Understanding the Use and Effects of Fear About Geert Wilders, Twitter and the Climate of Fear
Daniëlle de Boer
Graduate School of Communication - University of Amsterdam
Author Note
Master’s Thesis by Daniëlle de Boer (10200495), Graduate School of Communication, Master Political Communication, University of Amsterdam, Correspondence: danielle.deboer@student.uva.nl, Supervisor: Yphtach Lelkes, Date: June 24, 2015.
Abstract
Right-wing populist parties are thought to utilize perceived threats in society to create fear amongst citizens. These parties are thought to use fear appeals in order to gain political support. This study examined tweets by the Dutch politician Geert Wilders in order to find out how he contributes to the climate of fear in the Netherlands and to what extent he can create fear amongst citizens about the emerging topic of the Islamic State. Both a content analysis and an experiment were conducted to answer these questions. The findings indicate that Geert Wilders does not use fear appeals in his tweets despite the fact that they do contain a considerable amount of threats, which are mostly coming from the Islam or Islamic
terrorism. These threats, however, do not create fear amongst citizens and do not make people more likely to support Geert Wilders’ party, the PVV. Though, Geert Wilders does have a powerful Twitter network, which allows his messages to be spread amongst a lot of people and even the media show interest in his tweets. These tweets by themselves cannot create fear, but the considerable amount of attention given to his tweets and the reach of his tweets can contribute to the climate of fear in the Netherlands. These results, however, go against
previous findings, since other studies indicate that politicians successfully use fear in order to gain political support. Also, questions arise about the possible misuse of fear. Therefore, more research is necessary in order to get a better understanding of the use and effects of fear. The (mis)use of fear by politicians will keep its societal relevance for people around the world, as everyone has to deal with politics in one way or another.
Understanding the Use and Effects of Fear About Geert Wilders, Twitter and the Climate of Fear
In today’s society people are often confronted with fearful information, which makes them believe they live in a dangerous world (Altheide, 1997). Though, people do not base their feelings on personal experience but on indirect exposure to threats (Slone, 2000). The media can play a role in exposing people to threats, but politicians are also important in this process (Altheide, 2003; De Castella, McGarty, & Musgrove, 2009; Glassner, 2004).
Politicians who confront people with threats have the capacity to create fear amongst citizens (De Castella et al., 2009). Fear is often created in order to get political support, such as for retaliatory and military actions (Huddy, Feldman, Taber, & Lahav, 2005). It has already been proven that former president George W. Bush of the United States has successfully used fear in his communications in order to get political support after 9/11. The use of fear,
however, has only been found successful when politicians make use of the more traditional communication methods, like speeches (De Castella et al., 2009). Relatively new
communication channels politicians make use of, such as Twitter, have not been studied before despite the fact that Twitter is widely used by politicians to directly communicate with voters (Broersma, den Herder, & Schohaus, 2013).
In Europe, right-wing populist parties are thought to utilize perceived threats in society to create fear amongst citizens (Schmuck & Matthes, 2014). Some right-wing populist
politicians are immensely popular on Twitter, they have powerful Twitter-networks and their tweets have a lot of impact (Volkskrant, 2014). But this radical right-wing influence on Twitter has also raised some concerns amongst scholars (Blanquart & Cook, 2013), because the use of fear is not always well substantiated and occasionally misused by politicians for political gains (De Castella et al., 2009; Glassner, 2004). Though, it remains unclear if
and if Twitter is used by them to spread threats and fearful messages in order to get political support.
In this study, I will look into the use of fear by right-wing populist politicians and how Twitter can facilitate them in spreading fearful messages. Then I will look into theories about fear appeals, how fear is used and what the consequences of using fear are. After these general theories, the specific case of a Dutch right-wing populist, Geert Wilders, will be discussed. This study will analyze Wilders’ tweets of the last half-year in order to find out if he makes use of fear on Twitter, how these tweets influence people and if they can create fear and political support. The results of this study indicate that Wilders refers to threats in about a quarter of his tweets. However, his tweets do not contain fear appeals and they can neither create fear nor political support.
Theoretical Framework Right-Wing Populism and Fear
Right-wing populist parties have gained more electoral power since the last couple of decades (Rydgren, 2007; Schuermans & De Maesschalck, 2010). In Europe, there are a lot of different right-wing populist parties, like the Front National in France, the Lega Nord in Italy, the Freiheitliche Partei Österreichs in Austria, the Vlaams Belang in Belgium and the Partij Voor de Vrijheid in the Netherlands. These parties have two things in common. They are on the right end of the political spectrum and they are against the current establishment. These parties are not all equally extreme but their ideas correspond to each other; they are anti-EU, anti-immigration, anti-Islam (Akkerman, 2005) and aim at “strengthening the nation by making it more ethnically homogeneous and by returning to traditional values” (Rydgren, 2007, p. 242). In order to achieve these results, right-wing populists are thought to promote negative thoughts about minority groups and to utilize perceived threats in society in order to
get political support (Schmuck & Matthes, 2014). Schuermans and De Maesschalck (2010) even state “that it is generally accepted that extreme right-wing politicians play to everyday fears” (p. 248) and that fear is what the (Belgian) right-wing populist party is based upon.
Right-Wing Populism and Twitter
Twitter has become an important part of political communication (Enli & Skogerbø, 2013). Twitter is a micro-blogging platform created in 2006 and over the years more and more people started using the social media platform, resulting in 302 million monthly active users and 500 million tweets sent per day (Twitter, 2015a). Twitter can diffuse messages more directly than other social media platforms (Parmelee & Bichard, 2011), which could be the reason why Twitter is so popular amongst politicians. Politicians know that Twitter provides them with an unfiltered communication channel where they can directly
communicate with voters, which gives them more power than when they had to communicate through traditional media (Broersma et al., 2013). Twitter can be used to create visibility, to mobilize voters and to create a dialogue with them (Enli & Skogerbø, 2013).
