Intruders in the Land of Hospitality;The dichotomy between victim vs threat in Greek news frames, the mediating role of emotions and the effects on attitudes towards refugees in Greece.

The dichotomy between victim vs threat in Greek news frames, the mediating role of emotions and the effects on attitudes towards refugees in Greece.

University of Amsterdam

Graduate School of Communication

MA Journalism Globalisation and Media, Erasmus Mundus Master’s Thesis

Triantafyllia Ismailai

Student ID: 12906182

Supervisor: Laura Jacobs Date of Completion: 05-06-2020

Abstract

This study aims to enrich the current bibliography on news framing effects pertaining to migration and the role of emotions as mediators in the framing effects process. It builds upon representations of refugees in Greek newspapers. It investigates whether certain approaches in the news, namely the dichotomy of victim vs threat frame that accompanies refugee news content in Greece evokes emotions and leads to attitudinal changes towards refugees. The study hypotheses that the victim frame evokes positive emotions towards refugees and results in positive attitudinal formations, while the threat frame evokes negative emotions and results in negative attitudinal adoptions towards refugees in Greece. I applied a survey experiment (N= 497) to verify whether contact with news content yields the hypothesised effects. The results indicate that news frames lead to significant emotional effects, with the victim frame resulting in more positive and the threat frame in more negative emotions. Still, negative emotions are not significantly correlated with attitudes towards refugees and do not mediate the process, although positive emotions lead to less negative refugee attitudes. Lastly, the role of political orientation as a moderator was examined. The analysis indicated insignificant interaction effects between political beliefs (left-wing/right-wing) and refugee attitudes.

Introduction

“We have fallen victim of a collective effort to undermine the country.”

With these words, the Greek Minister of Development reacted to a refugee boat accident off the coast of Paxos island which counted 12 victims in January (“Georgiadis for Refugees: We have been the victim of a collective effort to undermine the country”, 2020). His statement resurged a heated media debate around the government’s approach towards the hundreds of refugees in the country and is only one of multiple instances that the issue occupies the social and political agenda. The situation escalated even further a month later, when Recep Tayyip Erdogan– the president of neighbouring Turkey- “opened the doors to” thousands of migrants and refugees, allowing their passage in Greece and subsequently the European Union (Stevis-Gridneff & Gall, 2020). As of this very moment, the refugee situation continues to bring complications within the country, its relationship with Turkey and the European Union.

The current refugee crisis is labelled as the biggest humanitarian crisis in Europe since WWII (Wirth, 2016). It has erupted into a heavily debated issue of European politics, where societal and political implications are intertwined with managing an adequate crisis response and balancing the humanitarian issue dimensions.

As the influx of refugees into Western Europe is undergoing, scholars have attempted to facilitate our understanding of the situation. Media research suggests that news content and framing can cause individual emotional reactions to the readers (e.g., Druckman and McDermott, 2008; Gross and D’Ambrosio, 2004). Thus, this research aims to investigate the connection between refugee news frames in Greece and the elicitation of emotions to the Greek audience. This study poses RQ1: To what extent does the dichotomy of victim vs threat frame pertaining to

The Syrian refugee crisis has ascended into a global phenomenon affecting the lives of millions, both refugees and host societies. It is mediated and transmitted through media (Boomgaarden & Vliegenthart, 2009); therefore, it is critical to investigate the connection between attitudes towards refugees and news content. There is an abundance of news frames studies and their role on attitudes towards immigrants, refugees and asylum seekers (e.g. Arlt & Wolling 2016; Bos et al., 2016; Bruneau et al., 2018; Eberl et al., 2018; Jacobs, Boukes & Vliegenthart, 2018; Jacobs, & Van der Linden, 2018, Igartua, Toranzo & Fernández 2011). In our attempt to bring forth new empirical knowledge, we examine the factors that influence framing effects.

Scholars have underlined the mediating role of emotions within the framing context (e.g. Lecheler, Schuck & de Vreese, 2013; Lecheler, Bos & Vliegenthart, 2015). This study is

predominantly interested in whether the attitudes of Greeks are mediated by positive or negative emotions deriving from the Greek news content. It presents RQ2: To what extent does the

dichotomy of victim vs threat frame pertaining to refugees in Greek news content affect the Greek public’s attitudes towards refugees and what role do emotions play in the framing process?

Research encourages that political orientation should be considered when examining attitudes towards minorities (Plener et al., 2017; Semyonov et al., 2006). We propose that political beliefs acts as a moderator in the news framing effects process. We expect the victim frame to lead to more positive attitudes for left-wing people and the threat frame to more negative attitudes for right-wing people. This study poses RQ3: What role does political

We conducted an online experiment and exposed participants (N=497) to two refugee frames. Literature has suggested that minority groups are presented either as threats or as victims (e.g. Van Gorp, 2005; Cheng et al., 2014). Initially drawing from Van Gorp’s work (2005), our study constructed two frames; the victim and the threat and used them to form two articles. The latter includes elements of Van Gorp’s intruder frame but also enriches it with symbolic and realistic threats and tests them in an experimental environment, in comparison to Van Gorp’s study.

The exploration of how the effects of two traditionally connected with migration frames can be mediated by the emotional responses the article evokes contributes to the novelty of the study, which measures both positive and negative emotions in an experimental environment. Moreover, Greece is expected to provide useful empirical knowledge, since it received 9,641 refugees in the first five months of 2020 only (Operation Portal- UNHCR, 2020). The country inevitably became a point of entrance due to its proximity with Turkey, which stands between East and West, and its extended and unmonitored coastlines. Examining the Greek media sphere and society, which is in closer proximity to the crisis than many other Western European

countries, is expected to shed a brighter light on framing migration and the factors that influence it.

Theoretical Framework Defining refugees.

Before this paper proceeds, a distinction between “a refugee”, “an asylum seeker” and “a migrant” must be drawn, since they are often confused within the context of the current crisis. A “refugee” is someone who “owing to a well-founded fear of persecution for reasons of race, religion, nationality, membership of a particular social group or political opinions, is outside the

country of his or her nationality and is unable or, owing to such fear, is unwilling to avail himself or herself of the protection of that country” (Douglas, Centron & Spiegel, 2019, p. 2). Asylum seekers are similar to refugees. However, they “await a decision on the application for refugee status under relevant international and national instruments” (Douglas et al.). The term migrant refers “to persons moving within or between countries to improve their economic and social conditions” (Douglas et al.). Since Greek news rarely makes distinctions between them, we conceptualise refugees combining refugees and asylum seekers. According to figures by UNHCR, 69% of the sea arrivals in the Mediterranean in 2020 consist of Syrian and Afghani nationalities (Operation Portal- UNHCR, 2020), leading us to choose the term “refugee” for this research but also use studies connected to immigration, since all other nationalities of arrivals in Greece fit its context.

This study aims to investigate how the dichotomy of the victim/ threat frame affects emotionally and attitudinally Greek readers and how those emotions influence the attitudes towards the refugee group. To answer the research questions, a review of literature around out-group and in-out-group attitudes, framing and emotions follows.

Integrated Threat Theory.

