setting research in more depth: “Visibility in the media, agenda-setting and framing are of obvious interest to scholars of migrants and minorities, but these topics have been insufficiently explored to dat e” (p.860). The present study therefore utilizes agenda-setting and framing as theoretical frameworks.

2.5 Agenda-setting and framing as cognitive media effects

Even though the term refugee has a clear-cut legal definition which is adopted by the German state, public opinion can diverge from that definition, among others, according to news media reporting. In her analysis of the discursive construction of asylum-seekers in Australia, Saxton (2003) has demonstrated how the dominant culture produces a certain image of asylum-seekers, e.g. referring to them as ‘people like that:’ “[t]he discursive management of the collective identity of asylum seekers by the dominant culture to construct a specific social reality [...]” (p.109). Similar studies conducted in the UK (Philips & Hardy, 1997) and Canada (Lacroix, 2004) have also shown that the term refugee is rather discursively constructed leaving the possibility to either modify or maintain the institutional frame . A large scale corpus analysis by Gabrielatos & Baker (2008) of UK refugee press coverage has even revealed that the media invent non-sense terms such as illegal refugee to create and maintain panic around refugees, a process that is commonly referred to as igniting a “moral panic” (McRobbie & Thornton, 1995). Thus it appears that the media agendas can diverge from the institutional and legal framework. The following sections therefore introduce the theoretical concepts of agenda-setting and framing.

2.5.1 Agenda-setting

The media, politicians, and other entities of the dominant culture can influence public opinion by means of a dominant discourse. As McCombs and Shaw (1972), the founding fathers of agenda-setting theory, put it: “Readers learn not only about a given issue, but also how much importance to attach to that issue from the amount of information in a news story and its position” (p.176). In their pioneering study on agenda-setting during the 1968 US presidential elections, McCombs and Shaw (1972) found that public perception of important topics correlates with mass media coverage of these topics. This effect has been labeled by the authors of the study as the agenda-setting function of mass media and has since been repeated in numerous studies that followed.

Some scholars argue, however, that in the diversified media landscape of the 21st century, agenda setting loses its importance as the public attention is scattered across various media with different agendas. Out of 137 articles published between 2000 and 2003 in social science journals on the topic of political communication, only 2.9% took an agenda- setting approach (Graber & Smith, 2005). However, McCombs (2005) presents strong empirical evidence for the continuing relevance of agenda-setting such as the high inter- media agenda similarity and the concentration of public attention on only a few news pages even on the World Wide Web.

As refugees and migration in general are transnational issues –refugees must cross an international border to be recognized as such– modern day agenda-setting studies might have to look beyond domestic borders. Already in 1984, Manheim and Albritton stated “[...] that external actors not only try systematically to influence the media-public-policy agenda system, but that they succeed” (p.656). Specifically, they analyzed the representation of foreign nation states in US newspaper coverage. More recently, Kiousis and Wu (2008) argued for example that news coverage of foreign nations potentially has strong agenda- setting effects on public opinion. Investigating the international agenda-building process, they found that media salience of foreign nations was linked to public attitudes regarding these states. RQ 3 has been derived following this line of thinking, i.e. that the media agenda might influence public opinion about refugees. Kaye (1994, 1998) has for example shown that in the UK there exists a reciprocal relationship between policy makers and media portrayals of refugees to alter and devaluate public perceptions of refugees. Similarly, Nikunen and Horsti (2013) found that in Finland the media agenda facilitates the emergence of anti-immigrant agendas in public debates, in turn strengthening nationalist politics. Empirically, they found for example that anti-immigrant actors increased in media coverage, labeling the anti- immigration movement as an agenda-setter.

Generally much has changed in our globalized world since the first agenda- setting studies. The transnational nature of the refugee debate might influence the media agenda in different ways than the classical example of domestic political campaigns. This notion is linked to debates about an emerging global public sphere: “Through this global and mediated publicity, the social and cultural values of society become externalized; they do not arise from inside the national community as a basis for or a result of deliberation, but are increasingly delivered from the outside through global media” (Hjarvard, 2001, p.20). As a result, discussions surrounding public affairs are increasingly constructed though global rather than national communication networks (Castells, 2008).

