Minorities and the EU
How cultural frames affect Austrian-Turkish EU attitudes
Sherin Quell 10602410
Master’s thesis
Graduate School of Communication Political Communication
Supervisor: Sophie Lecheler January 30, 2015
Abstract
A growing amount of research stressed the impact news frames have on people’s attitudes towards the EU. Although every EU citizen is affected by the EU’s future and decisions taken at the European level, studies about framing effects on EU attitudes neglect immigrants’ views on the EU and how media may affect them. In an online survey
experiment (N=91), this study investigates how cultural news frames affect EU attitudes of second generation Turkish immigrants in Austria. Participants were randomly assigned to read an article using arguments supporting Turkey’s accession to the EU (inclusive cultural frame) or an article, which argued against Turkish membership (exclusive cultural frame). Results showed that exposure to the inclusive cultural frame increased support of Turkey’s accession to the EU. Also, participants who were exposed to the inclusive cultural frame and were favourable towards religious diversity felt that they were part of the EU, which in turn made them identify stronger with the EU than participants who were not favourable towards religious diversity.
Minorities and the EU
How cultural frames affect Austrian-Turkish EU attitudes
Public opinion and support of all EU citizens is crucial to the EU’s future and success. Given the intangibility and abstractness of the EU, citizens need the news media in order to inform themselves and shape their opinion (Deflem & Pampel, 1996, de Vreese &
Boomgaarden, 2005; Maier & Ritterberger, 2008). In this regard, several scholars have stressed the crucial impact of media frames when investigating EU attitudes. Researchers have drawn attention to the effect of positively valenced news articles and different ways of framing the debate on Turkey’s possible accession on public opinion (de Vreese et al., 2011; de Vreese, van der Brug & Hobolt, 2011). However, studies investigating framing effects on EU attitudes focused on EU citizens but did not differentiate between immigrants and natives. Especially in Austria, where public support towards the EU in general and Turkey’s
accession in particular is low (Matzka, 2009), immigrants may have more favourable attitudes towards the EU. First, Austrian media tend to picture second generation Turkish immigrants (Austrian-Turks) as a group of outsiders that is marginalized from society (Zauner, 2008). Second, this image of Austrian-Turks is reflected within society given that Austrians compared to other EU citizens hold slightly more negative attitudes towards minorities (Thalhammer, Zucha, Enzenhofer, Salfinger & Orgis, 2001). Therefore, Austrian-Turks may turn towards the EU as a way of belonging in order to overcome the identity-dilemma and find a place where they have the feeling to be part of the group of insiders.
Communication scholars have found that different frames have different effects on different people (de Vreese et al. 2001; de Vreese et al. 2011). For instance, the economic frame emphasizing cost/benefit reasoning affects people on a more rational level and rather persuades them, whereas the cultural frame, which stresses the insider/outsider debate as well as religious arguments affects people on a more emotional level and are conditioned by
individual predispositions of religious diversity (de Vreese, et al., 2011). Taken individual predispositions and the potential search of a place of belonging into account, cultural frames regarding the debate around Turkey’s accession may affect Austrian-Turks’ attitudes towards the EU. Yet, research has widely ignored how immigrants perceive and feel about the EU.
This study aims to take this gap as a starting point and investigates how cultural frames in news media affect Austrian-Turks’ EU attitudes. An online survey experiment, conducted among second generation Turkish immigrants will shed light on the interplay between cultural framing effects and Austrian-Turks’ EU attitudes. I begin by discussing the relationship between Austria and the EU and how Austrian-Turks are perceived within society. Then, I embed my study within framing effects theory by discussing the relevant elements of the theory such as framing process, framing setting and the high impact frames have on EU attitudes. Furthermore, I emphasize the importance of mediators and moderators when discussing framing effects based on the assumption that framing effects differ across individuals (Entman et al. 2009; de Vreese et al. 2011). Specifically, this study addresses that these framing effects are not equal across individuals, but moderated by positive
predispositions towards religious diversity. Additionally, the article emphasizes the
multidimensional concept of EU attitudes outlined by Boomgaarden, Schuck, Elenbaas and de Vreese (2011). The current study thus adds to existing literature by investigating positively and negatively valenced cultural frames and by demonstrating how these may arouse the feeling of being part of the EU among Austrian-Turks, which in turn should affect their EU identity. Additionally this study evinces that these framing effects are not equal across individuals, but moderated by positive predispositions towards religious diversity.
Austrian-Turks and the EU
Austrians turned from being euro-enthusiastic to eurosceptic in the last 20 years since they entered the EU (Günay, 2008). This gradual change has several reasons. First, Austrians were proud to join the EU because they pictured it as a prestigious and “exclusive club” until their Eastern-European neighbours, which some Austrians perceive as poor and backward, entered in 2004 (Rauscher, 2008). Many Austrians had difficulties understanding why they joined the same club. Second, the government promised that EU membership and
enlargement will foster security and lead to economic advantages (Luif & Trauner, 2012), however the far-right party, FPÖ, stressed the disadvantages such as increasing crime rates and the loss of social cohesion. Third, tabloids, such as Kronen Zeitung (most popular
Austrian daily newspaper) picked up on these negative consequences of EU membership and further enlargements (Günay, 2008). Media depiction of the EU resonated within society as shown by Eurobarometer polls. In 2008, only 36% positively evaluated the EU (European Commission, 2008).
Regarding EU enlargement in 2004 with the first wave of Eastern countries entering and Turkey’s possible EU entry, a heated debate started in Austria. Among citizens, support for Turkey’s entry was quite low with only 16% of Austrians supporting it (Matzka, 2009). Researchers have demonstrated that a society’s unwillingness to support EU enlargement is tightly linked to a fear of other cultures (Mc Laren, 2002) and dread of immigrants (de Vreese & Boomgaarden, 2005).
However, Austrians were not against every EU enlargement as shown by strong public and political support for Croatia’s accession in 2004 (Günay, 2008). These different reactions among Austrians regarding Croatia’s and Turkey’s membership could be linked to the religious and cultural similarities between Austria and Croatia and the religious and
cultural differences between Austria and Turkey. In this regard, politicians, in particular those of the FPÖ, stressed in their anti-foreign and anti-Islamic slogans that the differences are of religious and cultural nature (Günay, 2008).
However, Turkish immigrants are part of Austria since the late 1950’s when a large wave of Turkish guest-workers came to Austria (Fassman, Münz & Seifert, 1997). In 2014, around 106,300 second generation Turkish immigrants lived in Austria (Glossy & Hametner, 2014). Research has been undertaken concerning the educational gap between children of these guest workers and natives (Fassman et al. 1997; Damelang & Haas, 2006) as well as their integration within their host countries (Ersan & Kirisci, 2004; Schalk-Soekar, van de Vijver & Hoogsteder, 2004). Despite the societal relevance, Austrian-Turks being one of the largest minorities in Austria, there is a lack of studies focussing on immigrants’ EU attitudes and what role the media plays. Given the negative sentiment of many Austrians towards the EU and immigrants, Austrian-Turks may feel ostracized from Austrian society and may have more positive attitudes towards the EU, depending on how the EU is depicted in the media.
