Master’s Thesis
BAUTZEN AND ITS PROBLEM OF RACISM
Differences between local and national press coverage of right-wing violence against refugees in the East-Saxon town of Bautzen
Graduate School of Communication
Erasmus Mundus Master in Journalism, Media and Globalization
Author: Gwendolin Güntzel Student-ID: 11896728
Supervisor: Jasper van de Pol
Abstract
Recently, German news media covered right-wing violent incidents against refugees in Germany intensively, using the town of Bautzen, with a population of 40,000, as an example to examine a larger problem of racism in the East of the country. Thus, it is interesting to find out to what extent local and national news media have differed in their coverage of violence between refugees and locals in Bautzen during the years 2016/17. Since media discourse influences perceived realities of local and national public spheres and thus, respective audiences, it is important to understand how local and national newspaper coverage differs when covering right-wing violent incidents of local origin. Therefore, I conducted a quantitative content analysis of 319 articles from three local and seven national newspaper outlets, finding meaningful differences between local and national newspaper discourses in regard to the intensity, content and function of coverage. This study found national press to appeal to morality and to overtake the watchdog role more intensively than local press. Local press rather stressed the attribution of responsibility for the problem of racism in town, while being caught between the roles of watchdog and local agent. This case study offers insights on some similarities but many differences between local and national newspaper coverage on right-wing violence against refugees in Bautzen during the years 2016/17.
BAUTZEN AND ITS PROBLEM OF RACISM
Differences between local and national press coverage of right-wing violence against refugees in the East-Saxon town of Bautzen
The large district town of Bautzen with its 40.000 inhabitants is located in the very East of Saxony, just a few kilometers from the German-Polish and the German-Czech border (Bautzen.de, 2018). Although a small city, Bautzen is not unknown throughout Germany. The prominence of Bautzen is not only due to the popular mustard or the famous pop band
Silbermond, but also caused by recent right-wing violent incidents against refugees.
Since news media have been watching the problem of right-wing violence in Saxony since the foundation of the right-wing populist movement PEGIDA (Salzborn, 2016), it was 1
not surprising that after an arson attack on future housing for asylum seekers in Bautzen on the 21st of February 2016, local as well as national media became aware of the tense
situation. This incident not only went viral because of the arson attack itself, but also because of a few people who were standing in front of the burning building, applauding and hindering the fire brigade’s work (Mut gegen rechte Gewalt, 2017).
When the so-called `refugee crisis´ came in early 2015, Bautzen had had only little experience with people from abroad. Even in the end of 2016, Bautzen counted only 1,919 foreigners, but 78% of them were from non-European countries (Sachsen.de, 2018). Nevertheless, the online platform Mut gegen rechte Gewalt (2017) has recorded around 20 right-wing violent incidents against refugees in Bautzen in the course of the years 2016 and 2017. These violent incidents do not only include physical attacks against asylum seekers but
Patriotic Europeans Against the Islamization of the the Occident, founded in the end of 2014 in Dresden, the
1
per-also “xenophobic demonstrations, assault, arson attacks, and miscellaneous attacks against refugee housing“ (Benček & Strasheim, 2016, p. 2).
Until now, little research has been conducted into the differences between local and national media coverage of right-wing violent incidents against refugees. Scholars have mostly focussed on specific cases dealing with particular right-wing violent incidents (Bates, 2017; Perthus & Belina, 2017), instead of analyzing the more general problem of racism within a town or a region. Whereas Bates (2017) had a closer look into differences between one local and one national newspaper outlet in regard to their coverage of a racist incident in the British town of Middlesbrough, Perthus and Belina (2017) put their main focus on police press releases, only slightly analyzing different local and national news outlets and their coverage on a violent incident between refugees and locals in Bautzen. Thus, academic research lacks a study focusing on differences between local and national coverage on a town’s broader problem of right-wing violence against refugees over time, which is based on multiple newspaper outlets.
By analyzing newspaper coverage of right-wing violence against refugees in the case of Bautzen, it is important to not only take national but also local newspapers into account. This is due to two reasons: First, local newspapers influence local audiences and thus, the respective political discourse on an issue of public concern (Oliver & Meyer, 1999). Second, local newspapers should not be underestimated in regard to their “relative comprehensiveness in covering local events (Oliver & Meyer, 1999, p. 42). And since most issues of national concern start as incidents of local origin, local news coverage should be considered to be as important for academic research as national news coverage (Oliver & Meyer, 1999).
Studying differences between local and national media coverage is essential, because both newspaper types either shape the local or the national public’s understanding and the
resulting meaning of a controversial issue circulating within society (Habermas, 2001). This is because news media operate in so-called public spheres, serving as a means of
communication between citizens and politicians (Habermas, 2001).
In this regard, albeit assuming that both local as well as national news media shape the discourse on right-wing violence between refugees and locals in the case of Bautzen, I expect local media coverage to differ from national coverage. Therefore, I formulate the following guiding research question:
RQ: To what extent have national and local newspapers differed in their coverage of right-wing violence against refugees in the East-Saxon town of Bautzen in the course of the years 2016/17?
Theory
The following section is divided into two parts: First, I will define the theoretical concepts paving the way to my analysis in order to answer my research question. Second, I will give an overview of former research dealing with differences between local and national media coverage on events of local origin, on which I will then base my expectations.
Theoretical Concepts
In this research study I aim to analyze local as well as national newspaper coverage in regard to their media discourse on right-wing violence against refugees in Bautzen and over time. Discourse can be defined here “as the ongoing comment on or discussion of a particular topic or bundle of related topics in the public arena within a community or
public domains“, such as local or national newspaper outlets, which are oriented towards their respective audiences (Luchtenberg & McLelland, 1998, p. 190).
I understand media coverage as discourse which is described by different aspects of newspaper coverage, including the frequency of coverage (intensity), the frames applied in coverage (content) and the extent to which newspapers overtake the media’s watchdog role (function).
So first, the frequency of coverage describes the intensity of media discourse on right-wing violence against refugees in Bautzen, which is based on the amount of articles published by local and/or national newspapers in the course of the years 2016/17. Thus, the frequency of coverage is an indicator of the amount of attention racism received in local and/or national media discourse over time.
Second and according to Entman (1993), I define framing “ 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). Hence, frames determine the focus set by different newspaper types and thus, the content of a news story. The content of a news story leads media discourse in a specific direction, in which one aspect will might be more discussed than another. If this is the case in Bautzen, framing decisions also indicate
differences between local and national newspaper discourses on right-wing violence against refugees.
Third, in modern democracies, news media function as accountability systems, so-called “watchdogs“, which are independent from political elites and their decisions
(Christmann, Kübler, Hasler, Marcinkowski, & Schermer, 2015). Watchdogs challenge people holding power, they are critical but objective in their coverage and they serve citizens’
Albæk, 2015). In the case of Bautzen, I then call local or national newspapers a watchdog when articles entail a variety of sources, criticism on political elites (e.g. including the tone of coverage), an objective view on the conflict and when articles report from a citizens’
perspective. These criteria can indicate further differences between local and national newspaper discourses on right-wing violence against refugees in Bautzen.