As Weaver (2007) further explains, there are two levels of agenda-setting. The first level of agenda-setting is concerned with the relative salience of objects. This variable is usually operationalized as perceived importance. In other words, the media differ in how much attention they pay to certain subjects. The second level of agenda-setting, on the other hand, focuses on the relative salience of attributes of issues (Weaver, 2007). The salience of attributes refers to the fact that each object on the media agenda has certain attributes, some of which are strongly emphasized while others are only mentioned in passing (McCombs, 2005). This can influence our understanding of the object in question. Studies on risk perception and decision making have for example frequently shown that the same information can be judged differently depending on how it is presented. The formulation of certain events in terms of either wins or losses has a strong impact on subsequent cognitive processing, independently of actual probabilities (Kahneman & Tversky, 1979, 1984; Tverksy & Kahneman, 1992). In other words, the way in which certain information is framed alters our judgment of the information. For the purpose of this study, no further distinction will be made between second level agenda-setting and framing.

2.5.2 Framing

As Semetko and Valkenburg (2000) explain, agenda-setting and framing have in common the focus on media news reporting of public policy issues and the public´s perception of these issues. Framing, however, extends what the public thinks to how to think about a certain reality (Pan & Kosicki, 1993; Semetko & Valkenburg, 2000). One of the most widely operated definitions of media framing derives from Entman (1993):

Framing essentially involves selection and salience. To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item described. (p. 52, italics in original)

The core premise of framing is that a subject can be viewed from multiple perspectives and accordingly be constructed as having implications for a variety of values and considerations (Chong & Druckman, 2007). More specifically, media framing involves the process whereby the media emphasize some aspects of an event while ignoring others and thereby try to convince the public to take a certain attitude towards the presented subject (cf. Entman, 1993). In any news reporting, the coverage takes a certain angle to represent reality, which in turn might influence how the public thinks about a certain subject. De Vreese (2005) suggests that this process happens in three stages. The outcome of the first stage, frame- building, is frames that are manifest in text. This process is shaped by internal as well as external factors. Internally, editorial policies and news values influence the process of frame- building. Externally, interactions between journalists, elites, and social movements shape this process. As Price et al., (1997) additionally explain with regards to the first stage, news values, i.e. beliefs about what would be a “good” story, do not only guide the choice of selecting which stories to cover, but also how to present these stories:

“Instead of explaining the general background and implications of issues, news reports emphasize only the most recent and attention-getting developments – for example, covering unemployment by focusing on vivid examples of people who have lost their jobs, while failing to link unemployment to any broader social, economic, or political processes. Social issues are thus mainly treated as discreet and isolated events.” (p.482-83)

The second stage is described by de Vreese as follows: “Frame-setting refers to the interaction between media frames and individuals’ prior knowledge and predispositions” (p. 52, italics in original). In other words, this stage involves the setting of the frames in regards to the information and attitude that members of the audience already have in regards to the topic. The final stage is the consequences associated with the first two stages. According to de Vreese, these can be assessed on the individual and the societal level. Altering one´s attitude after being exposed to a certain media frame is an individual consequence. Societal consequences include political socialization, decision-making, and collective actions. The framing process is visualized in Figure 2.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2: Integrated process model of framing (adopted from de Vreese, 2005, p. 52)

Previous research has identified the existence of specific frames routinely used in media reporting (Altheide, 1997; Semetko & Valkenburg, 2000; Valkenburg, Semetko, & de Vreese, 1999) which are partly used as a typology in the present study. These will be further explained in the methodology section. Earlier studies have further shown that private and public channels often differ in their coverage of the same issues, including the use of frames. The next section therefore briefly debates previous research addressing the gap between public and private news reporting in Germany.