Framing Effects and EU Attitudes Framing Effects Theory
Framing has been widely discussed in research and different elements have been distinguished (Scheufele, 1999; Entman, 1993; Scheufele & Tewksbury, 2007; de Vreese et al., 2001). In general, a frame is defined as “an emphasis in salience of certain aspects of a topic” (de Vreese, 2002, p.27). Thus, a frame has a selective function and emphasizes on certain aspects of a topic while pushing other aspects into the background (Lecheler, de Vreese & Slothuus, 2009).
The process of framing includes frame building as well as frame setting (Lecheler et al. 2009; de Vreese, 2002). According to de Vreese et al. (2012) frame building occurs in the
political realm as well as in the media realm. Politicians tend to frame issues in conformity with their ideological programme and in response to political opportunities (de Vreese et al., 2012). Besides, journalists are using the frames that are part of the repertoire of frames that we form in our culture to represent reality (van Gorp, 2005).
When investigating frames, a distinction between two kinds of frames is made: issue-specific news frames and generic news frames (de Vreese et al., 2001). The main difference lies in the range of issues which is included in the frame. Generic news frames are applied across a wide range of topics whereas issue-specific news frames are linked to a specific issue or news event (de Vreese et al., 2012). De Vreese et al. (2012), for instance, used issue-specific frames in their survey-embedded experimental study because they investigated economic and cultural news frames dealing with Turkish accession to the EU, which is a specific topic. Lecheler et al. (2009), on the other hand, used a generic news frame in their experimental study being the economic consequences frame because they conducted two experiments and argued that the use of generic news frames ensured that the results of the experimental manipulation did not arise from different frame constructions but from change in the issue.
In this study, I make use of issue-specific news frames because the aim is to present the issue at stake being Turkey’s accession to the EU through different arguments, which depict all the facets of the issue. An issue-specific approach allows for investigating a specific issue in great detail, which is crucial in the context of this study, whereas generic frames are broadly applicable to different topics (de Vreese, Peter & Semetko, 2001).
Next to this distinction, framing effect theory suggests that if the media presents an issue in a certain way, it is likely that this adopted way of framing will be reflected in public opinion (Entman, 1993; Scheufele & Tewksbury, 2007). In this regard, different types of framing effects are proposed. For instance, Druckman (2004) distinguished between
equivalency framing effect, which investigates how the use of different but rationally similar words or phrases affects people’s preferences, and emphasis framing effect, which presents an issue under a specific angle and makes people focusing on this angle when forming their opinion. Generally, an equivalency framing effect occurs when frames present the same information, but in a positive or negative way (e.g, 5% unemployment or 95% employment). The emphasis framing effect, on the contrary, occurs when frames focus on different
information in order to present an issue, but not in a logically identical way of making the same statement (Druckman, 2004). Sniderman and Thierault (2004), for instance, focused on emphasis framing effects and showed that when the government’s money spending for the poor is framed as increasing the chances that poor people can become successful, individuals tend to encourage increased spending. However, when it is framed as following from higher taxes, individuals tend to oppose increased spending.
Given that I expect that the exposure to the cultural frames, will differently affect Austrian-Turk’s EU attitudes depending on how Turkey’s accession to the EU is framed, this study is based on the emphasis framing effect. Chong (1993) suggested that these types of frames are regarded as “the essence of public opinion formation” (p.870). Besides, past research confirmed that news frames in general and cultural frames in particular do affect public opinion towards the EU (de Vreese et al., 2012; Koenig, Mihelj, Downey & Beck, 2006).
Cultural Framing Effects
When framing the EU in general and Turkey’s accession in particular, different kinds of issue-specific frames have been investigated (de Vreese et al. 2012; Schuck & de Vreese, 2006). For instance, de Vreese et al. (2012) stressed that cultural and economic news frames affect people’s attitudes towards Turkey’s accession to the EU. In their case, the content of
cultural frames emphasized religious arguments such as the Muslim population in Turkey and the positive or negative consequences for the EU in case of Turkey’s membership. They concluded that the positively valenced cultural frame increased support for Turkish membership, but only among people who favoured religious diversity.
Furthermore, when investigating the 2004 EU enlargements, Schuck et al. (2006) distinguished between the risk frame, which emphasized on negative economic consequences such as high costs in case of further EU enlargements, and the opportunity frame, which stressed positive consequences that went beyond economic issues such as the inclusion of new countries, the spread of democracy, freedom, and human rights. They found that participants who were exposed to the opportunity frame condition showed more support towards EU enlargement compared to those who were exposed to the risk frame condition (Schuck et al., 2006). These findings showed that frames do affect public opinion about the EU. However, de Vreese et al. (2012) and Schuck et al. (2006) exclusively compared frames with religious arguments (cultural frames) or economic arguments (economic and risk frames) even though studies have shown that news articles emphasize more than one argument when depicting the EU in general and Turkey’s accession in particular. With regards to this, Koenig, Mihelj, Downey and Bek (2006) distinguished several frames, which included different arguments to depict the EU, as they conducted a cross-country content analysis. The main frames that structured the media discourse around Turkey’s accession to the EU were the clash of civilization frame, which stressed the religious differences between Turkey and the EU, and the multiculturalist frame, which in contrast presented the EU as a Union that aims at unifying different civilizations (Koenig et al., 2006). In line with these findings, Torreblanca (2007) highlighted the presence of a range of arguments used when covering the debate about Turkish accession being of economic, legal, religious and cultural nature.
Based on the frames mentioned above and the importance to include several arguments when depicting the EU in general and Turkish membership in particular, the cultural frames used in this study include legal, economic and religious arguments when discussing Turkey’s accession to the EU. Besides, the study distinguished between a positively valenced cultural frame, namely the inclusive cultural frame and a negatively valenced cultural frame, namely the exclusive cultural frame.
These frames were partly adapted from de Vreese, Boomgaarden and Semetko’s (2011) issue-specific cultural threat frame and de Vreese et al.’s (2006) cultural frame. De Vreese et al’s (2011) cultural threat frame was distinguished as the most prominent news frame used when covering the debate about Turkey’s accession in Dutch media. Both studies showed that the positive or negative valence of the frames had an effect on participants’ opinion towards Turkish membership. Valenced frames assess political issues in either positive or negative terms (Schuck et al., 2006).
Furthermore, studies in the field of psychology or marketing have found considerable effects of valenced frames on perceptions, judgements and evaluations. However, they have not been extensively investigated in the field of political communication (Schuck et al., 2006).
Therefore, I emphasize valenced cultural frames and distinguish between the inclusive cultural frame and the exclusive cultural frame. The inclusive cultural frame depicts the EU as a multicultural Union that cherishes cultural and religious diversity and emphasizes the positive aspects of a possible Turkish EU membership. On the contrary, the exclusive cultural frame stresses the cultural and religious differences between Turkey and the member states and depicts the EU as a Union which includes countries with similar historical, religious and cultural backgrounds by emphasizing the negative aspects of Turkey’s possible accession to the EU. Based on the aforementioned findings regarding the effects of cultural frames on EU
attitudes, this study extends previous research not only by investigating cultural framing effects on EU attitudes, but also by focussing on immigrants, in particular second generation Turkish immigrants in Austria and how cultural frames affect their EU attitudes.
EU attitudes: a multi-dimensional concept
With regards to EU attitudes, I drew on Boomgaarden, Schuck, Elenbaas and de Vreese’s (2011) assumption that EU attitudes is a multi-dimensional concept. Several scholars came to this conclusion but have established different dimensions. For instance, Easton (1975) distinguished between two modes of support being specific and diffuse on one dimension and three objects of political support namely the community, the regime and the authorities on the second dimension.