When speaking about `refugees´ or `asylum seekers´, I refer to people, mostly Muslims who fled from Syria or other places in the Middle East and North Africa to the European Union in 2014/15 to date (Doomernik & Glorius, 2016). Thereby, I neglect their official status as either still waiting to be granted asylum, or as having already been allowed to stay in Germany (Mitchell, 2017).
European countries such as Germany have been increasingly confronted with right-wing violence against refugees, which reveals itself in “a groright-wing acceptance for antihuman attitudes“ (Strauß, 2017, p. 4). As a result, these antihuman attitudes lead to a rise of racism manifested by a “separation by ethnic categorization [which] is based on a notion of
difference that is both homogenizing and sociobiological in nature, looking at people only in terms of cultural/ethnical identity and not their subjectivity or individuality“ (Salzborn, 2016, p. 41). Racism and resulting right-wing violent attacks against refugees can be explained best by “Islamophobia“, which describes a hostile and undifferentiated condemnation of Muslims, who are often wrongly and exclusively associated with Islamist terrorists and Islamic
extremists, who make their entry to Europe (Druxes & Simpson, 2016). Expectations and Hypotheses
Even though academic research on differences between local and national media coverage has mainly focussed on national events, e.g. elections or climate change, some scholars analyzed differences between both types of media, when covering incidents of local
origin such as natural disasters, shootings and/or other (racially driven) attacks (Bates, 2017; Holody, Park and Zhang, 2013; Holt & Major, 2010; Molotch & Lester, 1975).
These studies found local press discussing local events more intensively than national press when considering frequency, time period and diversity of coverage. Molotch and Lester (1975) found differences in the focus of the news story set by local and national newspapers on one and the same issue. Holt and Major (2010) concluded that local press uses more diverse frames within coverage than national press but “similar system-based attributions on the crime“ (p.579). Holody et al. (2013) emphasized local newspapers’ quality by finding a greater intensity and diversity of local than of national press coverage.
Scholars mainly explained these differences by referring to the different audiences local and national news media need to serve. Holody et al. (2013) suggested that local media may have an urge to tell the broader story about a local incident to their audiences, whereas national newspapers would rather rely on a more general view on a crime, lacking local sources (Holody et al., 2013; Holt & Major, 2010).
To now bring forward my expectations for this study, I assume that both local as well as national newspapers covered right-wing violent incidents between refugees and locals in Bautzen in the course of the years 2016/17. But since right-wing violence against refugees is only one social problem within a “public arena“, it needs to compete with other issues of social concern (Hilgartner & Bosk, 1988, p. 55). Carrying capacities of those public arenas, i.e. local or national newspaper outlets, are limited, and sooner or later, more ‘newsworthy’ issues circulate within society (Hilgartner & Bosk, 1988). This aligns with agenda setting research, stating that the amount of attention paid depends on “the capacity of news media to cover public issues“ (Holody et al., 2013, p. 578). My hypotheses are thus as follows:
H1a: Newspaper coverage of right-wing violence against refugees in Bautzen has declined in the course of the years 2016/17.
Even though I expect a general decrease of media attention over time, research has shown that local press coverage lasts longer than national press coverage (Holody et al., 2013; Molotch & Lester, 1975). Local newspapers are often forced by a local community to discuss a conflict of local origin more intensively, whereas national newspapers cover local events only as long as they offer some contextual relevance for a broader public (Harry, 2001). Based on the principle of newsworthiness, news values lead national newspapers in particular to access the amount of attention a public event deserves (Bates, 2017). In light of this, I expect:
H1b: National newspaper coverage of right-wing violence against refugees in Bautzen had decreased earlier than local newspaper coverage.
By emphasizing some issues more than others, news media construct reality, which could in turn shape the perceptions of the public. When news media apply certain frames in their coverage, they give some sort of orientation to their audiences, stating an underlying basic idea or assumption (Denner & Peter, 2017).
I understand a frame as being created by an individual journalist who “connects the different framing and reasoning devices in a news article“ (Van Gorp, 2005, p. 487). In light of that, frames are often not used deliberatively but rather evolved on the basis of a cultural motive (Van Gorp, 2005). Therefore, and by stressing a specific frame, news stories can influence people’s evaluation and judgement of a specific incident (Holody et al., 2013).
So far, Perthus and Belina (2017) found criminalization of foreigners to play a central role within media discourse on violence between refugees and locals in Bautzen. Therefore, it is interesting to find out to what extent local as well as national newspapers portrayed
refugees as victims of right-wing violence or as intruders who threat the local environment. In this regard, I will look into the representation of asylum seekers in news stories, making use of the victim and the intruder-frame as developed by Van Gorp (2005).
When presenting asylum seekers as victims, the press raises awareness “that these people are in need of help“ (Van Gorp, 2005, p. 489). On the contrary and by presenting asylum seekers as intruders, the press strengthens the image of a criminal person, who takes advantage of social services and threats the cultural norm (Van Gorp, 2005).
Because of Germany’s history and its national-socialistic era, the problem of racism is still treated carefully and cautiously within the German society (Luchtenberg & McLelland, 1998). Therefore, I expect both national and local newspapers to present asylum seekers as victims rather than as intruders. I therefore hypothesize:
H2a: Both local as well as national newspapers applied the victim-frame more often than the intruder-frame in their coverage of right-wing violence against refugees in
Bautzen.
When now having a look into frames used by both local as well as national
newspapers to cover the situation in Bautzen and the town’s problem of racism, I relied on the attribution of responsibility-frame, the morality-frame and the economic consequences-frame as developed by Semetko and Valkenburg (2000).
The attribution of responsibility-frame searches for an individual, a (social) group or any level of government (Semetko & Valkenburg, 2000), who/which is responsible for the problem of right-wing violence against refugees in Bautzen. The morality-frame refers to a moral message, which is transmitted by news media (Semetko & Valkenburg, 2000). In the case of Bautzen, I expect these moral messages to transmit the moral duty to help refugees, to fight racism, xenophobia and discrimination and/or to appeal to humanity and tolerance, since all three `moralities´are logical argumentation threads to counterbalance against racism. The economic consequences-frame refers to any consequences for economy which result from the negative image Bautzen received due to right-wing violent attacks against refugees.
Research has shown that “National newspapers were significantly more likely to use the morality-frame than local newspapers“ when reporting about a crime of local origin (Holt & Major, 2010, p. 591). Local press was found to use the attribution of responsibility-frame more often than national press (Holt & Major, 2010). In the case of Bautzen, I therefore expect similar results regarding use of the morality- and the attribution of responsibility-frame.
