Welcome home : Valence frames, emotions, and attitudes towards the Syrian. refugee crisis. University of Amsterdam. Graduate School of Communication

Transcription

1 Welcome home : Valence frames, emotions, and attitudes towards the Syrian refugee crisis. University of Amsterdam Graduate School of Communication MSc Communication Science: Political Communication Master s Thesis Dimitrios Bairaktaris Samsarelos Student ID: Supervisor: Dhr. Dr. Andreas R. T. Schuck Date:

2 WELCOME HOME 2 Abstract The war in Syria has had significant consequences for Syrian and European citizens alike. While Syrian refugees flee from violence in their homeland, the reaction in European countries has been mixed. Reflecting this, media coverage of the crisis has exposed a vast range of opinions, using different news frames to explain events relating to the crisis. News frames can have a substantial impact in shaping public opinion by influencing audience emotions. To examine how news frames might impact attitudes toward Syrian refugees, an experiment was conducted with a total of 218 participants who were exposed to either positive or negative news frames. Results show that exposure to positive or negative news frames generate corresponding emotions, moderated by people s predispositions, which ultimately will lead to positive or negative attitudes toward Syrian refugees. Keywords: news frames, refugee crisis, emotions, attitudes.

3 WELCOME HOME 3 Introduction No one leaves home, unless home is the mouth of a shark Warsan Shire (Bausells, & Shearlaw, 2017). Since 2013, we can identify an ongoing refugee crisis in Europe. It is the largest humanitarian crisis in Europe since WWII (Wirth, 2016). The majority of the refugees come from Syria. The crisis in Syria began in 2011 when thousands of Syrian citizens protesting against their government. After a lot of protests, the war started, leading millions of Syrians to leave their destroyed country (Yagzan, Utku, & Sirkeci, 2015). Citizens were protesting for the right to a more democratic system. But the security forces guided by the government opened fire in one of the protests in March 2011 and as a result, more people took to the streets in protest (BBC, 2011). According to official data from Human Rights, more than 190,000 people were killed over the following three years of conflict (Price, Gohdes, & Ball, 2014). These conditions led the country into civil war, with various groups fighting with each other. The lack of stability in the country provided the conditions necessary for the Islamic state to take control in some areas of the country (BBC, 2015). Thus, millions of people started to leave their country in order to find a better future for them and their families, away from the war and their destroyed country. According to the UNHCR, (2016), we identify that more than 4.9 million Syrians left the country from the start of the war until today. Millions of refugees from Syria are trying to find a better future for them and their families, in the nearest peaceful continent that is Europe (Telegraph, 2017). Subsequently, increasing numbers of European citizens have been affected by the ongoing refugee crisis (De Swert, Schacht, & Masini, 2017).

4 WELCOME HOME 4 As the Syrian refugee crisis bears such influence on so many lives, and with most European citizens learning about the crisis via the media (Boomgaarden & Vliegenthart, 2009), research into how media coverage might shape public opinion may provide useful insights for managing the crisis effectively. In addition, according to Boomgaarden and Vliegenthart (2007), the amount of the topic related information in the news can potentially explain negative attitudes towards immigrants, and to some extent also toward refugees (Domke, McCoy, & Torres, 1999). The media shapes public opinion about certain events (de Vreese & Boomgaarden, 2003; Cheng, Igartua, Palacios, Acosta, & Palito, 2010), and therefore we believe that certain frames that are used by journalists will affect readers perceptions of Syrian refugee crisis. People are affected by the frames that are used by the media and news exposure can trigger different emotions according to the model of Schuck and Feinholdt (2015). According to this model, emotions are the mediators of the relationship between the frames that are used by journalists, and the attitudes that are the outcome of it. They also highlight the importance of people s predispositions of events and how they affect emotions and to some extent attitudes. We will utilize the above-mentioned model in our research to investigate valence news frames (Schuck, 2009; Schuck & de Vreese, 2006), that are used by journalists towards the Syrian refugee crisis. European citizens have either positive or negative attitudes towards the Syrian refugee crisis. However, we are interested if those attitudes are mediated by sympathetic or antipathetic emotions generated by the news. We add to the growing body of research on attitudes towards Syrian refugee crisis by applying the model of Schuck and Feinhold (2015) to explain the effects of news frames on attitudes mediated by emotions. As discussed above, frames,

5 WELCOME HOME 5 emotions, and pre-dispositions are expected to have a significant impact on people's attitudes. Ultimately, this study aims to provide insight to what might drive attitudes in European society toward refugees. Public opinion is often divided when it comes to events like the refugee crisis. This can be exacerbated by the media environment which varies from country to country (Yang et al., 2016). For instance, countries like Greece or Italy belong to Polarized Pluralist (Yang et al., 2016; Hallin & Mancini, 2004), which has been suggested to intensify the polarization process. Another possible contributor to shaping public opinion can be the abundance of evocative imagery in covering the refugee crisis. As people say; an image is worth a thousand words. The model of Schuck and Feinholdt (2015) provides space for interpretation in different ways, as the use of valence frames is not only text based but can also include visual material. Therefore, we will go a step forward to measure the impact of visual cues on people s attitudes toward the refugee crisis. Powerful imagery in the media can stimulate substantial emotional responses amongst their audiences in addition to text (van Schaik, 2016; Powell, Boomgaarden, De Swert, & de Vreese, 2015). Considering both text and visuals, we find it valuable to combine both aspects in order to identify their effects. RQ: To what extent does the way in which journalists frame the Syrian refugee crisis, as well as visual elements about the issue, affect people s emotions and attitudes? Theoretical Framework Frames Framing is a term that is used often the last few decades in communication research (Vliegenthart & van Zoonen, 2011), especially after Entman (1993) published his

6 WELCOME HOME 6 paper Toward classification of a structured paradigm. Entman s (1993) brought framing research to communication science and research of media effects (Scheufele, 1999). Framing refers to the observation on how media present a certain topic from a specific point of view or according to Entman (1993, p.52) To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described. From this definition one could say that in the way that media and journalists will choose to present a certain topic, affects the viewer s or reader s attitude (Jasperson, Shah, Watts, Faber, & Fan, 1998; Schuck & de Vreese, 2006). Or according to Gross and D Ambrosio (2004), the way that ordinary citizens understand and create an opinion about a political event is from the way that event is framed. As we read in the literature, a certain frame that will be followed is affected and driven by the values and the norms of the medium that journalists work for (Gross &Brewer, 2007). Thus, as different media outlets have different characteristics and values, they will frame differently the same event. Individuals that do not see the events with their bare eyes, and rely on the media to represent the story for them. News media are meant to provide sufficient information to the readers or viewers (Iyer, Webster, Hornsey, & Vanman, 2014; Chermak, & Gruenewald, 2006). This is another explanation about why the media have an impact on the readers or viewers of the news. Semetko and Valkenburg (2000), presented in their paper the different types of frames that are used by journalists when they write a story. News media provide far more information rather than the report of the facts (Gross & Brewer, 2007). Every event can be framed differently. For instance, crime stories use mostly a human-

7 WELCOME HOME 7 interest frame (Semetko & Valkenburg, 2000). The use of frames can help people understand and construct an opinion about an event (Schuck & de Vreese, 2006). Accordingly, below we will elaborate the importance of the visuals and how they affect people s perceptions. Valence Frames It is important for our study to introduce valence frames. As valence frames discuss events or issues in either positive or negative way (Schuck, 2009; Schuck & de Vreese, 2006; de Vreese & Boomgaarden, 2003; Isaac, & Poor, 2016; Nicolaou, Gunes, & Pantic, 2011). Frames, in general, are one tool that is used by journalists to interpret a story, as well as create assumptions on people and what they should think on a certain cause (Powell, Boomgaarden, De Swert, & de Vreese, 2015). Journalists will use different frames to emphasize points they consider newsworthy and what should and shouldn t be included in the article (Scheufele, 1999; van Gorp, 2007). Additionally, valence frames have been found that affect people perceptions in one way or another in different attitudes or solution (de Vreese & Boomgaarden 2003; Schuck & de Vreese, 2006), or in some experimental studies researchers observed that affect the decision-making of people (Isaac, & Poor, 2016; Peterson & Wilson, 1992; Buda & Zhang, 2000). Emotions Emotions in general, are considered a product of cognitive evaluations from a definite event (Gross & D Ambrosio, 2004). Thus, what people experience from an event, and how they evaluate it, can create a variety of emotions and can differ from person to person. There are different categories of emotions, for instance, valence and arousal emotions (Lerner & Keltner, 2000). In the case of this paper, we emphasize on the valence emotions. Based on the news frames we determine that different articles