This conceptualization was adapted by Lindberg and Scheingold (1970) to examine support for the EU and rather than specific and diffuse support, they named it utilitarian and affective. Utilitarian is related to EU support based on the costs and benefits of a membership (policy focused) and affective refers to the emotional dimension of EU support.
Boomgaarden et al. (2011) based their dimensions of EU attitudes on the former and distinguished between regime-specific attitudes and EU attitudes to the community. In the latter, they looked at utilitarian/specific modes of support including evaluations of the functioning of the EU and EU integration and affective/diffuse modes of support that took into account EU identity and emotional attachment to the EU. Related to these distinctions, Bruter (2003) focused on EU identity and how symbols and news media affect this identity. He measured his main concept by using three dimensions: the spontaneous , civic and cultural dimensions.
For this study, I defined the dependent variable EU attitudes based on these well-established dimensions of this concept. However, only three different dimensions have been
distinguished. The first one is EU citizenship referring to Bruter’s (2003) civic dimension, the relevance of European citizenship and the identification with a political system. In the case of Austrian-Turks, the focus lies in investigating how cultural frames (inclusive vs. exclusive cultural frame) affect their identification with the EU as a political system, especially their rights and obligation as EU citizens.
The second is EU identity, which is related to emotional attachment to the EU on an affective dimension (Boomgaarden et al., 2011). Concerning this dimension, I assume that the inclusive cultural frame will have a positive effect on Austrian-Turks’ EU identity because the depiction of the EU, which includes citizens of different religions and cultural backgrounds may evoke an emotional attachment to the EU, which may positively affect their EU identity.
The third dimension is EU enlargement focussing on Austrian-Turks’ opinion concerning further enlargements and their consequences (Boomgaarden et al., 2011). Regarding this dimension, the assumption is in line with de Vreese et al.’s (2006) findings that the inclusive cultural frame will positively affect Austrian-Turks’s attitudes towards EU enlargement whereas the exclusive cultural frame will have the opposite effect. Overall, the main hypothesis of this study, which includes the three dimensions outlined above, is as follows:
H1: Exposure to an inclusive cultural frame positively affects Austrian-Turks’ EU attitudes whereas the negative cultural frame has negative effects on Austrian-Turks’ EU attitudes.
Explaining Effects
In regard to the three dimensions of EU attitudes, I expect that the inclusive cultural frame will have an indirect effect on Austrian-Turks’ attitudes towards EU identity. This
indirect effect is assumed because of several reasons. First, extensive research on framing effects suggested that framing effects are mainly mediated by belief importance (Nelson, Clawson & Oxley, 1997). Nelson et al. (1997) argue that issue-frames affect opinion by “selectively enhancing the psychological importance, relevance, or weight accorded to specific beliefs with respect to the issue at hand” (p.1043).
Second, Slothuus (2008) found that frames affect belief content, which means that individuals make new connections between beliefs and the topic in question when exposed to a frame. These findings show that framing effects on opinion are mediated. In the case of this study, the assumed mediator between the inclusive cultural frame and EU identity is the feeling of belonging to the EU. This is argued in regard to the complexity of Austrian-Turks’ identity, which has been studied by several researchers (Gaitanides, 1996; Heckmann, 1992; Weiss, 2007). Concerning their identity and attachment to Austria, a quantitative study of a thousand face-to-face interviews with young second generationTurkish immigrants showed that only 67% perceive Austria as their home country (Weiss, 2007). However, the same percentage indicated that they feel at home in Austria to the same degree as they feel at home in Turkey. This would suggest that Austrian-Turks feel that they belong to Austria and Turkey without having to exclude one country.
This finding can be linked to Bruter’s (2003) conclusion that national identity does not exclude European identity and that they can coexist. Even though migration studies defined immigrants of the second generation as “the generation in dichotomy” (Gaitanides, 1996; Heckmann, 1992; Kastoryano, 2002; Weiss, 2007) only two levels – the parents’ host country and the parents’ country of birth – were taken into account in research; however, with the European Union a new level plays a role in immigrants’ identity. Furthermore, given the negative depiction of Austrian-Turks in Austrian media (Zauner, 2008) and their identity-dilemma (Weiss, 2007), they may feel excluded from mainstream society and may have the
feeling to belong to the EU after being exposed to the inclusive cultural frame, which in turn may positively affect their EU identity.
In this study, I extent previous research by drawing on studies, which demonstrated that framing effects are mediated and by combining these findings with the assumption that Austrian-Turks face an identity-dilemma. Thus the following hypothesis is formulated:
H2: The inclusive cultural frame will positively affect Austrian-Turks’ EU identity via the feeling of belonging to the EU.
Moderated Mediation
Next to studies that showed how framing effects can be mediated, recent research emphasized the importance of moderators of framing effects (Entman, et al., 2009; de Vreese, van der Brug & Hobolt, 2012). In particular, de Vreese et al. (2012) emphasized the role of attitudes towards religious diversity as a moderator between news frames and attitudes towards Turkey’s accession to the EU. Religious diversity dealt with people’s attitudes towards a multitude of religions in their country and if they perceived it as detrimental to or enriching for life. They concluded that people who were positively predisposed towards religious diversity became more positive towards Turkey’s accession through exposure to the cultural frame.
When looking at Austrian-Turks, Weiss (2007) found by means of a quantitative survey (N=282) that only 19% of young Turkish immigrants of the second generation are very devoutly religious, whereas 81% do not practice Islam. Weiss (2007) concluded that Austrian-Turks who are less devout, are more open to religious diversity. This may lead Austrian-Turks who are exposed to the inclusive cultural frame and have positive predispositions towards religious diversity to identify with the EU.
My model of framing effects builds on de Vreese et al.’s (2012) and Entman et al.’s (2009) assumption that framing effects are not equal across individuals and Weiss’ (2007) work on Austrian-Turks’ identity and their feeling of belonging. It advances their work by combining the mediator feeling of belonging to the EU and the moderator positive
predispositions towards religious diversity. Thus, the following hypothesis is formulated: H3: Exposure to the inclusive cultural frame positively affects Austrian-Turks’ EU identity via the feeling of belonging to the EU and is moderated by positive
predispositions towards religious diversity.
Method Design
To investigate the effects of cultural news frames on Austrian-Turks’ EU attitudes a post-test only, between subjects survey experiment was designed. The main advantage of an experimental design is the ability to establish causality. Furthermore, it allows manipulating frames that participants are exposed to and exclude possible alternative reasons for
differences between experimental groups (de Vreese et al. 2012). Even though the chosen design has disadvantages, such as a low external and ecological validity, the positive aspects, such as the ability to measure effects between the cultural news frame in the read article (independent variable) and EU attitudes (dependent variable) and the exclusion of other factors, were decisive. The experiment that was conducted online included two experimental groups and a control group.