But Twitter is also used to reinforce extremist ideas, while the social media platform can spread misinformation and hateful messages (Blanquart & Cook, 2013). Blanquart and Cook (2013) even state that “Twitter can re-enforce extremist policies as tweets allow authors to send information without explanation and directly control perception, thus assisting
extremists to transmit dogma without the accompanying reasoning” (p. 3). Therefore, statements on Twitter can be made without any justification, which makes Twitter an interesting medium to spread extreme ideas on, especially for (extreme) right-wing populist politicians (Gleason, 2013). It therefore seems like Twitter is the ideal platform for right-wing populists to “play to everyday fears” (Schuermans & De Maesschalck, 2010, p. 248) and to spread perceived threats about the EU, immigration and the Islam on. But it remains a
question if extreme-right populist politicians also use Twitter for this purpose.
Fear Appeals and the Extended Parallel Process Model
Fear is one of the eight primary emotions a person can experience (Plutchik, 1991). Fear can arise after people are confronted with threatening messages (Witte, 1992) and it is a negative emotion that can create unpleasant feelings (Coon & Mitterer, 2012). Fear can be used in the form of a fear appeal, which is defined as “a persuasive message that attempts to arouse the emotion fear by depicting a personally relevant and significant threat and then follows this description of the threat by outlining recommendations presented as feasible and effective in deterring the threat” (Witte, 1994, p. 114). In other words, it is a threatening message, which is created to change attitudes and behaviors (De Villiers, 2008).
There are different fear appeal theories, but they have caused a lot of inconsistent outcomes across studies. Therefore, Witte (1992) came up with her own version of a fear appeal theory, the Extended Parallel Process Model, which is now one of the most well-known fear appeal theories (De Villiers, 2008). The model explains two core components of a fear appeal; threat and efficacy. These two components consist of their own two elements. The threat consists of severity, “the magnitude of harm expected from the threat” and
susceptibility, “the degree to which one feels at risk for experiencing the threat” (Witte, 1994, p. 114). Efficacy consists of response efficacy, “one’s beliefs about whether the
recommended response works in averting the threat”, and self-efficacy, “one’s beliefs about his or her ability to perform the recommended response” (Witte, 1994, p. 114).
The Effects of Fear Appeals
Fear appeals are used in order to get attention and to change behaviors
could happen if they do not change their behavior or attitude. The effects of fear appeals, as predicted by the Extended Parallel Process Model, are two-sided. Fear appeals can influence the amount of fear, and the type of reaction (De Villiers, 2008). For a fear appeal to be successful it needs to satisfy two conditions. Firstly to create fear, there needs to be a high level threat that consists of both severity and susceptibility. Only a high level threat can create fear according to the Extended Parallel Process Model (Witte, 1992). Secondly, to influence the type of reaction, there needs to be efficacy, which must consist of both response efficacy and self-efficacy. As a result, the Extended Parallel Process Model predicts that high levels of threat and efficacy cause people to want more control over the threat, while lower levels of efficacy will cause people to ignore the threat (De Villiers, 2008).
The type of reactions to fear appeals are examined in different settings, like in health campaigning (De Hoog, Stroebe, & De Wit, 2005; Peters, Ruiter, & Kok, 2014). One of the best-known health campaigns that makes use of fear appeals is the anti-smoking campaign. Though, studies in this field find inconsistent results. While some authors find that people want to control the health threat, others find that people ignore the threat or even develop negative behaviors. (Borland et al., 2009; Brown & Locker, 2009; Hansen, Winzeler, & Topolinski, 2010).
In the political field, however, it is found that the use of emotional appeals can help politicians to achieve their goals (Brader, 2005). It is often found that fear appeals have a positive effect (the intended effect) on attitudes and behavior (De Castella et al., 2009). The most found effects show that people become more ethnocentric, xenophobic, prejudiced and intolerant. They are more receptive to information about the threat, are willing to sacrifice their civil liberties and to support government actions and policies against the threat.
(Feldman & Stenner, 1997; Huddy, Feldman, Capelos, & Provost, 2002; Huddy et al., 2005). This support has also been visible in the United States when U.S. citizens supported military
actions in order to deter a terrorist threat (Huddy et al., 2005). The ‘War on Terror’ under the Bush administration is sometimes seen as a strategy to create fear amongst citizens in order to get support for war related actions (De Castella et al., 2009). Huddy et al. (2005) study the example of September 11 and find that U.S. citizens who perceived a high threat were far more likely to support Bush’s antiterrorism policies than U.S. citizens who did not perceive a high threat. The authors explain that “as perceptions of threat increase, people become significantly more supportive of measures that restrict the rights of groups broadly associated with terrorism, and policies that limit civil liberties for all citizens more generally” (p. 605). But Bush is not the only politician to use fear appeals in his communication. De Castella et al. (2009) argue that John Howard, the former Prime Minister of Australia, may have
deliberately used fear appeals in his speeches in order to get political support. However, previous research has mainly focused on presidents. Therefore, it remains unclear if other politicians, like right-wing populists, can also create fear amongst citizens and if they can create political support.
Conservatism
The fear appeals used by politicians might have different effects on different people. These differences can be found between liberal and conservative people (Jost et al., 2007; Jost, Glaser, Kruglanski, & Sulloway, 2003). Therefore, political orientation is thought to play a role in how people react to threats. Conservatism is thought to be the “resistance to change and the tendency to prefer safe, traditional and conventional forms of institutions and behaviour” (Wilson, 2013, p. 4). Conservative opinions are based on uncertainty avoidance and threat management. Firstly, conservatives want to avoid uncertainties more often, are more intolerant of ambiguities, have more need for order, structure and closure and they are less open to new experiences than liberals are (Jost et al., 2007, see p. 989). Secondly,
conservatives want to manage the threat more often, they are more likely to believe that they live in a dangerous world and are more fearful about threats than liberals are. Both uncertainty avoidance and threat management can independently contribute to conservatism (Jost et al., 2007, 2003). Altemeyer (1998) and Duckitt and Sibley (2009) indeed also find that
conservatives more often feel that they live in a dangerous world than liberals do. This information corresponds to the findings of Jost et al. (2003), where conservatives were also more fearful about threats than liberals were. Therefore this study will take into account the possible differences in responses to threats between conservatives and liberals.
