Framing Public Opinion: the Application of Audience Frames to EU News Coverage
Kristian Gotthelf Schmidt Andersen E-mail: kristian.andersen@student.uva.nl
Student ID: 11081910
27 May 2016
Master’s Thesis University of Amsterdam Graduate School of Communication
Erasmus Mundus Masters in Journalism, Media & Globalisation Supervisor: Judith Möller
Abstract
News framing has been shown to affect public opinion on EU affairs, but often research relies on predefined frames that little reveal little about which frames citizens actually use to make sense of EU politics. This study advances a novel approach in framing studies that explores how audience frames of the EU are represented in EU news coverage. A quantitative content of Danish broadsheets (N=500) across five routine periods and five key EU events between 2000-2015 explored the prevalence of audience frames drawn from focus group data. It was found that the EU is framed predominantly as a diplomatic or legislative actor, which does reflect the diversity of the citizen perceptions. Furthermore, the study
investigated the effect of visibility of EU news on the prevalence of the audience frames and found that frames, which citizens evaluate negatively, are significantly more prominent during key events. I conclude with a discussion on news coverage of the EU’s diplomatic role, the relevance of a conceptual distinction between periods with high visibility of EU news and key events, and how the findings can point in new directions for research on the relationship between EU news coverage and public opinion dynamics.
Keywords: news framing, public opinion, EU visibility, EU perceptions, audience frames, issue-specific frames, media content analysis.
Framing Public Opinion: the Application of Audience Frames to EU News Coverage Introduction
A growing body of academic literature is concerned with how the European Union (EU) is covered in national news coverage. Media remains the most important source for information about the EU for European citizens, and various studies have shown that news framing affect public opinion on EU affairs (European Commission, 2014; Schuck & de Vreese, 2006; Maier & Rittberger, 2008; Vliegenthart, Schuck, Boomgaarden, de Vreese, 2010; Corbu, Ștefăniță, Oprea, & Udrea, 2015). In fact, due to a general lack of knowledge about EU politics among European citizens news about the EU is argued to have a
particularly strong effect on EU attitudes and even small fractions of information can play central role for citizens when forming opinions about it (Baccini, Sudulich, & Wall, 2013; Hobolt, 2006; Elenbaas, Boomgaarden, de Vreese, & Schuck, 2012).
Researchers have often used framing analyses to examine how citizens make sense of EU politics (Lecheler & de Vreese, 2010). However, little is still known about which frames news media make available in news coverage for citizens to make sense of. Rather, many of the studies that use framing to link media coverage with public opinion on EU affairs use frames designed by researchers as a point of reference for exploring a specific effect on news audiences (Boomgaarden & de Vreese, 2003; Maier & Rittberger, 2008; de Vreese, van der Brug, & Hobolt, 2012;). By only focusing on predefined frames, however, this type of studies overlook which alternative frames citizens actually use to make sense of EU politics, and which may not be part of the news coverage. One study on the naturalization of immigrants in Switzerland even found that the frame most commonly used by citizens to explain the issue was not at all present in the corresponding media coverage (Wettstein, 2012). For the present study, I am interested in how citizens conceive of the EU by and large, and in through a quantitative content analysis I will explore how these perceptions relate to media coverage
of EU affairs. The analysis will employ issue-specific frames, as they have been found to have a strong effect on not only whether citizens think positively or negatively about the EU, but also which topical considerations they highlight, when explaining their opinion about it (Lecheler & de Vreese, 2010).
The present study will add to the existing literature on the linkage between EU news frames and public opinion by exploring how citizen perceptions of the EU are represented in news stories. Public perceptions of the EU is concerned with how citizens understand the EU (Diez-Medrano, 2010). For the present study I will rely on real-world data to increase
external validity, and thus I am interested in public perceptions that have been defined by EU citizens themselves (Kinder, 2007).
Citizen perceptions of the EU have been to be highly contingent on country-specific historical experiences and variations are common between the member states (Diez-Medrano, 2003). As benchmark for the present study, I will rely on the results of focus group studies conducted with 157 citizens from 18 different member states in lieu of the Horizon EU project (see Timmerman, 2014). One drawback of using this study is that it has not been published in an academic journal. Still, it follows scientific criteria and the participants of the focus groups were sampled to establish a representative sample of European citizens taking into consideration socio-demographic criteria such as age, socio-professional status, gender, and nationality (see Timmerman, 2014, 1). Timmerman (2014) has the advantage that it provides a recent overview of both positive and negative perceptions of the EU generated by EU citizens. Alternative sources for citizen perceptions of the EU are few and are based on out-dated data (Diez-Medrano, 2003) or only focus on positive aspects (European
Commission, 2015).
In the study Timmerman (2014) listed the positive and negative aspects that the participants associated with the EU and found that citizens across the member states have six
positive and five negative perceptions of the EU. The positive aspect mentioned in most countries was the free movement of people (16 countries), followed by European values (14 countries), EU funding (11 countries), the free movement of goods, services, and capital and the single market (10 countries), and last the establishment of common standards and
regulations (7 countries), and economic and diplomatic advantages (7 countries). The negative aspects were inadequate overregulation (14 countries), limited successes (12 countries), lack of solidarity between EU member states (10 countries), bureaucratic EU (10 countries), and gap between EU and its citizens (8 countries). Based on these findings I will develop a set of frames and through a quantitative content analysis explore how the EU in the shape of citizen-defined constructs is presented to news audiences.
In addition, the present study considers the potential of contextual factors to affect how news coverage is framed. De Vreese (2003) has shown that the topical nature of key EU events influences which news frames are most prominent in the coverage. To obtain a reliable overview of EU coverage I include news from different types of events between the years 2000 and 2015. With the aim to investigate the prominence of the audience frames I ask the following exploratory research question:
RQ1: How is the EU framed in Danish newspapers between 2000 and 2015?
Denmark represents an interesting case, as the EU long has been a highly contested topic in the Danish public debate (Green-Pedersen, 2012). However, numerous studies have shown that how the EU is covered in the media is contingent upon from country you collect your data (Oppermann & Viehrig, 2009; Chaban & Bain, 2014). The results from the present study consequently do not apply to other EU member states.
A second central concept in the present study is visibility of EU news in the media. In previous studies two main trends emerge. First, EU as a news topic is only marginally
covered compared to other topics (de Vreese, 2001). Secondly, EU affairs are almost only covered during important EU-related events. In a cross-national study of news around three different EU events, De Vreese (2003) finds that coverage of EU affairs is cyclical peaking around the events, but hardly noticeable both before and after. Despite the scholarly interest in visibility of EU news, no study has investigated whether the variances in visibility has an effect on which frames are prominent in the news coverage. However, it is reasonable to assume that during key events more journalistic resources are allocated to covering the EU and that higher contestation among political parties will take place (Schuck, Xezonakis, Elenbaas, Banducci, & De Vreese 2011). Consequently, one can assume that a wider variety of issues will be discussed. Moreover, it is relevant to explore the relationship between visibility and news framing with respect to public opinion. First of all, citizens are more likely to discuss EU affairs, when visibility of the EU is high (Corbu et al., 2015). Secondly, with high visibility of a given political issue citizens are more likely to form resilient
opinions that are aligned with their fundamental beliefs (Hobolt, 2005). Consequently, it is important to understand how visibility of EU news affects news framing, and I ask the subsequent research question:
RQ2: How is the EU framed in routine periods and during key events?