Economic consequences may play a bigger role for the local community than for a national audience, because local industries such as the tourism or science branch are (in)directly affected by the problem of racism and the resulting negative image of Bautzen and/or Saxony. Therefore, I assume that the economic consequences-frame was applied more often by local than by national newspapers. So, I formulate the following hypotheses:
H2b: National newspapers applied the morality-frame more often than local newspapers.
H2c: Local newspapers applied the attribution of responsibility-frame and the economic consequences-frame more often than national newspapers.
In a representative democracy, news media coverage should ideally entail the monitoring of political elites and decisions affecting society (Ferree, Gamson, Gerhard, & Rucht, 2002). If they do so, news media thus act as watchdogs, representing citizens’ needs, raising the alarm for potential threats and reporting as objectively as possible, while including a variety of different sources and opinions (Kalogeropoulos et al., 2015). In the case of Bautzen, I expect local and national newspapers to take on the media’s watchdog role, but to differ in the intensity of performance.
A diversity of different sources, i.e. voices from different groups of society, indicates a newspaper’s balance in coverage, thus presenting a conflict as unbiased as possible. This is an important aspect of the media’s watchdog function. Whereas scholars found national
newspapers to rely mostly on official statements of national sources, local newspapers rather use a variety of local sources, which is potentially caused by better access to and/or
availability of interviewees (Holody et al., 2013; Holt and Major, 2010; Molotch & Lester, 1975). In the case of Bautzen, I therefore expect:
H3a: Local newspapers used a greater diversity of sources than national newspapers.
The media furthermore acts as watchdog when it criticizes political elites and their decisions rather than defending them. But research has shown that national press tend to support political elites (Holt & Major, 2010; Molotch & Lester, 1975), whereas local press serves as a town’s agent (Bates, 2017). Therefore, I expect both newspaper types to criticize
political elites on one another’s level of government, but that local newspapers tend to defend local/regional politicians and national newspapers defend political elites acting in a broader national public sphere. Therefore, I test:
H3b: National newspapers defended national elites and criticized local/regional elites more often than local newspapers.
H3c: Local newspapers defended local/regional elites and criticized national elites more often than national newspapers.
To furthermore test the watchdog criteria of criticism, it is interesting to have a look into the tone in which local and national press reported about the town of Bautzen. Bates (2017) found local newspapers to defend decisions or agendas of a town where right-wing violent incidents against refugees occurred, potentially caused by the feeling to
counterbalance against negative national press coverage. Based on that, I expect:
H3d: The tone applied differed between newspaper types. Whereas national newspapers reported more frequently in a negative tone about Bautzen than local newspapers, local press reported more frequently in a positive, neutral or ambivalent tone about the town than national press.
According to the watchdog function, it is essential that news media report objectively. Objective reporting is, inter alia, given when articles outline two or more sides involved in the conflict (Kalogeropoulos et al., 2015). Thus, I state:
H3e: Both local and national newspapers published more articles outlining multiples sides of the conflict than articles outlining only one side.
Watchdogs ideally report from the citizens’ perspective when monitoring political discourses (Christmann et al., 2015). Since local newspapers serve a local audience and thus, a local community, I expect them to monitor political discourses on a local rather than on a national level. In turn, I assume that national newspapers which serve a national and thus, a broader audience, report on political disagreements on a national level as well as between national and local/regional politicians. Based on these assumptions, I formulate:
H3f: Local newspapers reported more often about disagreements among local/regional elites, whereas national newspapers reported more often about
disagreements among national elites and/or between national and local/regional elites.
By serving citizens, it is also important for watchdogs to have a critical eye on the potential effects of a conflict. But since newspapers operating in smaller towns are more dependent on societal foundations than newspapers serving a whole nation (Harry, 2001), national press can be expected to have a more critical eye on potential effects of the problem of racism in Bautzen than local press.
H3g: National newspapers reported more often about effects on refugees as well as about effects on the town of Bautzen than local newspapers.
Methodology
In this chapter, I will now outline how I put my expectations into practice. In my case study on the East-Saxon town of Bautzen and its problem of racism, I focused on right-wing violent incidents between refugees and locals in the course of the years 2016/17. By
conducting a quantitative content analysis, I will be able to “meaningfully summarize [my findings] numerically“ which allows me to draw some generalizations (Reinard, 2006, p. 6).
My unit of analysis is press articles, because newspaper outlets still serve as so-called “leading media“ agenda setters determining the relevance and importance of a public event (Vollmer & Karakayali, 2017, p. 2). Intensive press coverage on Bautzen and its problem of right-wing violence began with an arson attack on asylum seekers’ future housing on the 21st of February 2016. The majority of right-wing violent attacks against refugees in Bautzen occurred in 2016; but there were still some incidents taking place in 2017 (Mut gegen rechte Gewalt, 2017). Therefore, and in order to see how press coverage has developed over time, I set the time period of analysis from the 21st of February 2016 to the 31st of December 2017. The newspapers chosen for data collection reflect a variety of opinion leading local as well as national, daily and weekly published press outlets with different political orientations. Based on these criteria, I decided upon three local newspaper outlets, namely Sächsische Zeitung (SäZ), Oberlausitzer Kurier (OK) and Bautz’ner /Bischofswerdaer Wochenkurier and seven national newspaper outlets, namely die Tageszeitung (TAZ), DIE ZEIT, Die Süddeutsche (SZ), Der Spiegel, Der Tagesspiegel, Frankfurter Allgemeine Zeitung (FAZ), Die Welt.
Data Collection
My unit of data collection equals my unit of analysis and corresponds to the whole population of newspaper articles that I could generate through the search criteria applied. My final data set consists of 319 newspaper articles (N=319).
First and foremost, data collection was made with the help of the online database LexisNexis, accessible through the UvA online library, and with the online archives of SZ and FAZ, to which I got access through the University of Jena, Germany. For sampling via
LexisNexis, I used a keyword search with the terms “Bautzen“ in headline and lead (to ensure that the article directly refers to the town of Bautzen) AND “Rassismus“ (“racism“, to
comprise racist incidents and/or right-wing violence against refugees with the help of a broader term) AND “Flüchtlinge“ (“refugees“) OR “Geflüchtete“ (“fleeing people“) OR “Asylsuchende“ (“asylum seekers“). Using these three synonyms was important to capture all articles dealing with refugees even though other words were used to refer to them. I applied the same sampling strategy within the online archives of SZ and FAZ, searching for
“Bautzen“ AND “Rassismus“ (“racism“) and for “Bautzen“ AND “Flüchtlinge“ (“fleeing people“). I then integrated the resulting overlap in my data of analysis. Articles from the local newspaper OK were collected manually in the city library of Bautzen while using the same selection criteria.
In a second step, I cleaned the data, excluding articles which did not discuss the situation in Bautzen and/or Saxony directly but only referred to Bautzen while drawing the line to a similar issue.