Pilot study
Beforehand a pilot study (N=32) was conducted to assess the comprehensibility of the articles and if participants correctly answered the content related questions depending which
frame they were exposed to. The pilot test showed that there was a difference between the groups concerning how interesting the participants found the article, F (2, 29) = 6.37, p=.005 and how comprehensible the participants assessed the articles, F (2, 29) =3.53, p =.042. Participants who were exposed to the inclusive cultural frame condition found the article the most interesting (M=5.77, SD=.83), and the most comprehensible (M=6.69, SD= .48) whereas those exposed to the control condition found it the least interesting (M=3.38, SD=2.61) and the least comprehensible (M=5.55, SD=1.75). Participants who were exposed to the exclusive cultural frame scored in between (M=3.45, SD=2.61), (M=6.63, SD=.74). LSD Fisher’s post-hoc test confirmed that there was a significant difference between the inclusive cultural frame condition and the exclusive cultural frame condition regarding participants’ interest for the articles, (Mdifference =2.31, p =.004) as well as for those exposed to the inclusive cultural
frame or assigned to the control group, (Mdifference =2.39, p =.007). The same was found
concerning the comprehensibility of the articles between the inclusive cultural frame and the exclusive cultural frame, (Mdifference =1.14, p =.020).
These findings can be explained by the hostile media effect. Scholars found that people who held strong opinions tend to be biased in their news media judgment (Gunther & Liebhart, 2006; Vallone, Ross & Lebber, 1985). In the case of this pilot study, the total mean showed that participants held overall positive opinion towards the EU in general (M=5.09, SD=1.41) and Turkey’s accession to the EU in particular (M= 5.00, SD=1.50). This could explain why those who were exposed to the inclusive cultural frame found the article more interesting and comprehensible than those who were in the two other conditions, because it was more in line with their opinion.
Furthermore, the pilot study included a manipulation check, which aimed at testing whether participants understood the arguments in the three conditions and their valence. The results of the manipulation check showed that the stimulus material was correctly understood
by the participants (see Appendix A for all the results) therefore the articles were considered appropriate to proceed in the main study and were not changed.
Sample
For the main study a sample of Austrians with Turkish migration background aged over18 years was used. The online survey experiment included two experimental groups and a control group with 30 Austrian-Turks in each group. The participants were recruited through the snowball method since the characteristics of the sample were specific and my entourage consisted only of few Austrian-Turks. Institutions such as the Türkische
Kulturgemeinde in Österreich, the Österreichischer Integrationsfond (ÖIF) and the African-Asian Institute (AAI) in Vienna were contacted in order to get access to this group of people. Additionally, people were contacted who knew Austrian-Turks that would be willing to participate in the experiment. They participated voluntarily in the online survey experiment. Even though the snowball method offers several advantages, such as an easier access to participants and the possibility to reach individuals with specific characteristics (in the case of this study: an Austrian citizenship but at least one parent who had the Turkish citizenship in the past), it can be seen as a biased sampling technique because it is not random and selects individuals based on social networks (Browne, 2005). However, the advantages were decisive and this method was most suitable to recruit participants for my study.
Stimulus material
For the study’s purposes newspaper articles were constructed. The length of the experimental groups’ articles was the same and each manipulated paragraph was built upon the same template. The title and the lead paragraph were identical for both articles. The article for the control group consisted only of the title and the lead paragraph. The
argument that only varied in their valence. Additionally, every paragraph included one quote of a fictitious expert. Given the design of the study, constructed articles were used because the focus lies on the cultural frame. Published news material usually mix different topics that are presented through different frames and therefore would not be suitable for this study. Scholars previously included the cultural frame in their studies arguing that it is an important frame used by the media to depict the EU (de Vreese et al., 2011; Koenig et al. 2006).
Therefore constructed newspaper articles still have a link to what media are reporting about. The article with the exclusive cultural frame addressed the restriction of human rights in Turkey and the legal improvements that still have to be undertaken (legal argument), the Muslim majority population that would create social tensions within Austrian society in case of Turkey’s accession to the EU (religious argument) and the financial burden the member states would have to carry in case of Turkey’s accession to the EU (economic argument).
To the contrary, the article with the inclusive cultural frame addressed the legal progress concerning human rights and democratic values that Turkey already undertook and it stressed that the accession to the EU would further stimulate it (legal argument).
Furthermore the article emphasized the enrichment of a Muslim population for the EU and the fact that the EU cherishes religious and cultural diversity (religious argument) and the minimal expenses for member states in case of Turkey’s accession to the EU (economic argument) (see Appendix B for the stimulus material).
Manipulation check
The main study included a manipulation check for the three groups. After the display of the stimulus material, participants were asked questions about the content of the articles. They had to indicate on a 7-point Likert scale (1= strongly disagree to 7= strongly agree) to which extent the article dealt with (a) high expenses for the member states in case of
democratic state. The results showed that there was a significant difference depending on the frame participants were exposed to and their answers concerning the content. For instance, participants who were exposed to the exclusive cultural frame (M=4.77, SD=2.28) noticed that the article dealt with high costs for member states in case of Turkey’s accession to the EU, whereas this was not the case for those who were in the inclusive cultural frame
(M=1.83, SD=1.60) and control condition (M=2.13, SD=1.64), F (2, 88) = 2.51, p<.001. The same result was found concerning Turkey’s depiction of a democratic state. Participants who were exposed to the inclusive cultural frame (M=5.66, SD=1.83) agreed with this statement, whereas those who were exposed to the exclusive cultural frame (M=2.50, SD=1.71) or to the neutral article (M=2.47, SD=1.77) disagreed that the articles they read included this
statement, F (2, 88) =31, 45, p<.001. Thus, the manipulation check showed successful manipulation for the experiment (see Appendix A for all the results).
Procedure
The survey was conducted using an online survey tool and was accessible during two weeks. Overall 474 persons started the survey, but after the two screening questions that asked whether they possess Austrian citizenship and whether at least one of their parents ever possessed Turkish citizenship, 302 were automatically rejected with a message stating that they did not fit into the sample. In the end, the sample consisted of 91 Austrian-Turks.
The youngest participant was aged 18 and the oldest 54 (M=27.5 SD= 7.5). Women were overrepresented in the sample (73.9 % women). The participants were randomly assigned to one of the two experimental groups or the control group. A randomization check regarding age, gender and education showed that there were no differences in age, F (2, 88) = 1.60, p = .207, gender, 𝛘² = 2.62, p = .163 or education, 𝛘² = 12.67, p = .696 between the groups. After participants read the stimulus material, an immediate post-test was conducted,
which included two open-ended questions followed by a manipulation check and questions measuring the dependent variable EU attitudes. At the end of the main study, participants were debriefed.
Measures
EU attitudes.The dependent variable EU attitudes was measured by showing
participants several questions for each dimension (see Appendix C for all the questions).An exploratory factor analysis was conducted on the 10 items for EU attitudes (see Appendix D for factor loadings and communalities). The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO=.74. An initial analysis was run to obtain
eigenvalues for each factor in the data. Three factors had eigenvalues over Kaiser’s criterion of 1 and in combination explained 58% of the variance.
The first factor had an eigenvalue of 3.62 and consisted of items relating to the EU as an institution and its democratic functioning and transparency. Furthermore, the item of trust in the EU also loaded on this factor. These items dealt with the relevance of European
citizenship and people’s identification with a political system. Linked to Bruter’s (2003) civic dimension, this factor represented the EU citizenship dimension of EU attitudes (M=3.29, SD=1.20).
The second factor had an eigenvalue of 1.63 and was labelled EU identity because it consisted of two items relating to identification with the EU and the feeling to be part of the EU (M=3.99, SD=1.36).