One study has identified specific events that have a strong effect on visibility of EU news: national referenda, elections for the European Parliament, installations of a new European Commission, enlargement rounds, EU Council summits in a country holding the presidency, and signing of EU treaties (Boomgaarden, Vliegenhart, de Vreese, & Schuck, 2010). Based on these findings I select five key events and five routine periods that include none of the events. Ideally these represent periods with respectively high and low visibility of EU news, however this assumption will be tested in the analysis.
Theoretical Framework Agenda Setting and Framing
Agenda setting. The speed, intensity and overload of information in the modern world put high demands on citizens to process information and require them to be selective when using information to make sense of political issues. In the context of public opinion formation Iyengar (1990) refers to an accessibility bias, which means that due to
psychological processes, information that is more easily retrieved from memory play a more significant role when forming judgments. Accessibility of information is a central part of agenda setting theory, which stipulates that issues that are more frequently covered in the media are more likely to be what the audience draws on when evaluating political affairs. (Scheufele & Tewksbury, 2007). Thus, the volume or visibility of a given issue in the media is a central aspect of opinion formation.
Second level agenda setting or framing. In contemporary studies framing has been coined as a concept that is an closely related to setting, so-called second level agenda-setting, where media not only makes certain issues more salient to the public, but where it also tells them how to think about them (McCombs et al., 1997; Lee et al., 2006). Recent literature argues that while the two theories deal with similar aspects of media studies they are conceptually different (Scheufele & Tewksbury, 2007; Weaver, 2007; Borah, 2011). Both second level agenda setting and framing are concerned with how certain issues are presented in the media (Weaver, 2007). Some scholars argue that rather than explaining media effects with the accessibility model, framing focuses on the applicability of media texts and relates to a wider range of cognitive processes including moral evaluations (Scheufele & Tewksbury, 2007; Weaver, 2007). Framing involves suggesting a connection between different concepts, and after exposure to the media text the audience will either accept or reject the suggested
connection depending on how well the information resonates with the audience’s pre-existing ideas (Nisbet, 2009). Theory on emphasis framing argues that media coverage, which
emphasises certain considerations of a political issue rather than others, leads audiences to link the issue at hand to these considerations (Borah, 2011). Connecting elements from agenda-setting literature with framing theory has practical relevance for understanding public perceptions of political issues, because an applicable construct (an audience frame) is more likely to be activated and thus shape judgments, when the issue at hand is accessible in media coverage (Scheufele, 2007). Along the same lines, Hobolt (2005) argues in a study on voting behaviour in EU referenda that when voters have access to more information about the policy issue, the more motivated they are to consider the consequences of their decision and vote according to their underlying attitude. This theoretical backdrop serves as a rationale to compare how the EU is represented in Danish news during key events and routine periods as well as to pay particular attention to how news frames align with public perceptions of the EU.
Framing Theory
Definition. What constitutes a frame? According to Entman’s (1993, 52) widely used conceptualization frames promote a particular “problem definition, causal interpretation, moral evaluation, and/or treatment recommendation”. If a communicative text presents information that reinforce one or all of the above four framing functions in a certain way, the recipients of that text are likely be guided by it. Entman’s conceptualization suggests that frames are implicitly evaluated, which aligns well with the present studies focus on aspects of the EU, which in Timmerman (2014) have been defined as either inherently negative or positive.
Media frames and audience frames. When discussing framing effects it is important to distinguish between media frames and audience frames (de Vreese, 2005). Simply put,
media frames relate to how the news is presented in the media setting, and audience frames determine how audiences make sense of news through schemas of interpretation (Chaban & Bain, 2014). In the present study I aim to integrate how citizens perceive different aspects of the EU (audience frames) with media frames found in news stories. Media frames need not to only deal with the factual content of news. Lee et al. (2006, 502) argue that news framing of a particular issue goes beyond the textual features and should be understood as a “process of organizing a news story, thematically, stylistically and factually to convey a specific story line.” This definition implies that a story about e.g. the EU is structured in a certain way that ascribes certain characteristics to the EU (Chaban & Bain, 2014).
Issue-specific frames. News framing studies generally distinguish between generic frames and issue-specific frames (de Vreese, 2005). Generic frames transcend the topical nature of an issue and can be found across different contexts and time periods. Issue-specific frames have the advantage that they allow the researcher to investigate the issue at hand at a high level of specificity (de Vreese & Lecheler, 2010). Issue-specific frames have been criticized for too easily leading researchers to find what they are looking for and for being hard to generalize and useless for theory-building (Hertog & McLeod, 2001). The present study is dedicated to issue-specific frames that look at how the EU coverage resonates with concepts that are generated from focus group data. By adapting concepts that have been identified by citizens into news frames instead of developing frames independently I decrease the risk of preconceived ideas leading me to find evidence that support them.
Inherently evaluated frames. Some previous framing of EU news have regarded frames as inherently neutral and then added a variable that allows them to code for either positive, negative, mixed, or lack of evaluation of the EU (see Schuck, Xezonakis, Elenbaas, Banducci, & de Vreese, 2010). In these studies evaluations were only coded as judgments that were ‘undoubtedly positive or negative’ such as ‘good’, ‘promising’, or ‘disappointing’
(Schuck et al., 2010, p. 47). Other studies do not separate evaluations from the frame itself and rather regard some frames to be inherently valenced meaning that the frames themselves carry positive or negative elements (de Vreese & Boomgaarden, 2003). These studies’ conceptualization of framing is in line with Entman’s (1993). Studies on valence framing are concerned with dichotomous frames, which the researchers find relevant to explore such as advantageous/disadvantageous (de Vreese & Boomgaarden, 2003) or risk/opportunity
(Schuck & de Vreese, 2006). My study relies on similar conceptual understanding of framing as inherently evaluated. By investigating frames, which audiences have predefined as either negative or positive, allows me to make a more detailed exploration of how EU news coverage conforms to public opinion that goes beyond how news texts present specific dichotomous pairs of valenced frames.
Visibility
This study aims to explore the prevalence of news frames across periods with low high and low visibility. Visibility of the EU in news has been commonly used in
communication studies (Boomgaarden, Vliegenhart, de Vreese, & Schuck, 2010;
Boomgaarden & de Vreese, 2015; Corbu, Ștefăniță, Oprea, & Udrea, 2015). However, often a clear definition of visibility itself is not explicated. As a conceptual substitute, I lean on Kiousis’ (2004) conceptualization of attention as an attribute of media salience (p. 74): ”From this perspective, salience is akin to media awareness of an object, usually gauged by the sheer volume of stories or space dedicated to topics in newspapers, television news, and so on.” This definition invites to connect the conceptual and operational framework of news visibility.
Literature Review
The present study builds on literature about both how the EU is framed in news coverage and how EU news frames relate to public opinion dynamics.
