Erasmus Mundus Master of Arts - Journalism, Media and Globalisation
[ joint degree ]
Uniting Audiences -
the Role of Television News Frames
on European Solidarity
Florence Kerstin Schulz Master’s Thesis Student ID: 5
Graduate School of Communication Master’s programme Communication Science Supervisor/Examiner: Dr. Katjana Gattermann
2
Abstract
News framing can strongly influence public opinions about EU policies and generate, for example, support or cynicism. However, virtually nothing is known about frame effects on solidarity. This study therefore compares how television news framed in terms of ‘common EU identity’ in comparison to ‘cost and benefit’ can affect levels of European social solidarity. Simultaneously, the role of belief importance as a possible mediator on those effects is tested. Solidarity, it turns out, is not influenced by the news frames and appears to be a constant, personal attitude, possibly driven by egalitarian values and cultural openness. Nevertheless, an indirect mediation influence of belief importance has been measured. This indicates that, while frames may not affect solidarity as such, they can alter the priority of considerations at play when thinking about solidarity issues. A majority of people seems to prefer identity-related considerations over cost and benefit calculus in this regard.
Introduction
The last decade has brought turbulent times for the European Union, leaving many doubting the success of an economic and political union of nations. Frictions between the so-called giver and receiver states began increasing as the Euro crisis unfolded in 2008, paving the way towards the EU’s status quo. Populism and a return to nationalist movements are currently sweeping across Europe (Bechtel, Hainmueller & Margalit, 2014; Ferrera, 2014). In his speech for the 60th anniversary of the Treaty of Rome, Pope Francis said the EU was facing
its end if it did not develop a new vision for itself: “Europe finds new hope in solidarity,
Word count: 7012
3 which is also the most effective antidote to modern forms of populism” (Bremer, Stabenow, & Schuller, 2017). Since its foundation, the ideal of solidarity has been a cornerstone of the European Union and also appears as a central theme in documents of the Lisbon Treaty. They refer to European solidarity as the crucial ingredient that keeps the Union together despite its ethnic heterogeneity and as a powerful tool to bolster the nation states’ problem-solving capabilities (Gerrits, 2014).
The challenges the EU needs to overcome are manifold, as Ferrera (2014) and Habermas (2001) point out: shaping a European civil society and shared political culture, strengthening the EU’s social dimension, upholding the institutional foundations of the national welfare systems as well as managing the redistributive tensions between the states. For Habermas, the existence of a European-wide public sphere is key to approach these issues. By stimulating and shaping the public discourse about the EU, the media thus take on a central role in the way citizens talk and feel about the EU. The way news on European issues are framed may therefore directly contribute to or erode solidarity among EU citizens. As previous research has shown, public support for EU policies and integration tends to be volatile and strongly dependent on media framing, oftentimes more than with national matters (de Vreese, 2003).
Given the powerful influence news coverage can exert on the public European discourse, this study aims to fill a substantial gap in the research field of framing effects. In an experiment employing two different news frames that depict a European issue, the influence of news framing on the levels of solidarity that Europeans feel for each other is measured. The research question “How do news frames of identity and cost/ benefit affect social solidarity among Europeans?” therefore seeks to explore how, to what extent and under which
moderating and mediating conditions social solidarity may be shaped through television news framing.
4 Research on solidarity, specifically on the European level, is scarce so far. Few empirical studies have been conducted based on survey data. By using Eurobarometer recordsand comparing support for welfare systems in fourteen European countries, Gelissen (2000) found that the type of welfare system had no impact on solidarity levels among the
population, while confidence in democratic institutions and commitment to the common good appeared to be strong predictors of solidarity levels. In 2009, Kankaras and Moors examined a more technical aspect by comparing measurements of solidarity within the EU by
conducting factor analyses on data from the 1999/2000 wave of the European Value Study. Important conclusions can be drawn from research done by Mau and Burkhardt (2009), who revealed a negative correlation between increased ethnic diversity and support for welfare state redistribution and inclusion of foreigners within states. This raises the question whether multiculturalism and free movement within the EU may lower feelings of community and solidarity. Bechtel, Hainmueller & Margalit (2014) assume the same and observe that eventually it is the degree of cultural openness which significantly conditions solidary support, in their case for Eurozone bailouts among a German public. A more nuanced approach to European solidarity is taken by Ciornei and Recchi (2017), who distinguish between trans- and international solidarity and explore the joint role of cross-border
interactions and political attitudes on the two types of solidarity. The influence of experiences abroad, they find, eventually depend on the individual commitment to egalitarian values.
While previous research has predominantly aimed at determining the causes and conditions of European solidarity in cross-country comparisons, virtually nothing is known about framing effects on solidarity. Moreover, Ciornei and Recchi (2017) see a lack of methodological approaches to better understand the concept, suggesting experimental research to be conducted. In line with this, the aim of this study it not only to add insights to
5 the field of television framing effects but to also amplify existing knowledge about the
support for European integration, in which solidarity may be absolutely necessary.
Dimensions of social solidarity and its driving factors
While solidarity among nation states generally appears in a political and economic context, solidarity between people differs in its nature. By referring to interpersonal ties, scholars tend to speak of social solidarity, for which Ciornei and Recchi (2017, p.2) offer the following definition: "At the societal level, solidarity entails networks of relationships that presuppose dependency, reciprocity and responsibility among the members of a group or a political community. Yet social solidarity is not charity or a gift, but an institutionalized system of mutual help of interdependent individuals." Taking this rather utilitarian understanding as a starting point, social solidarity seems to be primarily about sharing risks and resources to ensure security and prosperity for a defined community of people (Gelissen, 2000; Kankaraš & Moors, 2009). This can, however, happen on several dimensions, ranging from the micro to the macro level (see Figure 1). Since the aim of the experiment is to identify how news frames influence solidary feelings on the European level, the EU (macro) dimension is of most interest here; still the individual (micro), national (meso) and global level need to be defined to provide a larger context.
The first dimension of solidarity, as identified by Abela (2004), is located on the micro level. It refers to solidarity between individuals and their direct environment. In its more systematic form however, solidarity is usually referred to on the second dimensions, the meso level. Here it refers to the cohesive and redistributive functions of social protection systems in a nation-state. "In these terms welfare states are considered as concrete and generalized realizations of solidarity" (Kankaraš & Moors, 2009, p.558). Following this logic of
6 institutionalised solidarity, it becomes difficult to apply the concept to a third dimensions, the EU macro level. It is problematic as a solidary welfare system only functions within a
somehow defined, thus limited, community, making solidarity at the same time an inclusive and exclusive dynamic (Gerrits, 2014; Kankaraš & Moors, 2009). Since solidarity is tied to some idea of community, solidarity among EU citizens can therefore only exist where a national and European identity overlap in an individual, and the obligations felt towards those two political dimensions is successfully negotiated (Ferrera, 2014). The macro dimension of solidarity is manifested in two ways, first as horizontal solidarity, meaning the ties between EU citizens towards each other, and second as vertical solidarity, meaning from the individual person towards other EU nation-states in general (Witte, 2012). Finally, Abela (2004) also identifies a fourth dimension, which is a much broader expression of solidary feelings. It is directed at humankind in general, a global dimension. Karagiannis (2007) calls this exterior EU solidarity, mostly in terms of humanitarian relief measures.