Case Study: Geert Wilders
Geert Wilders, party leader of the Partij Voor de Vrijheid (PVV), is one of the most well known Dutch politicians on Twitter. He tweets about his ideas, which range from being anti-immigration and anti-EU to being against the Islam. He tweets provocative messages about these topics in order to get political support and does not sheer away from
discriminating content (Blanquart & Cook, 2013). Over the years Wilders got more support on Twitter, resulting in an enormous amount of re-tweets and more than 400.000 followers on his Twitter-network at this moment (Twitter, 2015b). This information indicates that Wilders has a lot of influence on Twitter, which raises concerns amongst scholars about the support for extreme right-wing populists (Blanquart & Cook, 2013).
Blanquart and Cook (2013) investigate how Geert Wilders makes use of Twitter. They make a distinction between topics and narratives of Wilders’ tweets. Out of all of the topics of his tweets, 43% is focused on national politics, 23% is focused on international politics, 19% is focused on anti-Islam content and 15% of his tweets are about his personal life or other topics. About 25% of the narratives in his tweets are about extremist topics, like terrorist generalizations or calls for action. A lot of his tweets are based on his spearhead ideas about
the Islam. These tweets promote concerning stories about the Islam, are hateful about the Islam and focus on extremist Islamic topics. The authors state that “Wilders uses Twitter to promote anti‐Islam ideology throughout Europe by aligning and promoting the fear of terrorist threats from Muslims throughout Europe” (p. 5). But these authors are not the only ones stating that Wilders promotes fear. Dutch media and Dutch politicians also state that Geert Wilders creates fear amongst Dutch citizens (AD, 2015; EenVandaag, 2015; Trouw, 2014; Volkskrant, 2015b). Therefore, this study will look into the specific case of the Dutch extreme-right populist Geert Wilders. It will examine how fearful Wilders’ tweets are and how they affect people’s fearfulness.
RQ1: To what extent does Geert Wilders contribute to the climate of fear in the Netherlands?
RQ2: To what extent can tweets by Geert Wilders create fear amongst Dutch citizens?
Hypotheses
Based on the above-mentioned theories and the Extended Parallel Process Model, six hypotheses were drawn. It is expected that high level threats will increase people’s levels of severity, susceptibly, fear, response efficacy, self-efficacy and support for Wilders’ party, the PVV. Moreover, it is expected that conservatives will be more fearful and supportive of the PVV than liberals.
H1: People who are confronted with a high level threat will find the threat more severe and susceptible, than people who are confronted with a low level threat.
confronted with a low level threat.
H3: People who are confronted with a high level threat will show higher levels of response efficacy and self-efficacy, than people who are confronted with a low level threat.
H4: People who are confronted with a high level threat are more supportive of the PVV than people who are confronted with a low level threat.
H5: Conservatives are more fearful about the threat than liberals are. H6: Conservatives are more likely to vote for the PVV than liberals are.
Study 1
This study was interested in two main questions. The first question was interested in how tweets by Geert Wilders contribute to the climate of fear in the Netherlands. Therefore Study 1 will examine to what extent Wilders makes use of fear appeals in his tweets. To answer this question a content analysis was conducted, which allowed the researcher to analyze Wilders’ tweets in a systematic and objective manner.
Method
Sample. While Wilders’ ideas range from being anti-immigration and anti-EU to being against the Islam, the Islam is the topic Wilders is most interested in (Vossen, 2011). Recently, Wilders’ ideas have been fueled by an Islamic organization, the Islamic State. A Google-Trends search with the search term “Islamic State” found a growing public interest in the Islamic State, which peaked September 2014 (Google, 2015). Though the data collection should have started at that exact point, Wilders’ tweets were only available from October 2014. Therefore, the data collection of this study ran from October 2014 until, exactly half a year later, the end of April 2015, when the analysis of the collected data started. Therefore, the largest amount of tweets as possible were included in this study. A codebook was
established in order to analyze Wilders’ tweets (see Appendix A). The analysis made use of the whole tweet, including the previews of the possible posted pictures and videos. All the tweets on Wilders’ account were analyzed, including re-tweets. The data was downloaded from the personal Twitter account of Geert Wilders (@geertwilderspvv), which led to 750 analyzed tweets (N = 750).
Operationalizations. The codebook was used to measure both general as well as fear appeal variables. This study was firstly interested in the type of tweet in order to determine what Wilders’ tweets look like (1 = Comment, 2 = Comment to news or online information, 3 = Comment on a tweet, 4 = Conversations, 5 = News or online information, 6 = Retweet, 7 = Hashtag only, 8 = None of the above). This question was prepared according to an
operationalization from Small (2011). Secondly, it was interested in the presence of images (1 = not present, 2 = present), videos (1 = not present, 2 = present), hashtags (1 = not present, 2 = present), the topic of the hashtags and the topic of the tweets (e.g. national politics, economy, health care, immigration and integration, etc.), and the amount of re-tweets and favorites the tweet got.
Next to these more general measures, the codebook contained questions about fear appeals, which was defined in this study as a “persuasive message that attempts to arouse the emotion fear by depicting a personally relevant and significant threat and then follows this description of the threat by outlining recommendations presented as feasible and effective in deterring the threat” (Witte, 1994, p. 114). Fear appeals were measured by looking into threats, severity, susceptibility, response-efficacy and self-efficacy according to the studies of Alharbi (2014) and Klein and Mattson (2009). All these variables were measured by their presence (1 = not present, 2 = present). The threat was defined as “a person or thing likely to cause damage or danger” (Oxford Dictionary, 2015) (1 = not present, 2 = present) and a further question was interested in the topic of the threat; where the threat was coming from (e.g.
national politics, economy, health care, immigration and integration, etc.). Only when a threat was present, the last four components of the fear appeal were measured (based on Witte, 1994). First, the severity of the threat, which referred to the beliefs about the significance or magnitude of the threat (1 = not present, 2 = present). Secondly, the susceptibility of threat, which referred to the (possible) harm of the threat (1 = not present, 2 = present). Thirdly, the response efficacy, which referred to all actions (that should be) taken in order to deter the threat (1 = not present, 2 = present). And fourthly, self-efficacy, which referred to a specific person or party that could help to deter the threat (1 = not present, 2 = present).