EU news and public opinion. The rich body of literature on EU news and public opinion has employed various approaches. Most studies investigate the effect of media framing on public opinion using an experimental setting (De Vreese & Boomgaarden, 2003; Schuck & de Vreese, 2006; Maier & Rittberger, 2008). Other studies take this into account and incorporate real-world data. One study linked news coverage of the EU and aggregate public opinion trends over time and found that while benefit framing has a positive
relationship with public support for the EU, conflict framing has a negative relationship (Vliegenthart, Schuck, Boomgaarden, & de Vreese, 2008). A mixed methods study compared the prominence of generic news frames in Romanian EU coverage to the frames employed by citizens to disciss the EU and found a strong effect on how citizens relate to the EU on
several levels; from which EU issues citizens are interested in and how important they find them to how they evaluate these issues, and which perspectives the use to explain them (Corbu, Ștefăniță, Oprea, & Udrea 2015). The above studies are all useful for explaining the connection between specific news frames and public opinion, but they do not inform about which perceptions of the EU citizens have in general. Based on this shortcoming I will contribute to the academic literature on this field.
EU in the news. A related area of framing studies is more interested in how EU affairs are covered in the media than the effects of frames. One line of studies in this area look at generic frames and find a prevalence of a responsibility frame (Semetko & Valkenburg, 2000), while others find that the conflict frame is most frequently used (de Vreese, 2003; Schuck, Vliegenthart, Boomgaarden, Elenbaas, Azrout, van Spanje, & de Vreese, 2013) In regard to my study, these generic frame studies form a poor basis for comparison, as they are concerned with how to package and present the news and not with which EU issues are in question. Another line of studies investigates the prevalence issue-specific EU frames in news coverage. In terms of focus these are closer related to the frames
in my study, but they typically either do not cover a sufficiently wide range of EU issues or they are too exclusively focused on a isolated event to make a meaningful comparison (see Lecheler & de Vreese, 2010; Schuck, Azrout, Boomgaarden, Elenbaas, van Spanje,
Vliegenthart, & de Vreese, 2011)
Evaluative direction of frames. There is no consensus in the literature on the
evaluative direction EU coverage. Results from previous studies vary from generally negative (Norris, 2000) and balanced evaluation (Schuck & de Vreese, 2006) to positive evaluations among some frames and negative evaluations among other frames (Schuck, Azrout,
Boomgaarden, Elenbaas, van Spanje, Vliegenthart, & de Vreese, 2011). My study will also investigate a balance between positive and negative frames, but the inconsistency of previous findings complicates forming a hypothesis.
In sum, the present study is distinct from previous studies in on framing of EU news in two ways: (1) it employs a novel approach to explore public opinion dynamics and (2) a set of frames that have not previously been investigated and based on the extant literature no meaningful theoretical expectations could be formulated.
Methods and Data Development of Codebook
The aim of my study is to find the prevalence of citizen perceptions of the EU in news coverage. In order to make this empirically feasible I made slight changes in the composition of the perceptions and adapted them into news frames. Regarding the positive aspects I split the aspect referring to ‘economic and diplomatic advantages’ in two, and then included the ‘economic advantage’ in the ‘free movement of services, goods, and capital’ aspect, as the economic advantage of being a member of the EU is often related to having access to the single market. For the negative aspects I combined ‘inadequate overregulation’ and ‘limited successes’ into one frame, as a pre-test revealed that an empirical distinction between the two
was not feasible. The new frame was labelled ‘Unfavourable consequences of EU policies and regulations’.
Ultimately, the codebook (see Appendix C) consisted of a total of ten audience frames. Six referred to inherently positive frames: a (1) free movement of people frame, a (2) free movement of goods, services, and capital frame, an (3) establishment of common
regulations, standards, and policies frame, an (4) EU funding frame, a (5) diplomatic clout frame, and a (6) European values frame. Additional four referred to inherently negative frames: an (7) unfavourable consequences of EU policies and regulations frame, a (8) bureaucratic EU frame, a (9) lack of solidarity between EU member states frame, and a (10) gap between EU and its citizens frame. Citizen statements from the focus groups explaining the reasoning behind the negative and positive perceptions of the EU form the basis for the indicators for each frame. Following Benoit’s (2011) recommendations on how to develop a content analysis the indicators were organised with the purpose of being as exhaustive and mutually exclusive as possible. This should allow for the construction of a cumulative scale for each frame. In a pre-test of 10% of the final sample I inductively added extra indicators, if I found them relevant for the overall frame and they were not represented by existing
indicators.
Newspaper Sample
I carried out a content analysis of EU coverage in two national Danish broadsheet newspapers, Politiken and Berlingske. The newspapers were selected based on the criteria that they were among the widest circulated broadsheets in Denmark (Mediawatch, 2014), they had different historical political affiliations, and are owned by different media
institutions. Although I recognise that only analysing broadsheet newspapers does not fully represent Danish EU coverage, this approach can give a reliable picture of the news
coverage, as broadsheets have been found to be agenda-setters for other news media (Kleinnijenhuis, 2003).
Periods of Study
The analysis focused on five routine periods and five key events between 2000 and 2015. Based on Boomgaarden, Vliegenhart, de Vreese, & Schuck (2010) I selected the
following key events: EU summits in Denmark during the Danish EU presidency in 2002, the installation of the Barosso Commission in 2004, the signing of the Nice treaty in 2007, European Parliament elections in 2009, and the referendum on the Danish opt-out on Home & Justice affairs in 2015. The events were selected to ensure an even distribution across all years and that the topical content of the events differs. News stories from four weeks
surrounding each event (or leading up to the event in cases of the 2009 EP elections and 2015 EU referendum) were included in the sample. In addition, news stories from five four-week routine periods in 2001, 2006, 2008, 2011, and 2013 was included (see appendix A)1. The routine periods were selected to ensure an even distribution over the entire time periods as well as across the different months of the year. The type of time period, routine period or key event, constitute the independent variable of this study.
Article Selection
Following Vliegenhart, Schuck, Boomgaarden, & de Vreese (2008) to increase the likelihood that the EU is prominent in the articles, I selected news stories that mention the EU or one of its institutions at least twice and with at least one of the references to the EU being in the headline or the lead of the article excluding opinion pieces, comments, and leading articles. Articles were collected using the online database Infomedia, which according to its own website, has Denmark’s largest media archive with all articles published in both Berlingske and Politiken since 1990 (Infomedia, 2016). Articles originally published in the print and online versions of the newspapers were both sampled.2
Coding Procedure
To obtain data on news frames during each time period 50 news stories were
randomly selected and coded. 50 articles have previously been found a sufficient number to give a reliable picture of the distribution of frames in a specific time period (Vliegenthart et al., 2008). This produced a total of 500 news stories, which were manually coded for the absence (0) or presence (1) of the frames. A subsample representing 10 % of the articles was coded by a Danish native speaker to assess inter-coder reliability. Across all indicators Krippendorff’s alpha ranged between .51 and 1 (see Appendix B). Krippendorff’s alpha could not be determined for two indicators, as they were absent from the entire sample. Dependent Variables
Presence of a specific frame was operationalized using between two and seven indicators for each frame. The number of indicators varies between the frames, as they represent the diversity of mutually exclusive examples citizens used to describe the frame in Timmerman (2014). A frame was defined to be present if any of the indicators were present, and the strength of the presence of a given frame increased with the number of present indicators. A maximum of three different frames were coded per article. As most of the indicators were mutually exclusive, the frame indicators were developed with the intent of forming a cumulative scale and were not expected to cluster into specific variables nor to correlate with each other. This expectation was corroborated by both correlation tests and tests for internal consistency.3 An overview of the share of indicators constituting each frame along with the count of articles in which the indicators were found present is presented in Table 1.