Having identified the dimensions of solidarity, the questions remains what factors actually produce social solidarity. Gelissen (2000) offers a quite clear categorisation of sources from which social solidarity (not necessarily European solidarity) derives. Concerning the
individual, he emphasises affectionate and emotional reasons on as well as a perceived Figure 1: Dimensions of social solidarity
7 common identityon . These culturally based convictions may evoke a moral obligation to serve the collective interest, he concludes. Besides that, people seem to apply a rational approach to solidary engagement and calculate their long-term interest in a costs and benefit equasion (e.g. the willingness to pay taxes for later support). Furthermore, Gelisses speaks of community control and coercion to contribute to the common good as source of solidarity, making it a necessary factor to prevent free-riding members in society. By combining these factors with quite similar patterns found by other scholars (Gerrits, 2014; Kankaraš & Moors, 2009; Witte, 2012), this study summarises three driving factors of European social solidarity:
- Cost/benefit calculus (utilitarian self-interest logic)
- identity related factors (perception of belonging to a defined community)
- egalitarian values (equal rights and commitment to the common good)
In order to explore the role these three factors play in solidarity framing effects and whether they might interact with each other, they first need to be examined more closely.
Cost benefit calculus and self-interest
With regard to the individual person, the rational side of solidarity is generally explained as self-interest logic (Ciornei & Recchi, 2017). Hechter's rational choice theory states that the more an individual depends on the welfare state, the higher their support for solidary systems will presumably be. This could only partially be supported by Gelissen (2000), who expected age, gender, income, chances on the job market, health issues etc. to have a rational,
moderating impact on solidarity. Interestingly though, he came to the conclusion that
confidence in democratic institutions was the most significant predictor of solidarity - people who evaluate their democratic institutions to appropriately exercise political power display higher support for solidary policies.
8 In this point, social solidarity appears to follow the same logic as trust in the EU, which has been researched by Harteveld, van der Meer and De Vries (2013). However, they conclude that the extent to which a rational assessment of institutional merits is objective may be irrelevant, as a subjective perception can still be rational. With regard to solidarity, Ciornei and Recchi (2017) have found rather cosmopolitan and well-educated citizens to express higher solidarity towards other nationalities. This, too, indicates the influence of rational considerations and to some degree of political sophistication as influential factors on solidarity.
Identity-related factors
In the analysis of attitudes on EU integration, identity has repeatedly been found to be a key moderator predicting support for the EU (Harteveld, van der Meer & De Vries, 2013; Risse, 2013; de Vreese, 2003). Simply put: the more people identify with a given institution, for example the welfare system, the more they are likely to support it. Within the EU, the notion
of a shared identity is a problematic concept though: the Union’s multilevel and supranational governance may be perceived as a threat, a devaluation of the national identity (Hooghe, & Marks, 2005). As long as other EU nations are perceived as "out group", people might emotionally prefer national governance over transnational institutions (Harteveld, van der Meer & De Vries, 2013).
Building on Allport's well-known intergroup contact theory, Karl Wolfgang Deutsch hypothesized already in 1957 that the more individuals are exposed to international
encounters, the more likely they are to support supranational governance and transnational solidary systems (Ciornei & Recchi, 2017). Assuming that such cross-border practices are consequently likely to increase support for transnational solidarity policies among EU
9 compared to another factor, cultural openness. Identification with the EU clearly fosters but doesn't prime solidarity, they conclude.
Egalitarian values
According to empirical results obtained by Ciornei and Recchi (2017), the extent to which citizens are willing to share social resources in a welfare system is largely explained through their political attitudes, namely how the relationship between individual, market and society is perceived in a national context. More specifically, it is the commitment to egalitarianism in society which, alongside the just mentioned cultural openness, causes feelings of
transnational solidarity. Egalitarianism also conditions how transnational experiences are perceived, more than rational or identity-related factors. While egalitarianism does not equate to solidarity as such, it can be conceptualised as the conviction that individual freedom and equality in all their ramifications are desirable social conditions as compared to a more self-enhancing mindset (Ciornei & Recchi, 2017). In his comparison of welfare policy support across the EU, Gelissen (2000) draws the same conclusion: despite the level of dependence on the welfare system, it is attitudes on equality and commitment to the common good which ultimately condition transnational social solidarity.
Frames, variables and hypotheses
By venturing into the unknown field of frame effects on EU social solidarity, some valuable insights can be borrowed from the extensive research on opinions about EU attitudes, such as support for EU integration, political trust and EU scepticism. One of them is the choice of frames, namely identity (ID frame) and cost/benefit (CB frame). Frames of economic
10 dominate news reporting about the EU. This indicates - and in turn provokes - that
considerations of costs and benefit take on a central role in the public reflection of EU issues (Hooghe & Marks, 2005; Van Klingeren, Boomgaarden & de Vreese, 2013). Identity-related frames are more present in national media coverage, however some scholars have found them to have strong framing effects on EU matters such as Euroscepticim (Van Klingeren,
Boomgaarden & de Vreese, 2013). Although neither identity nor cost/benefit-related considerations have so far been identified as actual source of solidarity, previous studies at least suggest that they are crucial multipliers of solidarity (Gelissen, 2000; Kankaraš & Moors, 2009). Egalitarianism and cultural openness, on the other hand, might indeed initiate social solidarity (Ciornei & Recchi 2017). They are chosen as moderators because the individual person’s commitment to these values presumably affects the interplay between frame and frame effects. Along with this, political sophistication is also included as possible moderator (see hypotheses).
When measuring the degree of solidarity after participants have been exposed to the stimuli, the measured outcomes is comparable between the two framing groups. However, an
individual might have had very high solidarity levels even before the experiment. It is therefore necessary to determine the relationship between frame exposure and frame effects. In framing theory, frames have been proven to mediate effects through two different
psychological processes: by adding new information and beliefs into an individual's available set of knowledge (belief content) and by "altering the weight of particular considerations” in the individual’s mind, referred to as belief importance (Lecheler & de Vreese, 2012, p. 187; Nelson, Clawson & Oxley, 1997). The experiment therefore also controls for belief
importance as a mediator of solidarity effects between the frames in comparison to each other.