The third factor had an eigenvalue of 1.19 and consisted of items dealing with further widening and deepening of the EU. It included items that approved support for the integration into one country (deepening) and the possible entry of Turkey and other countries (widening).
Therefore, this factor was labelled EU enlargement (M=3.12, SD=1.28) and was very similar to Boomgaarden et al.’s (2011) dimension of EU attitudes named EU strengthening.
Furthermore, a distinction can be made because this factor included items related to a general idea of enlargement and one item that dealt exclusively with Turkey’s accession to the EU. For the purposes of this study, I distinguished between EU enlargement in general (M=3.97, SD=1.57), including all items and EU enlargement in particular (M=3.09, SD=1.76), taken only the item about Turkey’s accession into account.
In a next step, scales were constructed based on the findings of the factor analysis. The EU citizenship scale had the highest reliability, Cronbach’s α = .80 and the scale EU enlargement had an acceptable reliability, Cronbach’s α =.60. However, the scale EU identity was the least reliable, Cronbach’s α =.50. This could be linked to the few items in this scale. According to Cortina (1993) the value of α depends on the number of items because the relationship between the number of items and α is curvilinear. For this reason, he suggested to look at the average correlation between items, which lies in the case of the EU identity scale above .30, which is respectable (Cortina, 1993). Therefore, the scale EU identity was taken into account in this study.
Mediator and moderator.This study included one mediator being feeling of belonging to the EU (M=3.74, SD=1.77) and one moderator being attitudes towards religious diversity (M=5.65, SD=1.74). The mediator was operationalized with the following question: “After you read this article, do you feel that you belong to the European Union” 1 means I feel that I belong to the EU and 7 I feel that I don’t belong to the EU. Participants were offered a 7-point scale in which only the two extremes were labelled. The moderator attitudes towards religious diversity was adopted from de Vreese et al.’s (2012) study about the effect of cultural and economic frames on support for Turkish accession. It was operationalized through the following question: “The next question concerns the number of different religions
in Austria. Which position describes your view on this matter?” The participants were given a 7-point scale, where 1 means that you feel that having many different religions is detrimental to the quality of life in Austria and 7 means that you feel that having many different religions enrich the quality of life in Austria.
Analysis
An Analysis of Variance (ANOVA) was conducted to test H1. The aim was to test if exposure to the inclusive cultural frame positively affects Austrian-Turks’ EU attitudes and if exposure to the exclusive cultural frame negatively affects Austrian-Turks’ EU attitudes. Given that I wanted to assess if there were any differences between the three groups
regarding Austrian-Turks’ EU attitudes, I opted for this analysis. I conducted three ANOVAs for the three dimensions of EU attitudes to see if there were differences between the inclusive cultural frame condition, the exclusive cultural frame condition and the control condition regarding participants’ attitudes towards these three dimensions.
In order to test the mediation and moderated mediation hypotheses – my H2 and H3 – I applied the method introduced by Preacher and Hayes (2004). These authors suggest using bootstrapping techniques, a nonparametric re-sampling approach to effect-size estimation, which makes no assumption about the shape of the distribution and improves the power of a model (Preacher & Hayes, 2004). Preacher, Rucker and Hayes (2007) advise bootstrap confidence intervals as the most powerful method to evaluate the significance of indirect effects. Applying this method enables me to generate 95% bias-corrected confidence intervals (95% BCa CI) on the basis of 1000 samples bootstrap samples for indirect effects. If the confidence interval does not contain zero, it means that the indirect effect significantly differs from zero (Preacher & Hayes, 2007; Preacher, Rucker & Hayes, 2008). I applied this method to my study context to assess if the inclusive cultural frame will positively affect
Austrian-Turks’ EU identity through the feeling of belonging (see Figure 1 for the model). Even though mediation analysis is often used in social sciences, Green, Ha, and Bullock (2010) stressed two important aspects, which have to be considered when conducting this type of analysis. The first aspect concerns omitted variables, meaning that researchers assume that the mediator (M) is correlated with the dependent variable (Y), though it could be that both are correlated with unobserved confounders (Green et al., 2010). The second aspect is that M is imperfectly measured, which may lead to an underestimation of M’s effect and concluding that factors other than M are responsible for the relationship between X and Y (Green et al., 2010).
Besides, I combined the mediation and moderation analysis, which has been referred to as mediated moderation or moderated mediation (Baron & Kenny, 1986). Based on these termini, Preacher et al. (2008) gathered such effects under the term conditional indirect effects and defined them as “the magnitude of an indirect effect at a particular value of a moderator” (p.186). In the context of my study, I wanted to assess at what value of predispositions towards religious diversity the inclusive cultural frame positively affects Austrian-Turks’ EU identity through the feeling of belonging to the EU (see Figure 2 for the model). To conduct these analyses, I used Hayes’ (2013) PROCESS macro in SPSS.
--FIGURE 1 ABOUT HERE--
Results
Main effects on EU attitudes’ dimensions
First, I predicted that the exposure to an inclusive cultural frame positively affects Austrian-Turks’ EU attitudes whereas the negative cultural frame has negative effects on Austrian-Turks’ EU attitudes. A one-way between subjects Analysis of Variance (ANOVA) was conducted to compare the effect of cultural news frames on EU attitudes’ dimensions in the inclusive cultural frame, the exclusive cultural frame and the control conditions (see Table 1 for all the means). Regarding the dimension EU citizenship the ANOVA showed no significant effect of cultural frames for the three conditions, F (2, 88) = .67, p = .512.
Furthermore, no significant effect was found regarding the three groups and the EU identity dimension, F (2, 88) = 1.27, p =.286. Finally, regarding the dimension EU enlargement the ANOVA showed no significant effect of cultural frames for the three conditions, F (2, 88) = 1.41, p = .250. However, when considering EU enlargement in particular, meaning Turkey’s possible accession to the EU, the ANOVA showed a marginal significant effect between the three groups, F (2, 88) = 2.90, p = .060. Thus, a post-hoc test was carried out. LSD Fisher’s test confirmed that there was a significant difference between participants who were exposed to the inclusive cultural frame (M = 3.72, SD=1.79) and those who were exposed to the exclusive cultural frame (M = 2.73, SD=1.68) regarding their attitude towards Turkey’s possible accession to the EU, (Mdifference =.991, p =.031).
Overall, the H1 was not confirmed, since the cultural frames did not have an overall effect on Austrian-Turks’ EU attitudes. Exposure to the inclusive cultural frame did not positively affect Austrian-Turks’ EU attitudes and participants who were exposed to the exclusive cultural frame were not negatively affected towards their EU attitudes.
Mediation and Moderated mediation
Second, I predicted an indirect positive effect of the inclusive cultural frame on Austrian-Turks’ EU identity via the belonging to the EU. As it can be noted in Figure 1, there was no direct effect between the independent variable, being the inclusive cultural news frame and the dependent variable, EU identity b = 0.41, p =204. Furthermore, the total indirect effect of the inclusive cultural frame on Austrian-Turks’ EU identity through the feeling of belonging to the EU, b = 0.415, BCa CI [-0.108, 0.358] was non-significant. Therefore, H2 has to be rejected. This means that Austrian-Turks’, who were exposed to the inclusive cultural frame did not have the feeling that they belong to the EU and therefore the inclusive cultural frame did not positively affect their EU identity.