Visibility Measure
Depending on the centrality of visibility as a concept in a given study, measures for visibility of EU in news coverage range from complex (see Boomgaarden et al., 2010) to
more simple (see Boomgaarden & De Vreese, 2015). For the scope of this study I will measure the external visibility by calculating the number of articles that mention the EU (or one of its institutions) as a share of the total number of articles in the same news outlet and time period (Boomgaarden & de Vreese, 2015). I will not calculate the internal prominence within the articles, which will make the final visibility measure less precise. Still, as I base my sampling of time periods on a robust study of when the EU is highly visible in the news (see Boomgaarden et al., 2010) and only aim to test their findings, the visibility measure of this study is satisfactory.
Scales to measure news frames
I developed three different measurements for the purpose of the analysis. I computed an unweighted scale to measure the overall occurrence of the frames, a weighted scale to measure the intensity of the presence of the individual frames, and I recoded the weighted scales of each frame into to two evaluated frame groups.
Unweighted scale. I recoded the aggregated sum of the indicators per frame into a binary present-absent variable measuring whether one or more of the indicators of a given frame were present or none were present. This allowed me to determine the share of the total sample in which the individual news frames were present.
Weighted scale. To take into account the strength at which the frames were present in an article I added up the he binary codes (‘yes’ = 1 or ‘no’ = 0) for each article and computed them into a weighted scale.4 As a maximum of three frame indicators were coded present in any news story, the values of each weighted scale ranged between .00 to 3.00. The higher the score, the stronger the presence of a frame.
Evaluated frame groups. Finally, I recoded the weighted scales into two evaluated frame groups, positive frames and negative frames according to the categorizations made by the participants in Timmerman (2014). This means that the free movement of people frame,
the free movement of goods, services, and capital frame, the common EU policies frame, the EU funding frame, the diplomatic clout frame, and the European values frame were collapsed into the positive frames group, while the unfavourable consequences frame, the bureaucratic EU frame, the lack of solidarity frame, and the gap between EU and citizens were recoded into the negative frames group. A weighted scale measured the two frame groups with values between .00 and 4.00. In all cases higher scores suggest a stronger prominence of the frame groups.
Results
In the following, I will present the results of the analyses. As a first step, I will analyse the visibility of EU news across all time periods and use this as the basis of the analysis. Then, I will provide an overview of the absolute distribution of the news frames in all time periods, and finally, I will compare the distribution of the frames between routine periods and during key events and between low and high visibility periods.
Visibility of EU News
The visibility of the EU in all time periods ranged from 0.28% to 1.9% of the total number of written articles. A t-test confirmed a significant difference in the level of EU visibility during key events (M=1.43, SD=0.3) and routine periods (M=0.78, SD=0.3); t(498)= −24.16, p = <.001. Four of the five key events generated higher levels of EU visibility than the routine periods with the 2009 EP elections as the highest. Contrary to initial expectations EU coverage during the Danish referendum in 2015 only represented 1% of the total coverage – lower than during the routine period of 2001, which produced a score of 1.13%. Since one predefined routine period had higher visibility of EU news than one of the key events, I was able to develop a control variable that allowed me to test whether the type of period or the visibility of EU news was the strongest predictor of variances in the dependent variables. I computed a new variable based on the actual scores of visibility of EU
news, which I split into two groups with the five periods with the lowest scores in one group and the five periods with the highest scores in the other group.
Distribution of Frames
Individual frame prevalence. Figure 1 shows the absolute distribution of the news frames in all time periods. To answer RQ1, in general news stories about the EU in Denmark predominantly highlight three different frames: the diplomatic clout frame mentioned in 43% of the coverage, the unfavourable consequences frame mentioned in 35% of the coverage, and the common EU policies frame mentioned in 33% of the coverage. A lack of solidarity frame is mentioned in 19% of the coverage, whereas the remaining frames all each are mentioned in less than 6% of the news stories.
[Figure 1 about here]
An analysis of the prominence of the frames using a weighted scale confirms the trends illustrated in Figure 1. Overall, the diplomatic clout frame was the most prominent (M=.52, SD=.65), followed by the unfavourable consequences frame (M=.42, SD=.64) and the common EU policies frame (M=.38, SD=.59). These three frames are very significant in Danish EU coverage. The lack of solidarity frame is less frequently used (M=.21, SD=.44), but is still a significant part of the coverage. The rest of the frames are much less prominent, and the free movement of people frame (M=.01, SD=.11), the free movement of goods, services, and capital frame (M=.02, SD=.14), and the bureaucratic EU frame (M=.01, SD=.08) are almost absent from the entire coverage.5 As shown in Table 1, a further
breakdown of the frames reveals that some frame indicators were mentioned more frequently than others. With regard to the most dominant frames, the vast majority of the times the diplomatic clout frame was mentioned, the article focused on the EU as an international actor in non-EU neighbourhood countries (62%), whereas 28% of the mentions presented the EU as an actor in EU neighbourhood countries. The Unfavourable consequences frame was
mostly mentioned as EU regulations, standards, or policies that are inadequate for solving a problem (31%) or in relation to a loss of national autonomy (25%). Finally, it is interesting to note that the lack of solidarity frame was almost only mentioned as a conflict between
member states or member states pursuing their own interests (87%). [Table 1 about here]
Audience frames vs. media frames. The prominence of the individual frames in the news coverage differs from how frequently they were mentioned in Timmerman (2014), at least among the positively evaluated frames. The positive audience frames mentioned in most countries (free movement of people, European values, free movement of goods, services, and capital, and EU funding) were the least prominent among the positive news frames and were hardly present at all in the dataset. In fact, the positive aspect mentioned most often in the focus groups, the free movement of people, was the least prominent of all positive frames in the news coverage. Conversely, among the negative frames the results from the focus group studies correspond to the prominence of the individual negative frames. However, the measure used in Timmerman (2014) is not directly comparable to prominence of a corresponding frame. Still, it suggests that there could be a discrepancy between what citizens think of the EU and which positive frames are highlighted in news coverage.
Evaluated frame groups. We see that positive frames (M=1.03, SD=.76) are more prominent than negative frames (M=.7, SD=.87) in the overall coverage. It should be noted, however, that the large difference in the mean scores between positive is mainly due to the strong prominence of the diplomatic clout frame. If the diplomatic clout was removed from the positive group the remaining positive frames are indeed much less prominent in the coverage (M=.51, SD=.73).
In sum, between 2000-2015 the EU was predominantly framed in order of prominence in terms of its diplomatic clout, unfavourable consequences of its policies, the establishment
of common regulations and policies, and to a lesser extent lack of solidarity between its member states. Interestingly, the most prominent positive frames differed greatly from the positive aspects most often mentioned by European citizens. Finally, positive frames were more prominent than negative frames.