11
Hypotheses
The first hypothesis refers to the expected effects produced by the identity versus the cost/benefit frame. Previous studies conducted on EU attitudes have come to conflicting results when comparing rational to identity factors: researching trust, Harteveld, van der Meer, De Vries (2013, p. 561) found that: "one’s identity is at best weakly related to trust in the EU: this relationship is not robust. [...] It is more calculus than sentiment that drives citizens’ trust". On the other hand, research by Van Klingeren, Boomgaarden and de Vreese (2013) has found national identity to play a significantly role in predicting Euroscepticim, more even than economic considerations. Similarly, in the support for EU integration of Turkey, identity turned out to be more important than utilitarian considerations (de Vreese, Boomgaarden & Semetko, 2011).
Klingemann and Weldon (2013), who have researched reasons for transnational dyadic trust see shared identity as superior to rational calculus because it encourages reciprocity and facilitates collective action and civic engagement. Similarly, Gerrits (2014, p.4) claims that financial assistance to other countries challenges rather than supports European solidarity due to a lack of mutual trust: “Community, identity and trust refer to a feeling of togetherness and similarity which is strong enough to feel a solidarity ‘that goes beyond mere short-term, instrumental calculations’". In line with this, Bechtel, Hainmueller & Margalit (2014) conclude that economic self-interest is not a major consideration by which Germans assess their support for Germany’s participation in Eurozone bailouts. In summary, a great share of research suggests that identity has a significant influence on EU attitudes as Euroscepticism, EU integration and dyadic trust. Following Gerrits (2014), the same can thus be assumed for solidarity, leading to the first hypothesis on frame effects:
12 H1: Exposure to the identity frame produces higher levels of social solidarity than exposure to the cost/ benefit frame.
While the feeling of a shared identity appears to be a reliable factor strengthening social solidarity, Harteveld, van der Meer and de Vries (2013) found trust to be surprisingly
detached from identity or rational factors. They see trust as largely unrelated to the EU per se and to be generated through individual predispositions instead, like a "trust syndrome”. With regard to Ciornei and Recchi's (2017) findings about the role of egalitarian values for
solidarity as well as Gelissen's claim that it is mainly the commitment to the common good which determines levels of solidarity, there is reason to assume that solidarity, too, is more of a value-based “syndrome”. Frame effects are therefore expected to be moderated by the individual degree of egalitarian values:
H2: Egalitarian values moderate the frame effects on social solidarity.
Alongside the role of egalitarianism, Ciornei and Recchi (2017, p.15) claim that "cultural openness can generate moral commitments among Europeans even more so than
identification with Europe or cross-border practices". They conceptualise this notion as a cosmopolitan outlook and open attitude towards globalization and immigration as opposed to the preference of cultural closure. Cultural openness significantly influences how
international experiences are perceived and is believed to be the strongest predictor for support of international mechanisms of redistribution by Bechtel, Hainmueller and Margalit, (2014). Therefore, cultural openness is also expected to moderate news frame effects.
H3: Cultural openness moderates the frame effects on social solidarity.
Another factor that has repeatedly been found to moderate EU attitudes is political sophistication, defined as the cognitive ability, engagement and thus knowledge of an individual in terms of political issues (de Vreese, Boomgaarden & Semetko, 2011).
13 Examining news frame effects on government confidence, Valentino, Beckmann and Buhr (2001, p.365) found that “the effects of the framing manipulation were powerfully and independently moderated by involvement and sophistication”. The same correlation was observed in Harteveld, van der Meer and de Vries’ (2013) research on trust, where political knowledge positively moderated the influence of rational considerations. The moderating effect of political sophistication on EU attitudes may therefore apply to solidarity, too:
H4: Political sophistication moderates the frame effects on social solidarity.
In order to measure whether the considerations that participants find important in terms of EU social solidarity are actually driven by the frames, a final hypothesis is to be tested:
H5: Belief importance mediates the frame effects so that participants evaluate social solidarity by prioritising the elements highlighted in the frame they were exposed to.
Data, operationalisations and method
Experiment design
The experiment was conducted online using a survey platform that can be assessed through a simple link. Before starting the experiment, participants had to agree to a statement of
informed consent which insured that no collected data were handed to a third party. They furthermore needed to confirm to be over 18 years of age and were informed that, by leaving their e-mail address, they would automatically participate in a draw for a 10€ voucher. Then, participants were required to fill out a pre-test which measured their scores on the three moderators egalitarianism, cultural openness and political sophistication. For the stimulus exposure, participants were randomly assigned to one of the two videos, followed by the measurement of solidarity. Finally, participants were debriefed and informed that the news
14 piece they had just seen was artificially created and has no news value whatsoever.
Sample
Because the cost/benefit frame used in this experiment needed to emphasise the possible consequences one defined country faces by supporting EU solidarity policies, the stimulus could only address people of the same nationality. Therefore, only German nationals were recruited using social media (N= 73, Mfemale = .59, SD= 0.5, Mage= 28.14, SD= 8.84). The sample does not claim to be representative for solidarity levels among the broad, German population. It is noteworthy that differences concerning the nationality of the sample could be expected – Germany as an economically wealthy and politically powerful EU state might have a population more prone to EU policy support than countries on the European periphery. As research by Bechtel, Hainmueller and Margalit (2014) and Gelissen (2000) has revealed, commitment to European welfare policies differs indeed between states.
Stimulus creation
As it has been done in similar experiments on frame effects, a fictional news piece was created in two versions. The topic was a fictional budget raise of the Erasmus + programme for European students which increases EU spendings about 18%. A high resemblance to real television news was ensured by using video material from the EU Commission's audio-visual platform "Europe by Satellite", the primary video source for television broadcastes. A typical news piece was thus constructed as it would be seen in public broadcasting news, lasting about 90 seconds. A voice-over read the information, the text was the only element that underwent manipulation. Differences between the identity (ID) and the cost/benefit (CB) frame were kept minimal to preserve internal validity (de Vreese, 2003; Dumitrescu, Gidengil & Stolle 2014). Apart from an additional mention of the 18% raise at the beginning of the CB
15 frame, only the last paragraph was altered. While both frames name that costs involved for Germany, the CB frame emphasises that Germany is also the country contributing most money to the programme and how much the German job market gains from this investment. The ID frame on the other hand ends with a fictional quote by the EU Commissioner for Education, Tibor Navracsics, who praises the importance of Erasmus + to foster shared values within the EU.