My unit of data analysis includes all types of newspaper articles from breaking news to editorials to journalistic opinion pieces in order to get an all-embracing image of press coverage about Bautzen’s problem of racism. I disregard letters to the editor or similar articles written by the readership. For data analysis, I applied a deductive approach, coding articles based on a previously prepared codebook. This codebook is based on existing literature and entails nominal and ordinal levels of measurements in order to test my hypotheses. The
codebook entails explicit notes for each operationalization, explaining the measurement applied.
Measures
At first, I measured the frequency of newspaper coverage (H1a, H1b) by counting articles on their respective dates of publication and set them in relation to time. This was done for the sum of articles published by both types of newspapers (N=319) as well as for local (N = 113) and national press outlets (N = 206).
Second, I measured different frames within articles published by both local as well as national newspapers (H2a-H2d). All items were measured with a dichotomous scale to indicate a pattern’s absence (No = 0) or presence (Yes = 1) and thereby, its frequency of occurrence (see Appendix A for detailed information).
Victim- and intruder-frames were measured according to Van Gorp’s (2005)
suggestions on potential coding items. Here, I chose two items per frame, the role of refugees and the problem definition. The victim-frame was then operationalized by the following two items: “refugees are portrayed as being in need of help“ and “the article explains violent incidents between right-wing extremists and refugees by referring to right-wing extremism“. The intruder-frame was operationalized by the following two items: “refugees resort to illegal practices“ and “the article explains violent incidents between right-wing extremists and refugees by referring to asylums seekers’ fundamentalist Islamic beliefs and/or their foreign attitudes“.
Even though a factor analysis showed no correlation among both items measuring the victim-frame (Cronbach’s α = - 0.26 ) and a very weak correlation among both items 2
measuring the intruder-frame (Cronbach’s α = 0.182), I decided to use this measurement for
my analysis. This decision is based on two reasons: At first, Van Gorp (2005) has already and successfully applied both items to measure refugees’ representation in news media. Second, these non-existing/weak correlations could be due to the fact that refugees were very rarely framed as people being in need of help, although newspapers declared almost in every article that refugees are victims of right-wing extremism. Besides, newspapers often described refugees as resorting to illegal practices, because police stated that refugees had an active role during the second violent incident. But here again, newspapers declared refugees as being a threat to the country only occasionally.
Items measuring attribution of responsibility-, morality- and economic consequences-frames were taken from Semetko and Valkenberg (2000) and customized to my research approach. In order to operationalize the responsibility-frame, I created four items specifying responsible persons or groups of persons for the problem of right-wing violence in Bautzen and/or Saxony. These items were: “government is responsible“, “right-wing extremists are responsible“, “refugees are responsible“, “society is responsible“. When at least one of these items was coded as being present, the responsibility-frame appeared within the respective article. Thereby, it was irrelevant whether all three items occurred within an article or one item only.
The same approach was used to measure the morality-frame, specifying the kind of morality the article appealed to. These items were: “appeal to humanity and tolerance“, “moral duty to help people in need (refugees)“ and “ importance to fight racism, xenophobia and/or discrimination“.
To measure economic consequences, I asked whether the article makes any reference to potential consequences for economy resulting from a negative image of Bautzen and/or Saxony which was caused by the problem of racism.
In order to measure differences between local and national newspapers in regard to their watchdog role (H3a-H3g), I had 18 dichotomous questions to indicate an item’s absence (No = 0) or presence (Yes = 1) and thereby, its frequency of occurrence. Different to the other items, “tone“ was measured with an ordinal level of measurement (0 = negative, 1 = neutral, 2 = ambivalent, 3 = positive). All items measured the following patterns of media discourse based on Kalogeropoulos et al. (2015): “variety of different sources“, “criticism/defense of local/national elites“/“tone of coverage “, “objective reporting“ and “serving citizens“. In order to capture the pattern “variety of different sources“, I summarized all items measuring “quotes“ to an index, containing all different sources, i.e. including quotes of local/national politicians, social actors, refugees and/or right-wing extremists (see Appendix A for detailed information).
Reliability
Reliability of the analysis was proven by an intercoder reliability process, in which 10% of the final sample (N=31) was coded by another coder who was trained beforehand. Intercoder reliability was demonstrated for each variable by Krippendorff’s alpha (KALPHA, 0.53 < α < 1, except outliers). Reliability was mostly but not always given since the lowest limit for conceivable results needs to be α ≥ 0.66 (De Swert, 2012). Within my results, I had few but two extreme outliers (see Appendix B for a detailed overview). Variables measuring “government is responsible“ and “moral duty to help people in need (refugees)“ scored low with α = 0.53, whereas variables measuring “refugees are responsible“ and “defense of national elites“ got an α = 0.00. These outliers could be due to the fact that “KALPHA returns low even with very few mistakes of the coders“ when using dichotomous items (De Swert, 2012). That is why only a few different codings were judged as big mistakes when calculating KALPHA for the variables mentioned above.
Analytical Strategy
In order to analyze the frequency of media coverage and to answer H1a, I ran a linear correlation calculation between the total amount of newspaper articles published in regard to the respective dates of publication (time). H1b was similarly tested but by running two linear correlation calculations between the dates of publication and the respective amount of articles published by local press as well as by national press. I also used graphs to visualize my findings.
All other hypotheses (H2a-H3g) were tested with the help of t-tests, comparing means of both newspaper types with the respective item I wanted to measure. In doing so, I tested whether there was a significant difference between local and national newspaper coverage. When having a closer look into proportions, I used descriptive statistics and calculated Chi-Squares to compare an item’s frequency of occurrence in both local and national newspapers.
Results
In this chapter, I will now report the results of this study. Frequency
As illustrated in Figure 1, a linear correlation calculation has shown a weak but significant negative correlation between newspaper coverage and time (r = - 0.26, p = .001), which means that newspaper coverage about right-wing violence against refugees in Bautzen has decreased in the course of the years 2016/17. Thus, I can confirm H1a.
A correlation between the total amount of articles published by national newspapers in relation to respective dates of publication (time) has shown a strong and significant negative association (r = - 0.35, p = .001), which shows that national newspaper coverage has
calculation between the total amount of articles published by local newspapers in relation to respective dates of publication (time) has shown a very weak negative and non-significant correlation (r = - 0.06, p = .608). So unexpectedly and as illustrated in Figure 2, local
newspaper coverage on right-wing violence against refugees has remained nearly constant in the course of the years 2016/17. Therefore, I need to reject H1b.
Figure 1. Intensity of press coverage in the course of the years 2016/17.
Frames
Victim- and intruder-frame. Data suggests that both newspaper types applied the victim-frame more often than the intruder-frame (H2a). As expected in H2a and as shown in Figure 3, it can be confirmed that both local as well as national newspapers portrayed refugees more often as victims than as intruders.
Furthermore, a t-test has shown a significant difference between local and national newspaper coverage in their usage of the victim-frame (t = - 6.71, df = 241, p = .000). Thus, national newspapers presented refugees significantly more often as victims than local
newspapers. By comparing the means of newspaper types using the intruder-frame, I could not find a significant difference between local and national newspaper coverage (t = - 1.30, df = 256, p = .194). Thus, national newspapers used the intruder-frame just as frequently as local newspapers.