Frame Prevalence by Type of Period
A main aim of this study was to explore whether the use if frames in the news differs with the type of period during in which the coverage took place.
Aggregated frame prevalence. Independent-samples t-tests were conducted to compare the mean scores of the frames during routine periods and key events. In general more frames were coded present during key events (M=1.77, SD=1.00) than in routine periods (M=1.67, SD=.98), but the difference is not significant, t(498)= −1.085, p =.279.
Individual frame prevalence. If we look at the frames individually as shown in Table 2, most frames are more prominent during key events than in routine periods except the common EU policies frame, the EU funding frame, and the diplomatic clout frame. Most of the changes between the periods are insignificant partly due to the fact that some of the frames were rarely found present. Only the lack of solidarity frame shows a statistically significant difference being more prominent during key events (M=.26, SD=.50) than during routine periods (M=.16, SD=.37), t(462)= −2.539, p=.011.6 One should note that the
diplomatic clout frame disguises statistically significant differences in the prominence of mentions of the EU as a diplomatic actor in neighbourhood countries and as an actor operating outside its neighbourhood. I recoded the indicators of diplomatic clout frame into two new frames representing respectively the EU as an actor in its neighbourhood, and the EU as an actor outside its neighbourhood. A t-test of the two new frames showed that
representations of the EU as a diplomatic actor in its neighbourhood were significantly more prominent during key events (M=.23, SD=.45) than in routine periods (M=.11, SD=.38,
t(485)= −3.227, p=.001).7 Conversely, representations of the EU as an actor outside its neighbourhood were significantly less prominent during key events (M=.27, SD=.51) than in routine periods (M=.42, SD=.53, t(496)= 3.269, p=.001).8
[Table 2 about here]
Evaluated frame groups. Positive frames were more prominent than negative frames during both types of time periods. As shown in Table 3 positive frames showed almost no difference in prominence between key events (M=.99, SD=.76) and routine periods (M=1.06, SD=.76, t(498)=1.060, p=.29). However, the prominence of negative frames in news about the EU varies significantly with the type of time period. Negative frames were significantly more prominent during key events (M=.78, SD=.87) than in routine periods (M=.61, SD=.86, t(498)=-2.226, p=.026).
[Table 3 about here]
I controlled the effect of the type of period on the prevalence of negative frames by testing whether low and high visibility of EU news caused a similar variance. As shown in Table 4, a t-test corroborated the finding that the type of time period rather than the level of visibility of EU news has a significant effect on the prominence of negative frames. In fact, when controlling for actual EU visibility the significant effect disappeared, t(498)=.567, p=.57.9
[Table 4 about here]
In sum, the type of period in which the EU is covered does predict a variance in the framing. Across routine periods and key events only the lack of solidarity frame showed a statistically significant variance and was found more prominent during key events. Recoding the diplomatic clout frame into two new separate variables revealed that the EU was
represented as a diplomatic actor in its neighbourhood more frequently during key events. Finally, negative frames were significantly more prominent during key events than in routine
periods, and control tests confirmed that the type of period rather than the level of EU visibility in the news explains variance of negative frames.
Conclusion and Discussion
The present study advanced a novel approach in framing studies that used citizen perceptions as a reference point for media content analysis to obtain new insights about how public understanding of the EU relates to news coverage. An analysis of Danish broadsheet coverage between 2000-2015 found that the EU was predominantly framed in terms of three frames that do not reflect the diversity of citizen perceptions. A subsequent analysis showed that the balance of positive/negative frames is contingent with the type of period in which it is covered, and that even though the coverage is dominated by positive frames, negative frames increase significantly during key events. Finally, the findings revealed the necessity to make a conceptual distinction between EU visibility and the type of period.
Prominence of Frames
Individual frames. A first conclusion concerns the use of the different frames (RQ1). The results highlight that the general coverage of the EU is fairly uniform and is dominated by three frames. The diplomatic clout frame was most prominently used followed by the unfavourable consequences of EU policies frame and the common EU policies frame. This implies that most coverage of the EU revolves around its political performance as either a foreign policy actor negotiating with global powers and EU neighbourhood countries or as a legislative actor governing EU member states and companies. The focus on the EU’s policy output (common EU policies frame) and an assessment hereof (unfavourable consequences frame) is somewhat in line with previous research that has found that most EU stories are of economic and technocratic character (de Vreese, 2003). Conversely, the predominance of the diplomatic clout surprising, and the first part of the discussion will focus on this frame only.
Diplomatic clout. Little scholarly work has focused on media coverage of the EU’s external relations. The few existing studies do not include research about how the EU’s diplomatic role feature in the news compared to other EU issues or any information that could help explain the reason behind the frame’s prominence (see Kandyla & de Vreese, 2011; de Vreese & Klausch, 2011). In fact, literature on news values in EU news would have predicted the opposite. Normally EU stories take on a national angle and journalists tend to overuse national frameworks to interpret EU issues (de Vreese, 2003; Statham 2008). Thus, it is unexpected that the EU’s diplomatic role that is the most prominent in my study, as it as a policy issue seems to be even further removed from the sphere of the nation-state than other EU issues. It lies beyond the scope of this study to fully explicate the causes behind this, and I can only hypothesize that a part of the explanation can be drawn from a media economy perspective. Reliance on news agencies is common in foreign news reporting in general and in covering EU affairs in particular as it cuts down production costs (Wu, 2000;
Raeymaeckers, Cosijn, & Deprez, 2007). One could expect that stories about the EU as an actor outside its neighbourhood, which would be costly to produce independently for an outlet and perhaps of less relevance for the national audience, are more likely to rely on reports from news agencies. This factor could help explain why articles about the EU as an actor outside its neighbourhood are significantly more prominent in routine periods, where editors are less inclined to invest resources in EU coverage and thus prefer pre-fabricated or easily produced content (Raeymaeckers, Cosijn, & Deprez, 2007). Further research could investigate further both the news values that apply to EU news and how media economic considerations may influence the coverage.
Media frames vs. audience frames. Another interesting finding was the apparent discrepancy between which positive frames were most mentioned in the news coverage and in Timmerman (2014). By and large, the positive aspects that were most frequently
mentioned in the focus groups were least prominent in the news coverage and vice versa. One important caveat should be acknowledged here regarding the issue of using secondary data from focus groups as benchmark for the evaluations of the frames. Timmerman (2014) did not provide information on how important the participants regard the individual aspects of the EU. This complicates making a qualified comparison between the media frames and the audience frames. However, a Eurobarometer survey support the assumption that the priorities of the media coverage differ from what citizens prefer: While 57% of Europeans consider the free movement of people, goods, services, and capital the most positive result of the EU, only 19% answer the same for the EU’s diplomatic influence (European Commission, 2015). Future studies should aim to explain the reason behind the discrepancy. The finding emphasises the relevance of drawing on audience frames for further research on the link between news coverage and public opinion on the EU.
More positive than negative coverage. The study found that positive frames were more prominent than negative frames both overall, in routine periods, and during key events. As the frames are investigated for a first time in the present study no direct comparison can be made. Whether EU news is positively or negatively slanted relies on which frames are investigated (Schuck, Azrout, Boomgaarden, Elenbaas, van Spanje, Vliegenthart, & de Vreese, 2011). It is fair to claim that the mainly positive evaluation found in the present study was mostly caused by the present study’s inclusion of EU’s diplomatic role as a positive frame.