Variables and measurements
The dependent variable social solidarity is measured with four five-point Likert scales asking to what degree participants agree to the respective statement (Cronbach’s α = .79). The items are leaning on Eurobarometer questions that Risse (2013) categorises as related to solidarity, for example the question on freedom of movement in the EU. The three other items indicate a general willingness to participate in costly EU policies for investment and debt relief, namely investment into Erasmus +, financial aid for member states and the introduction of a
European Fund for Strategic Investment. Similar items have been used by de Vreese (2003), who measures support for EU integration and by Van Klingeren, Boomgaarden and de Vreese (2013) to examine support of the EU accession of Turkey.1
Egalitarianism, supposedly a moderating variable (H2), is measured on a Likert scale as employed by Ciornei and Recchi. It measures the respondents’ degree of consent to prioritise equality and justice over freedom. "The question captures the trade-off between individual freedom and social equality and therefore is in line with previous arguments regarding the structuring of political orientations along the individualism vs. social equality principle" (Ciornei & Recchi, 2017, p. 6). Cultural openness (H3) is recorded using three scaled items
16 taken from the same study (Cronbach’s α = .71). The scales ask about attitudes towards globalisation and immigration and are then recoded to form an individual score of cultural openness ranging from 1 to 5.
The third moderator variable, political sophistication (H4), is measured with two factual knowledge questions about the EU as well as a five point Likert scale on interest in politics (Cronbach’s α = .55) as done by Harteveld, van der Meer and De Vries (2013). Knowledge is recoded into two dummy variables to which the value of interest in politics is added. Very similar to the formula of political sophistication used by de Vreese, Boomgaarden & Semetko (2011), the self-reported interest in politics is weighted less than the two knowledge questions as “interest” is a rather subjective concept2.
Finally, after stimulus exposure, the importance of solidarity-related considerations has to be measured in order to test the belief importance hypothesis (H5). As done by de Vreese, Boomgaarden and Semetko (2011) with regard to the Turkey accession, respondents are asked to rate the importance of four considerations for them personally when thinking about the issue reported in the video. Two considerations relate to each identity or cost/benefit factors respectively. Values were then recoded into two scores per participant, indicating how high the person ranked ID and CB considerations respectively (Nelson, Clawson, Oxley, 1997).3
2 The two knowledge questions are counted either 0 or 1 (correct answer) whereas the interest in politics is multiplied by 0.1, creating a scale for individual scores from .01 to 2.5 as the highest. For example, two correctly answered questions and an indicated political interest of 4 would result in a score of 2,4. De Vreese, Boomgaarden & Semetko (2011, p.202) justify the weighting of the items as follows: “We weighed the index since we believe that knowing a question more should be more strongly represented than moving up one point on political interest.”
3 Scores were calculated by counting the highest ranked consideration as 4, the second as 3 etc; afterwards the values for each identity or cost/benefit were summed up. A ranking of 1) identity; 2) identity; 3) cost/benefit; 4) cost/benefit therefore gives a score of identity belief importance 4+3=7 and cost/ benefit belief importance 2+1=3 on a scale from 3 to 7.
17
Analysis
In order to test the first hypothesis, an independent samples t-test is conducted to compare the means of solidarity among the two framing groups. Taking the moderator variables into consideration for H2, H3 and H4, separate ANCOVAS are run for the frames, as done by de Vreese (2003) and de Vreese, Boomgaarden and Semetko (2011) in testing frame effects on political cynicism. By testing H5, t-tests are carried out to compare the means of ID and CB belief importance between and within the two frames in order to assess whether the framing condition has had a significant influence on the thoughts people prioritise when considering solidary support for Erasmus +. In order to measure whether belief importance actually interacts as a mediator between frame and solidarity, two OLS Process regressions are run, each including the overall rating score of identity and cost/ benefit considerations
respectively as mediating variable. Spss’ Process regression uses bootstrapping procedures to provide a model of conditional indirect effects out of which a path model of mediation is then drawn (Otto, Maier, 2016; Lecheler, de Vreese, 2012).
Results
Before testing the hypotheses which were formulated under the research question – how the ID and CB news frames affect levels of social solidarity among European – some descriptive statistics have been conducted. Looking at the overall distribution of solidarity in the full sample (N=73; n ID frame= 39; n CB frame= 34), it appears that solidarity is strongly positively
skewed (M= 3.9, SD= .684, see Figure 2), which a Shapiro Wilk test confirms (p= .008). Nevertheless, Levene’s test for homogeneity of variances shows an equal distribution of solidarity among both framing conditions (p= .86), an important pre-condition for
18 significance testing. In terms of gender, solidarity levels between women (n= 43, M= 3.94, SD= .67) and men (n= 30, M= 3.83, SD= .71) are similarly high (t (71)= -.664, p = .509).
Contrary to the assumption that the ID frame produces higher levels of social solidarity as compared to the CB frame, no significant difference in solidarity levels was found and the hypothesis is rejected. Participants in the ID framing condition revealed solidary attitudes (M= 3.85, SD= .65) similarly high as those in the CB condition (M=3.96, SD= .73, p= .498). In order to test whether egalitarian values (M= 3.37, SD= .97), cultural openness (M= 3.33, SD= .53) or political sophistication (M= 1.86, SD= .71) function as moderators on the effects of news the frames, three one-way ANCOVAS were run. None of them revealed a significant influence on solidarity effects and the partial Eta Squared values are minimal. The
moderation effects were equally insignificant when being included into one ANCOVA controlling for the three covariates at the same time. Due to the lack of significant framing effects, post-hoc were not carried out. Since H2, H3 and H4 suggested to find moderating effects of these variables, they are equally rejected. Table 1 summarises the values for each moderator from the separately run ANCOVAS. The reason why ANCOVAS did not measure any significant effects might be partially explained because the residuals are not normally
19 distributed, as the Shapiro Wilk statistic has shown. A larger sample size might thus produce different values.
For the belief importance hypothesis (H5), the importance that participants gave to each ID and CB-related considerations, by ranking them in a hierarchy, has to be compared. It is assumed that those exposed to the ID frame would rank identity considerations higher than CB considerations. Therefore, the ranking scorers for both consideration types are compared within the whole sample as well as within each frame group.
Overall, identity considerations clearly dominate (M= 5.77, SD= 1.34, (t (72) = 36,6, p= >.001) among participants as compared to cost/benefit considerations (M= 4.23, SD= 1.33, (t (72) = 27.02, p= > .001) with a considerable mean difference of 1.54 units as shown in Figure 3. When first comparing frame effects on identity considerations across the two frames, identity considerations in the ID frame (M= 6.05, SD= 1.17) do not differ significantly from identity considerations in the CB frame (M= 5.44, SD= 1.46, t (71) = -1.981, p= .051). For cost/benefit belief importance, the same significance value (t (71) = 1.981, p= .051) appears when comparing the ID frame (M= 3.95, SD= 1.17) to the CB frame (M= 4.56, SD+ 1.46).