In order to understand how news framing might affect Greek people’s attitudes, this research draws on literature relevant to the Integrated Threat Theory (ITT). ITT in its original version suggests that negative attitudes towards out-groups derive from four types of threats: realistic and symbolic threat, negative stereotype and intergroup anxiety (Stephan & Stephan, 2000). Stephan, Ybarra, & Morrison amended the original version, so it includes two threats: realistic and symbolic (Stephan, Ybarra, & Rios Morrison, 2015). Realistic group theory argues that negative attitudes towards outgroups relate to intergroup competitions and struggle for

scarce resources (Blalock, 1967; Blumer, 1958; Quillian, 1995). It consists of either perceived threats to people's physical safety or fears that out-groups may endanger the in-group's economic and political power. Symbolic threats derive from the perceived conflict between norms and values of the out-group compared to the in-group (Stephan, Diaz-Loving & Duran, 2000). They can be traced back to symbolic racism theory, which argues that intergroup bias results in conflicting norms, values and belief (Kinder and Sears, 1981).

Research using ITT has been diverse, but it would be advantageous to sample populations such as the Greek one, which is experiencing evident pressures, both economically, politically and socially, within the context of the refugee crisis and as well as within the derivations of the financial crisis it “survived”.

We are certain that media operate as a “symbolic environment through which the process of in-group—out-group differentiation can be initiated” (Mastro, 2003, p. 100). We are also aware that mass media have been perceived as a contextual cause of negative intergroup attitudes (Boomgaarden and Vliegenthar, 2009), essentially arguing that they carry frames that influence attitudes towards minorities. We take all the above mentioned into account while we explore the dichotomy of refugee frames, their effects on attitudes and the role of emotions.

Framing Theory

Framing theory has been conceptualised through interdisciplinary approach and is rooted both in psychology and sociology. The political communication approach usually relies on the definition by Entman (Vliegenthart & Roggeband, 2007). For him “to frame is to select some aspects of perceived reality and make them more salient in a communicating text, in such a way as to promote a particular definition of a problem, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (Entman, 1993, p.52). There are four

functions of framing; they define problems, diagnose causes, make moral judgements and suggest remedies (Entman, 1993, p. 52). In this study, we will conceptualise frames as “distinct patterns of news media coverage that highlight certain aspects of an issue over others, thereby making a selection of relevant aspects.” (Lecheler et al., 2015, p. 814).

Issue- Specific Frames and Emotions.

Frames can be divided into multiple categories. However, often literature distinguishes between generic frames and issue-specific frames (e.g Semetko & Valkenburg, 2000). Generic frames encapsulate various topics while issue-specific frames are related to certain topics (e.g. de Vreese, Boomgaarden, & Semetko, 2011). Research often examines the coverage of migration through issue-specific frames (Gabrielatos & Baker, 2008;Cheng et al., 2014; Horsti, 2013; Roggeband & Vliegenthart, 2007; Saharso & Lettinga, 2008; Van Gorp, 2005), therefore issue-specific frames are employed for the purpose of this research.

There is substantial empirical proof that when immigrants are framed, they are depicted through the lens of a threat frame (e.g. Dunaway et al., 2011; Van Gorp,2005; Eberl et al., 2018) or through a frame which underlines the humanitarian aspects (Parrott et al., 2019). For the threat frame, as mentioned above, we draw on the elements of ITT’s realistic and symbolic threats. The victim frame perceives refugees as “victims of persecution and other kinds of harm and who are forced to leave their country” (Van Gorp, 2005, p. 491).

Research has reported that emotions are “integral components of news frames” (Lecheler et al., 2013, p. 189) and that individuals experience different emotional reactions to the news frame they are exposed to (e.g., Druckman and McDermott, 2008; Gross and D’Ambrosio, 2004). Lecheler et al. (2015) mention that limited studies measure the audience's emotional

responses to immigration frames (Lecheler et al., 2015, p. 813), while framing out-groups literature argues that frames cause strong intergroup emotions (e.g., Mackie et al., 2000). According to Gross, an experience will result in an emotion (Gross, 2008). For that matter, we can argue that when a reader is exposed to a news item, it will trigger one or multiple emotional reactions on them, something that Lecheler et al. proved in their study (Lecheler et al., 2013, p. 202). Nabi conceptualises emotions as “internal, mental states representing evaluative, valanced reactions to events, agents, or objects that vary in intensity” (p. 226). We expect the victim frame, which is positively inclined to cause more positive emotions and the threat frame to evoke more negative emotions. (Lecheler et al., 2013).

To bridge the gap of literature and answer RQ1 this study hypotheses the following: H1a: The victim frame will evoke more positive emotions to the readers.

H1b: The threat frame will evoke more negative emotions to the readers. Framing Effects on Attitudes and Emotions as Mediators.

News frames are associated with having effects since they suggest a way to perceive and interpret an issue. This process is known as “framing effect” is connected, among others, to effects on attitudes and opinions (e.g., Druckman & Nelson, 2003; Jacoby, 2000). For most of the audience, media is the only gateway to connect with and define the perception of events. Literature indicates framing can have effects on perception and attitudes towards migrant groups (e.g Boomgaarden & Vliegenthart, 2009). Drawing from Bobo (1997) we accept that “images of racial minority groups in the media at once reflect and shape the attitudes and beliefs others will hold toward those groups”. In order to avoid confusion, we employ the following definition.

Attitudes are “a psychological tendency which comes to surface through positive or negative

associated with refugees will influence attitudes of the Greek public towards the depicted refugee groups.

The majority of framing effects studies have directed their focus on cognitive and

attitudinal effects, however ‘‘they have said little about whether news frames can influence what citizens feel about politics’’ (Gross & Brewer, 2007, p. 124). Following years of “cognitive bias” (Lecheler et al. 2015, p.812), current research has picked up a significant development and focuses on the role of mediating processes on framing effects.

Emotions have been identified as one of the leading mediators of the process (e.g., Aarøe, 2011; Gross, 2008; Gross & D’Ambrosio, 2004; Holm, 2012; Nabi, 2003; Lecheler et.al, 2013; Lecheleret al., 2015), going against the traditional approach, where emotions were often

discarded (e.g., Price, Tewksbury, & Powers, 1997). Mediators are “underlying psychological processes that can explain why and how a framing effect takes place.” (Lecheler et al., 2015, p.816). Schuck and Feinholdt proved that emotions act as mediators between the frames and the attitudes that surface (Schuck and Feinholdt, 2015). Lecheler et al. (2013) tested for the

mediating role of two sets of emotions which were previously used in political communication science and can be viewed as expressions of different patterns of mediators: positive, enthusiasm and contentment and negative emotions, anger and fear. They found that anger and enthusiasm mediate a framing effect, contentment and fear do not (Lecheler et al., 2013, p. 202).

In a later study, Lecheler et al. (2015), examined both positive and negative emotions as mediators, since they argue that “positive emotions have been neglected in this literature” (Lecheler et al., 2015, p. 827) and are essential determinants in opinion formation (Griskevicius Shiota & Neufeld, 2010). Their study provided substantial empirical proof that indeed both negative and positive emotions mediate the process of framing effects. Framing studies usually

focus on the range of negative emotions, which the current research will measure for (Fear,

Worry, Irritation, Anger, Sadness), but Lecheler’s et al. results showcased the effect of

compassion and enthusiasm on the victimisation frame, which leads us to test and include four positive emotions as well (Hope, Empathy, Happiness, Excitement). Verkuyten (2004) proved that news frames result in positive emotional reactions, strengthening our reasoning behind their inclusion. To complete the scale of emotions, indifference is added as the tenth and last emotion.