A Change of Frames
Routine periods vs. key events. The most interesting finding when comparing frames across types of time periods (RQ2) was that during key events the negatively
evaluated frames are significantly more prominent than in routine periods. De Vreese (2003) claims that an increase in the volume of EU news coverage may not be beneficial for public
support, as intensified exposure could invoke negative stereotypes that citizens have about the EU. The present study corroborates this by showing that frames, which citizens associate with negative aspects of the EU, are in fact significantly more frequently covered during key events, and adds that it is not merely an issue of increased coverage activating specific EU stereotypes but that the news stories themselves highlight more negative aspects. Previous research shows that EU key events are ‘abundantly’ framed in terms of conflict (de Vreese, 2003). This corresponds well with my finding that the lack of solidarity frame, which also accentuates conflict, is significantly more prominent during key events. Conflict-driven news has also been found to depress public support for EU policies, as it emphasises disagreement and the inability of the EU to operate (Vliegenthart et al., 2008).
EU visibility or type of period. The study demonstrated the need to make a conceptual distinction between EU visibility and type of period. Initially I assumed that all key events would have higher EU visibility than routine periods. However, the results showed that only the type of period and not EU visibility caused significant variance in the prominence of news frames. Even though the EU is often more visible in news during key events the concepts should not be used interchangeably. The distinction is important, because it implies that just because the EU is more visible in the news does not necessarily the content of the coverage engages with alternative aspects of the EU. Rather, the reason behind the stronger effect of the type of period could be that politicisation of the EU increases during key periods (Statham & Trenz, 2013). Hutter & Grande (2014) argue that not only do a wider variety of political actors and opinion leaders take part in the debate but they also more likely to frame EU questions differently. They found that with higher levels of EU politicisation political parties are more likely to frame the EU in cultural terms rather than economic terms (Hutter & Grande, 2014). Since elite contestation is an important news value for journalists, the media coverage is likely to reflect the change in framing (Schuck, Xezonakis, Elenbaas,
Banducci, & De Vreese, 2011). In regard to the present study, I also find that a majority of the frames that increase in prominence during key events (free movement of people; free movement of goods, services, and capital; European values; bureaucratic EU; gap between EU and its citizens) are closer related to the constitutive elements of the EU and less
concerned with its diplomatic performance and legislative output that otherwise dominate the news coverage. It should be noted, however, that these frames remain among the least
prominent even during key events.
Visibility, framing, and public opinion. With respect to the relationship between visibility and public opinion, the present study was dedicated to see whether the EU was framed differently during key events, where citizens were more likely to be guided by the news frames (see Scheufele & Tewksbury, 2007). The results did indeed reveal that key events are framed differently with a significant increase in negative frames. However, it was also found that this does not imply that key events are overall negatively slanted. In which way the change of frames relate to public opinion is beyond this study to say with any certainty. Framing effects rely on the weight and importance that citizens ascribe to certain considerations regarding a political issue (Nelson & Oxley, 1999). The lack of information about how the focus group participants in Timmerman (2014) assess the importance of the different positive and negative aspects of the EU in their opinion formation makes it hard to develop further on the issue here. Future studies could advance knowledge in this field by using a multi-methodological design that uses detailed input about how citizens perceive the EU and how they rank their perceptions and integrate it into a content analysis.
Limitations
The study represents a novel way for linking media representations with citizen perceptions and may ideally serve as a basis for further exploration of this type of research. Several limitations apply to this study, and consequently the findings should be interpreted with
caution, as they represent only initial steps for integrating citizen perceptions into news frame studies.
Categorizing qualitative data. The novelty of this study, to use focus group data as a baseline for exploring media framing of the EU, also represents a main limitation in terms of reducing the diversity of citizen perceptions. Sorting statements into broader categories may influence the variety of the actual perceptions, and thereby put an arbitrary spin on the issues at hand. To develop the audience frames for this study, I modified already categorised
statements, which increases the risk of letting researcher subjectivity influence the categories. Retrospectively, one potential caveat of the categorizations used in this study was to include the EU’s diplomatic relations with its neighbourhood countries and the rest of the world into one frame. It is not clear from Timmerman (2014) that EU citizens necessarily incorporate the EU’s neighbourhood policy into its diplomatic role, even though the policy areas were merged with the Lisbon Treaty. Considering the dominance of the EU’s diplomatic role overall and how the subcategories show different variances across time periods, a distinction between the EU’s relations with its neighbourhood and with the rest of the world could have been helpful. Future research should rely on a refined measure of preserving perception diversity, and preferably integrates gathering of information on citizen perceptions into the study design.
Sample size. The total sample of the study was 500 articles. However, as some of the variables were hardly present in the sample, very low mean scores made it difficult to make meaningful comparisons between the presences of the frames across different time periods. A larger sample would have made it possible to conduct a more detailed analysis of the
difference between the individual frames.
Country sample. The singular focus on one country in the study affects the
greatly between countries, and it is important to underline that the findings of this study apply to Denmark only (Schuck, Azrout, Boomgaarden, Elenbaas, van Spanje, Vliegenthart, & de Vreese, 2011). Moreover, how citizens perceive the EU also depend on country-specific factors (Diez-Medrano, 2010). Thus, extending future research to include more countries would deepen the understanding of the link between audience frames and media frames and could also yield different results than what has been found in this study.
Medium sample. It is highly likely that only using broadsheets in the study have affected the results. Even though broadsheets are often agenda-setters for other media types, it is likely that the EU coverage in television and in tabloids is distinct (de Vreese, 2003). Television coverage in particular could be relevant for investigating audience frames, as TV remains the most important source for citizens to acquire information about the EU
(European Commission, 2015). Practical Implications
These limitations notwithstanding, some practical lessons can be drawn from the study. First of all, the evidence of this study suggests that citizens to some extent are interested in different aspects of the EU than those made available to them in the media. Often EU journalists complain about the public’s lack of interest in and knowledge about European affairs (see Raeymaeckers, Cosijn, & Deprez, 2007). A remedy to raise public interest could be that journalists put more emphasis on the constitutive aspects of the EU and tone down the predominant focus on the EU as legislative and diplomatic actor. In order to strengthen European democracy public interest is essential, and this requires a covering a wide range of topics and includes continuous debate of not only the policy output but also the basic principles and ideas behind the European collaboration. Secondly, the study
demonstrated that during key events, when citizens are more likely to form opinions about the EU, negative frames became more prominent. For pro-European policy-makers this is a
call to action to try and affect the EU debate rather than try to silence it, as has been a common reaction in many countries (see de Wilde & Zürn, 2012). Politicization of key EU events is a fact, and in recent years Eurosceptic parties have been able to define the debate on their terms (Hutter & Grande, 2014). My study shows that citizens appreciate aspects of the EU, which are not covered in the media. By drawing more heavily on the positive frames that are not covered in the media pro-European actors could perhaps counter the framing of key EU events.