20 Comparing the belief importance means within each frame also shows clear differences in consideration preference: as expected by H5, identity considerations outweigh cost/benefit considerations in the ID frame (MID= 6.05, SD= 1.17; MCB= 3.95, SD= 1.17). Surprisingly, identity considerations also dominate within the CB framing group (MCB= 4.56, SD= 1.46; MID= 5.44, SD= 1.46), although cost/benefit considerations are, as expected, higher than in the ID frame4. The fact that belief importance scores significantly differs within each frame lends
partial support to H5, according to which each frame should induce thoughts related to the respective frame. In order to test whether belief importance really acts as a mediator on solidarity levels, an OLS Process regression is carried out.
As done by Lecheler & de Vreese (2012) as well as Nelson, Clawson & Oxley (1997), the regression coefficients are used to construct a path model (see Figure 4). The regression controlling for the overall score of identity belief importance as a mediator depicts of course the same relation between frame and ID belief importance as the initially conducted t-test (b= .61, t (71)= 1.98, p= .051). The b path however, between mediator and dependent variable,
21 reveals a significant effect of belief importance on solidarity levels (b= .26, t (70) = 4.67, p < .001). The total measured effect (c) of the model is insignificant again (b= - .1, t (71) = - .68, p= .49) and so is the direct effect of framing on solidarity when adjusting for the indirect mediator effect (c’) (b= - .27, t (70)= - 1.82, p= .07). Since the indirect effect of the model has a confidence interval that does not include zero (b= .16; 95% CI [ .04 - .33]), a significant mediation effect of belief importance on solidarity can be attested, thus supporting H5. This means that belief importance functions as an indirect mediator on solidarity.
The model looks strikingly similar when the overall score of cost/benefit considerations is tested for mediation effects, showing the same highly significant relation between the mediator and dependent variable on the b path (b= - .26, t (70)= 4.67, p < .001). In the cost/benefit model, CB belief importance also turns out to have a strong mediating effect on solidarity as the indirect effect does not contain zero (b= .16; 90% CI [ .05 - .33]). As assumed in H5, the model shows that the framing process was significantly mediated by belief importance when thinking about solidarity issues.
22
Discussion and conclusions
Contrary to the initial assumption that solidarity levels among an audience would differ depending on the news frame (H1), both experimental groups have displayed elevated levels of solidarity (M= 3.9, SD= .684) with almost no difference between the framing conditions. This already constitutes a counter argument to Gerrits (2014), who doubts solidarity can exist on the EU macro level because of a lack of essential ingredients, such as trust and a feeling of a shared identity. While the skewed distribution of solidarity might be specific to this sample and thus lack external validity, the question remains why no difference was yielded by the frames within the sample. De Vreese (2003), who has not found significant framing effects on support for EU enlargement, concludes that this might be due to the personal non-
obtrusiveness and rather uncontroversial nature of EU policy issues. Choosing a more emotion-loaded frame topic would, however, have sharply increased the risk of having unidentified z-variables influencing the outcome. Generally in framing effect research,
scholars have found mediated framing effects to vary significantly across topics, depending on the importance an issue has on the media agenda or for the individual person (Lecheler & de Vreese, 2012).
One possible reason why both framing groups displayed the same solidarity levels might be that participants simply followed a self-interest logic: it is well possible that within the
comparatively young sample (Mage= 28.14, SD= 8.84) participants (e.g. because they might be students or at the beginning of their professional career) are more dependent on state support like the Erasmus stipend than older participants would be (Gelissen, 2000). Logically, general support for an increase of financial aid for young people could be expected for both framing groups. It is furthermore known from research on frame effects that negative information tends to be more salient and memorable in people’s mind. Since both frames in this study have
23 a positive valence, any strong effects may have been extenuated (de Vreese, Boomgaarden & Semetko, 2011).
The fact that the level of egalitarian values and cultural openness have not shown to be significant moderators of frame effects contradicts results found by Ciornei and Recchi (2017), who claim that equality and cultural openness are the pillars of transnational solidarity. Similarly, ()) have found cultural openness to be a significant covariate in analyzing support for policies of international redistribution among the German public. The specific role of egalitarian values for solidarity is quite nuanced in Ciornei and Recchi's (2017) explanations: they found egalitarian values to be mediated by the amount of
international experiences of the individual person, which made people in turn prioritize more specific policy arrangements. Linking this to Bechtel, Hainmueller and Margalit’s (2014) results, it is thinkable that the young German sample is guided by a relatively high cultural openness - being from a wealthy country that is massively benefitting from globalization, most participants have probably enjoyed a great share of intercultural experiences. In addition to the already discussed self-interest logic, both framing groups are therefore likely to be culturally open to any kind of common European project as Erasmus. The sample’s strikingly high mean score of cultural openness supports this explanation (M= 3.33, SD= .53 on a five-point scale) in relative comparison to the mean Ciornei and Recchi (2017) obtained with an older sample, using the same items of measurement.5
In terms of political sophistication, the lack of significant influence is in line with numerous previous studies on framing effects. Some researchers assume that the more politically sophisticated are also more susceptible to framing effects, others expect the opposite (de
5 Cultural openness in the sample of Ciornei & Recchi: M= 1.89, SD= .63 (on a scale ranging 0.44-3.02) with
24 Vreese, Boomgaarden & Semetko, 2011; Valentino, Beckmann & Buhr, 2001). This result indicates that fundamental and personal dispositions such as solidarity might not be much affected by knowledge about EU politics after all, as Harteveld, van der Meer and De Vries (2013) also conclude with regard to trust.
Although the two news frames have not yielded any differences in solidarity directly, testing for belief importance as a mediator has revealed a significant indirect effect, particularly on the b-path between belief importance and solidarity levels. This result is in line with research conducted on mediated frame effects on policy support as well as political trust (de Vreese, Boomgaarden & Semetko, 2011; Nelson, Clawson & Oxley, 1997; Otto & Maier, 2016). This mediation effects can be interpreted so that news frames can indeed prime specific
considerations that are available to a person and thus influence the judgment process of an issue – even if no direct relationship between frame and solidarity has been measured.