Intercoder Reliability. In order to evaluate the reliability of the codebook, the intercoder reliability was measured. A student of the University of Amsterdam, who was familiar with coding, was asked to help coding tweets for the intercoder reliability. She was first explained the aim of the study alongside with an extended explanation about the main variables. After this introduction to the study, the student and the researcher herself both coded about ten percent (n = 77) of the tweets out of the total sample. After a first inspection of the intercoder reliability, three questions were adjusted, where after additional coding was done in order to obtain good level of reliability, with an average of kalpha = .79 (see
Appendix B).
Results
A content analysis was conducted to determine how Wilders makes use of Twitter and fear. While reading the results, it should be noted that Wilders is popular on Twitter, with over 400.000 followers (Twitter, 2015b), and a total of 66.951 re-tweets and 30.900 tweets marked as favorites in only half a year (excluding re-tweets and favorites from re-tweets).
Mostly comments or comments to (online) news or information. Wilders makes use of different types of tweets. The most used types are “comments” (18.7%) and “comments
to (online) news or online information” (29.5%). These tweets contained a general comment or a comment to (online) news or online information that included a link or image referring to the news. It became clear that Wilders uses Twitter to give comments and does not use
Twitter to have conversations with other Twitter-users, while this type of tweet only takes up 1.9% of all tweets. However, the re-tweet is also a popular type of tweet (37.9%), which indicates that more than one-third of all tweets are not originally written by Wilders himself, but are written by other tweeters.
Little use of hashtags and videos, much use of images. Wilders is not a fan of using
hashtags, while they are only used in 15.3% of all tweets. Though, the hashtags used are
mostly referring to the Islam (22.6%), with the frequently recurring “#nomoreislam”. Wilders’ tweets contained relatively little videos that were directly placed in the tweet (4.1%), though, more than half of the tweets did contain an image (58.5%).
Most tweets about Islam or Islamic terrorism. The tweets had a variety of topics, but most referred to the Islam or Islamic terrorism (33.4%). Other somewhat more used topics were “national politics” (10.9%), “elections” (7.7%) and “immigration and integration” (4.1%). But interestingly, Wilders does not mind tweeting about himself or his actions, while “Geert Wilders” was the topic of 19.7% of his tweets.
Threats but no fear appeals. A big interest of the content analysis was the presence of fear appeals. The study was interested in the presence of a threat and the four components of a fear appeal, severity, susceptibility, response efficacy and self-efficacy. In 25.6% of all tweets a threat was present. Out of the tweets that referred to a threat, 42.7% referred to the severity of the threat, 43.8% referred to susceptibility, 28.6% referred to response efficacy and 5.7% referred to self-efficacy. These numbers are mostly coming from the Islam or Islamic terrorism, because this topic contains almost four-fifth of the all threats used in the tweets. However, none of the tweets with a threat contained all the elements of a fear appeal
(severity, susceptibility, response efficacy and self-efficacy). Most threats consisted of one (42.2%) or two (30.7%) elements. The tweets therefore never contained a full fear appeal. More threats in safety and Islamic topics. A strong association was found between the topics of the tweets and the presence of a threat (V = .574). 24 different z-tests for the different topics revealed that there were two topics that contained more threats than the other topics. After a Bonferroni correction, the topics “safety” (p < .001) and “Islam or Islamic terrorism” (p < .000) showed to contain more threats than expected (see Figure 1).
Figure 1. Percentage of Threats per Topic
The four elements of fear appeals were also assessed in order to find out if certain topics consisted of more severity, susceptibility, response efficacy and self-efficacy than expected on the basis of coincidence. However, there was a weak association found between severity and the topics of the tweets (V = .249), susceptibility and the topics of the tweets (V = .221), response efficacy and the topics of the tweets (V = .238) and self-efficacy and the topics of the tweets (V = .183). Also, z-tests revealed that there were no topics that contained significantly more severity, susceptibility, response efficacy or self-efficacy.
0 10 20 30 40 50 60 70 80 90 100 PVV
Islam or Islamic Terrorism Immigration and Integration Health care Safety Defense International Politics Government National Politics Geert Wilders
Study 2
The results of study 1 indicate that Geert Wilders does not make use of fear appeals in his tweets, but they do contain a considerable amount of threats. Though there is a lack of fear appeals, the use of threats in tweets can also affect people. The second question of this study is therefore interested in the effects of tweets by Geert Wilders. This second research question will be answered through an experiment, where the stimuli material can allow the researcher to make conclusions about the effects of Wilders’ tweets.
Method
Sample. Respondents were asked to fill in an online survey in the online survey tool Qualtrics (see Appendix C). There were no extended criteria for the selection of the
respondents, but participation was only possible when the respondent had the Dutch
nationality. The respondents were recruited through Facebook and were randomly assigned to one out of four conditions. A total of 118 people participated in the experiment, divided over condition 1 (n = 35), condition 2 (n = 17), condition 3 (n = 29) and condition 4 (n = 37).