Today, the EU finds itself in the midst of several unfolding crises, and public opinion across Europe seems to favour political disintegration rather than deeper
collaboration. With none of the problems likely to disappear in the near future, it is important to both understand which perceptions European citizens have of the EU, but also how these perceptions are made available through news coverage, as the way EU affairs are framed have implications for the public understanding of the EU. My study has demonstrated that Danish EU coverage presents a fairly uniform picture of the EU as a diplomatic and
legislative actor, which does not recognize the diversity of perceptions among citizens, and it points to the relevance of going beyond news coverage, when investigating how citizens understand the EU. Moreover, the results show that the EU is framed differently during key events with an stronger emphasis on frames that citizens evaluate negatively, and it adds to the extant literature on how public opinion dynamics relate to media coverage.
References
Baccini, L., Sudulich, M. L., & Wall, M. T. (2013). Lost in Transmission: Evaluating Internet Effects on Citizens’ Attitudes Towards the European Union in Times of Crisis
(Working Paper MWP 2013/03). Retrieved from European University Institute website: http://cadmus.eui.eu/handle/1814/26616
Benoit, W. L. (2011). Content analysis in political communication. In E. P. Bucy & R. L. Holbert (Eds.), The sourcebook for political communication research: methods, measures, and analytical techniques, (pp. 268-279). New York: Routledge
Boomgaarden, H. G., Vliegenthart, R., de Vreese, C. H., & Schuck, A. R. (2010). News on the move: Exogenous events and news coverage of the European Union. Journal of European Public Policy, 17(4), 506-526.
Boomgaarden, H. & de Vreese, C. H. (2015). Do European Elections create a European Public Sphere? In W. Van der Brug and C. H. de Vreese. (Un)intended consequences of European Elections. Oxford: Oxford University Press.
Borah, P. (2011). Conceptual issues in framing theory: A systematic examination of a decade's literature. Journal of communication, 61(2), 246-263.
Chaban, N. & Bain, J. (2014). Framing the EU in a Time of Crisis: Media Reflections from EU Strategic Partners in Asia-Pacific. In N. Chaban & M. Holland
(Eds.), Communicating Europe in times of crisis: External perceptions of the European Union (pp. 118-142). Palgrave Macmillan.
Corbu, N., Ștefăniță, O., Oprea, D., & Udrea, G. (2015). The European Union–From Agenda Setting and Media Frames to Citizens’ Conversations. Romanian Journal of
Communication and Public Relations, 17(1), 41-57
De Vreese, C. H. (2001). Europe in the News A Cross-National Comparative Study of the News Coverage of Key EU Events. European Union Politics, 2(3), 283-307.
De Vreese, C.H. (2003). Framing Europe. Television News and European Integration. Amsterdam: Aksant Academic Publishers.
De Vreese, C. H. (2005). News framing: Theory and typology. Information design journal+ document design, 13(1), 51-62.
De Vreese, C. H., Banducci, S., Semetko, H., & Boomgaarden, H. (2006). The News Coverage of the 2004 European Parliamentary Election Campaign in 25 countries. European Union Politics, 7(4), 477-504.
De Vreese, C. H., & Boomgaarden, H. (2003). Valenced news frames and public support for the EU. Communications, 28(4), 361-381.
De Vreese, C. H., & Klausch, T. (2011). The salience of frames and their effects on support for CFSP and supranational policy allocation. In Oppermann, K., & Viehrig, H. (Eds.), Issue salience in international politics, 118-136.
De Vreese, C. H., & Lecheler, S. (2012). News Framing Research: An Overview and New Developments. In H. A. Semetko & M. Scammell (Eds.), The SAGE handbook of political communication (pp. 292-306). London: SAGE
de Vreese, C. H., van der Brug, W., & Hobolt, S. (2012). Turkey in the EU?: How cultural and economic frames affect support for Turkish accession. Comparative European Politics, 10(2), 218-235.
De Wilde, P., & Zürn, M. (2012). Can the Politicization of European Integration be Reversed? JCMS: Journal of Common Market Studies, 50(s1), 137-153.
Elenbaas, M., De Vreese, C. H., Boomgaarden, H. G., & Schuck, A. R. (2012). The impact of information acquisition on EU performance judgements. European Journal of
Political Research, 51(6), 728-755.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of communication, 43(4), 51-58.
European Commission. (2014). Eurobarometer 82. Brussels: EC DG Press and Communication.
European Commission. (2015). Eurobarometer 83. Brussels: EC DG Press and Communication.
Gattermann, K. (2013). News about the European Parliament: Patterns and external drivers of broadsheet coverage. European Union Politics, 14(3), 436-457.
Green-Pedersen, C. (2012). A Giant Fast Asleep? Party Incentives and the Politicisation of European Integration. Political Studies, 60, pp. 115-130.
Hobolt, S. B. (2005). When Europe matters: The impact of political information on voting behaviour in EU referendums. Journal of elections, public opinion & parties, 15(1), 85-109.
Hobolt, S. B. (2006). How parties affect vote choice in European integration referendums. Party Politics, 12(5), 623-647.
Hobolt, S. B., & Brouard, S. (2011). Contesting the European Union? Why the Dutch and the French rejected the European constitution. Political Research Quarterly.
Hertog & McLeod (2001). A Multiperspectival Approach to Framing Analysis: A Field Guide. In Framing Public Life: Perspectives on Media and Our Understanding of the Social World (pp. 141-162). New Jersey: Lawrence Erlbaum Associates.
Hutter, S., & Grande, E. (2014). Politicizing Europe in the national electoral arena: A
comparative analysis of five West European countries, 1970–2010. JCMS: Journal of Common Market Studies, 52(5), 1002-1018.
Infomedia. (2016). Retrieved April 27 2016 from http://infomedia.dk/mediearkivet/ Iyengar, S. (1990). The accessibility bias in politics: Television news and public
Kandyla, A. A., & de Vreese, C. (2011). News media representations of a common EU foreign and security policy. A cross-national content analysis of CFSP coverage in national quality newspapers. Comparative European Politics, 9(1), 52-75.
Kinder, D. R. (2007). Curmudgeonly advice. Journal of Communication, 57, 155-162. Kiousis, S. (2004). Explicating media salience: A factor analysis of New York Times issue
coverage during the 2000 US presidential election. Journal of Communication, 54(1), 71-87.
Kleinnijenhuis, J. (2003) ‘Het Publiek Volgt Media die de Politiek Volgen’, in Raad Voor Maatschappelijke Ontwikkeling (ed.), Medialogica, The Hague: SDU, pp. 151–212. Lecheler, S., & De Vreese, C. H. (2010). Framing Serbia: the effects of news framing on
public support for EU enlargement. European Political Science Review, 2(01), 73-93. Lee, S. T., Maslog, C. C., & Kim, H. S. (2006). Asian Conflicts and the Iraq War A
Comparative Framing Analysis. International Communication Gazette, 68(5-6), 499-518.
Maier, J. & Rittberger, B. (2008). Shifting Europe's Boundaries: Mass Media, Public Opinion and the Enlargement of the EU. European Union Politics, 9 (2), 243-267.