8 WELCOME HOME 8 create different opinions to the readers (Price, Tewksbury, & Powers, 1997). Different news frames can trigger specific emotions towards a certain event. Thus, coverage of the same event can stimulate different emotional responses dependent on the news frame adopted (Lecheler, Bos, & Vliegenthart, 2015; Nelson, Lecheler, Schuck, & de Vreese, 2012). Readers sense different emotions based on the information that they receive (Gross & D Ambrosio, 2004; Lecheler, Schuck, & de Vreese, 2013). Emotions and Valence Frames Emotions are divided into different categories. As this study focuses on refugee crisis and a lot of stories were written on this, we have selected the emotions we want to research. The valence frames used can produce emotions in both positive or negative perspective (Lecheler, Schuck, & de Vreese, 2013). Lecheler, Bos and Vliegenthart (2015) found that immigration is most often framed negatively in the media. Islam as a threat (Vliegenthart, 2007) and negative economical and fear (Bleiker, Campbell, Hutchison & Nicholson, 2013) are the most commonly used frames from journalists in their articles about immigration. On the other hand, a lot of studies avoid to research positive emotions, as sympathy (Lecheler, Bos, & Vliegenthart, 2015). One explanation could be the fact that negative news in general are more newsworthy in comparison with positive events (Kamhawi & Grabe, 2008). Thus, in our study, we will research both aspects of valence frames and how they affect emotions. As a journalist could portray a story in both ways (van Gorp, 2005), we argue that we will identify positive or negative emotions to the readers. As it is mentioned in the paper of Schuck and Feinholdt, (2015), valence frames are part of the two-dimensional theory of emotions, as in a psychological perspective all the possible emotions that one feels can be categorized between arousal or valence (Nicolaou, Gunes, & Pantic, 2011; Mehrabian, & Russell, 1976).

9 WELCOME HOME 9 On the other side, recent studies (Bakker, van der Voordt, Vink, & de Boon, 2014; Fontaine, Scherer, Roesch, & Ellsworth, 2007; Bakker, van der Voordt, Vink, & de Boon, 2014), support that the space of emotions is not two dimensional only. Additionally, when one is conducting research about valence frames, we are not only interested in the emotions elicited but other potential consequences (Schuck & de Vreese, 2006). With all of this in mind, we can hypothesize that a story can have a different impact on reader emotions based on the frame used. Valence frames and attitudes Valence frames have an effect on emotions as we discussed above. Lecheler and de Vreese (2011) found that peoples attitude changes based on the frames used, but it has also been proven that peoples knowledge and predisposition can have a mediating effect on the frames used (Scheufele, 1999). For example, if one person has political views to a far-right wing party, it is difficult for him to have positive attitudes towards refugees. Similar results about people s predispositions found in the study of Stroud (2007), that the political beliefs can even are even related to the media that they want to expose themselves. The study of Domke, Perlmutter and Spratt (2002), found that people can actually be convinced by images and news frames, but the effect will be pronounced from the viewers existing attitudes, pre-dispositions and experiences. Therefore, considering prior attitudes is essential when measuring the impact of news frames (Bertrand & Mullainathan, 2001). To conclude the above and for further testing we formulated the hypotheses below. H1a: A journalistic article that is framed positively about the refugee crisis, will lead to more sympathetic emotions towards refugees. H1b: More sympathetic emotions regarding refugees will lead to more positive attitudes towards refugees.

10 WELCOME HOME 10 H1c: People s predispositions about Syrian refugee crisis will moderate the indirect effect between a positive journalistic article via sympathetic emotions to attitudes towards Syrian refugees. H2a: A journalistic article that is framed negatively about the refugee crisis, will lead to more antipathetic emotions towards refugees. H2b: More antipathetic emotions regarding refugees will lead to more negative attitudes towards refugees. H2c: People s predispositions about Syrian refugee crisis will moderate the indirect effect between a negative journalistic article via antipathetic emotions to attitudes towards Syrian refugees. Visuals As people say, a picture is worth a thousand words. Nowadays, most news articles, in addition to text, feature a photograph or a visual representation of the story (Boomgaarden, Boukes, & Iorgoveanu, 2016). Visual materials can impact on people s attitude and opinion of a certain topic (Zillmann, Knobloch, & Yu, 2001). Wanta and Roark (1993) discuss how photographs impact people s opinion and found that visuals were more readily remembered by media audiences. Wanta and Roark (1993) expressed that emotional pictures should be used with causation by journalists as they can have a sizable impact on readers (Brantner, Lobinger, & Wetzstein, 2011). To further support the potential power of visual information, Gadarian (2010) found that images of the 9/11 attacks stimulated audience emotions and subsequently led to a change in attitude in foreign policy. Also, Papacharissi and de Fatima Oliveira (2008), sought to identify how newspapers in the US and in the UK framed the 9/11 terrorist attack in the United States. Indeed, in their study, they recognized differences between the two countries.

11 WELCOME HOME 11 Therefore, it could be said that the way that journalists frame an issue is affected by the geographical location. When one takes a deeper look in the research about the effects of pictures or photographs in a political communications context, one can identify how powerful the visual representations are. For instance, in the research of Iyer et al. (2014), the participants that were exposed to the photographs of terrorists had stronger opinions about the dangerousness of the terrorists. Also, similar results support the salience of visuals in comparison with the text (Geise & Baden, 2014). Powell et al., (2015), found that images have a stronger effect in comparison with text in terms of what readers perceive. In the same study, visuals in general were identified to connect better the reader to the story (Powell et al., 2015). Similar results are indicated in the study of Domke, Perlmutter, and Spratt (2002), peoples' attitudes can be changed because of an image, additionally visual representation of a story can activate peoples mind for a faster decision-making process. The articles mentioned above provide explanation why in the case of the refugee crisis the recent event of the young boy, Alan Kurdi, that was found dead on the coast of Turkey (Smith, 2015), influenced a lot of people to want to know more about the refugee crisis. Therefore, we argue that a photograph, positive or negative, in a story will cause stronger positive or negative emotions when it's accompanied by the text. H3: A positive journalistic article about the refugee crisis with a positive photograph will lead to more sympathetic emotions towards the refugees than a journalistic article without a photograph. H4: A negative journalistic article about the refugee crisis with a negative photograph will lead to more antipathetic emotions towards refugees than a journalistic article without a photograph.

12 WELCOME HOME 12 Conceptual model Ultimately, we are concerned with comparing the effects of the positive emotions in contrast with the negative. In terms of marketing and advertising, previous research has shown that positive emotions produced more favorable attitudes towards the advertisement in contrast to those that elicited negative emotions (Lau-Gesk & Meyers-Levy, 2009). Soroka, Loewen, Fournier, and Rubenson, (2016) that used news articles about the war in Afghanistan as stimuli material, found that photographs can shape people s attitudes toward military intervention. The way that media frame each photograph generates different attitudes among people. Further experimental research on news framing effects (Lecheler, Bos, & Vliegenthart, 2015) discusses positive valence frames and the stronger effect that they have in comparison with the negative frames (Bos, Lecheler, Mewafi, & Vliegenthart, 2015). According to Lecheler, Bos, and Vliegenthart (2015) and Bos et al. (2015) when researching framing effects on immigration news coverage, found that positive frames may have stronger effects on opinion than negative frames. As the context of our study is about the Syrian refugee crisis we gain confidence about the theoretical implications of Lecheler, Bos, & Vliegenthart (2015), that is about immigrants. To test the effect of valence frames we formulated the following hypothesis. H5: The positive journalistic article will lead to stronger emotions than a negative journalistic article, and this effect will be more pronounced when a photograph is present. Based on the theory that is mentioned above, as well as the hypotheses, two conceptual models were created for better understanding. The general model (Schema 1) indicates the relationship between the independent, dependent and mediator variables. In addition, one can identify similarities to the study of Schuck and