Figure 3. Frequency of frames applied by local and national newspapers. Victim and intruder-frame.
Morality-, responsibility- and economic consequences-frame. In relation to H2b, a t-test has shown a significant difference between local and national newspapers in regard to the application of the morality-frame (t = - 3.630, df = 214.838, p = .000). As illustrated in Figure 4, national newspapers applied the morality-frame within their coverage about right-wing violence against refugees significantly more often than local press outlets. Thus, I can confirm H2b.
This result is mainly due to the strong and significant difference between local and national newspaper coverage in regard to the application of the item measuring “the importance of fighting racism, xenophobia and discrimination“ (t = - 4,661, df = 229, p = . 000). National newspapers stressed the importance of fighting racism, xenophobia and discrimination significantly more often than local newspapers.
Figure 4. Frequency of frames applied by local and national newspapers. Morality-, responsibility- and economic consequences-frame.
Even though data also suggests that local newspapers applied the attribution of responsibility-frame more often than national newspapers, the difference between both
newspaper types is very small and not significant (t = 0.99, df = 317, p = .320). Nevertheless, national newspapers attributed responsibility significantly more often to any level of
government than local press (t = - 2.23, df = 258, p = .027), whereas local press attributed responsibility significantly more often to right-wing extremists than national newspapers (t = 3.731, df = 185, p = .000). According to the second part of H2c, a t-test confirmed a
significant difference between local and national newspaper coverage in regard to usage of the economic consequences-frame (t = 2.21, df = 185, p = .028). Even though local newspapers reported significantly more often about potential consequences for economy, both newspaper types used this frame rarely in their coverage. Nevertheless, H2c can partly be confirmed. Watchdog Role
Contrary to H3a, national newspapers used a significantly greater diversity of sources than local newspapers (t = - 4.16, df = 317, p = .000). Within their coverage of right-wing violence against refugees in Bautzen, national press included a bigger variety of direct and indirect quotes than local press as presented in Figure 5. Therefore, I must reject H3a.
National politicians were quoted significantly more often by national newspapers, with a probability of 28.2%, than by local newspapers, which quoted national elites with a probability of only 2.7% (t = - 7.31, df = 283.69, p = .000). The same holds true for quotes of right-wing extremists; national newspapers quoted right-wing extremists significantly more often than local newspapers (t = - 3.62, df = 303.94, p = .000). There were no significant differences between local and national press coverage regarding quotations of local politicians (t = - 0.49, df = 317, p = .625), social actors (t = - 0.31, df = 317, p = .759) and refugees (t = - 1.14, df = 284.86, p = .256). Thus, local politicians, social actors and refugees were quoted by
local newspapers as often as by national newspapers. In comparison to all other groups of sources, refugees were quoted very little by both local (3.5%) and national newspapers (6.3%).
Data suggests that national newspapers defended national elites and criticized local/ regional elites more often than local newspapers (H3b). But whereas national press coverage was significantly different from local press reportage when defending national elites (t = - 2.48, df = 205, p = .014), it was not significantly different to local press coverage when criticizing local/regional elites (t = - 1.10, df = 237.31, p = .273). Local newspapers did not defend national elites at all, and national newspapers defended them rarely, with a probability of 2.9% only. Local elites were criticized more often by national newspapers (39.8%), but not much less by local newspapers (33.6%). Based on these findings, I can only partly confirm H3b.
In turn, I expected local newspapers to defend local/regional elites and to criticize national elites more often than national press (H3c). Here, data does not show a significant difference between coverage from the two newspaper types when defending local/regional elites (t = - 0.65, df = 317, p = .513) but when criticizing national elites (t = - 2.30, df = 282.08, p = .022). But contrary to my assumption, national elites were significantly more often criticized by national newspapers (19.9%) than by local newspapers (10.6%). Thus, I need to reject H3c.
The tone used by local newspapers when reporting about the town of Bautzen is significantly different to the tone applied by national newspapers (t = 7.59, df = 198.33, p = .000). As illustrated in Figure 6, in 42% of all articles, local newspapers reported in a neutral tone about the town of Bautzen. This is very different to national newspapers in which only 19.9% of all articles reported neutrally about Bautzen. Besides, local newspapers applied more often an ambivalent (20.5%) or positive (18.8%) tone when speaking about the town of Bautzen, whereas national newspapers reported most frequently in a negative tone (64.6%). Therefore, I can confirm H3d, stating that local newspapers reported more often in a positive, neutral or ambivalent tone about Bautzen than national newspapers, which in turn, reported most frequently in a negative tone about the town.
Multiple sides of the conflict were mentioned in 64.6% of all articles published by local press and in 84% of all articles published by national press. In this regard, both
newspaper types published more objective articles than unobjective articles, so I can confirm H3e.
Nevertheless, articles published by national newspapers entailed outlines of two or more sides involved in the conflict significantly more often than articles published by local newspapers (t = - 3.73, df = 185.12, p = .000).
Considering media’s task of reporting from citizens’ perspective, local newspapers reported significantly more often about disagreements among local/regional politicians than national newspapers (t = 2.31, df = 192.67, p = .022), whereas national newspapers reported significantly more often about disagreements among national elites and/or between national and local/regional elites than local newspapers (t = - 3.24, df = 307, p = .001). Thus, I can confirm H3f.
According to my expectations, national newspapers discussed effects on refugees significantly more often than local newspapers (t = - 2.91, df = 256, p = .004). But there was no significant difference in regard to the frequency in which local and national newspapers discussed effects on the town of Bautzen and its citizens (t = - 0.25, df = 317, p = .802). Thus, both newspaper types reported as much as the other about the impact of right-wing violence against refugees on the town of Bautzen. Therefore, H3g can only partly be confirmed.
Conclusion and Discussion
In Germany, right-wing violent incidents of local origin are not only of importance for a town or a region, but also for the whole nation. Therefore, both local as well as national newspapers discuss right-wing violent incidents against refugees but differ in their discourses. In the following, I state four main findings of this study.
First, newspaper coverage about right-wing violence is decreasing over time, albeit local press coverage remains more or less constant when covering an incident of local origin.
Second, within press coverage, newspapers often use the narrative of refugees as victims of right-wing extremism; the image of refugees as intruders, who threaten the German culture and/or exploit social services, is almost nonexistent in German press coverage.
Regardless of the attributed role, it is conspicuous that refugees were absolutely
underrepresented. Only very rarely does an article quote and/or reflect a refugee’s opinion. Third, even though local as well as national newspapers discuss to some extent the matter of morality, the attribution of responsibility and potential consequences for economy, both newspaper types differ in the focus set on the issue of right-wing violence against refugees. Whereas national press coverage focuses on the importance of fighting racism, xenophobia and discrimination, local press coverage concentrates on the facts surrounding the situation and thus, on the matter of responsibility. Potential economic consequences, resulting from a negative image and caused by the problem of racism, are more often part of local media discourse but did not set the main focus in either local or national newspaper coverage.