To understand the mediating effect of emotions on attitudes, we employ the premise of the Cognitive Appraisal Theory of the frame (Kühne, 2012). Kühne suggests that framing exposure results to accessibility and applicability effects, which lead to a cognitive appraisal of the frame. This appraisal ignites an emotional response, as established at the beginning of this paper, which in response influences attitudes, possibly holding a mediating role in the process of framing effects. In relation, Holm (2012) argues that emotions are leading mediating elements of framing. Taking everything into account, from the ITT theory to the framing theory and the empirical proof on the mediating role of emotions, this study aims to answer RQ2: “To what

extent does the dichotomy of victim vs threat frame pertaining to refugees in Greek news content affect the Greek public’s attitudes towards refugees and what role do emotions play in the framing process?” by hypothesising as follows:

H2a: Positive emotions deriving from the victim frame, will mediate the relationship between exposure to the frame and refugee attitudes, in the sense that positive emotions will result in positive refugee attitudes.

H2b. Negative emotions deriving from the threat frame will mediate the relationship between exposure to the frame and refugee attitudes, in the sense that negative emotions will result in negative refugee attitudes.

The moderating role of political ideology

Sheufele proved that people’s predispositions could influence the used frames’ effects (Scheufele, 1999). Research on media and political behaviour has demonstrated that

confirmation biases influence the messages the audience perceives (e.g. Knobloch-Westerwick, Mothes & Polavin, 2017). Early on, Festiger (1957) explained this throughout the theory of cognitive dissonance, which states that confirmation bias derives from individuals’ need to find information that corresponds with their prior attitudes and beliefs. Caprara and Zimbardo (2004) underlined as well that a frame is most effective when it is in line with an individual’s

disposition. Domke, Perlmutter and Spratt (2002) discovered that although the audience is indeed influenced by news frames, the effect is also impacted by existing attitudes and predispositions, which adds to the existing literature (e.g Plener et al., 2017; Semyonov et al., 2006) that finds political orientation to be determining attitudes. Considering all the above this study aims to add to the limited literature by arguing that political beliefs moderate the effects of the refugee frames on refugee attitudes. In order to answer RQ3- What role does political orientation play in

the framing effects on attitudes of refugee stories in Greek media? - it presents the third and last

hypothesis:

H3a: The positive effect of the victim frame on refugee attitudes will be moderated by political ideology, so that the positive effect will be less strong for right-wing people.

H3b: The negative effect of the threat frame on refugee attitudes will be moderated by political ideology, so that the negative effect will be less strong for left-wing people.

Based on the theoretical framework, as well as the hypotheses, a conceptual model with all the variables and their relationships is presented in Figure 1.

Figure 1. Conceptual model

Method

This study uses a survey-embedded experimental approach, following the footsteps of current framing research (Lecheler & de Vreese, 2011; Schuck & de Vreese, 2006; Schuck & Feinholdt, 2015; de Vreese & Boomgaarden, 2003). Experiments are suitable when one is aiming to make causal claims. Survey-embedded experiments are appropriate for studies that have individuals as units of analysis and aim to measure attitudes of a large population (Babbie, 2005, p. 252). They provide a great amount of flexibility, low cost, high speed of returns (Fink, 2017) and are most suitable for research that examines from a distance, which fits our case, given the Covid-19 restrictions. We should not underestimate the downsides of experimental approaches however, such as the absence of “ecological validity” (Wenzel & Żerkowska-Balas, 2019, p.50) which makes it hard to generalise the findings to a broader natural environment. Although framing experiments are criticized for their lack of external validity as well (Barabas & Jerit, 2010), it remains the most suitable methodological approach for this study. We have taken

Political ideology Victim frame Threat frame Positive emotions Negative emotions Refugee attitudes H1b H1a H3a H3b

precautions to ensure that both the stimuli content and situation (i.e., exposure to an online news story) closely resemble real-life situations.

Main Study

The experiment took place online amongst Greek citizens between May 2 and May 9, 2020. To recruit subjects, an anonymous link was created and distributed via Facebook and Instagram. The survey followed a non-random method of snowball sampling.Online surveys in particular are beneficial due to their high response rates (Verhoeven, 2010). The experimental approach was chosen to avoid selection bias, through random selection in different groups of the experimental design. In total, 687 people began the survey and 608 completed it (i.e., a

completion rate of 89.3%). Ultimately, after cleaning the data for irregular and missing values and excluding respondents that did not pass the exposure check, which had a successful rate of 85.1%, 497 participants are included in the final sample. The sample consisted of Greeks, 57.3% of which lived within the country’s borders and 42.7 % outside (SD= .497). Participants were on average 30,5 years old (SD = 9.15), while 56.9% of the sample was female, 41,2% male, and two participants responded “Other” and seven “Prefer not to say”. Most participants were highly educated, with 40% having a Bachelor’s, 40,8% a Master’s degree and 6,6% a PhD. Finally, a 7-point scale from 1 (“hard to cope”) to 7 (“very comfortable”) measured how comfortable they can live considering the current level of income of their household and the respondents scored rather comfortably on average (M=4.63, SD= 1.33).

Procedure

This study used Qualtrics software to design and distribute the survey. At the beginning, participants were informed about the study’s goal, which was decepted; respondents were briefed that the experiment measures perceptions of social and political news by the Greek audience.

Since we did not want to influence participants, the real scope of the survey was partially concealed to avoid priming effects. In accordance, they were informed about their rights, terms of confidentiality and were directed to the responsible channels for further communication. After signing the declaration of consent, in which they stated that they comprehend all the above and decide to partake in the study voluntarily, the survey asked for basic socio-demographic questions, before proceeding to the exposure in one of the conditions.

Stimuli

The conditions themselves consisted of two external stimuli and one control group1_{. One}

of them represented refugees as victims and the other, portrayed refugees as a threat. In both cases, the article was based on an already published article in a local outlet in Greece, adding to the study’s external validity. The core of it, pertaining to a story about the local community of Serviwn, in Kozani, Greece was similar as well as its length and a neutral picture. In the victim condition, the Mayor welcomed refugees in the area, underlining the humanitarian aspects and how these people “should be approached with compassion and interest”. In the threat version, the Mayor blocked their arrival and stressed that “people with such different ideas, culture and frankly morals and ways of conduct will not fit well with our traditional and Greek way of life” (see details in Appendix A).

Concluding the survey, participants received a thorough debriefing, explaining the real scope of the research. They were informed that the article was built upon a real published version

1 _{The control group variable was excluded from the main analysis, which focused on the} differences between the exposure to the two frame groups.

and was not entirely true as it was presented to them and directed to the authors’ email for further communication and questions.