13 WELCOME HOME 13 Feinholdt (2015), but in our case the model is based on the topical context of the Syrian refugee crisis. Schema 1 The conceptual model with all the relationships between the variables as they are described in the theory is presented in the Schema 2. Schema 2 Methodology Design. A lot of studies about frames and their effects are using experimental designs for their research (Powell et al., 2015; Lecheler & de Vreese, 2011; Schuck & de Vreese, 2006; Esses, Veenvliet, Hodson, & Mihic, 2008; Bos et al., 2016; de Vreese

14 WELCOME HOME 14 & Boomgaarden, 2003; de Vreese & Kandyla, 2009; Schuck & Feinholdt, 2015). The design that this study will follow is a 2 by 2 between subjects, true experimental design and will contain four experimental and one control group. We will have a pretest and a post-test in all the five conditions. The pre-test will measure the attitudes of the participants about the refugee crisis and identify their perspectives and predispositions about this topic. In the post-test, we will measure the emotions and attitudes of participants after the manipulation. The goal is to identify how the manipulations affect individual attitudes. In the experiment, participants will be randomly assigned to five different conditions regarding different stimuli that participants will be exposed to. The factors in this paper will be the valence frames as it is our independent variable. The first condition will include a positive article, the second condition will be the same positive article with a photograph. The third condition will be a negative article about the refugee crisis that we expect to cause negative, antipathetic emotions about refugees. The fourth condition will be the same article but with a photograph that will accompany it. Our last and fifth condition will be the control group (neutral article without photograph). On the schema 3 below, one can identify the full experimental design. The need of a control group is to check that our manipulation indeed worked and it serves as a baseline to compare the other conditions to (Field, 2013). t 0 t 1 Group I R O 1 X 1 O 2 (positive article) Group II R O 1 X 2 O 2 (positive article with photo) Group III R O 1 X 3 O 2 (negative article) Group IV R O 1 X 4 O 2 (negative article with photo) Group V R O 1 O 2 (control condition)

15 WELCOME HOME 15 Schema 3 It is important for this study to select two photos that are significantly different from each other, in terms of how they will spark emotions to the viewers. The same applies to the text that is used for this experiment. Therefore, we will have a pre-test of our stimuli material to test the effectiveness. Four photographs and six articles that are positive and negative towards refugees will be tested in the pre-test. Type of journalistic articles Photograph Antipathetic (negative) Sympathetic (positive) Present Absent Table 1 Syria crime wave + rapefugees not welcome photograph Syria crime wave Integration means you are Syrian and now you're Irish as well + refugees welcome photograph Integration means you are Syrian and now you're Irish as well In table 1 above we have information about our independent variables which are the journalistic articles and the photographs. Our dependent variables are the attitudes towards refugees, with emotions as mediators of the effects of media coverage on people s attitudes to refugees. Also, there is a description of the stimulus material that was used, after it was tested as part of our pre-test. In our experiment the respondents were assigned to five different conditions (positive article (N=44, 20.2%); positive article with photograph (N=40, 18.3%); negative article (N=46, 21.1%); negative article with photograph (N=42, 19.3%); control group (N=46, 21.1%)). Sample. The participants of this study were recruited via the social media network of Facebook with the use of snowball sampling as for convenience. The questionnaire

16 WELCOME HOME 16 was created with the use of Qualtrics software and was online between till In this period 218 responses were collected, with N=77, 35.3% male and N=141, 64.7% female respondents. The average age of the respondents was M=28.89, SD= The majority of them (N=83, 38.1%) were residents of Greece, followed by residents of The Netherlands (N=48, 22%) and Germany (N=25, 11.5%). Our sample was also well educated as 182 out of the total of 218 hold a Bachelor s or a Master s degree (N=182, 83.5%). The majority of the participants were students (N=98, 45%), followed by full-time employed (N=62, 28.4%) and part-time employed (N=41, 18.8%). Our sample was mostly university students, which is considered acceptable given that according to Druckman (2001) the behavior of students does not systematically differ from a non-student population. In terms of political orientation with 0=Left wing, 10=Right wing, our sample considered rather left wing (M=4.04, SD=2.29) and not that much religious (M=2.57, SD=1.82) on a scale of 0=Not religious at all to 7=Very religious. The last demographic question was about "How would you describe your interest in international affairs" with a value of 1=Not interested at all to 7=A lot interested, with the majority of the respondents expressing an interest in what is happening around them M=5.52, SD=1.24. Pre-test. In order to be certain about our stimuli material, before the actual study a pre-test is needed. We tested three negative and three positive articles, from real events that happened the last few years in Europe. Accordingly, we tested two positive and two negative photographs to indicate how photographs are perceived by people. Each one of the respondents was exposed to one positive and one negative photograph as well as, one positive and one negative article. We followed this procedure, because we did not want the respondents to be exposed to all the articles, and avoid biased answers. The survey was distributed via Facebook during the first

17 WELCOME HOME 17 week of May 2017, and the goal was to achieve a snowball effect in order to gather enough responses. In the end, we gathered 79 responses, but since some people did not know about the ongoing refugee crisis, we only considered 68 valid responses, thus we had a response rate of 86%. Among these respondents, 76.5% are female and 23.5% are male, with the 95.6% of them being in the age groups between In terms of the country that these people live in, the majority of them live in The Netherlands (63.2%), followed by Greece with 11.8%. As of political self-placement on a scale of 0=Left wing to 10=Right wing, our respondents sympathize with leftwing opinions, attitudes and perceptions (M=3.84, SD=1.85, N=68). In order to understand how people perceive different articles in our pre-test, we asked them Do you find the article as positive or negative? with possible answers, 1=Extremely positive to 5=Extremely negative. Respondents perceived the first article as more positive (see Appendix A, positive article 1) M=1.36, SD=.49, the second one had a value of M=1.43, SD=.60, and the third article had a value of M=1.5, SD=.74. Thus, no significant difference was identified. On the other hand, for the article that we wanted to trigger antipathetic emotions, we see that the first article (see Appendix A, negative article 1) had a higher value on our scale with M=4.68, SD=.48, and the second and third article had a value of M=4.45, SD=.74, and M=4.64, SD=.73, accordingly. No significant difference between the negative articles was identified, and our selection will be based on the other variables of the questionnaire (see Appendix B). Therefore, for our study we will use the first positive article as it is the one that respondents found more hopeful, accurate and they were enthusiastic about it. On the other hand, the selection of the negative article is considered as more complex. We will select the third article for our stimulus material, because respondents

18 WELCOME HOME 18 acknowledged that it is the most intense, and they felt angry and afraid when they were reading it. Finally, we will also use a neutral article (see Appendix 1, neutral article) for the control group, that describes the meaning of refugees, immigrants and other terms related to refugees. For the selection of the photograph that we will use in this study, the same 5- point scale was used. The first positive photograph in our questionnaire (see Appendix A, positive photograph 1), is considered as more positive (M=1.26, SD=.51) rather than the second (M=1.48, SD=.67). In terms of the negative photograph, both options had the same mean value (M=4.64), but the second one was indicated with a standard deviation (SD=.6), in comparison with the first one (SD=.65). Thus, it would make no difference which negative photograph we will choose. For more information about the rest of the variables see Appendix B. To summarize the above as for the selection of visual content we will use the first positive photograph from the pre-test as it is considered the most sympathetic towards Syrian refugees and seen as more accurate. For the choice of the negative photograph, we chose the second option. Respondents indicated this photograph to be more accurate. Main study. For the main study, we created questions in order to measure people's perceptions, emotions, and attitudes regarding the Syrian refugee crisis. In order to measure people s attitudes before the stimulus material, respondents had to answer a set of questions about the anti-refugees sentiments (De Vreese & Boomgaarden, 2005). After we performed the appropriate recoding in the some of the questions, we identified that there is one component with an Eigenvalue of 2.98 that explains 59.59% of the total variance, this scale is reliable with Cronbach s alpha =.83.