Design. A 2x2 between-subjects design was constructed to get insights into the effects of Wilders’ tweets (see Table 1). The stimuli material used in this study was based on the Extended Parallel Process Model by Witte (1992). The model states that people will not get fearful without a threat. Therefore the stimuli were based on a “threat” or “no threat” basis. But the experiment also had to take into account the effect Wilders himself could have on respondents. Wilders has a reputation in the Netherlands of being provocative and
discriminating and has been prosecuted because of his comments (NRC, 2014). Therefore this study was also interested in what kind of effect Wilders himself had on people, possibly even without the use of a threat. Therefore the experiment will consist of four conditions, a
condition. While Wilders is anti-Islam, the threat in the “threat condition” is coming from a topic were Wilders as well as Dutch citizens have a growing interest in, the Islamic State (Google, 2015; Twitter, 2015b). Two tweets by Wilders that contained a threat from the Islamic State, were chosen to be the “threat stimuli” and two tweets that did not contain a threat were chosen to be the “no threat stimuli” (see Appendix C). Decisions about these choices were based upon the fear appeal measurements as operationalized in study 1. All the respondents had the same chance to be assigned to one of the four conditions. And though the respondents were randomly assigned to one of the four conditions, they received the exact same questions to fill in.
Table 1. 2x2 design: four conditions
Threat No Threat
Wilders Wilders Threat Wilders No Threat
No Wilders No Wilders Threat No Wilders No Threat
Operationalizations. Before starting the survey respondents were asked to agree with a form of consent, explaining their anonymity and regulations as set by the Amsterdam
School of Communication Research (ASCoR). It also provided the respondents with the email address of the researcher in case of any questions. When respondents agreed to this form, they were directed to the first general questions regarding their gender, age, religion, level of education, voting decision during the last elections and political ideology (e.g. level of conservatism/liberalism). All these questions included a “prefer not to say” option. The survey then led the respondents to the stimuli material, where they were randomly assigned to either the “Wilders threat”, the “Wilders no threat”, the “no Wilders threat” or the “no
to look at the pictures thoroughly. After the stimuli material was shown, questions regarding their fear levels were asked.
The fear appeal, as mentioned before in study 1, was defined as a “persuasive message that attempts to arouse the emotion fear by depicting a personally relevant and significant threat and then follows this description of the threat by outlining recommendations presented as feasible and effective in deterring the threat” (Witte, 1994, p. 114). After respondents were exposed to the pictures, statements regarding the components of the fear appeal (severity, susceptibility, response efficacy and self-efficacy) were given. They were prepared according to the study of McMahan, Witte and Meyer (1998). Unlike in study 1, the operationalizations of these components in study 2 were more focused on the respondents’ feelings. Severity was defined as “the magnitude of harm expected from the threat”, susceptibility was defined as “the degree to which one feels at risk for experiencing the threat”, response efficacy was defined as “one’s beliefs about whether the recommended response works in averting the threat” and lastly, self-efficacy was defined as “one’s beliefs about his or her ability to perform the recommended response” (Witte, 1994, p. 114). Respondents were confronted with three different statements about each component. These statements ranged from “The threat of the Islamic State is severe” (severity), “I could be at risk because of threats by the Islamic State” (susceptibility), “The government is taking the right actions in order to prevent a terrorist attack by the Islamic State” (response efficacy) to “I feel confident that the Islamic State can’t hurt me” (self-efficacy). For all these statements a 7-point Likert scale was used ranging from “Strongly Disagree” (1) to “Strongly Agree” (7).
After the respondents completed these questions, both fear and support for the PVV were measured. Fear was operationalized as the arousal of fearful emotions (McMahan, Witte & Meyer, 1998). Respondents were given three statements, like “I am frightened”. Again a 7-point Likert scale was used ranging from “Strongly Disagree” (1) to “Strongly Agree” (7).
The last variable, support of the PVV, was defined as the likeliness to vote for the PVV in upcoming elections. The question “How likely is it for you to vote for the following parties?” was asked and respondents could indicate their likeliness to vote on a 7-point Likert scale ranging from “Very Unlikely” (1) to “Very Likely” (7). The voting options were based on all political parties that participated during the last Dutch elections in 2012. After this last question, the respondents were thanked for their participation and were given the option to leave a comment if necessary (see Appendix C).
Results
Next to the use of fear appeals in Geert Wilders’ tweets, the experiment was interested in how tweets by Wilders influenced its readers. 118 people participated in the experiment, divided over condition 1 (n = 35), condition 2 (n = 17), condition 3 (n = 29) and condition 4 (n = 37). The respondents were mostly women (80.5%), but their ages were very diverse, they were between the ages of 18 and 24 (53.4%), 25 and 34 (13.6%) and 55 and 64 (18.6%). Most of these people were catholic (15.3%), protestant (14.4%) or not religious (64.4%). Around half of the respondents were higher educated (51.7%) or had a postgraduate education (21.2%). The political parties D66 (26.3%) and VVD (22.9%) got the most votes from the respondents during the last elections, and also PvdA did quite well (10.2%). The respondents’ political ideology was mostly liberal (23.7%) or somewhat liberal (31.4%), while only 12.7% of them indicated to be somewhat conservative or conservative. A considerable amount of people (27.1%), however, indicated not to be liberal or conservative (M = 3.51, SD = 1.41).
Factor analyses. Three statements measured the fear appeal components and three statements measured fear. To analyze the fear appeal components and fear, principal components factor analyses were conducted in order to test whether the statements given to the respondents would fit into a scale per component. These factor analyses with Varimax
rotation indicated that in all cases there was one factor measuring the variables severity, susceptibility, response efficacy, self-efficacy and fear. For all variables there was one component found with an Eigenvalue higher than 1 and there were no negative factors. All factors were above .45 and the total explained variances were 90.42% (severity), 79.81% (susceptibility), 80.45% (response efficacy), 62.14% (self-efficacy) and 84.90% (fear). To assess the reliability of the scales, a Cronbach’s Alpha was calculated. The Cronbach’s Alpha indicated that all the scales were reliable (Cronbach’s Alpha > .69). Therefore, the scales of these variables were composed and used in the next analyses.