Measures

Dependent variable: Anti-refugee attitudes

After exposure to the stimuli, participants filled in a couple of questions about their attitudes towards refugees and asylum seekers. The items drew on symbolic and realistic threats and were designed to combine different types of threat that the respondents may experience when it comes to refugees in Greece. We include 10 items from fears pertaining to culture and identity to positive evaluations of refugees on humanitarian grounds. Answers were provided on a 7-point scale (1= “completely disagree” to 7 = “completely agree”). All items were codeed, so that higher values represent more negative attitudes toward refugees. They consisted of

statements such as “Refugees abuse the public services of my country because they benefit more than they contribute.”; “The identity of my country is threatened because there are too many refugees coming to Greece.” (all items are in Appendix C). To test if the 10 items formed a reliable scale, a principal axis factor analysis (PCA) using Oblimin Rotation took place; to this end, three items were re-coded so that they are worded negatively and fit the others. Two

components had an eigenvalue higher than one, which explained 59,2% of the total variance and the scree plot indicated a well-defined point of inflection after the components. Two items which stated, “I pity refugees because they are also humans” and “I should help refugees because they are also humans”, scored low on both factors, even when they were negatively computed; therefore, they were excluded. Without them, a reliability check indicated a high Cronbach’s alpha (.90) for the remaining 8 items. We therefore re-coded a new variable which encapsulates negative attitudes towards refugees (M = 2.75, SD = 1.48).

Mediator: Emotions

Lecheler et al. (2013) measured emotions such as contentment, compassion, enthusiasm, hope, anger, fear, and sadness using a 7- point scale (1 = “not at all”, 7 = “very much”). They asked their subjects “To what extent did you feel one or more of the following emotions while reading the article?” (Lecheler et al., 2015 p.92). This survey followed their design and asked respondents to think about what they just read and assess to what extent it made them feel any of these emotions; fear, hope, worry, irritation, empathy, indifference, happiness, anger, excitement, sadness. To verify whether the items construct one scale that can be applied to measure

participants emotions a principal axis factor analysis (PCA) using Oblimin rotation was conducted. It showed the 10 items form two scales (i.e., one for positive emotions and one for negative emotions), explaining 59,3% of the total variance. Two components with an eigenvalue above 1 and there is a clear point of inflection after them in the scatterplot. Reliability analysis shows that with a high Cronbach’s alpha of .88 the five negative emotions form a reliable scale. Likewise, the remaining positive emotions also have a good reliability score

(

α = .89). One item,

indifference, scored low on both factors and was excluded. We computed new variables out of the two scales, namely positive emotions (M = 3.10, SD = 1.67) and negative emotions (M = 3.00, SD =1.67).

Moderator: Political orientation.

Participants were asked to rate their political orientation on a 10-point scale, where 0 equals ‘very left-wing’ and 10 ‘very right-wing’. Political ideology is perceived as a moderator in this research design, since it has been argued to influence the effect of the condition treatment on the dependable variable (Wenzel & Żerkowska-Balas, 2019, p. 53). We therefore expect a

difference in the perceived effects on the stimuli based on the political orientation. Results indicate that the sample is relatively left-wing. Please refer to Table 1 for a detailed overview.

Table 1

Descriptive statistics of the political belief variable

n Min. Max. M SD

Political Beliefs 497 0 10 3.90 2.399

Valid N (listwise) 497

Manipulation check

During a pretest, the two stimuli materials were tested, carrying the frames the study focuses on and distributed via Facebook for three days. In total, 52 participants started the study and 49 completed it (94,2% completion rate). After excluding responses with missing cases in the variables, 44 cases were included in the final sample. We randomized Qualtrics to make sure the participants were exposed to one of the frames; Half of them were given the victim (N = 20) and the others the threat frame (N = 24). To understand how participants perceived the stimuli we employed a manipulation check, asking them to think about what they just read and answer these two questions on a 7-point-scale: “To what extent did the article present refugees as victims? And “To what extent did the article present refugees as a threat?”. A mean comparison between the two conditions took place, to examine how the respondents perceived it. Indeed, there was a significant difference between participants in the victim and threat frame on how respondents perceived it to portray refugees as victims; F(1,42) = 53.3, p = .000. Same applies for the threat frame and how they perceived refugees as threat; F(1,42) = 133.2, p = .000. Respondents in the victim frame stated predominantly that the stimuli represented refugees as

victims (M = 5.63; SD =1.67) compared to representing them as a threat (M=2.28, SD= 1.37). Participants in the threat condition perceived significantly more often the stimuli as portraying refugees in a threatening way (M = 6.08; SD=1.67) compared to presenting them as victims (M=1.53, SD= .964). Hence, the stimuli and manipulations were successful.

Randomization check

We checked whether the participants in the various conditions do not significantly differ in terms of a set of relevant sociodemographic variables, such as age, gender, education, comfort of life and place of residence based on their income and political orientation. Table 2 displays the results. Overall, these findings clarify that the randomization was successful. Therefore, we conduct our analyses without covariates.

Table 2

Randomization of non-experimental factors.

Source F

Df

p Between Groups Within Groups

Age .727 2 337 .484 Gender 1.048 2 494 .381 Education .966 2 494 .381 Income .402 2 494 .669 Residence .447 2 494 .640 Political beliefs .014 2 494 .986

Main effects

To test the H1a, we compared means. Subjects in the victim condition report significantly more often positive emotions (M = 4.06, SD = 1.72), compared to participants in the threat condition (M = 2.12, SD = 0.82), F(1,337) = 173.4, p = .000 and η2 _{= 0.34. We accept that indeed}

the victim frame evokes more positive emotions to the readers. H1b argues that the threat frame, which presents refugees in Greece in a negative way, will lead to experiencing negative

emotions. The analysis illustrates that indeed participants in the threat condition experienced more negative emotions (M = 4.07, SD = 1.44) compared to the respondents in the victim condition (M = 1.96, SD = 1.15), with F(1,337) = 220.7, p = 0.000, η2_{= 0.396. We accordingly}

accept H1b.

H2a and H2b argued the mediating role of emotions in the framing effects process. To test them, we ran multiple regression analyses. First, a simple linear regression analysis took place to determine the effects of the frame variable (victim/threat) on refugee attitudes. An insignificant equation was found for the frame variable (F (1,337) = .050, p = .823 with and R2

=.000. Therefore, we conclude that the frames do not predict attitudes, essentially accepting that without a significant effect and a direct relationship, there is nothing to mediate. However, we proceed to examine further the evocation of emotions from the stimuli and the mediating effects; if any.

Table 3

Results of regression analysis of frame on attitudes

Dependent variable: negative attitudes B SE B β p

Constant 2.766 .252 .000

Note. N=339 A second simple linear regression was conducted to determine the effects of the frames on emotions (= mediators) the participants experienced after exposure. For the negative

emotions, a significant regression equation was found (F (1,337) = 220.7, p = .000, with an R2 _of

.396. The coefficient of threat frame has a significant positive effect on negative emotions enhancing the results of the previous comparison of means for H1b. A significant regression equation was also found for the positive emotions (F (1,337)= 220.7, p = .000, with an R2 _of

.340. The coefficient of the threat frame indicates a negative relationship; thus, the threat frame leads to less positive emotions, mirroring the results for H1a above.

After we established the effects on attitudes and emotions, we conducted a final linear regression analysis, to check the mediating role of emotions on refugee attitudes while including the frame variable as well. A significant regression model was found (F (3,335)= 29.6, p = .000 with R2 _{= .122. However, the coefficients for the negative emotions result in an insignificant}

relationship, Therefore, we have to reject H2b. We observe though a direct negative relationship between positive emotions and negative attitudes. This finding mirrors H2a which states that: Positive emotions deriving from the victim frame partially mediate the relationship between exposure to the frame and refugee attitudes, in the sense that positive emotions result in positive refugee attitudes. We can accept H2a. Table 4 includes the separate regression analyses.