19 WELCOME HOME 19 After the stimuli material, a few questions regarding the emotions and attitudes towards Syrian refugees were asked to participants. In order to create scales of questions about the sympathetic emotions, we performed PCA analysis using Varimax Rotation with six items and it provided one component with an Eigenvalue of 3.97 that explains 66.1% of the total variance, also this scale is reliable with Cronbach s alpha =.89. For the antipathetic emotions, the same analysis was used with two items and one component has an Eigenvalue of 1.67 that explains 83.46% of the total variance, this scale has a Cronbach s alpha =.80. For the attitudes of people towards the Syrian refugees after the manipulation a set of six questions was included in the PCA analysis with Varimax Rotation. In the end, one component with Eigenvalue=3.81 that explains 63.54% of the variance, with Cronbach s alpha=.88. Randomization check. It is important to check the if participants of the experiment were randomly assigned successfully to the five conditions. In order to perform a randomization check, we did a chi-square test with the experimental conditions as independent and gender as a dependent variable. This relation is not significant (χ " (4)=1.59 p=.811), thus we have successful randomization in terms of gender. For other relevant variables, we tested them using one-way ANOVA with the different conditions as an independent variable. Indeed we have successful randomization for age, as our results are not significant (F(4,213)=.9, p=.466), also for education level, (F(4,213)=1.9, p=.112), political orientation (F(4,209)=.9, p=.465), and religiousness (F(4,213)=.3, p=.879). Thus, all our participants were successfully assigned randomly to the five conditions of the experiment. Manipulation check. The question to identify if the participants read the article was: What was the article about?, with possible answers: Integration of refugees that

20 WELCOME HOME 20 is the correct answer for the positive articles. The people that were exposed to the positive article answered it correctly, (90.7%), same applies to the positive article with a photograph (97.4%). Crimes that refugees committed which is the correct answer for the negative articles. In the case of the negative article 85% answered correctly the question, and 87.2% answered correctly in the case of the negative article with a photograph. Lastly, for the control group the correct response was the last one None of the above with 55.6% of the people that were exposed in the condition answered correctly, but the 28% answered the Integration of refugees. Thus, we should be conscious with the results about the control group. The next question to check if the manipulation worked, we asked participants to evaluate if the article that they read was considered as positive or negative on a 7 point Likert scale, with 0=Extremely Negative to 7=Extremely Positive. The positive conditions (first and second) had a value of M=5.3, SD=1.36 and M=5.33, SD=1.1, thus participants evaluated them as positive. In the negative conditions (third and fourth) the results were M=2.3, SD=1.66 and M=1.98, SD=1.44, this indicates that the participants of the questionnaire identified them as negative articles. In the last condition and control group the participants identified the article Neither positive nor negative (M=4.04, SD=.99), as expected. Also one-way ANOVA indicates that there is a significant difference between the groups F(4,213)=61.4, p<.001. To summarize all of the above, we safely assume that the manipulation worked as intended. Procedure. To gather enough responses for our experiment, we distributed it with an anonymous link via the social media Facebook and Twitter. Also, the same link was sent by s. After the participants clicked on the link they were informed about the anonymity of their response as well as general information about the experiment. After they agreed on the terms of the experiment, they were exposed to a set of

21 WELCOME HOME 21 demographic questions. The next step of the experiment was the randomization of the participants to one of the five conditions. The stimulus material, manipulation check and an open question about their general attitude towards Syrian refugees were the questions that participants had to answer. Lastly, a set of different emotions, positive and negative, and attitudes about the Syrian refugees were asked. At the end of the questionnaire, participants were asked if they wanted to participate in a contest for a 20 Euro voucher. After they filled in their s, respondents were debriefed about the study and the experimental condition that they were assigned to. The debriefing informed them that the article they read was factually true but used for the purpose of a media effect study and was published in the online version of a newspaper. Results In this study, we want to identify the effects of news frames regarding the Syrian refugee crisis. Positive and negative articles, with or without a photograph were our stimulus material in this experiment. To test the first hypothesis, we will use PROCESS MACRO (Hayes, 2013) model 4, with the condition of the positive article recoded (positive article=1, negative article=0, ELSE=missing values) as an independent variable, attitudes after the manipulation as dependent variable and positive emotions after the manipulation as mediator of this relationship. The regression model with the article as independent variable and emotions as dependent, is not significant F(1,88)=.93, p=.149, R 2 =.02, thus there is no direct effect between positive articles and positive emotions. Although, there is a significant effect between positive emotions and people s attitudes F(2,87)=71.37, p<.001, R 2 =.62, with the coefficient of positive emotions to have a positive effect on attitudes (b=.99, SE=.08, t(87)=11.85, p<.001, 95%[.82, 1.15]). Additionally, there is no direct effect between the positive article and the attitudes F(1,88)=.93, p=.337, R 2 =.01. Thus, we have to

22 WELCOME HOME 22 reject H1a, as we did not identify a significant effect between our key variables, a positive article does not necessarily mean that it will spark positive emotions among the readers. On the other hand, we can accept H1b, as there is an enhancing effect between positive emotions on attitudes of people towards Syrian refugees. The higher people score on positive emotions scale they have more positive attitudes towards refugees, meaning that emotions indeed affect the attitudes that people have. However, as we have seen before, the positive stimulus material did not affect positive emotions. With model 7 of PROCESS (Hayes, 2013) using 5000 bootstrap samples to reduce bias, we can add a moderator variable to our regression model and test the third scale of our first hypothesis. In our case, the use of the variable refugees sentiments (De Vreese & Boomgaarden, 2005), that describes the attitudes of people before the manipulation and the higher score to indicate more negative attitude, is considered as a moderator for the relation between the stimulus material and emotions. To test this moderator variable, with the positive article as an independent variable, positive emotions as a dependent variable, we identify that the model is significant F(3,86)=28.04, p<.001, R 2 =.49, with the coefficient of attitudes before the manipulation to have a negative effect on the positive emotions b=-.55, SE=.1, t(86)=- 5.55, p<.001, 95%[-.74, -.35], although the positive article did not predict the positive emotions b=.8, SE=.48, t(86)=1.68, p=.096, 95%[-.15, 1.75]. Additionally, the regression model about attitudes after the manipulation is significant as well, F(2,87)=71.37, p<.001, R 2 =.62, the coefficient of the positive article did not predict the attitudes of respondents after the manipulation b=-.06, SE=.19, t(87)=-.29, p=.776, 95%[-.44,.33], but the positive emotions successfully predicts the attitudes and have a positive effect on them (b=.99, SE=.08, t(87)=11.85, p<.001, 95%[.82, 1.15]). Thus,

23 WELCOME HOME 23 people s predispositions have a buffering effect on emotions of the people towards refugees. Likewise, the positive emotions have a positive effect on people s attitudes. Therefore, we accept H1c. For the second hypothesis, we will perform the same type of analysis using as independent variable the condition of the negative article recoded (positive article=0, negative article=1, ELSE=missing values), dependent variable will be again the attitudes after the manipulation and mediator will be the variable of antipathetic (i.e., negative) emotions. The overall regression model did not provide a significant direct effect between the negative article and the attitudes F(1,88)=.93, p=.337, R 2 =.11. Albeit, the model predicts attitudes F(2,87)=24.2, p<.001, R 2 =.36, based on the coefficient of emotions b=-.55, SE=.08, t(87)=-6.86, p<.001, 95%[-.72, -.4]. However, negative emotions are not predicted by the negative article F(1,88)=.59, p=.44, R 2 =.01. To summarize the above we reject H2a, as negative article does not lead to antipathetic emotions. However, we do accept H2b, because the higher people score in negative emotions the more negative attitudes they have towards Syrian refugees. Nonetheless, as with positive emotions, likewise negative emotions are not affected by the stimulus material. Thus, emotions do matter for subsequent attitudes, however, the stimulus material used in our study did not trigger these emotions. In order to test H2c, we used the recoded condition of the negative article as independent variable, attitudes after manipulation as dependent variable, negative emotions as mediator and the refugees sentiments as moderator. This regression model for the negative emotions after the manipulation is significant, F(3,86)=33.11, p<.001, R 2 =.54, but the coefficient of the negative article does not predict the negative emotions b=1.07, SE=.6, t(86)=1.78, p=.079, 95%[-.13, 2.27], although the refugees sentiments that is our moderator, successfully predicts the negative emotions (b=1.01,