Fourth, analysis has shown that the national press can be called the `better watchdog´, because national newspaper discourse entails a greater diversity of sources, more objective reports and more criticism of national as well as local elites than local newspaper discourse.
Luchtenberg and McLelland (1998) saw the intense media discourse on racism in Germany as being related to its ‘Nazi era’. Differences between national and local newspaper coverage can be explained by agenda setting research which says that “the type of media determines the level of issue competition“ (Holody et al., 2013, p. 578).
National newspapers often start reporting on an event of local origin when a “conflict escalates“ (Harry, 2001, p. 428). But since national newspapers serve a broader audience, they have more issues to report about. Hence, issue competition is tougher for national news coverage than for local news coverage and one possible reason for the decrease of attention over time (Holody et al., 2013). On the contrary, local newspapers are forced by the local community to discuss a local conflict intensively. Thus, local news coverage often lasts longer than national news coverage (Harry, 2001).
That national newspapers are the better watchdogs could be caused by the fact that news media operating in smaller towns, i.e. in Bautzen, often rely on a consensus which is set by the community they serve (Harry, 2001). Thus, for local press it is sometimes hard to report as critical as national press due to their dependence on so-called “primary definition-makers“ such as local politicians (Belina & Perthus, 2017, p. 251).
But it could also be argued that there is a need in public discourse to blame something or somebody to ease complex issues, which was mainly detected within national press
coverage, stressing the matter of morality (Holt & Major, 2010). Although local press reported more often in a positive tone about Bautzen than national press, local newspapers reported most often in a neutral tone about the town, whereas national newspapers did not hesitate to shade a negative light on Bautzen.
Spokesperson for Bautzen, A. Wucht (personal communication, April 11, 2018), confirmed that he also had the impression that journalists from national news outlets came to
Bautzen only with the task of finding racism on Bautzen’s streets. U. Schönbach (personal communication, April 17, 2018), editor-in-chief of the Sächsische Zeitung, regional edition Bautzen, agreed that Bautzen was used by national news media to make an example of right-wing extremism in Saxony without seeing any necessity to further investigate actual
circumstances.
Limitations and Future Research
Since content analysis is descriptive in nature, my findings give information about what has happened is newspaper coverage, but they do not offer deeper insights into the reasons why local or national newspapers reported as they did (Reinard, 2006).
The study’s validity is given by the theoretical framework outlined above and by the fact that I did not need to draw a sample but studied the entire population (N=319). But since my research still focuses on one specific case only and my data solely refers to right-wing violent incidents between refugees and locals in Bautzen, generalizations regarding
differences between local and national newspaper coverage should be treated carefully and can be adapted best and confidently to most similar cases.
Since I only analyzed newspaper articles, my findings are not easily applicable for other media outlets, e.g. television or radio. Besides, the non-existing correlation between items measuring the victim- and intruder-frame are another weakness of this study, which should have been measured and changed before analyzing. If multiple items had correlated, a frame would have been more present within coverage. Discrepancies in coding could be due to the translation of the original codebook from English into German. The translated
codebook was used by the other coder, whereas I used the original one in English.
Future research should look for different but similar cases of right-wing violence against refugees and analyze differences between newspaper outlets with varying political orientations or different types of news media such as television or radio. Because I expect that differences between local and national media coverage about right-wing violence can shape a society’s feeling of being used as scapegoat to examine a broader problem, research would profit if scholars intended to apply a more qualitative approach, e.g. interviews, to investigate effects on small towns such as Bautzen, caused by differences between local and national media coverage. Results could then be used to prevent local societies from counterbalancing against a negative image, which ultimately urges more than the investigations of the actual problem. So far, this study has enriched media research on differences between local and national media discourses on right-wing violence against refugees by analyzing the whole population of newspaper articles dealing with a town's problem of racism over time.
References
Bates, D. (2017). The “red door” controversy—Middlesbrough’s asylum seekers and the discursive politics of racism. Journal of Community and Applied Social Psychology, 27(2), 126–136. https://doi.org/10.1002/casp.2300
Bautzen.de (2018). Zahlen und Fakten. Retrieved from https://www.bautzen.de/buerger-rathaus-politik/stadtportraet/zahlen-fakten/
Benček, D., & Strasheim, J. (2016). Refugees welcome? A dataset on anti-refugee violence in Germany. Research & Politics, 3(4), 205316801667959. https://doi.org/
10.1177/2053168016679590
Braun, R., & Koopmans, R. (2010). The diffusion of ethnic violence in Germany: The role of social similarity. European Sociological Review, 26(1), 111–123. https://doi.org/
10.1093/esr/jcp056
Christmann, A., Kübler, D., Hasler, K., Marcinkowksi, F., & Schemer, C. (2015). New
Regionalism—Not Too Complex for the Media Watchdog. Urban Affairs Review, 51(5), 676-707. https://doi.org/10.1177/1078087414549549
De Swert, K. (2012). Calculating inter-coder reliability in media content analysis using Krippendorff’s Alpha. Retrieved from https://www.polcomm.org/wp-content/uploads/ ICR01022012.pdf
Denner, N., & Peter, C. (2017). Der Begriff Lügenpresse in deutschen Tageszeitungen. Publizistik, 62(3), 273–297. https://doi.org/10.1007/s11616-017-0354-4
Doomernik, J., & Glorius, B. (2016). Refugee migration and local demarcations: New Insight into European Localities. Journal of Refugee Studies, 29(4), 429–439. https://doi.org/ 10.1093/jrs/few041
Druxes, H., & Simpson, P. A. (2016). Pegida as a European Far-Right Populist Movement. German Politics and Society, 34(4), 1–16. https://doi.org/10.3167/gps.2016.340401 Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of
Communication, 43(4), 51. Retrieved from http://proxy.uba.uva.nl:2048/docview/ 1302274657?accountid=14615
Ferree, M. M., Gamson, W., Gerhard, J., & Rucht, D. (2002). Four Models of the Public Sphere in Modern Democracies. Theory and Society, 31(3), 289-324. Retrieved from http://www.jstor.org/stable/658129?seq=1#page_scan_tab_contents
Habermas, J. (2001). Why Europe needs a constitution. New Left Review, (11), 5-26. Retrieved from https://newleftreview.org/II/11/jurgen-habermas-why-europe-needs-a-constitution
Harry, J. (2001). Covering Conflict: A Structural-Pluralist Analysis of how a Small-Town and a Big-City Newspaper Reported an Environmental Controversy. Journalism & Mass Communication Quarterly, 78(3), 419-436. https://doi.org/doi.org/
10.1177/107769900107800302
Hilgartner, S., & Bosk, C. (1988). The Rise and Fall of Social Problems: A Public Arenas Model. American Journal of Sociology, 94(1), 53-78. Retrieved from http://
www.jstor.org.proxy.uba.uva.nl:2048/stable/2781022
Holody, K. J., Park, S. Y., & Zhang, X. (2013). RACIALIZATION OF THE VIRGINIA TECH SHOOTINGS: A comparison of local and national newspapers. Journalism Studies, 14(4), 568–583. https://doi.org/10.1080/1461670X.2012.726499
Holt, L., & Major, L. (2010). Frame and Blame: An Analysis of How National and Local Newspapers Framed the Jena Six Controversy. Journalism & Mass Communication Quarterly, 87(3-4), 582-597. https://doi.org/10.1177/107769901008700309
Kalogeropoulos, A., Svensson, H. M., Van Dalen, A., De Vreese, C., & Albæk, E. (2015). Are watchdogs doing their business? Media coverage of economic news. Journalism, 16(8), 993–1009. https://doi.org/10.1177/1464884914554167
Krueger, A., & Pischke, J. (1997). A Statistical Analysis of Crime against Foreigners in Unified Germany. The Journal of Human Resources, 32(1), 182-209. Retrieved from http://www.jstor.org/stable/146245
Luchtenberg, S., & Mclelland, N. (1998). Multiculturalism , Migration and Racism: the role of the media. A comparative study of Australian and German print media. Journal of Intercultural Studies, 19(2), 187–206. https://doi.org/10.1080/07256868.1998.9963463 Mitchell, H. (2017, August 17). The distinction between asylum seekers and refugees.