Table 4

Summary of multiple regression analyses predicting negative attitudes towards refugees

Dependent variable B SE B β p B SE B β p B SE B β p Negative emotions Constant -.150 .223 .503 Threat frame 2.108 .142 .629 .000 Positive emotions Constant 6.005 .232 .000 Threat frame -1.944 .148 -.583 .000 Negative attitudes. Constant 5.034 .412 .000 Threat frame -.778 .213 -.264 .000 Negative Emotions .004 .059 .004 .950 Positive Emotions -.378 .056 -.428 .000 Note. N=339

H3a and H3b concerned the moderating role of political beliefs in the effects of framing on attitudes. Multiple regressions were conducted in this case as well. First, a regression was done between frame condition and attitudes which did not yield any significant effects. Then we ran a second regression analysis, including the frame condition and political belief as

independent variables. A significant regression equation was found with a p = .000 and R2 ₌

.323. No significant effect was found for the condition variable on attitudes. Still, there was a significant effect of political belief on attitudes, with right-wing people scoring more negative attitudes. Running the same regression analysis with interaction effects lastly, did not show any direct effect of interaction term on attitudes This means that the effects of frame on attitudes do not differ for people with different political beliefs. We therefore reject H3a and H3b.

Table 5

Summary of two individual regression analyses predicting negative attitudes towards refugees

Model 1 Model 2 Dependent variable B SE B β p B SE B β p Negative attitudes. Constant 1.409 .234 .000 1.369 .394 .001 Threat frame -.037 .132 -.012 .781 -.010 .253 -.003 .970 Political Beliefs .350 .028 .568 .000 .360 .086 .585 .000 Interaction -.007 .055 -.020 .900 Note. N=339 Discussion- Conclusion

This study set out to determine the effects of the victim/ threat news frame concerning refugees in Greece on the emotions and attitudes the Greek public has towards the refugee group. Results indicated that the frames lead to emotional responses in the audience, with the victim frame evoking more positive emotions and the threat frame more negative emotions amongst the survey participants. Examining RQ2 did not provide the full spectrum of hypothesized results. The analysis proved an insignificant relationship between the frames and Greek public’s attitudes. However, there was a direct negative relationship between positive emotions and negative attitudes, verifying H2a. Contrary to the hypothesized association between political beliefs and refugee attitudes, results indicated that political beliefs do not moderate the effect of the according frames on refugee attitudes.

For H1a and H1b this study drew on literature that suggests that media news content leads to emotional reactions to the readers (e.g., Druckman and McDermott, 2008; Gross and

D’Ambrosio, 2004). Results of the study indeed proved that the victim frame leads to experiencing more positive feelings, while the threat frame evokes more negative feelings between the respondents, findings which are in line with current literature that states that

emotions are fundamental elements of news frames (Lecheler et al., 2013, p. 189). Although the difference between the means of the conditions was minor, the threat frame evoked slightly more negative emotions than the positive ones deriving from the victim frame. This finding coincides with the ones from Dunaway et al. (2011), who underlined how negative minority stories have a greater impact than positive ones.

The second research question examined the influence the two frames had on the reader's attitudes and embarked on the exploration of the mediating role of emotions in the framing effect process. In opposition to the hypotheses, no empirical proof linking the frames and attitudes was found. A possible interpretation of the results could derive from the fact that the frame did not convey the message as aimed. However, the manipulation check indicated the exact opposite; frames did carry the significance we set out to. A second immediate explanation would draw on the randomisation effect of the experiment; however, in our case all respondents and their characteristics were properly randomised, so we must reject that notion. Ultimately, the insignificant relationship might be explained by the fact that the refugee issue is a heavily debated and widely covered topic in Greek news and society. Therefore, respondents could not be influenced only by a brief article carrying one of the two most prominent frames; they possibly carry their own predispositions on the topic that affect them more and are already exposed in various content before encountering our stimuli.

Regardless of the insignificant findings on the relation between frames and attitudes, RQ2 dealt with emotions and how they influence the effect of the frames on refugee attitudes.

Kühne (2012) argued that emotions act as mediators between news frames and opinion formations. Although literature has primarily investigated negative emotions undermining the role of positive ones (Lecheler et al., 2015, p. 827), this research examined both. Results further support current media research that perceives positive emotions as integral elements in the process of forming opinions and attitudes (e.g Griskevicius et al), since only positive emotions were found to predict refugee attitudes, whereas the relationship between negative emotions and attitudes was insignificant, going against the above mentioned current literature that proves the effect of both negative and positive emotions (e.g Lecheler et al., 2015). This finding consists of useful empirical knowledge, since it corresponds to a Greek sample, which has its own unique characteristics; from the geopolitical positioning of the country to the pressures it has been experiencing within the refugee crisis. Drawing a link between how positive emotions can influence refugee attitudes could be proven useful for media and political communication research which should focus more on positive emotions and bridge further the empirical gap this study aimed to see through.

Lastly, RQ3 concerned the moderating role of political beliefs on refugee attitudes, which ranged from very left-wing to very right- wing. We hypothesized the connection based on

literature findings that indicates that certain predispositions and attitudes impact the effect that the frames have on individuals (Domke et al, 2002) and argued that political beliefs consist of such strong predispositions. Contrary to the hypothesised connection, results of the experiment showed that the interaction of political beliefs and frame condition was not significant, indicating that the effects of the frames on attitudes did not differ depending on the respondent’s political beliefs. Again, this could derive from other factors influencing the process, or simply because the audience has already been exposed to an abundance of content connected to refugees. Although

not initially within the scope of this research, results also reveal a significant effect of political beliefs on attitudes, with more right-wing people exhibiting more negative attitudes towards refugees. This discovery could potentially contribute to the understanding of ideology and

attitudes towards minorities, especially within the context of the resurge of right- wing politics in Europe. Drawing from the results, we can argue that this study could be useful both to media research scholars but also for journalists and other professionals within the media sphere.

This study partially achieved some of the objectives that drove its RQ and

experimentation. However, its limitations should be addressed. It is not evident how it is limited based on its sample, which was properly randomized and could be argued to represent the native Greek population. However, one could assume that the respondents might have not read the article as carefully as intended, especially since it was connected to a topic they are familiar with. The study imposed a 30 second timer and constructed an article that could be read comfortably within that interval of time; however, we cannot measure if they were indeed exposed to it fully. Survey-embedded experimental designs in general pose certain restrictions when it comes to real-life generalisations (e.g., Kinder, 2007). Still, they provide more consistent sizes of effects than field studies (e.g., Jerit, Barabas, & Clifford, 2013), and were chosen as the most suitable methodological approach.

Lastly, this study was constructed within the premise of literature that argued the causal relationship between emotions and attitudes. We were only able to prove that news frames can evoke a range of emotions depending on their positive or negative content and that only positive emotions partially mediate the effect of the frames on attitudes towards refugees. Although the core of the study concerned attitudinal effects, we found an insignificant relationship between

them and the frames. For future purposes, it would be better to measure their attitudes before and after exposure to the frames.