24 WELCOME HOME 24 SE=.13, t(86)=7.99, p<.001, 95%[.76, 1.26]). Additionally, the regression model about the attitudes after people were exposed to the negative article is significant, F(2,87)=24.21, p<.001, R 2 =.36, the coefficient of the negative emotions has an negative effect on the negative attitudes (b=-.56, SE=.08, t(87)=-6.86, p<.001, 95%[-.72, -.4]), but the coefficient of the negative article (b=-.16, SE=.25, t(87)=-.63, p=.53, 95%[-.66,.34]), does not predict the negative attitudes. Thus, as the model is found to be significant, we accept H2c. Therefore, the relationship between a negative journalistic about Syrian refugee crisis and antipathetic emotions it is moderated by people s predispositions. This information is considered as valuable, because we understand that people s pre-dispositions have an impact on their negative emotions. If one scores high on the refugees' sentiments scale, it means that one has more antipathetic emotions towards Syrian refugees after he will read a negative article about the issue. In order to test the third hypothesis, regression analysis was performed. As we argue that a positive article with a photograph will lead to stronger emotions than a positive article without a photograph. The variable of positive emotions after the manipulation was our dependent variable and the condition of positive article with a photograph that was recoded as a dummy variable (positive article=0, positive article with photograph=1, ELSE=missing values), was used as independent variable. The ANOVA table indicates that the model is not significant F(1,82)=.01, p=.927, R 2 =.000, thus we have to reject our third hypothesis and we gain confidence in the alternative one. A positive article with a photograph does not mean that will produce stronger positive emotions rather than an article without a photograph. For the fourth hypothesis, the same procedure as described above was conducted. Negative emotions was our dependent variable and the recoded condition

25 WELCOME HOME 25 of the negative article with photograph (negative article=0, negative article with photograph=1, ELSE=missing values), as independent variable. The model is not significant as the ANOVA table illustrates F(1,86)=2.26, p=.137, R 2 =.03. Therefore, the fourth hypothesis should be rejected. We believe that an article that is negatively framed with a photograph does not necessarily generate more antipathetic emotions rather than a negative article without a photograph. For the last hypothesis, we want to identify if the effects of emotions are stronger in the positive article rather than the negative and this effect we expect to be pronounced when a photograph is present. Thus, we conducted one-way ANOVA with dependent variables the positive and negative emotions and as independent variable the recoded variable of the positive condition (positive article=1, negative article=0). We did not identify significant differences between the two conditions, for the positive emotions F(1,88)=2.12, p=.149, η " =.02, nor for the negative emotions F(1,88)=.6, p=.443, η " =.01. Thus as we did not identify significant differences between the groups we can say with certainty that we need to reject our hypothesis. There is no stronger effect between positive or negative emotions, regarding the article. Discussion and conclusion In this study, the effects of valence frames in the context of Syrian refugee crisis were examined. As we have seen in the theory, multiple publications (Lecheler, Schuck, & de Vreese, 2012; Schuck & de Vreese, 2006; van Gorp, 2005) assume the relationship between the different frames that are used by journalists and the emotions that they spark among readers. Accordingly, we have seen that emotions are not only based on the news articles that people read, but also on what people already believe and experience (Schuck & Feinholdt, 2015). Different studies also imply the importance

26 WELCOME HOME 26 of predispositions towards different subjects as they affect people (Domke, Perlmutter, & Spratt, 2002; Bertrand & Mullainathan, 2001; Stroud, 2007). For our research, we try to identify how and what are the effects of valence frames on citizens towards Syrian refugees. In this study, people s predispositions and general attitudes toward Syrian refugees, have a significant effect on their emotions. De Vreese and Boomgaarden s, (2005) refugee sentiment scale was utilized in our experiment. The information provided from this scale when it was used as a moderator in our theoretical model (Schema 2), implied that the existing attitudes affect the emotions in either positive or negative way. Thus, our model whilst using the above-mentioned scale as moderator between the news frames and emotions is significant. Consequently, one of the theoretical implications of the findings of this experiment was that people generate emotions not only based on the text or article that they read, but also on their general beliefs and experiences that they already have about a specific subject, in this case about the Syrian refugee crisis. A similar path was found by Kühne (2012), with emotions mediating the relationship between news frames and opinion formation. Likewise, the study of Lecheler, Bos and Vliengenthart (2015), discusses that valence emotions mediate the relationship between different news frame types used for covering immigrants and policy opinion. It is important to mention our finding that emotions have a significant impact on people s attitudes. People base their attitudes on the emotions that are produced when reading journalistic articles. Another aspect of our study worth mentioning, is that no difference was found between the emotions elicited when a photograph was present or absent in combination with the article. As shown in the literature (Powell et al., 2015; Iyer et al., 2014) visual cues can strengthen emotions in comparison to

27 WELCOME HOME 27 when they are absent in news articles. Similar results were identified by Wanta and Roark (1993), but our findings contrast the findings of this study and others, and is a potential limitation of this paper. It also puts emphasis on the fact that the impact of visuals might depend on the topical context and/or the specific selection of visual material being used in a study. Recent studies have shown (Bos, Lecheler, Mewafi, & Vliegenthart, 2015; Lecheler, Bos, & Vliegenthart, 2015), that the positive news frames have a stronger effect than negative news frames. However, in our analysis similar results were not established. Thus, a potential limitation might be the articles that were selected. A further limitation considered is the sample of this study. The experiment was designed carefully but it may be that a lot of people when they read a news article tend to search for keywords and not actually pay attention to the full content of the text (Liu, 2005). Also, the same information may be perceived differently per person. It could be that people s pre-dispositions which may be formed by what they have already read about the topic, affects their attitudes and emotions. In general, the timespan required to read the article was not very lengthy. Thus, it is likely people based their answers on their feelings and what they already knew on the topic. The articles were a key element of this study but in this case, they were not able to stimulate different emotions. It is possible the articles used as our stimuli material were not strong enough to be able to identify stronger effects and emotions. For this study, a convenience sample was used, appropriate for the purposes of this paper, but means that our sample was likely not representative of the population, and thus our results may have been affected. It was distributed to natives of multiple countries, many of which were not native English speakers. It is considered as more difficult for a non-native speaker to understand fully a journalistic article that is not in

28 WELCOME HOME 28 his/her native language and therefore is likely to have a lesser effect on their emotions. Therefore, this is likely the main limitation to our study has considerable implications for our findings. For future research, that could solve some of the issues in our study, is to conduct this experiment in a more controlled environment. Ultimately, a larger selection of more emotional images and articles to identify stronger effects and a more representative sample may improve the validity of research results. This paper provides practical implications for journalists and producers of content. News articles can influence people s attitudes and perceptions. Therefore, professionals should be aware of the potential effects of their work and should use this influence wisely. Secondly, people say that a picture is worth a thousand words, but our findings in this study do not reflect this. They indicate that visual content that accompanies a story is not necessarily a story by itself. Thus, journalists should focus on the story and not only visual materials. Thirdly, an article that uses provocative language in terms of vocabulary, is obviously more attention grabbing, but a journalist cannot spark emotions always based only on that. This is due to people s existing experiences and beliefs which influence people s emotions, and to some extent their attitudes. To summarize, the content of an article is important and professionals should be conscious that they can influence attitudes. This study provides insights about the importance of valence frames used by journalists. To conclude, we have found that people have certain predispositions and attitudes towards Syrian refugees, and to an extent the crisis in Europe. Also, sympathetic or antipathetic emotions affect attitudes towards Syrian refugees. However, it seems that ultimately the effect of an article may be reliant on what people have already been exposed to regarding the subject and the attitudes they have

29 WELCOME HOME 29 already formed. A journalistic article can spark people s emotions in relation to the attitudes they already hold towards refugees. Our study adds to the growing body of literature on refugee news coverage and allows us to have a better understanding of the effects of news frames about the ongoing Syrian refugee crisis and the moderating influence of individuals predispositions.