However, the belief importance process may be more complex than making “top-of-the-head” judgments and involve a deliberate integration of considerations and how they are weighted in decision-making (Nelson, Clawson and Oxley, 1997, p. 578). The reason for this is that frames may supposedly strengthen the emphasis on specific considerations, however they probably do not alter the relative order of consideration preferences as they are set in a person’s mind. Wiht regard to the results of this study, it seems as if identity considerations are dominant in most people’s judgements. Contrary to the rational choice theory, according to which participants in the CB frame should have justified their support based on
cost/benefit reasons, more participants still prioritised identity considerations (MCB= 4.56, SD= 1.46; MID= 5.44, SD= 1.46). Although cost/benefit belief importance was relatively raised in this frame, the result suggests that considerations are not prioritised merely
25 issue, hence thinking in terms of identity and less about costs or financial benefit. Thus, identification with the EU or its citizens may not cause solidarity per se, instead it functions as a multiplier of individual values and considerations regarding solidarity, a conclusion also drawn by Ciornei and Recchi (2017).
These underlying values behind consideration preference can be assumed to be quite stable, according to Nelson, Clawson and Oxley (1997). By measuring tolerance for KKK rallies, they discovered that the general value of free speech largely resisted their framing conditions. News frames may thus alter the perceived importance of considerations and therefore create support for a specific issue, nevertheless they barely touch on profound values. This applies as long as no other, competing value comes into play, they conclude. For this study, this means that many participants might have been indirectly influenced to judge the Erasmus investment in terms of cost and benefit (although identity considerations remained dominant), thus showing support for this specific matter, but the frames most likely have not affected
individual levels of solidarity. This would presumably only happen with news stories in which another profound value overshadows the values causing solidarity, e.g. news of the
accommodation of refugees in local communities, where xenophobia could actually suppress solidarity.
All points concluded, this experiment has offered a few insights into European solidarity: It seems as if news frames have no direct effect on solidarity, which turns out to be a robust attitude. Instead, news frames can prioritise certain considerations through indirect mediation, thus influencing the perception of a news piece. It appears that a majority of people rely on identity-based consideration when facing solidarity issues, a phenomenon which can be expected to be found also in samples with a less skewed distribution of solidarity. The role of belief importance for solidarity may of course go either way – news with a positive valence,
26 emphasising a shared identity or profit, could generate support for specific solidary policies without making a person more solidary in general. On the other hand, a negative frame might also decrease such support, e.g. by focussing on potential financial losses or by invoking in-group feelings and a threat from “the others”. Which considerations dominate and at what point support or rejection of European solidarity policies is created is certainly highly dependent on the issue and a person’s deeply engrained values. Although no effects of egalitarianism and cultural openness have been measured in this study, they remain the primary suspects to be the values causing European social solidarity (Bechtel, Hainmueller & Margalit, 2014; Cionei & Recchi, 2017; Gelissen, 2000). Further research in this field should therefore focus on the potential sources and moderators of European solidarity, which could also include the degree of identification with the EU, feelings of moral obligation to the common good or trust in EU institutions (Ferrera, 2014; Gelissen, 2000; Gerrits, 2014; Mau & Burkhardt, 2009).
When interpreting the results of this study, its limitations must be kept in mind. In terms of experimental set-up, only a small sample was used which displayed a skewed distribution of solidarity, which might in return have affected some analyses and decreases external validity. Furthermore, generalisations about frame effects on solidarity have to be handled cautiously because only one issue and two frames were tested, therefore reactions to different frames are uncertain. The general robustness of solidarity against news framing can still be assumed though. To what degree the results from other framing effect studies can be compared to this experiment remains uncertain, as the majority has used newspaper articles as stimuli; only de Vreese (2003) and Nelson, Clawson and Oxley (1997) have also employed television news for their framing purposes. Finally, a much more faceted picture of European solidarity could be obtained by including a wider range of moderators, e.g. education, political leaning, income,
27 post-materialism or international experiences (Ciornei & Recchi, 2017; Gelissen, 2000). Despite a lack of external validity, this experiment is the first conducted on solidarity on the macro EU-level, offering new insights into the role of news framing in creating or
undermining solidarity. Considering the current development of nationalism and populist movements in European politics, this knowledge might be of use to all journalists covering news about the EU.
References
60 Jahre Römische Verträge: „Es wird auch einen 100. Geburtstag der EU geben (2017, March 25th), Frankfurter Allgemeine Zeitung. Retrieved May 30th, 2017, from http://www.faz.net/aktuell/politik/ausland/eu-jubilaeumsgipfel-in-rom-hat-begonnen-14941892.html
Abela, A. (2004) ‘Solidarity and Religion in the European Union: A Comparative Sociological Perspective’, in P. Xuereb (ed.) The Value(s) of a Constitution for Europe, pp. 71–101. Malta: European Documentation and Research Centre, University of Malta.
Bechtel, M. M., Hainmueller, J., & Margalit, Y. (2014). Preferences for International
Redistribution: The Divide over the Eurozone Bailouts. American Journal of Political Science, 58(4), 835-856. doi:10.1111/ajps.12079
Bremer, J., Stabenow, M., & Schuller, K. (2017, March 24). Der Papst und die EU:
28 http://www.faz.net/aktuell/politik/ausland/papst-macht-regierungschefs-mut-fuer-neuanfang-14941032.html
Ciornei, I., & Recchi, E. (2017). At the Source of European Solidarity: Assessing the Effects of Cross-border Practices and Political Attitudes. JCMS: Journal of Common Market Studies, 55(3), 468-485. doi:10.1111/jcms.12507
De Vreese, C. (2003). Framing Europe: Television news and European integration. Amsterdam: Aksant.
De Vreese, C. H., Boomgaarden, H. G., & Semetko, H. A. (2010). (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.
doi:10.1177/0093650210384934
Dumitrescu, Delia, Gidengil, Elisabeth, & Stolle, Dietlind. (2015). Candidate Confidence and Electoral Appeal: An Experimental Study of the Effect of Nonverbal Confidence on Voter Evaluation. Political Science Research And Methods,3(1), 43-52.
Ferrera, M. (2014). Solidarity in Europe after the Crisis. Constellations, 21(2), 222-238. doi:10.1111/1467-8675.12091
Gelissen, J. (2000). Popular support for institutionalised solidarity: a comparison between European welfare states. International Journal of Social Welfare, 9(4), 285-300. doi:10.1111/1468-2397.00140
Gerrits A.W.M. (2014), Solidarity and the European Union: From the Welfare State to the Eurocrisis. In: Hillebrand E., Kellner A.M. (Eds.) Shaping a Different Europe. Contributions to a Critical Debate.. Berlin: Dietz Verlag. 63-71.
Habermas, Jurgen. (2001). Why Europe needs a constitution. New Left Review, (11), 5. Harteveld, E., Meer, T., & Vries, C. (2013). In Europe we trust? Exploring three logics of
trust in the European Union. European Union Politics, 14(4), 542-565.