No effects on severity and susceptibility. This study was interested in how people responded to the threat stimuli material. Four two-way ANOVA’s were calculated in order to find out how Geert Wilders’ tweets can influence people’s thoughts about the severity of the threat, their susceptibility, the response efficacy of the government and their self-efficacy. First, a two-way ANOVA was conducted to find out if the different threat conditions influenced people’s thoughts about the severity of the threat. However, there was no significant effect found of being confronted with a “Wilders tweet” on people’s thoughts about the severity of the threat, F (1, 110) = .153, p = .696. People who saw a “Wilders tweet” (M = 5.04, SD = 0.20) did not significantly differ from people who saw a “no Wilders tweet” (M = 4.94, SD = 0.17). There was also no significant effect found of being confronted with a “threat tweet” on people’s thoughts about the severity of the threat, F (1, 110) = .103, p = .749. People who saw a “threat tweet” (M = 4.95, SD = 0.17) did not significantly differ from
people who saw a “no threat tweet” (M = 5.03, SD = 0.20). The test further revealed that there was no significant interaction effect between the (no) Wilders tweet and a (no) threat tweet on the severity of the threat, F (1, 110) = .000, p = .998. Overall, people seem to think the threat is somewhat severe (M = 4.98, SD = 1.31).
A second one-way ANOVA was conducted to find out if the different threat
conditions influenced people’s thoughts about their susceptibility to the threat. However, there was no significant effect found of being confronted with a “Wilders tweet” on the
susceptibility of the threat, F (1, 112) = .583, p = .447. People who saw a “Wilders tweet” (M = 3.99, SD = 0.22) did not significantly differ from people who saw a “no Wilders tweet” (M = 4.20, SD = 0.18). There was also no significant effect found of being confronted with a “threat tweet” on the susceptibility of the threat, F (1, 112) = .946, p = .333. People who saw a “threat tweet” (M = 4.23, SD = 0.18) did not significantly differ from people who saw a “no threat tweet” (M = 3.96, SD = 0.21). The test further revealed that there was no significant interaction effect between the (no) Wilders tweet and a (no) threat tweet on the susceptibility to the threat, F (1, 112) = .381, p = .539. Overall, people seem to think they are neither
susceptible nor not susceptible to the threat by the Islamic State (M = 4.10, SD = 1.43). People in the “threat tweet” condition did not significantly found the threat more severe or found themselves more susceptible to the threat (see Figure 2). Therefore, hypothesis 1, “people who are confronted with a high level threat will find the threat more severe and susceptible, than people who are confronted with a low level threat”, was rejected.
No effects on fear. Though people in the Wilders threat condition did not show higher levels of severity or susceptibility, a two-way ANOVA was conducted to find out if the different threat conditions influenced people’s fearfulness about the threat. However, there was no significant effect found of being confronted with the “Wilders tweet” on being fearful about the threat, F (1, 114) = 361, p = .549. People who saw a “Wilders tweet” (M = 3.31, SD = 0.21) did not significantly differ from people who saw a “no Wilders tweet” (M = 3.48, SD = 0.18). There was also no significant effect found of being confronted with a “threat tweet” on the amount of fear people have about the threat, F (1, 114) = 2.087, p = .151. People who saw a “threat tweet” (M = 3.60, SD = 0.18) did not significantly differ from people who saw a
“no threat tweet” (M = 3.20, SD = 0.21). The test further revealed that there was no significant interaction effect between the (no) Wilders tweet and a (no) threat tweet on the fearfulness about the threat, F (1, 114) = .2.071, p = .275. Overall, people are somewhat un-fearful about the threat of the Islamic State (M = 3.45, SD = 1.43). Therefore, hypothesis 2, “people who are confronted with a high level threat are more fearful, than people who are confronted with a low level threat.”, was rejected. All these results are shown in Figure 2.
Figure 2. Levels of Severity, Susceptibility and Fear by Condition Type.
No effects on response efficacy and self-efficacy. Next, it was assessed whether people in the Wilders threat condition significantly differed in their reactions about the
response efficacy and their self-efficacy. A two-way ANOVA was conducted to find out if the different threat conditions influenced people’s thoughts about the response efficacy in order to deter the threat. However, there was no significant effect found of being confronted with a “Wilders tweet” on people’s thoughts about the response efficacy, F (1, 104) = 1.227, p = .270. People who saw a “Wilders tweet” (M = 4.23, SD = 0.21) did not significantly differ from people who saw a “no Wilders tweet” (M = 3.94, SD = 0.18). There was also no
significant effect found of being confronted with a “threat tweet” on people’s thoughts about the response efficacy, F (1, 104) = 1.640, p = .344. People who saw a “threat tweet” (M =
1 2 3 4 5 6 7
Severity Susceptibility Fear
Severity, Susceptibility and Fear by Condition
Wilders Threat Wilders No Threat No Wilders Threat No Wilders No Threat
4.21, SD = 0.18) did not significantly differ from people who saw a “no threat tweet” (M = 3.96, SD = 0.21). The test further revealed that there was no significant interaction effect between the (no) Wilders tweet and the (no) threat tweet on the response efficacy of the threat,
F (1, 104) = .000, p = .990. Overall, people do not have an outspoken opinion about the
response efficacy of the government (M = 4.08, SD = 1.34).
Another two-way ANOVA was conducted to find out if the different threat conditions influenced people’s self-efficacy. There was no significant effect found of being confronted with a “Wilders tweet” on people’s self- efficacy, F (1, 114) = 0.081, p = .777. People who saw a “Wilders tweet” (M = 3.06, SD = 0.16) did not significantly differ from people who saw a “no Wilders tweet” (M = 3.00, SD = 0.13). There was also no significant effect found of being confronted with a “threat tweet” on people’s self-efficacy, F (1, 114) = 0.107, p = .745. People who saw a “threat tweet” (M = 3.06, SD = 0.13) did not significantly differ from people who saw a “no threat tweet” (M = 2.99, SD = 0.16). The test further revealed that there was no significant interaction effect between the (no) Wilders tweet and the (no) threat tweet on self-efficacy, F (1, 114) = .802, p = .372. Overall, people felt somewhat low self-efficacy (M = 3.01, SD = 1.06). People in the Wilders threat condition did not get significantly more response efficacy or self-efficacy than people in the other conditions (see Figure 3). Therefore, hypothesis 3, “people who are confronted with a high level threat will show higher levels of response efficacy and self-efficacy, than people who are confronted with a low level threat“, was rejected.