30 WELCOME HOME 30 References Bakker, I., van der Voordt, T., Vink, P., & de Boon, J. (2014). Pleasure, Arousal, Dominance: Mehrabian and Russell revisited. Current Psychology, 33(3), Bausells, M., & Shearlaw, M. (2017). Poets speak out for refugees: 'No one leaves home, unless home is the mouth of a shark'. the Guardian. Retrieved 14 April 2017, from speak-out-for-refugees- BBC (2011, March 18). Middle east unrest: Three killed at protest in Syria. BBC Middle East. Retrieved from BBC (2015, December 2). What is Islamic state? BBC Middle East. Retrieved from BBC (2016, March 11). Syria: The story of the conflict. BBC Middle East. Retrieved from Bertrand, M., & Mullainathan, S. (2001). Do People Mean What They Say? Implications for Subjective Survey Data. American Economic Review, 91(2), Bleiker, R., Campbell, D., Hutchison, E., & Nicholson, X. (2013). The visual dehumanisation of refugees. Australian Journal Of Political Science, 48(4), Boomgaarden, H., & Vliegenthart, R. (2007). Explaining the rise of anti-immigrant parties: The role of news media content. Electoral Studies, 26(2),

31 WELCOME HOME 31 Boomgaarden, H. G. (2007). Framing the others: news and ethnic prejudice Boomgaarden, H., & Vliegenthart, R. (2009). How news content influences antiimmigration attitudes: Germany, European Journal Of Political Research, 48(4), Boomgaarden, H.G., Boukes, M., Iorgoveanu, A.,. (2016). Image versus text: How newspaper reports affect evaluations of political candidates. International Journal of Communication: IJoC, 10, 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, Brantner, C., Lobinger, K., & Wetzstein, I. (2011). Effects of Visual Framing on Emotional Responses and Evaluations of News Stories about the Gaza Conflict Journalism & Mass Communication Quarterly, 88(3), Buda, R., & Zhang, Y. (2000). Consumer product evaluation: the interactive effect of message framing, presentation order, and source credibility. Journal Of Product & Brand Management, 9(4), Cheng, L., Igartua, J., Palacios, E., Acosta, T., & Palito, S. (2010). Framing Immigration News in Spanish Regional Press. International Migration.

32 WELCOME HOME 32 Chermak, S., & Gruenewald, J. (2006). The Media s Coverage of Domestic Terrorism. Justice Quarterly, 23(4), Davies, G. (2016). Seven migrants arrested after a homeless man was set on fire. (2016). Mail Online. Retrieved 9 May 2017, from De Swert, K., Schacht, L., & Masini, A. (2015). More than Human Tragedy? A Quantitative Comparison of Newspaper Coverage on Irregular Migration and Lampedusa in Five European Countries. Italian Studies, 70(4), De Vreese, C., & Boomgaarden, H. (2005). Projecting EU Referendums. European Union Politics, 6(1), De Vreese, C., & Kandyla, A. (2009). News Framing and Public Support for a Common Foreign and Security Policy. JCMS: Journal Of Common Market Studies, 47(3), De Vreese, C., & Boomgaarden, H. (2003). Valenced news frames and public support for the EU. Communications, 28(4). Domke, D., McCoy, K., & Torres, M. (1999). News Media, Racial Perceptions, and Political Cognition. Communication Research, 26(5), Domke, D., Perlmutter, D., & Spratt, M. (2002). The primes of our times?. Journalism, 3(2),

33 WELCOME HOME 33 Druckman, J. (2001). On the Limits of Framing Effects: Who Can Frame?. The Journal Of Politics, 63(4), Entman, R. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal Of Communication, 43(4), Esses, V., Veenvliet, S., Hodson, G., & Mihic, L. (2008). Justice, Morality, and the Dehumanization of Refugees. Social Justice Research, 21(1), Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage. Fontaine, J., Scherer, K., Roesch, E., & Ellsworth, P. (2007). The World of Emotions is not Two-Dimensional. Psychological Science, 18(12), Gadarian, S. (2010). The Politics of Threat: How Terrorism News Shapes Foreign Policy Attitudes. The Journal Of Politics, 72(2), Geise, S., & Baden, C. (2014). Putting the Image Back Into the Frame: Modeling the Linkage Between Visual Communication and Frame-Processing Theory. Communication Theory, 25(1), Gross, K., & Brewer, P. (2007). Sore Losers: News Frames, Policy Debates, and Emotions. The Harvard International Journal Of Press/Politics, 12(1),

34 WELCOME HOME 34 Gross, K., & D'Ambrosio, L. (2004). Framing Emotional Response. Political x Psychology, 25(1), Hall, A. (2016). Germany's migrant crisis: Refugees committed more than 200,000 crimes last year. Express.co.uk. Retrieved 9 May 2017, from Germany-migrant-crisis-last-year Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and politics. Cambridge university press. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press. Hamilton, M. (2016). Hundreds of Syrians in UK arrested over string of offences including rape and child abuse. (2016). The Sun. Retrieved 9 May 2017, from Hosford, S. (2017). 'Integration means you are Syrian and now you're Irish as well'. TheJournal.ie. Retrieved 9 May 2017, from Jan2017/ Isaac, M., & Poor, M. (2016). The sleeper framing effect: The influence of frame valence on immediate and retrospective judgments. Journal Of Consumer Psychology, 26(1), Iyer, A., Webster, J., Hornsey, M., & Vanman, E. (2014). Understanding the power of the picture: the effect of image content on emotional and political responses

35 WELCOME HOME 35 to terrorism. Journal Of Applied Social Psychology, 44(7), Jasperson, A., Shah, D., Watts, M., Faber, R., & Fan, D. (1998). Framing and the Public Agenda: Media Effects on the Importance of the Federal Budget Deficit. Political Communication, 15(2), Kamhawi, R., & Grabe, M. (2008). Engaging the Female Audience: An Evolutionary Psychology Perspective on Gendered Responses to News Valence Frames. Journal Of Broadcasting & Electronic Media, 52(1), Kühne, R. (2012, May). Political news, emotions, and opinion formation: Toward a model of emotional framing effects. In Annual Conference of the International Communication Association (ICA), Phoenix, AZ. Lau-Gesk, L., & Meyers-Levy, J. (2009). Emotional Persuasion: When the Valence versus the Resource Demands of Emotions Influence Consumers Attitudes. Journal Of Consumer Research, 36(4), Lecheler, S., & de Vreese, C. (2011). Getting Real: The Duration of Framing Effects. Journal Of Communication, 61(5), Lecheler, S., Bos, L., & Vliegenthart, R. (2015). The Mediating Role of Emotions. Journalism & Mass Communication Quarterly, 92(4),

36 WELCOME HOME 36 Lecheler, S., Schuck, A., & de Vreese, C. (2013). Dealing with feelings: Positive and negative discrete emotions as mediators of news framing effects. Communications - The European Journal Of Communication Research, 38(2). Lerner, J., & Keltner, D. (2000). Beyond valence: Toward a model of emotionspecific influences on judgement and choice. Cognition & Emotion, 14(4), Liu, Z. (2005). Reading behavior in the digital environment. Journal Of Documentation, 61(6), London, K. (2017). Why do refugees and migrants come to Europe, and what must be done to ease the crisis?. Telegraph.co.uk. Retrieved 25 April 2017, from Mehrabian, A., & Russell, J. (1976). An approach to environmental psychology (1st ed.). Cambridge, Mass.: M.I.T. Press. Mustefa, Z. (2017). Syrian refugees get a warm welcome in Scotland. Aljazeera.com. Retrieved 9 May 2017, from Nelson, T. E., Lecheler, S., Schuck, A. R., & De Vreese, C. (2012, March). Framing effects on values, emotions, and attitudes. In Trabajo presentado en 2012