Hooghe, L., & Marks, G. (2005). Calculation, Community and Cues: Public Opinion on European Integration. European Union Politics, 6(4), 419-443.
Kandyla, A.A. & De Vreese, H. (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.
Kankaraš, M., & Moors, G. (2009). Measurement Equivalence in Solidarity Attitudes in Europe. International Sociology, 24(4), 557-579. doi:10.1177/0268580909334502 Karagiannis, N. (2007). Solidarity wthin Europe/ Solidarity without Europe. European
Societies, 9(1), 3-21.
Klingemann, H., & Weldon, S. (2013). A crisis of integration? The development of
transnational dyadic trust in the European Union, 1954–2004. European Journal of Political Research, 52(4), 457-482.
29 Lecheler, S., & De Vreese, C. (2012). News Framing and Public Opinion. Journalism & Mass
Communication Quarterly,89(2), 185-204.
Mau, S., & Burkhardt, C. (2009). Migration and welfare state solidarity in Western Europe. Journal of European Social Policy, 19(3), 213-229.
doi:10.1177/0958928709104737
Nelson, T. E., Clawson, R. A., & Oxley, Z. M. (1997). Media framing of a civil liberties conflict and its effect on tolerance. American Political Science Review, 91(3), 567-583.
Otto, L., & Maier, M. (2016). Mediated and moderated effects of personalized political communication on political trust. Communications, 41(1), 21-45.
Risse, T. (2013). Solidarität unter Fremden? Europäische Identität im Härtetest. KFG Working Paper Series, No. 50, May 2013, Kolleg-Forschergruppe (KFG) “The Transformative Power of Europe“ Freie Universität Berlin.
Valentino, N., Beckmann, M., & Buhr, T. (2001). A Spiral of Cynicism for Some: The Contingent Effects of Campaign News Frames on Participation and Confidence in Government. Political Communication, 18(4), 347-367.
Van Klingeren, M., Boomgaarden, H., & De Vreese, C. (2013). Going Soft or Staying Soft: Have Identity Factors Become More Important Than Economic Rationale when Explaining Euroscepticism? Journal of European Integration, 35(6), 689-704. Witte, F. (2012). Transnational Solidarity and the Mediation of Conflicts of Justice in
Europe. European Law Journal, 18(5), 694-710. doi:10.1111/j.1468-0386.2012.00624.x
Appendix
I Online survey and video stimulus
Dear participant,
Thank you for participating in this study on public opinions regarding the Erasmus + programme, conducted for a Master thesis within the University of Amsterdam.
Your help is very much appreciated.
In the following, you are first going to be asked to answer a few questions relating to your personal attitudes and political interest. Then, you will be asked to watch a video of about one minute, please listen carefully to the content. Afterwards, some final questions about the Erasmus + programme will follow. The survey contains eleven questions all in all and will take about 5-10 minutes to complete.
As a motivation for participants to take part in the survey, a reward in the form of a 10€ Amazon voucher can be won. Taking part in the raffle is only possible after the completion of the entire survey and by giving your e-mail address on the last page. This information will not be given to any third party and only be used to contact the winner of the raffle.
Please note:
As this research is being carried out under the responsibility of the ASCoR, University of Amsterdam, we can guarantee that:
1) Your anonymity will be safeguarded, and your personal information will not be passed on to third parties under any conditions, unless you first give your permission for
this.
2) You can refuse to participate in the research or cut short your participation without having to give a reason for doing so. You also have up to 24 hours after participating to withdraw your permission to allow your answers or data to be used in the research.
3) Participating in the research will not entail your being subjected to any appreciable risk or discomfort and you will not be exposed to any explicitly offensive material.
4) No later than five months after the conclusion of the research, we will be able to provide you with a research report that explains the general results of the research.
For more information about the research, you are welcome to contact me at any time via e-mail: florence.schulz@student.uva.nl
Should you have any complaints or comments about the course of the research and the procedures it involves as a consequence of your participation in this research, you can contact the designated member of the Ethics Committee representing ASCoR, at the following address: ASCoR Secretariat, Ethics Committee, University of Amsterdam, Postbus 15793, 1001 NG Amsterdam; 020‐525 3680; ascor‐secr‐fmg@uva.nl. Any complaints or comments will be treated in the strictest confidence.
On the following page, please agree to the declaration of informed consent before beginning the survey. Thank you very much!
Kind regards, Florence Schulz
Declaration of informed consent
I hereby declare that I have been informed in a clear manner about the nature and method of the research, as described in the email invitation for this study.
I am aware that the minimum age for participation is 18 years and that I may not take part in the survey if I am underage. Furthermore, I am informed that in order to participate, German nationality is required. As I fulfil these conditions, I agree, fully and voluntarily, to take part in this research study.
With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the survey at any time. If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely safeguarded. My personal data will not be passed on to third parties without my express permission.
If I wish to receive more information about the research, either now or in future, I can contact Florence Schulz at florence.schulz@student.uva.nl. Should I have any complaints about this research, I can contact the designated member of the Ethics Committee representing the ASCoR, at the following address: ASCoR secretariat, Ethics Committee, University of Amsterdam, Postbus 15793, 1001 NG Amsterdam; 020‐525 3680; ascor‐secr‐fmg@uva.nl.
I understand the text presented above, and I agree to participate in the research study.
Debriefing
Thank you for participating in this study!
Please not the following – the intention of this study was not, as you have been told in the beginning, to test participants’ opinion about the Erasmus + programme. Instead, the actual focus was on a broader concept of which Erasmus + is a mere example: this study measures the levels of social solidarity that are produced among EU citizens when exposed to certain news frames. In this case, you have either seen a video framing the extension of the Erasmus + budget as a “cost and benefit” or as an “EU identity” -related issue. The news report you have seen is merely fictional and was constructed for this survey – the
EU Commission has not announced any raise of the Erasmus + budget. The quotation from EU
Commissioner Tibor Navracsics was taken out of a different context and the raise of 18% of the budget is not true. However, the Erasmus + budget until 2020 is indeed 14.8 bn Euros.
This procedure of misinforming you about the intention of the study is a standard procedure in this research field. It is necessary because if you knew beforehand that the research is about European solidarity, it could affect your behaviour and attitudes when answering the questionnaire.
Please also note that everything else you were told at the outset remains true:
Your responses will be confidential and no identifying information such as your age, nationality or email address will be collected. Also, your personal information will not be shared with any third party. To help protect your confidentiality, all the data will be analysed collectively. Furthermore the reward, a 10€ Amazon voucher, will indeed be given out in a raffle.
Please do not reveal any information about this study to anyone else. Do you have any questions or comments?
Did you talk to anyone beforehand who had participated in the study?