2.6 Private and public broadcasting in Germany

As Esser and Hemmer (2008) put it with regards to the German media landscape: “Public broadcasting is expected to provide more and better-quality public affairs coverage and to offer higher standards of informational, educational, and recreational programs” (p.289). Labeling the division between public and private channels a dual system of broadcasting, they continue to argue that public channels are obliged to contribute to a free individual and public opinion formation according to the State Treaty on Broadcasting in Germany. According to the State Media Treaty, public channels are further committed “to ‘principles of objectivity and impartiality of coverage’ as well as diversity of opinion, and coverage has to be ‘independent and factual’” (Esser & Hemmer, 2008, p.292).

Private channels however are less regulated by the state and have no duty to inform or educate (Esser & Hemmer, 2008). Analyzing the TV news coverage of ARD and ZDF (public) and RTL and Sat.1 (private) of the 1998 elections in Germany, Semetko and Schoenbach (1999) found that “[t]here was a clear difference between the public and private channels in the emphasis on the campaign in the news. The private channels devoted far more attention to the campaign and the opinion polls than the public channels, and correspondingly less to the substantive issues over which the campaign was being fought” (p.85). Semetko and Schoenbach stress however that they also found “strong similarities” between public and private channels in their descriptive and neutral portrayal of politicians during the campaign.

As Hafez and Richter (2007) explain, ARD and ZDF consider themselves role models in news reporting about questions surrounding the multicultural living together. Investigating the portrayal of Islam in ARD and ZDF, they found however that these two channels do not necessarily fulfill this self-proclaimed role. The news reporting by ARD and ZDF was found to be unilateral in their thematic choices and thereby increasing the audience’s willingness to engage in stereotyping (Hafez & Richter, 2007). Studies analyzing ARD and ZDF separately are scarce. Rather, most studies analyze them in unity as they are sought to have very similar agendas (cf. Hafez & Richter, 2007; Brosius, 1998; Weiderer, 1993).

RTL is part of the RTL Group – Europe’s largest broadcasting enterprise with 42 TV- and 31 radio-channels in ten countries (Aumüller, 2011). RTL describes itself as having five central values: Diversity, innovation, consistency, quality and relevance (Förster, 2011). Its emphasis lies upon non-fictional entertainment, fictional formats, and information with a focus on assembling a distinct emotional attachment (cf. Förster, 2011). Sat.1 is part of the ProSiebenSat.1 Group. Sat.1 has the second highest market share of all private broadcasters after RTL (Media Data Southwest, 2016). News reporting by Sat.1 has previously been labeled as News- Shows, highlighting the mix between information and entertainment (Brosius, 1998).

Studies analyzing differences in public and private news reporting often group together ARD and ZDF on the one hand and RTL and Sat.1 on the other (Brosius, 1998). RQ 2 has been derived following this direction. Investigating the effect of different media systems in various European countries on the audience’s knowledge of public affairs, Curran et al., (2009) found for example a connection between patterns of news coverage and levels of public knowledge. As their data suggest, market-driven private channels tend to lead to lower awareness of public affairs issues as compared to public services. In theoretical terms, this has been labeled by McQuail (1992, p.11) as the “media performance in the public interest.”

3. Research Methods

Following the clustering of RTL and Sat.1, and ARD and ZDF, the present study analyzed news reports published by these four channels using quantitative framing analysis. The exact method of data analysis will be further explained in section 3.2. Generally quantitative framing analysis is the process in which “[a] previously defined framing scheme may [...] be applied to quantify the occurrence of particular frames in a sample” (Carver et al., 2013, p.457). Riffe et al., (2014) even claim that quantitative analysis is the only logical way to assess mass media content: “Only this information-gathering technique enables us to illuminate patterns in communication content reliably and validly” (p. 13). The purpose of quantitative framing analysis for the present study was to identify quantitative differences in the framing of refugees between public and private channels. The following sections therefore move on to describe how the data for the present study have been conducted, followed by the approach to data analysis, and finally the coding protocol for the framing analysis.