37 WELCOME HOME 37 Meeting of the Western Political Science Association. Portland, ee. uu.: Western Political Science Association. Nicolaou, M., Gunes, H., & Pantic, M. (2011). Continuous Prediction of Spontaneous Affect from Multiple Cues and Modalities in Valence-Arousal Space. IEEE Transactions On Affective Computing, 2(2), Papacharissi, Z., & de Fatima Oliveira, M. (2008). News Frames Terrorism: A Comparative Analysis of Frames Employed in Terrorism Coverage in U.S. and U.K. Newspapers. The International Journal Of Press/Politics, 13(1), Peterson, R., & Wilson, W. (1992). Measuring customer satisfaction: Fact and artifact. Journal Of The Academy Of Marketing Science, 20(1), Powell, T., Boomgaarden, H., De Swert, K., & de Vreese, C. (2015). A Clearer Picture: The Contribution of Visuals and Text to Framing Effects. Journal Of Communication, 65(6), Price, M., Gohdes, A., & Ball, P. (2014). Updated statistical analysis of documentation of killings in the Syrian Arab Republic. Human Rights Data Analysis Group, Geneva, Price, V., Tewksbury, D., & Powers, E. (1997). Switching Trains of Thought. Communication Research, 24(5),

38 WELCOME HOME 38 Refugee doctors to help staff shortages in North East hospitals. (2017). ITV News. Retrieved 9 May 2017, from 18/refugee-doctors-to-help-staff-shortages-in-north-east-hospitals/ Ruz, C. (2015). The battle over the words used to describe migrants. (2015) BBC News. Retrieved 19 May 2017, from Scherer, K. (2005). What are emotions? And how can they be measured?. Social Science Information, 44(4), Scheufele, D. (1999). Framing as a theory of media effects. Journal Of Communication, 49(1), Schuck, A. R. T. (2009). Referendum campaign dynamics: news media, campaign effects and direct democracy Amsterdam: The Amsterdam School of Communications Research (ASCoR) Schuck, A., & de Vreese, C. (2006). Between Risk and Opportunity. European Journal Of Communication, 21(1), Schuck, A. R., & Feinholdt, A. (2015). News framing effects and emotions. Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource. Semetko, H., & Valkenburg, P. (2000). Framing European politics: a content analysis of press and television news. Journal Of Communication, 50(2),

39 WELCOME HOME 39 Smith, H. (2015, September 2). Shocking images of drowned Syrian boy show tragic plight of refugees. The Guardian. Retrieved from Soroka, S., Loewen, P., Fournier, P., & Rubenson, D. (2016). The Impact of News Photos on Support for Military Action. Political Communication, 33(4), Stroud, N. (2007). Media Use and Political Predispositions: Revisiting the Concept of Selective Exposure. Political Behavior, 30(3), UNHCR 2016: Syria Regional Refugee Response Retrieved from php?id=122 Van Gorp, B. (2005). Where is the Frame?. European Journal Of Communication, 20(4), Van Gorp, B. (2007). The Constructionist Approach to Framing: Bringing Culture Back In. Journal Of Communication, 57(1), Van Schaik, R. (2016). Visual representations of (Syrian) refugees in European newspapers. Vliegenthart, R. (2007). Framing immigration and integration: Facts, parliament, media and anti-immigrant party support in The Netherlands. Amsterdam, The Netherlands: Free University

40 WELCOME HOME 40 Vliegenthart, R., & van Zoonen, L. (2011). Power to the frame: Bringing sociology back to frame analysis. European Journal of Communication, 26(2), Wanta, W., & Roark, V. (1993). Cognitive and affective responses to newspaper photographs. Wirth, T. (2016). The european Union s framing of the european refugee crisis. Elon Journal. 45. Yang, J., Rojas, H., Wojcieszak, M., Aalberg, T., Coen, S., & Curran, J. et al. (2016). Why Are Others So Polarized? Perceived Political Polarization and Media Use in 10 Countries. Journal Of Computer-Mediated Communication, 21(5), Yazgan, P., Utku, D. E., & Sirkeci, I. (2015). Syrian crisis and migration. Migration Letters, 12(3), Zillmann, D., Knobloch, S., & Yu, H. (2001). Effects of Photographs on the Selective Reading of News Reports. Media Psychology, 3(4),

41 WELCOME HOME 41 Appendix A Pre-Test stimulus material Positive text 1 Integration means you are Syrian and now you're Irish as well' A welcome reception will held for Syrian refugees in Blanchardstown today, with its organiser saying both sides can help the integration process. Shaykh Dr. Umar Al-Qadri, Head Imam of the Islamic Centre, said that the Irish nation is known for its generosity and warm welcome but this can change if the refugees do not integrate in the society. He told TheJournal.ie that Irish people can welcome refugees in practical ways. Practically speaking, there are already some in Ireland like community groups and church groups, who are doing great things. I think local people could start meetings in churches, GAA halls, community centres you could welcome them into your house for dinner or show them the local shopping centre. Al-Qadri says real integration can make Irish society stronger and ward off extremism. If people wanted to be proactive, they could contact us here in the mosques or the refugee centres, because that s where refugees spend their days. That is important because integration doesn t happen in centres, it happens in the streets. Personally, I m not aware of the government integration procedure and that is worrying. They should have engaged with the likes of ourselves. We want these people to know they are welcome here. Irish society is a plural society and we want that to continue, but we also want them to know that they have to give back to that community that is welcoming them.

42 WELCOME HOME 42 I think seeing that welcome will inspire refugees and show them that integration does not mean you have to leave your identity behind. It means you are Syrian and now you re Irish as well. I try to teach them that the Prophet says you must love your homeland and love the country that has welcomed you twice as much. The event will take place at 1pm in the Islamic Centre of Ireland today. Each Syrian refugee family will also be introduced to a host family who will support them in their resettlement and integration in Ireland. The Syrian refugees will also be presented with gifts and donations. Source: Hosford, S. (2017). 'Integration means you are Syrian and now you're Irish as well'. TheJournal.ie. Retrieved 9 May 2017, from Positive text 2 Syrian refugees get a warm welcome in Scotland Glasgow and Isle of Bute, Scotland - For newly arrived Syrians in Scotland, Munir Emkideh is a man to know. He knows how to get things done. Whether delivering free furniture or filling out forms, the 52-year-old is always on call to help the refugees settle into their new homes. In the last three years, he has become renowned for setting up a small network through which he distributes donated items to refugees and assists them with everyday tasks. After arriving in Scotland in 2012, Emkideh, who is from the Syrian city of Deraa, received his refugee status and has been at the core of many community-based projects since.

43 WELCOME HOME 43 "When I first came here, I didn't know anything when I wanted to do simple day-today things, so that was the moment I decided to help others," he explains. "Small things can make the biggest difference." Last year, the charity launched "Family Nest", a project aimed at bringing Syrian families together from across Scotland to help them get back on their feet again. "We try to promote creative writing and other opportunities to engage them. They come from well-educated backgrounds and we have to find out how we can help them use their skills," Sherifi says. The project consists of weekly workshops aimed at mixing Scottish and Syrian cultures together through food, arts and dance. Most recently, MIN hosted a "Burns Supper Night" to commemorate the famous Scottish poet Robert Burns. A key element of the event was the trademark haggis dish, with a Syrian twist. "We had Syrian haggis, Algerian haggis and even Lebanese haggis and everyone loved the various versions of the dish," she says. "All members of the community came together. Cooking is a huge part of Arab culture, so it was amazing to make that connection through cuisine." Sherifi believes Scotland will always be more welcoming than elsewhere. "If the war stops, we would all love to go back and till we can, I would like to thank the Scottish people for their warm welcome and hospitality," he says. "They are very nice people." Source: Mustefa, Z. (2017). Syrian refugees get a warm welcome in Scotland. Aljazeera.com. Retrieved 9 May 2017, from html

44 WELCOME HOME 44 Positive text 3 Refugee doctors to help staff shortages in North East hospitals A refugee doctor who was forced to flee Afghanistan is now set to work in hospitals in the North East. It is part of a new scheme which aims to fill staff shortages in the NHS. Until recently, Ahmad Zia Baluch was working as a doctor in a children's hospital in Kabul, the capital of war-torn Afghanistan. While completing his medical training as a specialist in paediatric orthopaedics, Ahmad supplemented his income as an interpreter for NATO forces - and that put him in the Taliban's line of fire. Targeted as a spy, the 35-year-old fled with his wife and three small children, making his way to England last summer. After claiming asylum in London, Ahmad and his family were bussed to a hostel in Birmingham before being allocated accommodation in Middlesbrough. Nine months on, he is part of a pilot project on Teesside with two main objectives: to provide humanitarian support for refugee doctors; and to bolster the under pressure NHS. Ahmad is one of the first cohort of 12 health care professionals to be given places on the Refugees Programme for Overseas Doctors a pioneering partnership between North Tees and Hartlepool NHS Foundation Trust, and a charity called Investing in People and Culture. At the same time, Professor Jane Metcalf, Deputy Medical Director of North Tees and Hartlepool NHS Foundation Trust, had identified a need to support refugees who had a background in healthcare in their own countries.