Future research should also focus on enhancing the mediation models, with multiple variables from different dimensions, in order to test to what extent they showcase significance. Due to the limited time and resources within this master’s dissertation, we measured 10 emotions that have been correlated with framing studies. However, following the cited literature which acknowledges it as well (e.g Lecheler et al.), we could always enhance the scales of emotions and experiment within their correlations, something that future research should not overlook. This study measured attitudes of the native Greek population. In the process of the data gathering, we observed and discarded responses that belonged to what in Greece is called a “second generation” immigrant, people whose parents had migrated in Greece and they were born within its borders, without resuming Greek citizenship, due to the strict and debated Greek citizenship law. Future research could review and test differences of migration frames to both native and minority populations.

To conclude, this study offers an understanding about the role of the victim/threat frame in Greek refugee stories. We found that the positively inclined victim frame induces more positive emotions than the negative threat frame, which results to more negative emotions. However, we were unable to create a correlation between negative attitudes towards refugees and the frame conditions, a process which was only partially mediated by positive emotions leading to positive attitudes. Lastly, the moderating role of political beliefs was rejected as hypothesized in the premise of the theoretical framework. We can safely argue that frames evoke emotions, however the mediating and moderating processes should be further examined. This study aimed to contribute to the current literature pertaining to framing migration and the role of emotions. It

provided insights on the effects of the Greek coverage of refugees, which can be proven useful in order to contextualize and comprehend the multiple and complex dimensions of the

unprecedented crisis.

References

Aarøe, L. (2011). Investigating frame strength: The case of episodic and thematic framing.

Political Communication, 28, 207-226.

https://doi.org/10.1080/10584609.2011.568041

Arlt, D & Wollin, J. (2016). The Refugees: Threatening or Beneficial? Exploring the Effects of Positive and Negative Attitudes and Communication on Hostile Media Perceptions.

Global Media Journal: German Edition, 6(1).

https://doi.org/10.1002/9781118541555.wbiepc217

Babbie, E. (2005). The basics of social research (3rd). Belmont, CA: Thomson Wadsworth. Chapter 9: Surveys, 250-292.

Barabas, J., & Jerit, J. (2010). Are survey experiments externally valid? American Political

Science Review, 104, 226-242.

https://doi.org/10.1017/S0003055410000092

Bobo, L. (1997). Race, Public Opinion, and the Social Sphere. Public Opinion Quarterly, 61(1), 1–15.

https://doi.org/10.1086/297784

Boomgaarden, H., & Vliegenthart, R. (2009). How news content influences anti- immigration attitudes: Germany, 1993-2005. European Journal Of Political Research, 48(4), 516-542. https://doi.org/10.1111/j.1475-6765.2009.01831.x

Blalock, H. M. (1967). Toward a theory of minority group relations. New York: Capricorn. Blumer, H. (1958). Race Prejudice as a Sense of Group Position. Pacific Sociological Review,

1(1), 3–7. Retrieved from http://search.proquest.com/docview/1310932368/ Bos, L., Lecheler, S., Mewafi, M., & Vliegenthart, R. (2016). It’s the frame that matters:

Immigrant integration and media framing effects in the Netherlands. International

Journal of Intercultural Relations, 55, 97–108.

https://doi.org/10.1016/j.ijintrel.2016.10.002

Bruneau, E., Kteily, N., & Laustsen, L. (2018). The unique effects of blatant dehumanization on attitudes and behavior towards Muslim refugees during the European “refugee crisis” across four countries. European Journal of Social Psychology, 48(5), 645–662. https://doi.org/10.1002/ejsp.2357

Caprara, G. V., & Zimbardo, P. G. (2004). Personalizing politics: A congruency model of political preference. American Psychologist, 59(7), 581–594.

http://doi.org/10.1037/0003-066X.59.7.581

Cheng, L., Igartua, J., Palacios, E., Acosta, T., & Palito, S. (2014). Framing Immigration News in Spanish Regional Press. International Migration, 52(6), 197–215.

https://doi.org/10.1111/j.1468-2435.2010.00647.x

Cohen J. Safe in our hands? A study of suicide and self-harm in asylum seekers. J Forensic Leg

Med. 2008;15(4):235-44.

https://doi.org/10.1016/j.jflm.2007.11.001

De Coninck, D., Matthijs, K., Debrael, M., De Cock,R., & D’Haenens, L.(2019). Unpacking Attitudes on Immigrants and Refugees: A Focus on Household Composition and News

Media Consumption. Media and Communication, 7(1), 43–55 https://doi.org/10.17645/mac.v7i1.1599

De Vreese, C. H., & Boomgaarden, H. (2006). News, political knowledge and participation: The differential effects of news media exposure on political knowledge and participation. Acta

Politica, 41(4), 317-341.

https://doi.org/10.1057/palgrave.ap.550016

De Vreese, C. H., Boomgaarden, H., & Semetko, H. (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.

https://journals.sagepub.com/doi/10.1177/0093650210384934

Domke, D., Perlmutter, D., & Spratt, M. (2002). The primes of our times?. Journalism, 3(2), 131-159.

http://dx.doi.org/10.1177/146488490200300211

Douglas, P., Cetron, M., & Spiegel, P. (2019). Definitions matter: migrants, immigrants, asylum seekers and refugees. Journal of Travel Medicine, 26(2).

https://doi.org/10.1093/jtm/taz005

Druckman, J. N., & McDermott, R. 2008. Emotions and the framing of risky choice. Political

Behavior, 30, 297–321.

https://doi.org/10.1007/s11109-008-9056-y

Druckman, J., & Nelson, K. (2003). Framing and Deliberation: How Citizens’ Conversations Limit Elite Influence. American Journal of Political Science, 47(4), 729–745.

https://doi.org/10.1111/1540-5907.00051

Media coverage and public opinion on Latino immigration in post-Katrina Louisiana. Social

Science Quarterly, 92(4), 917-937.

Eagly, A., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth: Harcourt Brace Jovanovich.

Eberl, J.-M., Meltzer, C. E., Heidenreich, T., Herrero, B., Theorin, N., Lind, F.,Strömbäck, J. (2018). The European media discourse on immigration and its effects: a literature review.

Annals of the International Communication Association, 42(3), 207–223.

https://doi.org/10.1080/23808985.2018.1497452

Entman, R. M., & Rojecki, A. (1993). Freezing out the public: Elite and media framing of the U.S. anti-nuclear movement. Political Communication, 10(2),155–173.

https://www.tandfonline.com/doi/abs/10.1080/10584609.1993.9962973

Esses, V., Dovidio, J., Jackson, L., & Armstrong, T. (2001). The Immigration Dilemma: The Role of Perceived Group Competition, Ethnic Prejudice, and National Identity. Journal of

Social Issues, 57( 3), 389–412.

https://doi.org/10.1111/0022-4537.00220

Festinger, L. (1957). A theory of cognitive dissonance. Evanston, Ill: Row, Peterson.

Fink, A. (2017). How to conduct surveys: a step-by-step guide (Sixth edition.). Los Angeles: SAGE.

Gabrielatos, C., & Baker, P. (2008). Fleeing, Sneaking, Flooding: A Corpus Analysis of

Discursive Constructions of Refugees and Asylum Seekers in the UK Press, 1996-2005.

Journal of English Linguistics, 36(1), 5–38.