3.1 Data Collection

The news reports collected for the study covered a time span of six month, i.e. three months before the mass sexual assault in Cologne and three months after (October 2015 – March 2016). The texts were collected from the online archives of the two largest private (ARD and ZDF) and public (RTL and Sat.1) news channels. The channels were selected based on their respective overall TV market shares in 2015 (Media Data Southwest, 2016, see Figure 3).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3: Market shares in percentage of selected German TV channels 2015 (adopted from Media Data Southwest, 2016 : http://www.mediendaten.de/fileadmin/Fernsehen/Marktanteile/P DF/MA_BRD.pdf)

As Figure 3 shows, there are quite strong differences in channel preference between the various age groups. While ARD and ZDF are most popular in the 50+ age category, Pro7 is by far the most popular in the 14-29 category followed by RTL. As the present study is however not concerned with age differences between members of the audience, it is a reasonable choice to select the two largest private and public broadcasters based on their overall market share.

To uncover news about refugees published by these channels, the search terms refugee (Flüchtling), refugee crisis (Flüchtlingskrise) and refugee debate (Flüchtlingsdebatte) were entered into the four online databases. This search generated an initial sample of nearly 4000 articles. Narrowing down this initial sample to the six months relevant for the present study, 1172 articles remained. Of these 352 (30%) were excluded due to irrelevance. Irrelevant articles were defined as articles that did not primarily deal with refugees. These articles included previews to movies and TV series and articles about politics that did not mention the word “refugee” directly. This concerned mostly articles about politics that were linked to articles about refugees and therefore brought up by the search. This led to a sample of 840 articles for the relevant time period, equaling 4.62 articles a day. To reduce this sample to a manageable size, only every other article was considered for further analysis, leading to the final sample of 420 articles. These articles serve as a proxy for all articles published during the time period relevant to the present study (October 2015 – March 2016).

3.2 Data analysis

The present study took the framing analysis approach which Matthes and Koring (2008) refer to as manual holistic approach: “[...] frames are first generated by a qualitative analysis of some news texts and then are coded as holistic variables in a manual content analysis. [...] After that, these frames were defined in a codebook and coded in a subsequent quantitative content analysis” (p.280). This approach was chosen to combine the strengths of both quantitative as well as qualitative framing analysis.

In a first step, 40 articles were randomly chosen from the final sample, including ten articles of each channel (five from before the Cologne mass sexual assault and five from after the incident). These articles were then thoroughly read to identify frames. Frames were identified using the definition by Entman (1993), i.e. making a pattern of aspects more salient to promote a certain definition, interpretation, and evaluation of the issue at hand. Using this technique, eleven frames were identified which were repeatedly used.

As Matthes and Koring (2008) point out, one pitfall of this technique is that it might be hard to observe the emergence of new frames once the initial coding scheme is developed. To account for this, the category “other” was created for articles not using any of the initially identified frames, carefully checking for the emergence of new frames. However, no further frames emerged within this category, meaning that no similar pattern of aspects was repeatedly made salient over a variety of articles. Furthermore the category “other” reduced the risk of forcefully trying to press articles into an already existing framing category, another caveat highlighted by Matthes and Kohring.

The identification of frames was based on inductive as well as deductive measures. Five frames (human interest, responsibility, economic consequences, conflict, and morality) were derived deductively from existing literature on the framing of European politics (Semetko & Valkenburg, 2000; Valkenburg, et al., 1999). The remaining six frames (victim, political, statistical, integration, problem/threat, and witness/source) were derived inductively. Using these frames, all articles included in the final sample were manually coded and the data were entered into SPSS. Other variables in the data set were the channel from which the article was derived (and whether it was a private or public channel); the month in which the article was published (and whether it was before or after the incident); and if it was a positive, neutral, or negative article. All variables were set to nominal measure.

For agenda-setting purposes the salience of the refugee subject in the media was measured as the frequency with which articles were published. Furthermore the occurrence of certain frames at certain times was researched, trying to find any underlying patterns of the media agenda. These data were then compared to public opinion data about the perceived importance of political issues as well as evaluations, measured in opinion polls about voter preferences. All in all, the analytical procedure followed the three-step model described by Scheufele (2000, see Figure 4). Finally, an in-depth content analysis was conducted for eight articles, i.e. four positive and four negative articles.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4: Three step agenda-setting research model (adopted from Scheufele, 2000, p.302)

[...]

---

^* http://www.bundestag.de/bundestag