If so, did you hear from them about what they watched or answered questions about?
Dear participant, thank you again for taking part in this study!
If you would like more information about your participation or about the study itself, please do not hesitate to contact me: florence.schulz@student.uva.nl.
I have read and understood the above information. Kind regards,
Florence Schulz
If you want to participate in the raffle for a 10€ Amazon voucher, please type in your e-mail address below – it will be used only to contact the winner and neither stored, nor given out to a third party.
Stimulus voice-over (frame manipulations in bold face) Voice-over ID frame
For the 30-year anniversary of the Erasmus + programme for European students, the EU Commission intends to raise the programme’s budget until 2020.
The step was announced last week at a meeting of the EU Commission for the creation of an EU solidarity corps and has now been decided: As Tibor Navracsics, EU Commissioner for education declared this week,
the budget for Erasmus + will be increased from 14.8 to 17.5 bn Euros. This corresponds to a raise of
18%, supplied by the common EU budget.
Since the beginning of the exchange programme in the 1980s, it has enabled about nine million EU citizens to study or do an internship in another EU country. The most popular destinations last year were Spain, Germany and the UK. Over 84.000 German students have received financial aid through Erasmus + in 2016.
The announced increase of the Erasmus + budget will raise German contributions to 92 mio Euros per year. EU Commissioner Tibor Navracsics stressed that the programme was one of the best examples of what Europe could do for its citizens. “The exchange of European students is essential to preserve our common values and our European identity as a solidary state union.” Erasmus + provides equal conditions for all EU students, Navracsics said.
Voice-over CB frame
For the 30-year anniversary of the Erasmus + programme for European students, the EU Commission intends an 18% raise of the programme’s budget until 2020.
The step was announced last week at a meeting of the EU Commission for the creation of an EU solidarity corps and has now been decided: As Tibor Navracsics, EU Commissioner for education declared this week, the budget for Erasmus + will be increased from 14.8 to 17.5 bn Euros, supplied by the common EU budget. Since the beginning of the exchange programme in the 1980s, it has enabled about nine million EU citizens to study or do an internship in another EU country. The most popular destinations last year were Spain, Germany and the UK. Over 84.000 German students have received financial aid through Erasmus + in 2016.
The announced increase of the Erasmus + budget will raise German contributions to 92 mio Euros per year. This makes Germany the main depositor for the exchange programme among all EU countries. However, as a study conducted by the German Economic Institute revealed in January, Erasmus + is quite a profitable investment for Germany: the influx of highly qualified graduates of the Erasmus programme contributes around 320 million Euros per year to the German job market, the institute found.
II Variable measurements and scale reliabilities
Factor items
Dependent variable solidarity
On the following scales, please indicate to what extent you support the following statement:
- „Germany should invest more money into a common European education network as Erasmus +, even if it is not just German students benefitting from it.”
- “All European citizens should have the right to live, work and study in any EU country of their choice.”
- “I support the introduction of a common European Fund for Strategic Investments which will give loans to private businesses all across Europe to strengthen investment in the EU.”
- “In times of crisis, it is desirable for Germany to give financial help to another EU member state facing severe economic and financial difficulties.”
1 (not at all) - 5 (highly agree)
Moderator egalitarianism
On the following scale, please indicate to what extent you agree to the following statement:
„We need more equality and justice in the EU even if this means less freedom or wealth for the individual.“ 1 (not at all) - 5 (highly agree)
Moderator cultural openness
On the following scale, please indicate to what extent you agree to the following statements:
1. “Globalization is primarily an opportunity for economic growth.” 2. “Globalization means a threat to small, private businesses.” 3. “Germany benefits a lot from its immigrants.”
1 (not at all) - 5 (highly agree)
Moderator political sophistication
Please tick the right answer to the two following questions according to your best knowledge - and without further research:
1. The president of the European Commission is:
a) Jean-Claude Juncker; b) Antonio Tajani: c) Donald Tusk; d) I don’t know 2. How many member states is the EU comprised of:
a) 17; b) 52; c) 28; d) I don’t know
On the following scale, please indicate to what extent you are interested in politics. 1 (not interested at all) - 5 (highly interested)
Mediator belief importance
Please rank the following sentences in the order of personal importance to you (the first one being the most important one):
Germany should invest more money into the Erasmus + programme because… a) Of the economic profit for the German job market.
b) Of our common European values which we foster through education and exchange.
c) We are all European. Therefore we should support and welcome European students into Germany if they want to study here.
d) A higher Erasmus + budget also means German students receive more financial help.
Factor loadings and scale reliability for variables > 2 items Factor Total Eigenvalue Explained variance in % Cronbach's α Solidarity 2,03 50,81 .79 Cultural openness 1,345 44,833 .71 Political sophistication 1,494 49,817 . 55
III Statistical results
Descriptive statistics M SD Scale Female 0,59 0,50 0-1 Age 28,14 8,84 > 18 Solidarity 3,90 0,68 1-5 Moderators Egalitarianism 3,37 0,96 1-5 Cultural Openness 3,33 0,53 1-5 Political Sophistication 1,86 0,71 0.1-2.5 Mediators ID belief importance 5,77 1,34 3-7 CB belief importance 4,23 1,34 3-8
Regression results
Scores of identity (ID) and cost/benefit (CB) belief importance within framing conditions (H5)
ID frame CB frame
M SD M SD
ID belief importance 6,05 1,17 5,44 1,46
Scores of identity (ID) and cost/benefit (CB) belief importance within, across and regardless of framing conditions
Process regression on belief importance
ID belief importance unstandardized
b-coefficient t p Lower level CI Upper level CI a path .6101 1.98 .0514 0.1 1.12 b path .26 4.67 < .001 0.16 0.35 c path (total effect X on Y) - .12 - 0.68 .49 - 0.38 0.16 c' path (direct effect X on Y) - .27 - 1.82 . 07 - 0.51 - 0.02 c-c' path (indirect effect X on Y) .156 - - 0.04 0.33
ID belief importance unstandardized
b-coefficient t p Lower level CI Upper level CI a path - .6101 -1.98 .0514 - 1.12 -0.1 b path - .26 - 4.67 < .001 - 0.35 - 0.16 c path (total effect X on Y) - .12 - 0.68 .49 - 0.38 0.16 c' path (direct effect X on Y) - .27 - 1.82 . 07 - 0.51 - 0.02 c-c' path (indirect effect X on Y) .156 - - 0.05 0.33
Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00
t ID belief importance df p t CB belief importance df p Across ID x CB frame -1,98 71 ,051 1,98 71 ,051 Within ID frame 32,34 38 < .001 21,10 38 < .001 Within CB frame 21,72 33 < .001 18,20 33 < .001