Georgiadis for Refugees: We have been the victim of a collective effort to alter the country. (2020, January 11) CNN News. Retrieved from

https://www.cnn.gr/news/politiki/story/203681/georgiadis-gia-prosfyges-exoyme-pesei-thyma-omadikis-prospatheias-alloiosis-tis-xoras

Griskevicius, V., Shiota, M. N., & Neufeld, S. L. (2010). Influence of different positive emotions on persuasion processing: A functional evolutionary approach. Emotion, 10, 190-206. https://doi.org/10.1037/a0018421

Gross, K. (2008). Framing persuasive appeals: Episodic and thematic framing, emotional response, and policy opinion.' Political Psychology, 29(2), 169-192.

https://doi.org/10.1111/j.1467-9221.2008.00622.x

Gross, K., & Brewer, P. (2007). Sore Losers: News Frames, Policy Debates, and Emotions. The

Harvard International Journal of Press/Politics, 12(1), 122–133.

https://doi.org/10.1177/1081180X06297231

Gross, K., & D’Ambrosio, L. (Gros2004). Framing emotional response. Political Psychology, 25, 1–29.

https://doi.org/10.1111/j.1467-9221.2004.00354.x

Holm, E. M. (2012). Emotions as mediators of framing effects. Aarhus, Denmark

Horsti, K. (2007). Asylum Seekers in the News: Frames of Illegality and Control. Observatorio (OBS*), 1,145–161.Retrieved from

https://dialnet.unirioja.es/servlet/oaiart?codigo=5576976

Jacoby, W. G. (2000). Issue framing and public opinion on government spending. American

Journal of Political Science, 44, 750-767.

Mackie, D., Devos, T., & Smith, E. (2000). Intergroup Emotions: Explaining Offensive Action Tendencies in an Intergroup Context. Journal of Personality and Social Psychology, 79(4), 602–616.

https://doi.org/10.1037/0022-3514.79.4.602

Nabi, R. L. (2003). Exploring the framing effects of emotion. Communication Research, 30, 224-247.

https://doi.org/10.1177/0093650202250881

Schemer, C., Wirth, W., Matthes, J., & Matthes, J. (2012). Value Resonance and Value Framing Effects on Voting Intentions in Direct-Democratic Campaigns. American Behavioral

Scientist, 56(3), 334–352.

https://doi.org/10.1177/0002764211426329

Stevis-Gridneff, M. & Gall, C. (2020, February 29). Erdogan Says, ‘We Opened the Doors,’ and Clashes Erupt as Migrants Head for Europe. The New York Times. Retrieved from https://www.nytimes.com/2020/02/29/world/europe/turkey-migrants-eu.html

Wirth, T. (2016). The European Union’s framing of the European refugee crisis. Elon Journal. 45.

Eberl, J.-M., Meltzer, C. E., Heidenreich, T., Herrero, B., Theorin, N., Lind, F.,Strömbäck, J. (2018). The European media discourse on immigration and its effects: a literature review.

Annals of the International Communication Association, 42(3), 207–223.

https://doi.org/10.1080/23808985.2018.1497452

Igartua, J., Moral-Toranzo, F., & Fernández, I. (2011). Cognitive, Attitudinal, and Emotional Effects of News Frame and Group Cues, on Processing News About Immigration.

Journal of Media Psychology: Theories, Methods, and Applications, 23(4), 174–185.

https://doi.org/10.1027/1864-1105/a000050

Jacobs, L., Boukes, M., & Vliegenthart, R. (2019). Combined Forces: Thinking and/or Feeling? How News Consumption Affects Anti-Muslim Attitudes through Perceptions and Emotions about the Economy. Political Studies, 67(2), 326–347.

https://doi.org/10.1177/0032321718765696

Jacobs, L., & Van der Linden, M. (2018). Tone matters: Effects of exposure to positive and negative tone of television news stories on anti-immigrant attitudes and carry-over effects to uninvolved immigrant groups. International Journal of Public Opinion Research ,3 0(2), 211–232.

https://doi.org/10.1093/ijpor/edw036

Kinder, D., & Sears, D. (1981). Negative attitudes and politics: Symbolic racism versus racial threats to the good life. Journal of Personality and Social Psychology, 40, 414–431. https://doi.org/10.1037/0022-3514.40.3.414

Knobloch-Westerwick, S., Mothes, C., & Polavin, N. (2020). Confirmation Bias, Ingroup Bias, and Negativity Bias in Selective Exposure to Political Information. Communication

Research, 47(1), 104–124.

https://doi.org/10.1177/0093650217719596

Kühne, R. (2012, May 24-28). Political news, emotions, and opinion formation: Toward a model

of emotional framing effects. Paper presented at the Annual Conference of the

International Communication Association (ICA), Phoenix, AZ.

Lecheler, S., Bos, L. & Vliegenthart R. (2015) The Mediating Role of Emotions: News Framing Effects on Opinions About Immigration, Journalism & Mass Communication Quarterly,

Vol. 92(4) 812–838.

https://doi.org/10.1177/1077699015596338

Lecheler, S., Schuck, A., de Vreese, C., & Political Communication & Journalism. (2013). Dealing with feelings: positive and negative discrete emotions as mediators of news framing effects.The European Journal Communication Research, 38(2) ,189–209. https://doi.org/10.1515/commun-2013-0011

Mastro, D. E. 2009). “Effects of Racial and Ethnic Stereotyping.” In Media Effects: Advances in

Theory and Research, 3rd ed., ed. Jennings Bryant and Mary Beth Oliver, 325–341.

Hills- dale, NJ: Lawrence Erlbaum Associates.

Operational Portal, Refugee Situations, UNHCR, Retrieved from

https://data2.unhcr.org/en/situations/mediterranean/location/5179

Plener, L. P., Groschwitz, C. R., Brähler, E., Sukale, T. & Fegert, M. J. (2017). Unaccompanied refugee minors in Germany: attitudes of the general population towards a vulnerable group. European Child & Adolescent Psychiatry, 1-10.

https://doi.org/10.1007/s00787-017-0943-9

Quillian, L. (1995). Prejudice as a response to perceived group threat: Population composition and anti-immigrant and racial prejudice in Europe. American Sociological Review, 60(4), 586–611.

https://doi.org/10.2307/2096296

Saharso, S., & Lettinga, D. (2008). Contentious citizenship: Policies and debates on the veil in the Netherlands. Social Politics: International Studies in Gender, State & Society, 15, 455- 480.

Scheufele, D. A. (2000). Agenda-setting, priming, and framing revisited: Another look at cog- nitive effects of political communication. Mass Communication and Society, 3, 297-316. Semetko, H., & Valkenburg, P. (2000). Framing European politics: a content analysis of press

and television news. Journal of Communication, 50(2), 93–109. https://doi.org/10.1111/j.1460-2466.2000.tb02843.x

Semyonov, M., Raijman, R., & Gorodzeisky, A. (2006). ‘The rise of anti-foreigner sentiment in European societies, 1988–2000.’ American Sociological Review, 71(3), 426–449.

https://doi.org/10.1177/000312240607100304

Stephan, W., Diaz-Loving, R., & Duran, A. (2000). Integrated Threat Theory and Intercultural Attitudes: Mexico and the United States. Journal of Cross-Cultural Psychology, 31(2), 240–249.

https://doi.org/10.1177/0022022100031002006

Stephan, W. G., & Stephan, C. W. (2000). An integrated threat theory of prejudice. Reducing prejudice and discrimination, 23-45.