Migration Watch UK. Retrieved from https://www.migrationwatchuk.org/briefing-paper/70
Molotch, H., & Lester, M. (1975). Accidental News: The Great Oil Spill as Local Occurrence and National Event. American Journal of Sociology, 81(2), 235-260. Retrieved from http://www.jstor.org.proxy.uba.uva.nl:2048/stable/2777377
Mut gegen rechte Gewalt (2017). Karte flüchtlingsfeindlicher Vorfälle. Retrieved from https:// www.mut-gegen-rechte-gewalt.de/chronik-karte
Oliver, P. E., & Meyer, D. J. (1999). How Events Enter the Public Sphere: Conflict, Location, and Sponsorship in Local Newspaper Coverage of Public Events. American Journal of Sociology, 105(1), 38–87. https://doi.org/10.1086/210267
Perthus, S., & Belina, B. (2017). Policing the Crisis in Bautzen. Soziale Probleme, 241–259. https://doi.org/10.1007/s41059-017-0035-6
Publications. Chapter 1, “Using statistics to conduct quantitative research”, 3–16. Retrieved from uva blackboard.
Sachsen.de (2018). Regionaldaten Gemeindestatistik Sachsen. Gemeindestatistik 2016 für Bautzen, Stadt. Retrieved from https://www.statistik.sachsen.de/Gemeindetabelle/jsp/ GMDAGS.jsp?Jahr=2016&Ags=14625020
Salzborn, S. (2016). Renaissance of the New Right in Germany? A Discussion of New Right Elements in German Right-wing Extremism Today. German Politics and Society, 34(2). https://doi.org/10.3167/gps.2016.340203
Semetko, H. a., & Valkenburg, P. M. (2000). Framing European Politics. A Content Analysis of Press and Televison News. Journal of Communication, 50(September 1998), 93– 1009. https://doi.org/10.1111/j.1460-2466.2000.tb02843.x
Strauß, H. (2017). Comparatively Rich and Reactionary: Germany between “Welcome Culture“ and Re-established Racism. Critical Sociology, 43(1), 3–10. https://doi.org/ 10.1177/0896920516683643
Van Gorp, B. (2005). Where is the Frame? European Journal of Communication, 20(4), 484– 507. https://doi.org/10.1177/0267323105058253
Vollmer, B., & Karakayali, S. (2017). The Volatility of the Discourse on Refugees in Germany. Journal of Immigrant and Refugee Studies, 2948, 1–22. https://doi.org/ 10.1080/15562948.2017.1288284
Appendix A Codebook Unit of data collection.
News articles retrieved from three local newspapers:
Sächsische Zeitung/ Regionalausgabe Bautzen (SäZ), Oberlausitzer Kurier (OK) and Bautz’ner und Bischofswerdaer Wochenkurier (BBW).
News articles retrieved from seven national newspapers:
die Tageszeitung (TAZ), Die Zeit (Zt), Die Süddeutsche Zeitung (SZ), Der Tagesspiegel (Tag), Der Spiegel (Sp), Die Frankfurter Allgemeine Zeitung (FAZ), Die Welt (W).
Coder ID. 1= Gwendolin
2= Maximilian Schochert (intercoder reliability) Article ID.
The code should follow the format: newspaper abbreviation/author’s initials/day/month/year, e.g. SäZ/MS/21/02/2016.
If there is no author mentioned, please use the shortcut “NA“ for not available. If there are two or more than two articles per day which are written by an unknown author, start counting and add the number to the shortcut, e.g. “NA2“.
Date of publication. Please note the date of publication: ___________________ Source. Please choose the newspaper outlet from which the news article was retrieved:
1 = Sächsische Zeitung/ Regionalausgabe Bautzen (SäZ) 2 = Oberlausitzer Kurier (OK)
3 = Bautz’ner und Bischofswerda Wochenkurier (BBW) 4 = Tageszeitung (TAZ)
5 = Die Zeit (Zt)
6 = Die Süddeutsche Zeitung (SZ) 7 = Der Tagesspiegel (Tag)
8 = Der Spiegel (Sp)
9 = Die Frankfurter Allgemeine Zeitung (FAZ) 10 = Die Welt (W)
1. Victim and intruder-frame.
1a. Passive role (victim). Does the article refer to refugees/ asylum seekers as being in need of help?
1 = Yes 0 = No
1b. Active role (intruder). Does the article refer to refugees/ asylum seekers as being criminals who resort to illegal practices?
Note: The story refers to refugees’ complicity in violent incidents in Bautzen.
1 = Yes 0 = No
1c. Problem definition (victim). Does the article explain violent incidents between right-wing extremists and refugees by referring to right-right-wing extremism, racism and anti-immigrant sentiments?
Note: The story refers to one of these aspects mentioned above in order to explain violent incidents in Bautzen.
1 = Yes 0 = No
1d. Problem definition (intruder). Does the article explain violent incidents between right-wing extremists and refugees by referring to asylums seekers’ fundamentalist Islamic beliefs and/or their “foreign“ (or different) attitudes?
Note: The story refers to one of these aspects mentioned above in order to explain violent incidents in Bautzen.
1 = Yes 0 = No
2. Attribution of responsibility-frame.
2a. Does the article suggest that any level of government is responsible for the problem of racism in Bautzen?
Note: The story considers at least one time that any level of government is responsible for the problem of racism in Bautzen.
1 = Yes 0 = No
2b. Does the article suggest that refugees/ asylum seekers are responsible for the problem of racism in Bautzen?