Third, I predicted that exposure to the inclusive cultural frame positively affects Austrian-Turks’ EU identity via the feeling of belonging to the EU and is moderated by positive predispositions towards religious diversity. Table 2 provides the conditional indirect effects of the inclusive cultural frame on EU identity through the feeling of belonging to the EU at one standard deviation above the mean of the moderator variable (M =7.00; “positive attitudes towards religious diversity”), directly at the mean (M = 5.57; “neutral attitudes towards religious diversity”) and one standard deviation below the mean (M = 3.89; “negative attitudes towards religious diversity”). For the analysis the independent variable was coded as followed: inclusive cultural frame=1, exclusive cultural frame=0 and control group = system missing.
The results showed that attitudes towards religious diversity moderates the indirect effect of the inclusive cultural frame on EU identity via the feeling of belonging to the EU for participants who had positive attitudes towards religious diversity b = 1.106, 95% CI [0.435, 1.780]. However, when participants do not held strong attitudes towards religious diversity there is a non-significant negative relationship between the indirect effect of
exposure to the inclusive cultural frame and EU identity through the feeling of belonging to the EU, b = 0.496, 95% CI [-0.361, 1.021]. Additionally, attitudes towards religious diversity did not moderate the indirect effect of the inclusive cultural frame on EU identity via the feeling of belonging to the EU for participants who were not in favour of religious diversity, the b = -0.219, 95% CI [-0.884, 0.490]. Thus, the H3 is confirmed.
TABLE 2 ABOUT HERE
--Discussion
On a political level decisions concerning the future of all EU citizens are often made by elites. However, extensive research has shown how important it is to take public opinion into account. As the EU is often perceived as intangible and abstract by the majority of
citizens, media play a crucial intermediary role between the EU and citizens (Boomgaarden et al. 2011; Deflem & Pampel, 1996; de Vreese et al., 2012; Maier & Rittberger, 2008). This study not only supports the importance of the interplay between media and public opinion regarding the EU, but went a step further by showing that it is highly important to dig deeper and not only including native European citizens but also considering immigrants because they are affected to the same extent by decisions made by the EU.
This study contributes to the literature about framing effects on EU attitudes in the following three ways. First, I showed how cultural news frames affect Austrian-Turks’ EU attitudes. Contrary to expectations, I found that neither participants’ exposure to the inclusive cultural frame positively affected their EU attitudes nor participants’ exposure to the
exclusive cultural frame negatively affected their EU attitudes. This absence of main effects could be explained by the fact that I primarily focused on Turkey’s accession to the EU in the
articles. This could have led participants to concentrate solely on this issue without taking the big picture into account, that is, the overall depiction of a more tolerant EU that cherishes cultural and religious differences versus a more intolerant EU depicted as a Union of like-minded countries that are against a mix of cultures and religions.
Second, participants who were exposed to the inclusive cultural frame had a more positive attitude towards Turkey’s accession to the EU. This could be explained by the specific issue the stimulus article dealt with: Turkish membership. Furthermore, this effect can be explained by the positive arguments presented in this condition and shows that the combination of legal, religious and economic arguments supporting Turkey’s accession does affect Austrian-Turks’ opinion on this issue.
This finding is in line with Schuck et al.’s (2006) study which concluded that the opportunity frame, including different positive aspects of EU enlargement, positively affected people’s attitudes towards EU enlargement. Furthermore, it is in accordance with the
emphasis framing effect, which considers that a specific angle of an issue’s depiction make people emphasize this angle when forming their opinion (Druckmann, 2004). However, this finding is inconsistent with de Vreese et al.’s (2006) results which indicated that the cultural frame was not as effective as they predicted. They explained this through the context of their study, which was conducted in 2010 during the economic crisis across Europe. Therefore, participants were more affected by the economic frame and by the fact that their cultural frame merely emphasized religious arguments. Given that I included legal, religious and economic arguments in the inclusive cultural frame, participants were exposed to aspects that went beyond religious ones, which could explain the strong positive effect of the cultural frame on EU enlargement in particular, thus Turkey’s accession to the EU.
Third, my results complete recent communication science studies on framing effects by emphasizing the impact of individual moderators. I built on Hobolt et al.’s (2012) and de
Vreese et al.’s (2006) studies that highlighted the importance of taking individual
predispositions towards religious diversity into account when investigating framing effects on EU attitudes. However, I also included a mediator – the feeling of belonging to the EU – in my model and combined the moderation analysis with the mediation analysis. I expected an indirect effect of exposure to the inclusive cultural frame on EU identity via the feeling of belonging to the EU for participants who supported religious diversity. Indeed, my finding confirmed the presence of a conditional indirect effect. This is particularly important due to the absence of an indirect effect of the inclusive cultural frame on EU identity through the feeling of belonging to the EU. Besides, for those who were negatively or indifferently predisposed towards religious diversity, no conditional indirect effect was found. This is in line with de Vreese et al.’s (2012) study, which confirmed that not all individuals are affected in the same way by cultural frames. However, by combining a mediator and a moderator, my study goes a step further and suggests that a cultural frame, which depicts the EU as a
multicultural Union that cherishes diversity, gives Austrian-Turkish citizens with positive predispositions towards religious diversity the feeling to be part of it, which in turn may increase their identification with the EU. Therefore, this study adds evidence that feelings and individual predispositions play a role in the relationship of framing effects and Austrian-Turks’ EU identity.
Even though the findings give valuable insights into the interplay between framing effects and Austrian-Turkish EU attitudes, I acknowledge that my study has several shortcomings. Regarding the experimental set-up, I cannot be certain if the results are
affected by the Austrian context, given the rather negative sentiments Austrians hold towards immigrants compared to other EU citizens (Thalhammer, 2001). Besides, I only focused on Turkish immigrants of the second generation in Austria but the findings may be different when focusing on the same target group in, say, Germany or France. Therefore it would be
interesting to extend the study and take Turkish immigrants of the second generation of different EU countries into account and compare these findings with mine. Consequently, I cannot make inferences from my sample to any “real” population. With regards to the experimental design, there is an ecological validity problem: the situation in which the participants were exposed to the stimulus material was artificial and I cannot be certain that they would have paid as much attention to the content of the articles in their every-day lives.
Concerning the measures, the EU identity dimension of EU attitudes was taken into account in my study although its two item scale was not reliable with a Cronbach’s α of .50. Nevertheless, I included it and argued based on Cortina’s (1993) assumption that the
relationship between items and the value of α is curvilinear and therefore that the correlation is more meaningful when dealing with two item scales. Given that the inter-correlation value was respectable, I kept the scale. Nevertheless, this has to be considered when interpreting my findings.
Future research could extend these findings by including different moderators in order to investigate how cultural frames can affect Austrian-Turks’ EU attitudes in general and their EU citizenship, EU identity and attitudes towards EU enlargement in particular. For instance, Turkey’s current political situation could be included as a moderator. In this study, participants were asked to express their opinion in one sentence concerning Turkey’s
accession to the EU and anecdotal evidence showed that some participants were against Turkish membership as long as President Erdogan remains in power. They stressed his conservative religious views and his undermining of basic human rights. Therefore, strong negative opinions towards Turkey’s political leadership could have lessened the framing effects on Austrian-Turks’ attitudes towards Turkey’s accession to the EU. Given that this assumption is not scientifically founded, future research could extend these findings and take Austrian-Turks’ predispositions towards the Turkish government into consideration.