Stephan, W. G., Ybarra, O., & Rios Morrison, K. (2015). Intergroup threat theory. In T. Nelson (Ed.), Handbook of prejudice (pp. 255–278). Mahwah, NJ: Lawrence Erlbaum.

Schuck, A., & de Vreese, C. (2006). Between Risk and Opportunity. European Journal Of

Communication, 21(1), 5-32.

http://dx.doi.org/10.1177/0267323106060987

Schuck, A. R., & Feinholdt, A. (2015). News framing effects and emotions. Emerging Trends in

the Social and Behavioural Sciences: An Interdisciplinary, Searchable, and Linkable Resource.

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.

https://doi.org/10.1177/0267323105058253

Verhoeven, N. (2010) Wat is onderzoek? Den Haag: Boom Lemma Uitgevers.

Verkuyten, M. (2004). Emotional Reactions to and Support for Immigrant Policies: Attributed Responsibilities to Categories of Asylum Seekers. Social Justice Research, 17(3), 293– 314.

https://doi.org/10.1023/B:SORE.0000041295.83611.dc

Vliegenthart, R. & Roggeband, C. (2007). Framing Immigration and Integration. Relationship between Press and Parliament in the Netherlands. The International Communication

Gazette, 69(3), 295-319.

https://doi.org/10.1177/1748048507076582

Wenzel, M., & Żerkowska-Balas, M. (2019). Framing Effect of Media Portrayal of Migrants to the European Union: A Survey Experiment in Poland. East European Politics & Societies

and Cultures, 33(1), 44–65.

APPENDIX A

Stimuli Materials in English

Victim Frame

The mayor of Serviwn welcomes the refugees to the community.

Source; kozanimedia.com

The mayor of Serviwn, in Kozani has announced that refugees and migrants will go to a hotel in the area and be treated as victims with compassion, on the occasion of the upcoming visit of Caesarea Mayor Davelli Mehmet Tsamur.

Mr. Eleftheriou stressed that the visit of the Turkish mayor is an opportunity for "a journey back in time and history, a tribute to our refugees from Cappadocia and Pontus, but also a tribute to the Anonymous refugee that extends to those of today”. He said that the refugee crisis in Greece often causes concern, but these people" with a painful and anxious appearance must be approached with compassion and interest. "

Mr. Papathanassiou, a representative of the local business sector of the community, stressed that they will facilitate the integration of refugees as soon as they settle in the hotel. "We owe them an apology from history and a promise to help every persecuted person who might be in the same position as they are now. I am convinced that we have chosen the policies of an open society in the best possible way, without discrimination or exclusion. We will support these people. "

Davili Mehmet Chamur will meet with local officials next Tuesday. The refugees themselves will be relocated to their new residence by the end of the month.

Threat Frame

The mayor of Serviwn blocks the arrival of refugees in the community.

Source; kozanimedia.com

The mayor of Serviwn, in Kozani, has announced that refugees and migrants who were to be accommodated in a hotel in the area will not find "open doors" after the local negative reaction

of the community and on the occasion of the upcoming visit of the mayor of Caesarea, Daveli Mehmet Chamur.

Although Mr. Eleftheriou stressed that the visit of the Turkish mayor is an important opportunity "for a journey back in time and history, a tribute to our refugees from Cappadocia and Pontus" he said that the refugees are causing concern to the local community, which seems reluctant to host them. "People with a painful and anxious appearance need to be approached with interest and caution," he said.

Mr Papathanassiou, a spokesman for the local business sector, said the news of the refugees' arrival had caused panic throughout the community. "Many members of our community are worried and afraid of what will happen if we bring refugees into our area. When one tries to survive they will do everything. We have a safe community here whose security may be at stake. " He stressed that the arrival of "people with so many different ideas, cultures and ways of behaving will not fit well with our traditional way of life".

Davili Mehmet Chamur will meet with local officials next Tuesday. The refugees themselves were to be relocated to their new accommodation at the end of the month.

Source:

Appendix B

Online Survey

Introductory Text

Dear Madam or Sir,

This is an invitation to participate in a study for the Amsterdam School of Communication Research (ASCoR) at the University of Amsterdam (The Netherlands). I am conducting research, as part of my master’s degree, to collect insights on perceptions of political news stories in Greece.

Participating only takes 10 minutes. Please be assured that all your responses and data will be protected, and you have the right to withdraw at any time. Greek natives aged 18 and above, living in Greece or abroad can participate in this survey.

1. Your anonymity will be safeguarded, we will not collect personal information such as names, addresses, IP-addresses, photos, videos etc. Fully anonymised research data can be shared with other researchers.

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 7 days 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 you're being subjected to any appreciable risk or discomfort, and you will not be exposed to any explicitly offensive material.

Should you have any complaints or comments about the course of the research and the procedures it involves as a consequence of your participation, you can contact the designated member of the Ethics Committee representing ASCoR:

ASCoR Secretariat, Ethics Committee University of Amsterdam, PO Box 15793 1001 NG Amsterdam; 020-525 3680; ascor-secr-fmg@uva.nl.

A confidential treatment of your complaint or remark is guaranteed. For more information about the research and the invitation to participate, you can send an email to the project leader Laura Jacobs (l.c.n.jacobs@uva.nl) or contact me directly at triantafyllaism@gmail.com.

Thank you in advance for your assistance with this research.

With kind regards,

Triantafyllia (Roza) Ismailai

Informed consent

I hereby declare that I have been informed in a clear manner about the nature and method of the research.

I agree, fully and voluntarily, to participate 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 experiment 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.

I have read and understood the above text:

• I Agree to participate in the research study

• I do not wish to participate in the research study

Socio-demographic questions

Before we start, we’d like to ask you some general questions about yourself. Please keep in mind that you cannot go back and change your answers as the survey continues.

1. How old are you? (Please fill in the number) 2. Which gender do you identify with?

∘

Masculine

∘

Other

∘

Prefer not to say

3. What is the highest degree or level of education you have completed? If you are currently studying, please indicate the one from the list.

∘

No formal education

∘

High school diploma

∘

College degree

∘

Vocational training

∘

Bachelor’s degree

∘

Master’s degree

∘

Professional degree

∘

Doctorate degree

∘

Other

4. How comfortable can you live taking your current household income level into account

1. Difficult

to cope to 2 3 4 5 6 comfortable 7. Very

∘

∘

∘

∘

∘

∘

∘

∘

Greece

∘

Other, namely:

6. Do you currently live in the country of your nationality?

∘

Yes

∘

No

7. Choose the option “none of the above” as a response to show you are paying attention (attention check)

∘

Left

∘

Right

∘

Center

∘

None of the above

∘

No position

Now a question on politics follows. In political terms, people talk about “the left” and “the right”. What is your position? Please use any number on a scale from 0 to 10, where 0 means “left” and 10 is “right”. Where do you place yourself?

0.

Left 1 2 3 4 5 6 7 8 9 Right 10.

∘

∘

∘

∘

∘

∘

∘

∘

∘

∘

∘

Bridge to stimulus

To continue, you will now be presented with a short news article that has been published in the media recently. The article will remain on the screen for a minimum of 30 seconds before