McCombs, M. E., Shaw, D. L., & Weaver, D. H. (1997). Communication and democracy: Exploring the intellectual frontiers in agenda-setting theory. Psychology Press. Mediawatch. (2014). Oplag: Hverdagsaviser H2 2014. Retrieved February 3 2016 from
http://mediawatch.dk/Statistik/Oplagstal_aviser/article7496701.wece
Medrano, J. D. (2003). Framing Europe: Attitudes to European Integration in Germany, Spain, and the United Kingdom. Princeton University Press.
Medrano, J. D. (2010). Unpacking European Identity. Politique européenne, (1), 45-66. Nelson, T. E., & Oxley, Z. M. (1999). Issue framing effects on belief importance and
Nisbet, M. C. (2009). Communicating climate change: Why frames matter for public engagement. Environment: Science and Policy for Sustainable Development, 51(2), 12-23.
Norris, P. (2000). A virtuous circle: Political communications in postindustrial societies. Cambridge University Press.
Oppermann, K., & Viehrig, H. (2009). The public salience of foreign and security policy in Britain, Germany and France. West European Politics, 32(5), 925-942.
Peter, J. & de Vreese, C. H. (2002). A faceless European Union: A cross-national comparative investigation of the television news coverage of EU representatives. Paper presented at the annual meetings of the International Communication Association, ICA, Seoul, Korea.
Raeymaeckers, K., Cosijn, L., & Deprez, A. (2007). Reporting the European Union: an analysis of the Brussels press corps and the mechanisms influencing the news flow. Journalism Practice, 1(1), 102-119.
Scheufele, D. A. (2000). Agenda-setting, priming, and framing revisited: Another look at cognitive effects of political communication. Mass Communication & Society, 3(2-3), 297-316.
Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of communication, 57(1), 9-20. Schuck, A.R., Azrout, R., Boomgaarden, H., Elenbaas, M., van Spanje, J., Vliegenthart, R.,
& de Vreese, C. (2011). Media Visibility and Framing of the European Parliamentary Elections 2009: A Media Content Analysis in 27 Countries. In Maier, M., Strömbäck, J., & Kaid, L. L. (Eds.), Political Communication in European Parliamentary
Schuck, A. R., Xezonakis, G., Elenbaas, M., Banducci, S. A., & De Vreese, C. H. (2010). EES (2009) Media Study Data Advance Release Documentation, 31/03/2010. (www.piredeu.eu)
Schuck, A. R., Xezonakis, G., Elenbaas, M., Banducci, S. A., & De Vreese, C. H. (2011). Party contestation and Europe on the news agenda: The 2009 European Parliamentary Elections. Electoral Studies, 30(1), 41-52.
Schuck, A. R., Vliegenthart, R., Boomgaarden, H. G., Elenbaas, M., Azrout, R., van Spanje, J., & De Vreese, C. H. (2013). Explaining campaign news coverage: How medium, time, and context explain variation in the media framing of the 2009 European parliamentary elections. Journal of Political Marketing, 12(1), 8-28.
Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A content analysis of press and television news. Journal of communication, 50(2), 93-109.
Statham, P. (2008). Making Europe news: How journalists view their role and media performance, Journalism, 9(4), 398-422.
Statham, P. & Trenz, H.-J. (2013). How European Union Politicization can Emerge through Contestation: The Constitution Case, JCMS: Journal of Common Market Studies, 51(5), 965-980.
Timmerman, V. (2014). How do citizens see the European Union. Synthesis of Horizon EU: European Citizenship, a Horizontal Development, Notre Europe – Jacques Delors Institute, November 2014; Optem network. Retrieved from:
http://www.institutdelors.eu/media/howcitizensseeeu-timmerman-ne-jdi-nov14.pdf?pdf=ok
Vliegenthart, R., Schuck, A. R., Boomgaarden, H. G., & De Vreese, C. H. (2008). News coverage and support for European integration, 1990–2006. International Journal of Public Opinion Research, 20(4), 415-439.
Weaver, D. H. (2007). Thoughts on agenda setting, framing, and priming. Journal of communication, 57(1), 142-147.
Wettstein, M. (2012). Frame Adoption in Referendum Campaigns The Effect of News Coverage on the Public Salience of Issue Interpretations. American Behavioral Scientist, 56(3), 318-333.
Wu, H. D. (2000). Systemic determinants of international news coverage: A comparison of 38 countries. Journal of Communication, 50(2), 110-130.
Notes 1
The routine period of 2011 was extended to five weeks in order to sample a sufficient number of articles meeting the criteria of 50 articles per time period.
2
Despite several requests Infomedia refused to share how they collect the articles for their media archive for the purpose of this study.
3
A principal components analysis with Varimax rotation on 34 frame items was conducted, and as expected the Kaiser-Meyer-Olkin measure for sampling adequacy suggested that the correlations do not cluster among specific variables (KMO=.482). All frames were tested for their scale reliability confirmed low internal consistency between the different items of each frame and gave Cronbach’s alphas between -.043 and .37. A scale analysis of all indicators across frames revealed that all possible sets of pairs among the indicators have low internal consistency, and none have a Cronbach’s alpha higher than .3
4
Earlier framing studies divide the aggregated value for each article by the number of frame items (Semetko & Valkenburg, 2000; Schuck & de Vreese, 2006). As the frame items for my study varied between two and seven, this approach would not provide a comparable measure for the prominence of the frames.
5
The presence of the different frames correlated only weakly with associations ranging between r=.13 and -.36.
6
Levene’s test indicated unequal variances (F=26.6, p=<.011), and the degrees of freedom were adjusted from 498 to 462.
7
Levene’s test indicated unequal variances (F=35.8, p=<.001), and the degrees of freedom were adjusted from 498 to 485.
8
Levene’s test indicated unequal variances (F=19.4, p=<.001), and the degrees of freedom were adjusted from 498 to 496.
9
Also, no correlation between EU visibility and the presence of negative frames was found, r=.06, p=.198. In general the prominence of the individual frames did not correlate with EU visibility (all relationships scored r=<.1) and the associations were non-significant. Only the European values frame differed, r=.11, p=.01.
Tables
Table 1. Frame indicators per frame in absolute count of indicator presence in all EU news stories (N=500) and percentage of indicator presence of all indicators in the frame.
Frame name and indicators Count Share (%)
Free movement of people Students
6
1 17%
Labour 5 83%
Traveling / Residing 0 0%
Free movement of goods, services, and capital Effect consumers Effect business Economic advantage 10 7 3 0 70% 30% 0% Common EU policies Regulating companies Monitoring governments Common solutions 192 66 41 85 34% 21% 44% EU funding
Funding businesses & sectors
23
9 39%
Funding member states 11 48%
Funding redistribution 3 13% Diplomatic clout International actor 258 161 62% Funding non-EU 13 5% EU membership motivation 11 4% Actor in EU neighbourhood 73 28% European values Nature of EU 25 2 8%
Guardian cultural diversity 5 20%
Guardian fundamental rights 4 16%
Democratic state of EU 14 56% Unfavourable consequences Over-regulation 210 26 12% Inadequate 65 31%
Loss national autonomy 52 25%
EU immigration 12 6%
EU funding 16 8%
Non-EU immigration 21 10%
Cause of new problems 18 9%
Bureaucratic EU
Bureaucratic administration
3