No effects on support. A two-way ANOVA was conducted to find out if the different threat conditions influenced people’s support for the PVV. However, there was no significant effect found of being confronted with a “Wilders tweet” on supporting the PVV, F (1, 112) = 1.339, p = .250. People who saw a “Wilders tweet” (M = 2.13, SD = 0.30) did not
no significant effect was found of being confronted with a “threat tweet”, F (1, 112) = 0.035,
p = .852. People who saw a “threat tweet” (M = 2.32, SD = 0.25) did not significantly differ
from people who saw a “no threat tweet” (M = 2.39, SD = 0.30). The test further revealed that there was no significant interaction effect between of the (no) Wilders tweet and a (no) threat tweet on people’s support for the PVV, F (1, 112) = 1.163, p = .283. People in the Wilders threat condition did not have significantly more intention to vote for the PVV than people in the other conditions (see Figure 3). Overall, people were unlikely to vote for the PVV (M = 2.33, SD = 1.97). Therefore, hypothesis 4, “people who are confronted with a high level threat are more supportive of the PVV than people who are confronted with a low level threat.”, was rejected. People in the different conditions also did not significantly differ in their intentions to vote for any other Dutch political party (p > .05).
Figure 3. Levels of Response Efficacy, Self-Efficacy and Support by Condition Type.
No differences between conservatives and liberals and fear. Lastly this study was interested in how conservatives responded to the different threat conditions. A three-way ANOVA was conducted in order to find out if conservatism contributed to the fearfulness of participants in the Wilders threat condition. There were no significant differences found between conservatives and liberals, F (1, 73) = 3.363, p = .348. Conservatives (M = 3.22, SD
1 2 3 4 5 6 7
Response Efficacy Self-Efficacy Support
Response Efficacy, Self-Efficacy and Support by Condition
Wilders Threat Wilders No Threat No Wilders Threat No Wilders No Threat
= 1.43) were not significantly more fearful about the threat than liberals were (M = 3.43, SD = 1.45). No interaction effects were found between the (no) Wilders tweets and conservatism on people’s fearfulness, F (1, 73) = 0.028, p = .868, and between the (no) threat tweets and conservatism on people’s fearfulness, F (1, 73) = 3.073, p = .084. The test further revealed that there was no significant interaction effect between the (no) Wilders tweet, the (no) threat tweet and conservatism on people’s fearfulness, F (1, 73) = .176, p = .676. Therefore,
hypothesis 5, “conservatives are more fearful about the threat than liberals are”, was rejected (see Figure 4).
Small differences between conservatives’ and liberals’ support of the PVV. Also, a three-way ANOVA was conducted in order to find out if conservatives support the PVV more often in the Wilders threat condition. There were no significant differences found between conservatives and liberals, F (1, 71) = 0.559, p = .457. Conservatives (M = 2.80, SD = 1.78) were not significantly more supportive of the PVV than liberals (M = 2.05, SD = 1.71). A very weak interaction effect was found between the (no) Wilders tweet conditions and conservatism, F (1, 71) = 4.008, p = .049, η2 = 0.05. Conservatives who saw a “no Wilders tweet” were the most likely to vote for the PVV. They were more likely to vote for the PVV (M = 3.63, SD = 0.60) than liberals were who saw a “no Wilders tweet” (M = 2.00, SD = 0.31), than other conservatives who saw a “Wilders tweet” (M = 1.50, SD = .92) and than liberals who saw a “Wilders tweet” (M = 2.24, SD = .31). A post hoc test, however, found that there were only significant differences between conservatives and liberals in the “no Wilders tweet” condition. Conservatives were more likely to support the PVV than liberals were (Mdifference =
-1.627, p < .05). Furthermore, no interaction effect was found between the (no) threat tweets and conservatism, F (1, 71) = .003, p = .956. The test further revealed that there was no significant interaction effect between the (no) Wilders tweet, the (no) threat tweet and conservatism on people’s support of the PVV, F (1, 71) = 2.267, p = .137.
Though one interaction effect was found between the conservatives and liberals in the (no) Wilders tweet condition, overall conservatives were not more supportive of the PVV than liberals and the level of support remains fairly low. Therefore, hypothesis 6,
“conservatives are more likely to vote for the PVV than liberals are”, was rejected (see Figure 4).
Figure 4. Overall Fear and Support Levels by Political Ideology.
Discussion
This study was interested in two main research questions. The first question was “To what extent does Geert Wilders contribute to the climate of fear in the Netherlands?”, and the second question was “To what extent can tweets by Geert Wilders create fear amongst Dutch citizens?”. In an effort to answer these questions, both a content analysis and an experiment were conducted.
The findings of this study indicate that Geert Wilders refers to threats in about a quarter of his tweets. These threats are mostly found in topics about the Islam or Islamic terrorism. Wilders, however, does not use fear appeals in his tweets. Despite of these results, Wilders’ influence on Twitter is great, with over 400.000 followers and around 67.000 re-tweets in only half a year. Based on the power of this Twitter network and the impact of his political tweets, Wilders was found to be one of the most influential tweeters of 2014. This
0 2 4 6
Fear Support
Fear and Support Levels Conservatives and Liberals
Conservatives Liberals
powerful network allows Wilders to spread his messages to an enormous amount of followers and even the media have interest in his tweets (Broersma & Graham, 2013). His tweets therefore have the power to spillover from the alternative media (Twitter) to the more traditional media (Mathes & Pfetsch, 1991), which allows his messages to be seen by even more people. The considerable amount of attention given to his tweets and the reach of his tweets, can therefore contribute to the climate of fear in the Netherlands.