Furthermore, this study considered immigrants when dealing with media effects on EU attitudes, which should be emphasized in further research because native citizens, as Bruter (2003) confirmed in the context of the UK and de Vreese et al. (2012) in the context of the Netherlands, can identify with their home country as well as with the European Union. In the context of Austria, as this study showed, Austrian-Turks identify as well with the EU but under some conditions: the way of framing the issue and their individual predispositions towards religious diversity. The complex identity construction of immigrants has been investigated extensively in migration studies (Gaitanides, 1996; Heckmann, 1992; Weiss, 2007) but experiments, as the one conducted in this study, add insights regarding framing effects on their EU attitudes and should be pursued in the future.
My study demonstrated the importance to include all EU citizens – natives and immigrants – when discussing EU related topics given that they affect all of us. In
multicultural societies such as those of many European countries, studies concerning framing effects on EU attitudes should not neglect immigrants. In the case of Austria, the sentiment towards the EU in general and Turkish membership in particular is negative (Matzka, 2009; Thalhammer et al. 2001). This study, however, showed that this may be tightly linked to the EU’s depiction in the media and individuals’ predispositions.
References
Austria, S. (2013). Kommission für Migrations-und Integrationsforschung der
Österreichischen Akademie der Wissenschaften, 2013. Migration & Integration: Zahlen, Daten, Indikatoren.
Boomgaarden, H. G., Schuck, A. R., Elenbaas, M., & de Vreese, C. H. (2011). Mapping EU attitudes: Conceptual and empirical dimensions of Euroscepticism and EU support. European Union Politics, 12(2), 241-266.
Browne, K. (2005). Snowball sampling: using social networks to research non‐heterosexual women. International Journal of Social Research Methodology, 8(1), 47-60.
Bruter, M. (2003). Winning Hearts and Minds for Europe: The Impact of News and Symbols on Civic and Cultural European Identity. Comparative Political Studies, 36(10), 1148– 1179.
Chong, D. (1993). How people think, reason, and feel about rights and liberties. American Journal of Political Science, 867-899.
Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of applied psychology, 78(1), 98.
Damelang, A., & Haas, A. (2006). Arbeitsmarkteinstieg nach dualer Berufsausbildung: Migranten und Deutsche im Vergleich (No. 2006, 17). IAB-Forschungsbericht.
Deflem, M., & Pampel, F. C. (1996). The myth of postnational identity: Popular support for European unification. Social Forces, 75(1), 119-143.
De Vreese, C. H. (2006). Media Message Flows and Interpersonal Communication: The Conditional Nature of Effects on Public Opinion. Communication Research, 33(1), 19– 37.
De Vreese, C. H., & Boomgaarden, H. G. (2005). Projecting EU Referendums Fear of Immigration and Support for European Integration. European Union Politics, 6(1), 59-82.
De Vreese, C. H., & Boomgaarden, H. G. (2006). Media Effects on Public Opinion about the Enlargement of the European Union*. JCMS: Journal of Common Market Studies, 44(2), 419–436.
De Vreese, C. H., Boomgaarden, H. G., & Semetko, H. a. (2011). (In)direct Framing Effects: The Effects of News Media Framing on Public Support for Turkish Membership in the European Union. Communication Research, 38(2), 179–205.
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.
Druckman, J. N. (2004). Political preference formation: Competition, deliberation, and the (ir) relevance of framing effects. American Political Science Review, 98(04), 671-686. Easton, D. (1975). A re-assessment of the concept of political support. British Journal of
Political Science 5, 435–457.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of communication, 43(4), 51-58.
Erzan, R., & Kirisçi, K. (2004). Turkish immigrants: their integration within the EU and migration to Turkey. Turkish Policy Quarterly, 3(3), 61-68.
EUCE. (2008). Turkey’s quest forEU Membership (pp. 1–8). Retrieved from
http://europe.unc.edu/wp-content/uploads/2013/08/Brief4-0803-turkeys-quest.pdf European Commission, (2008). Eurobarometer 69 Public Opinion in the European Union
(1st ed.). Retrieved from
Fassmann, H., Münz, R., & Seifert, W. (1997). Die Arbeitsmarktposition ausländischer Arbeitskräfte in Deutschland (West) und Österreich. Mitteilungen aus der Arbeitsmarkt-und Berufsforschung, 30(4), 732-745.
Gaitanides, S. (1992). Probleme der Identitätsfindung der zweiten Einwanderergeneration (pp. 32–40).
Glossy, F., & Hametner, M. (2014). Die türkische Community in Österreich. Der Standard. Retrieved from http://derstandard.at/2000002072258/Die-tuerkische-Community-in Oesterreich
Green, D. P., Ha, S. E., & Bullock, J. G. (2010). Enough already about “black box” experiments: Studying mediation is more difficult than most scholars suppose. The Annals of the American Academy of Political and Social Science, 628(1), 200-208. Günay, C. (2008). Austrian stakeholders in the EU-Turkey debate. Talking Turkey in Europe:
Towards a Differentiated Communication Strategy. Rome, Italy: Quaderni IAI, 59-84. Gunther, A. C., & Liebhart, J. L. (2006). Broad Reach or Biased Source? Decomposing the
Hostile Media Effect. Journal of Communication, 56(3), 449–466.
Hayes, Andrew, F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis: A regression-based approach (p. 507). Guilford Press.
Heckmann, F. (1992). Ethnische Minderheiten, Volk und Nation: Soziologie inter-ethnischer Beziehungen. Stuttgart: Ferdinand Enke.
Kastoryano, R. (2002). Negotiating Identities: States and immigrants in France and Germany. Princeton University Press (p. 240). Princeton University Press.
Koenig, T., Mihelj, S., Downey, J., & Gencel Bek, M. (2006). Media Framings of the Issue of Turkish Accession To the Eu. Innovation: The European Journal of Social Science Research, 19(2), 149–169.
Lecheler, S., de Vreese, C., & Slothuus, R. (2009). Issue Importance as a Moderator of Framing Effects. Communication Research, 36(3), 400–425.
Lindberg, L., & Scheingold, S. (1970). Europe’s would be-polity: Patterns of Change in the European Community. New Jersey: Prentice Hall.
Lubbers, M. (2005). Political versus Instrumental Euro-scepticism: Mapping Scepticism in European Countries and Regions. European Union Politics, 6(2), 223–242.
Luif, P. & Trauner F. (2012): Österreich: In: Weidenfeld, Werner/Wolfgang Wessels (eds): Jahrbuch der Europäischen Integration 2012, Baden-Baden: Nomos, 439-442.
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. Matzka, C. (2009). Austria and Turkey: Their burden of histories. Geographical diversity,
163-169.
McLaren, L. M. (2008). Public Support for the European Union: Cost/Benefit Analysis or Perceived Cultural Threat? The Journal of Politics, 64(02), 551–566.
Nelson, T. E., & Oxley, Z. M. (1999). Issue framing effects on belief importance and opinion. The Journal of Politics, 61(04), 1040-1067.
Preacher, K., & Hayes, A. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Bahvior Research Methods, Instruments Computers, 36, 717– 731.
Preacher, K., Rucker, D., & Hayes, A. (2008). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Behavior Research Methods, 40, 879–891.