Note: The story considers at least one time that refugees/ asylum seekers are responsible for the problem of racism in Bautzen.
1 = Yes 0 = No
2c. Does the article suggest that right-wing extremist individuals or groups are responsible for the problem of racism in Bautzen?
Note: The story considers at least one time that right-wing extremist individuals or groups are responsible for the problem of racism in Bautzen.
1 = Yes 0 = No
2d. Does the article suggest that society is responsible for the problem of racism in Bautzen?
Note: The story considers at least one time that society is responsible for the problem of racism in Bautzen.
1 = Yes 0 = No 3. Morality-frame.
3a. Does the article consider the moral duty to help refugees?
Note: The story emphasizes the necessity and the important role of the civil society to support refugees and help them to integrate into society.
1 = Yes 0 = No
3b. Does the article consider the importance of fighting racism, xenophobia and discrimination?
Note: The story emphasizes the importance to stand and speak against general racism and/or xenophobia and/or discrimination of refugees.
1 = Yes 0 = No
3c. Does the article entail any appeal to humanity and tolerance?
Note: The story appeals to humanity and/or tolerance when mentioning and/or referring to one of those values at least one time.
1 = Yes 0 = No
4. Economic consequences-frame. Is there any reference to economic consequences due to the negative image of Bautzen caused by right-wing violent incidents?
Note: The story makes at least one time reference to possible consequences for Bautzen and/or Saxony due to a negative image caused by the problem of racism in Bautzen..
1 = Yes 0 = No
Newspapers as watchdogs.
5a. Defense/Criticism. Does the article entail any defense of local/regional politicians?
Note: The story defends local/regional politicians when supporting their arguments, reactions or suggestions. Local politicians are responsible persons acting in the municipality of Bautzen and regional politicians are responsible persons acting on the federal state level of Saxony.
1 = Yes 0 = No
5b. Defense/Criticism. Does the article entail any criticism of local/regional politicians?
Note: The story criticizes local/regional politicians when questioning their arguments, reactions or suggestions. Local politicians are responsible persons acting in the municipality of Bautzen and regional politicians are responsible persons acting on the federal state level of Saxony.
1 = Yes 0 = No
5c. Defense/Criticism. Does the article entail any defense of national politicians?
Note: The story defends national politicians when supporting their arguments, reactions or
suggestions. National politicians are responsible persons acting on the national level of the federal republic of Germany.
1 = Yes 0 = No
5d. Defense/Criticism. Does the article entail any criticism of national politicians?
Note: The story criticizes local/regional politicians when questioning their arguments, reactions or suggestions. National politicians are responsible persons acting on the national level of the federal republic of Germany.
1 = Yes 0 = No
6a. Sources. Does the article entail quotes from local/regional elites?
Note: Quotes are recorded opinions or expressions of sources. Local/regional elites are politicians acting on the federal state or municipality level of governance (Saxony or the municipality of Bautzen). The code applies when local/regional politicians are either directly or indirectly quoted.
1 = Yes 0 = No
6b. Sources. Does the article entail quotes from national elites?
Note: Quotes are recorded opinions or expressions of sources. National elites are politicians acting on the national level of governance (Federal Republic of Germany). The code applies when national politicians are either directly or indirectly quoted.
1 = Yes 0 = No
6c. Sources. Does the article entail quotes from social actors?
Note: Quotes are recorded opinions or expressions of sources. Social actors are citizens of the town of Bautzen who either work in social initiatives, projects or other social measures, aiming to fight racism and xenophobia in Bautzen as e.g. churches, police, street worker, youth center managers, artists/
actors, etc. The code applies when social actors are either directly or indirectly quoted. References to police press conferences do not count as quotes, whereas statements of police men do count as quotes.
1 = Yes 0 = No
6d. Sources. Does the article entail quotes from refugees/ asylum seekers?
Note: Quotes are recorded opinions or expressions of sources. Refugees/ asylum seekers are people who flew from the Middle East or North Africa due to persecution, violence and/or war and seek for asylum in Germany/ Bautzen. The code applies when refugees/ asylum seekers are either directly or indirectly coded.
1 = Yes 0 = No
6e. Sources. Does the article entail quotes from right-wing extremists?
Note: Quotes are recorded opinions or expressions of sources. Right-wing extremists are people who are more than critical towards Germany’s asylum policy (“welcome culture“) and/ or have
discriminating and/or xenophobic and/or racist attitudes towards refugees/asylum seekers. Right-wing extremists want refugees/ asylum seekers to leave Germany or - at least - the town of Bautzen. In case of Bautzen, right-wing extremists showed themselves willing to use violent and/or criminal offenses in order to threat refugees/ asylum seekers. The code applies when right-wing extremists are either directly or indirectly coded.
1 = Yes 0 = No
7a. Objectivity. Does the article refer to two sides or to more than two sides being involved in the conflict in Bautzen?
Note: The story outlines arguments from two reproaching sides or more than two reproaching sides involved in the conflict in Bautzen. Reproaching sides are not only right-wing extremists against refugees/ asylum seekers but also politicians, social actors and/or others (i.e.“the media“) who have different opinions/attitudes/approaches on how to solve the conflict.
1 = Yes 0 = No
7b. Serving citizens. Does the article report about disagreements among national elites and/or between national and local/regional elites?
Note: The story entails critical discourse on a national level when discussing disagreements among politicians acting on the national level and/or between national and local/regional politicians acting on the federal state level or on the municipality level in regard to the accusation of responsibilities/ competencies/policies and/ or possible solutions etc.
1 = Yes 0 = No
7c. Serving citizens. Does the article report about disagreements among local/ regional elites?
Note: The story entails critical discourse on a local/regional level when discussing disagreements among local/regional politicians in regard to the accusation of responsibilities/competencies/policies and/ or possible solutions etc.
1 = Yes 0 = No
7d. Serving citizens. Does the story report on any broader effects on refugees/ asylum seekers due to the problem of racism in Bautzen?
Note: The story considers any effects on refugees/ asylum seekers due to the problem of racism in the town of Bautzen, e.g. being afraid/homesick, need to move etc. The code does not apply when articles purely summarize what has happened, i.e. when refugees were attacked. But the code applies when the article stresses and outlines several attacks on refugees, e.g there was a chase, afterwards, refugees were hit by stones, etc.
1 = Yes 0 = No
7e. Serving citizens. Does the story report on any broader effects on Bautzen and its citizens due to the problem of racism in Bautzen?
Note: The story considers any effects on the town Bautzen and/or on its citizens due to the problem of racism in the town of Bautzen, e.g. more police controls, stricter rules, etc.
1 = Yes 0 = No
8. Criticism/Tone. How is the overall tone of the article when speaking about the town of Bautzen?
Note: The overall tone of an article is the final impression you get from the news article, summing up, i.e. positive or negative frames, quotes, portrays and/or arguments.
0 = Negative 1 = Neutral 2 = Ambivalent 3 = Positive
Appendix B