45 WELCOME HOME 45 Following talks with Professor Namita Kumar, Postgraduate Dean for Health Education England in the North-East and Cumbria, funding was found to support the experimental programme by covering examination fees and administration costs. Bini Araia, co-founder of Investing in People and Culture, came to Britain 16 years ago as a political refugee from Eritrea, in northeast Africa, and is similarly optimistic about the value of the pilot. He said: It is very positive because it helps those doctors who have fled their own countries, and it is helping the NHS too. It is a win-win." Source: Refugee doctors to help staff shortages in North East hospitals. (2017). ITV News. Retrieved 9 May 2017, from 18/refugee-doctors-to-help-staff-shortages-in-north-east-hospitals/ Negative text 1 Seven migrants arrested after a sleeping homeless man was set on fire on Christmas Eve at a Berlin subway station Seven migrants have been arrested after a homeless man was set on fire on Christmas Eve at a Berlin subway station. The men, aged 15 to 21, are suspected of torching the victim as he slept on a platform bench at Schönleinstraße station. Six of the suspects are Syrian and one is Lybian and an attempted murder investigation has been launched. Thomas Neuendorf, the vice chief of the press office at Berlin Police, told Bild that detectives believe the 21-year-old to have been the main perpetrator. He, along with his alleged accomplices, were arrested on yesterday. After torching the homeless man, the suspected assailants hopped onto a train to make their

46 WELCOME HOME 46 getaway. CCTV footage captured them watching and laughing as the man was burning on the platform. It appeared they were celebrating, according to local press. Remarkably, the victim escaped almost unscathed, thanks to the rapid intervention of several witnesses. Source: Davies, G. (2016). Seven migrants arrested after a homeless man was set on fire. (2016). Mail Online. Retrieved 9 May 2017, from Negative text 2 Germany's migrant crisis: Refugees committed more than 200,000 crimes last year Refugees in Germany committed 92,000 more offenses in the country last year than in 2014, according to official figures. The leaked report comes just weeks after New Year's Eve sex attacks in the country, which has taken in more than one million people during the current migrant crisis. Officials also warned of a "growing trend" of terror suspects mingling with genuine refugees. The data comes from 13 of the 16 states which make up the Federal Republic. Eighteen percent of the crimes committed by migrants were assaults and robberies - with less than one percent sex crimes. The number of criminal offenses rose to 208, more than in the previous year.

47 WELCOME HOME 47 The nature of the offenses appears to be mostly minor: 28,712 cases of riding on public transport without paying the fare, 52,167 incidents of forging paperwork in a bid to get money, 85,035 cases of theft, mostly shoplifting - nearly double those of Assaults, robberies and what Germany classes as "predatory extortion" and "offenses against person freedom," including threatening behavior, doubled over 2014 with 36,010 cases, accounting for 18 percent of the crime total. Logged last year were 1,688 cases of sexual offenses, including against children, including 458 rapes or acts of "sexual coercion." Hamburg, Bremen and North Rhine-Westphalia, three states governed by centre-left SPD parliaments, did not deliver refugee criminal data to the BKA. That means the events in Cologne on December 31, when hundreds of women were sexually molested or robbed by marauding gangs of immigrant men, are not included in these statistics. The report goes on to state that there were 240 attempted murders by immigrants in and in two-thirds of all cases, perpetrators and victims were of the same nationality. One German was murdered, 27 immigrants were killed by other immigrants. Syrians are officially listed as making up the bulk of asylum seekers - 48 percent - with them being suspected of 24 percent of the crime. As to terror suspects hiding among genuine refugees, the report stated that there were 266 instances individuals suspected of being "fighters and members of terrorist organizations abroad." Eighty were ruled out, 186 cases are still being probed. The report called the infiltration of militants into the country "a growing trend."

48 WELCOME HOME 48 Source: Hall, A. (2016). Germany's migrant crisis: Refugees committed more than 200,000 crimes last year. Express.co.uk. Retrieved 9 May 2017, from Negative text 3 SYRIA CRIME WAVE Hundreds of Syrians in the UK arrested over string of offenses including rape and child abuse Hundreds of Syrians in Britain were arrested last year for a string of offenses including rape, death threats, and child abuse. Police forces have revealed the figures which show 897 were held in England and Wales. Our revelation comes after Northumbria Police were last month accused of a cover-up over an alleged attack by Syrians on two teenage girls in Newcastle. One of the suspects was being followed by the BBC for a report on how his family had settled in the UK. In Northumbria, the force admitted that last year the latest available figures they arrested a total of 16 Syrians. This tally of arrests included two alleged sex offenses, six claims of violence assaults and three of public disorder. In London, 238 arrested included two for rape, three for death threats and seven for sex offenses. And Nottinghamshire cops nicked one for sexual activity with an underage girl.

49 WELCOME HOME 49 There have been concerns that refugees allowed to settle in the UK can find it difficult to assimilate into the country. The issue has been highlighted in Europe, where Germany has been rocked by attacks including the machete murder of a woman and a suicide bombing outside a restaurant carried out by Syrian refugees. Source: Hamilton, M. (2016). Hundreds of Syrians in UK arrested over string of offences including rape and child abuse. (2016). The Sun. Retrieved 9 May 2017, from Neutral text The battle over the words used to describe migrants The word migrant is defined in Oxford English Dictionary as "one who moves, either temporarily or permanently, from one place, area, or country of residence to another". It is used as a neutral term by many media organisations but there has been criticism of that use. News website Al-Jazeera has decided it will not use migrant and "will instead, where appropriate, say refugee". An online editor for the network wrote: "It has evolved from its dictionary definitions into a tool that dehumanises and distances, a blunt pejorative." A Washington Post piece asked if it was time to ditch the word. There are some who dislike the term because it implies something voluntary but that it is applied to people fleeing danger. A UN document suggests: "The term 'migrant' should be understood as covering all cases where the decision to migrate is taken freely by the individual concerned, for reasons of 'personal convenience' and without intervention of an external compelling factor."

50 WELCOME HOME 50 "Migrant used to have quite a neutral connotation," explains Alexander Betts, director of the Refugee Studies Centre at Oxford University. "It says nothing about their entitlement to cross that border or whether they should be." But some people believe that the word has recently developed a sour note. It is being used to mean "not a refugee", argues Betts. A refugee, according to the 1951 Refugee Convention, "is any person who, owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his/her nationality and is unable, or owing to such fear, is unwilling to avail himself/herself of the protection of that country". But there would be many people who would be wary of labeling someone a refugee until that person has gone through the legal process of claiming asylum. In the UK, and other places, claims for "refugee status" are examined before being either granted or denied. "The moment at which they can officially say whether they are refugees or economic migrants is the moment at which the EU state that is processing their claim makes its decision," says Tim Stanley, historian, and columnist for the Daily Telegraph. "I am not questioning the validity of their narrative, I am not saying that anyone was lying about it. I am saying that it is down to the state in which they have arrived to define what they are." Asylum seeker refers to someone who has applied for refugee status and is waiting to hear the result of their claim. But it is also often used about those trying to get to a particular country to make a claim. The word asylum is very old indeed having first been used in 1430 to refer to "a sanctuary or inviolable place of refuge and protection

51 WELCOME HOME 51 for criminals and debtors, from which they cannot be forcibly removed without sacrilege". Source: Ruz, C. (2015). The battle over the words used to describe migrants - BBC News. [online] BBC News. Available at: [Accessed 19 May 2017]. Positive photo 1 Positive photo 2

52 WELCOME HOME Negative photo 1 Negative photo 2 52