Schalk-Soekar, S. R. G., van de Vijver, F. J. R., & Hoogsteder, M. (2004). Attitudes toward multiculturalism of immigrants and majority members in the Netherlands. International Journal of Intercultural Relations, 28(6), 533–550.
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of communication, 49(1), 103-122.
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., & De Vreese, C. H. (2006). Between Risk and Opportunity News Framing
and its Effects on Public Support for EU Enlargement. European Journal of Communication, 21(1), 5-32.
Slothuus, R. (2008). More Than Weighting Cognitive Importance: A Dual‐Process Model of Issue Framing Effects. Political Psychology, 29(1), 1-28.
Sniderman, P. M., & Theriault, S. M. (2004). The structure of political argument and the logic of issue framing. Studies in public opinion: Attitudes, nonattitudes, measurement error, and change, 133-65.
Thalhammer, E., Zucha, V., Enzenhofer, E., Salfinger, B., & Orgis, G. (2001). Attitudes towards minority groups in the European Union: A special analysis of the
Eurobarometer 2000 survey. Vienna.
Torreblanca, J. I. (2007). European public opinion and Turkey's accession: Making sense of arguments for and against (Vol. 16). CEPS.
Van Gorp, B. (2005). Where is the frame? Victims and intruders in the Belgian press coverage of the asylum issue. European Journal of Communication, 20(4), 484-507. Weiss, H. (2007). Leben in zwei Welten. Zur sozialen Integration ausländischer
Jugendlicher der zweiten Generation, Wiesbaden.
Zauner, K. (2008). Einstellungen von ChronikjournalistInnen österreichischer
Tageszeitungen zu den Themen Migration und mediale Integration von MigrantInnen im Kontext ihres Rollenverständnisses. Universität Wien.
Tables and figures Table 1
Means of the three conditions on EU attitudes’ dimensions EU attitudes EU citizenship EU enlargement (general) EU enlargement (in particular) EU identity
Groups n Mean n Mean n Mean n Mean
Inclusive cultural frame 29 3.41 29 3.72 29 3.72 29 4.52 Exclusive cultural frame 30 3.39 30 2.73 30 2.73 30 4.37 Control group 32 3.09 32 2.84 32 2.84 32 3.91 Total 91 3.29 91 3.09 91 3.09 91 4.25 Note. EU enlargement (general) refers to all items included in the EU enlargement scale, whereas EU enlargement (in particular) only includes one item concerning Turkey’s accession to the EU
Figure 1. Indirect effect of X on Y through M1. Note. This model is based on Model 4 in PROCESS (Hayes, 2013). The confidence interval for indirect effect is a BCa bootstrapped CI based on 1000 samples. Inclusive cultural news frame (x) EU Identity (y) Feeling of belonging to the EU (M1) b =0.67, p = .135 b =0.60, p = .000 Direct effect, b = 0.41, p =204 Indirect effect, b = 0.41, 95% CI [-0.108. 0.358]
Figure 2. Conditional indirect effect of X on Y through M1. Note. Direct effect of X on Y= c'.
This model is based on Model 7 in PROCESS (Hayes, 2013).
c ' Inclusive cultural news frame (x) EU Identity (y) Feeling of belonging to the EU (M1) Attitudes towards religious diversity (W)
Table 2
Conditional indirect effect of the inclusive cultural frame on EU identity through the feeling of belonging to the EU
BC 1000 BOOT
Moderator Mediators Point
estimate
SE LL95 UL95
EU identity
Negative attitudes towards religious diversity Belonging to EU -.2195 .3533 -.8842 .4901 Neutral attitudes towards religious diversity Belonging to EU .4964 .2642 -.0361 1.0214 Positive attitudes towards religious diversity Belonging to EU 1.1064 .3339 .4538 1.7803 Note: Negative attitudes towards religious diversity -1SD from mean (3.89); Neutral attitudes towards religious diversity: mean (5.57); Positive attitudes towards religious diversity: +1SD from mean (7.00). Significant indirect effect is determined by the BCa 95% confidence interval (CI) which does not include a zero. Unstandardized effect sizes. Independent variable coded as 1=inclusive cultural frame, 0=exclusive cultural frame, Syst. miss. = control group.
Appendix A
Results manipulation check pilot study Table A1
Manipulation check – Analysis of variance (ANOVA) between the inclusive cultural frame group (n=13), the exclusive cultural frame group (n=11) and the control group (n=8)
Question df F η² P General interest for Austrians 2 3.26 .17 .052 Opinion changed 2 2.33 .06 .115 Understandable 2 3.53 .08 .053 Interesting 2 6.37 .41 .005 Quality journalism 2 3.83 .19 .033 General interest for Turkish migrants of the second generation 2 2.36 .14 .112
Note. The displayed results show differences between groups. Significant at the p<.0.05 level
Table A2
Manipulation check on the articles’ content – Analysis of variance (ANOVA) between the inclusive cultural frame group (n=13), the exclusive cultural frame group (n=11) and the control group(n=8)
Question The article that you read mentions…?
Df F η² p
High costs for
member states 2 25.06 .10 .000
Turkey’s
Turkey as a democratic state 2 29.94 .92 .000 EU as a Christian Union*² 2 1.15 .06 .329 Cultural differences between member states 2 4.10 .21 .027
Low costs for
member states 2 22.38 .03 .000 Freedom of religion and tolerance within the EU 2 7.81 .35 .002
Results manipulation check main study Table A3
Manipulation check for the main study Question
The article that you read mentions…?
Df F η² p
High costs for
member states 2 22.51 .29 .000 Turkey’s accession 2 2.40 .01 .096 Turkey as a democratic state 2 31.45 .37 .000 Fear of Islamisation of the EU 2 23.96 .33 .000 Large cultural differences between member states 2 46.68 .54 .000
Low costs for
member states 2 22.38 .22 .000 Freedom of religion and tolerance within the EU 2 40.47 .47 .000
Appendix B Stimulus material Article: Exclusive cultural frame (German)*
*English translation
New Think Thank Study on Turkey’s EU-Entry Spurs Interest
among Experts
Brussels/Vienna – The Brussels-based Think Tank “Friends of Europe” presented a new study on Turkey‘s much debated EU-Entry during a press conference last Wednesday. The study spurred great interest among the Austrian experts that were present. In their study, „Friends of Europe“ mention the popular argument regarding limited basic rights in Turkey. Austrian human rights expert, Thomas Kern, agrees with the latter. “Turkey is not ready for the EU. The cultural and legal differences are simply too big. Turkey still has a long way ahead to become a fully democratic state that respects and protects human rights”. „Admitting a country with a population that is Muslim in majority can lead to social
conflicts“, says the Think Tank’s study. Peter Unterhauser, a sociologist at the University of Innsbruck, shares concerns about an Islamization of Europe. Unterhauser highlights the increasing intercultural conflicts between Turks and Austrians in Austria and argues that Turkey’s EU-entry would augment such conflict in his soon to be published book „European Future”
According to the Brussels-based Think Tank’s calculations EU member countries would experience high costs if Turkey were to fully enter the EU. Experts agree: “The country is so poor that 16, 5-27, 5 billion Euros would have to flow into Turkey. If Turkey were to enter the EU, Austria, too, would have the bear financial costs”, says EU economist Manuel Henn. According to European Commission’s new president, Jean-Claude Juncker, negotiations regarding Turkey’s as well other EU accessions will be resumed in Brussels during the spring. (APA, 09.10.2014)