These tweets, however, do not make people fearful about a very current Islamic topic, the Islamic State. Both the threat about the Islamic State as well as Wilders himself did not cause any effects. The answer to the question “To what extent can tweets by Geert Wilders create fear amongst Dutch citizens?” is therefore simple. Tweets by Geert Wilders do not create fear, not even when they contain a threat. The tweets also did not have any effects on the individual components of fear appeals; severity, susceptibility, response efficacy and self-efficacy, and none of the tweets made people more likely to vote for Wilders’ party, the PVV.
Although the expected results were not found, this study tried to take a different approach as apposed to earlier studies. Firstly, it was unknown if right-wing populist politicians, like Geert Wilders, made use of fear, how they made use of fear, how Twitter facilitated them in this use and how these tweets affected people. Both the use of fear by politicians and the use of Twitter by politicians were studied before. A combination, however, was never examined before. Therefore, this study tried to take the first steps into getting a better understanding about the use and effects of fear by right-wing populist politicians on Twitter. Secondly, fear in Wilders’ tweets were measured by a manual content analysis. 750 tweets were analyzed and were examined for four different fear appeal elements; severity, susceptibility, response efficacy and self-efficacy. This made it possible to thoroughly examine to what extent Wilders makes use of fear in his tweets. Thirdly, a very recent topic was chosen to investigate. The Islamic State is a topic that is getting more and more attention
(Google, 2015). Therefore the topic of this study comes close to reality and is very relatable for a lot of people. Lastly, this study included two different methods in order to answer both research questions. Therefore, statements could not only be made about the use of fear on Twitter, but it could also make statements about the effects of these tweets. This made it possible for the researcher to get a broader and more thorough view about the research topic.
Previous Findings
The findings of this study do not completely correspond to the findings of previous studies. However, they were somewhat similar about the use of fear by politicians. Some scholars think politicians create a culture of fear by over exaggerating threats and by keeping terrorist plots a trending topic. An example of this exaggeration is the exposure of outdated terrorist threats, which happened under the Bush administration in the U.S., but also in other countries (Mythen & Walklate, 2008). In this present study it is also found that Wilders’ referred to a considerable amount of threats in his tweets, mostly about the Islam and Islamic terrorism. Therefore, not only Bush, but also Geert Wilders exposes citizens to a lot of
threatening information. Therefore multiple politicians are thought to contribute to the climate of fear in their countries.
Though, the effects of fear appeals in this present study differ from previous studies. De Castella et al. (2009) state that president Bush used and created fear after 9/11 in order to gain political support for antiterrorism policies and actions. Other findings indicate that Bush has successfully used fear after 9/11. U.S. citizens who perceived a high threat were far more likely to support Bush’s antiterrorism policies than U.S. citizens who did not perceive a high threat (Huddy et al., 2005). Another politician that has used fear is Howard, former president of Australia. He is thought to have used fear appeals in his speeches to get political support from Australian citizens (De Castella et al., 2009). However in this present study, there is no
evidence that Wilders can also create fear and support like Bush and Howard were able to do. Though previous studies found that politicians were successful in creating fear amongst citizens and in creating political support, findings of the present study point to the opposite. However, these contradicting findings might be due to different timeframes. Previous studies examined the effects of fear appeals right after one of the biggest terrorist attacks in history, 9/11 (Huddy et al., 2005). These results were therefore based on a recent national attack, while Geert Wilders could not base his tweets on a recent Dutch attack. Therefore, Wilders does not “play to everyday fears” (Schuermans & De Maesschalck, 2010, p. 248) as other politicians were able to do (De Castella et al., 2009).
Limitations
The findings of this study indicate that people were not affected by Wilders or the threats about the Islamic State. These findings, however, might be the result of the
misdistribution of the respondents within this study. The sample contained relatively many females, not-religious people and liberals. This might influence the external validity of this study, while the results are not representative for the whole Dutch population. Next to the difficulties with the external validity, the findings might be influenced by the lack of full fear appeals in Wilders’ tweets. These messages, with a maximum of 140 characters, could be too short to contain all the elements of a fear appeal. Therefore, in contrast to the expectations, Twitter might not be the right channel to distribute fear appeals. Though, it can also explain the lack of fear resulting from Wilders’ threat tweets. The stimuli materials were chosen to contain a threat by the Islamic State, but these short messages might have included too little information about the threat to affect people, while they were only able to see a maximum of 280 characters about the threat (two tweets).
Next to these short messages, external factors could also have influenced the results. The Islamic State is a topic often discussed by the media, and Dutch citizens also show more interest in the Islamic organization (Google, 2015). Therefore, possible additional information may have influenced the respondents in the experiment. Since the end of 2014 and the
beginning of 2015 the Islamic State has been associated with different terrorist attacks around the world. First in Australia, where the Islamic State was associated with a hostage taking in a café, secondly, in Paris where the Islamic State was associated with the attack on Charlie Hebdo and thirdly the Islamic State was associated the shooting in Copenhagen. These events are of course also thoroughly discussed by the media (e.g. NRC, 2015b; RTL Nieuws, 2014; Volkskrant, 2015a). Therefore people might have had a lot of additional information to base their opinions on, other than the information given in the stimuli material.
Next to the effects of these recent events, Geert Wilders also has a reputation of being provocative and is known for his anti-Islam statements. His statements and tweets are often discussed by the Dutch media (e.g. NRC, 2015a). Possibly, people are “used” to these kinds of statements by Wilders, because Wilders is too well known for his judgments and ideas.
Future Research
This study has made the first steps into understanding how a right-wing populist like Geert Wilders makes use of Twitter to spread fearful messages, how these affect people and how he contributes to the culture of fear in the Netherlands. But the findings of this study, combined with previous findings, raise important questions. While previous studies focused on presidents, the present study focused on a politician that is currently not even in the Dutch government. Future research could take into account the power that (right-wing populist) politicians have and why certain politicians are successful in creating fear and support amongst citizens, while others are not.
Other studies also took into account different communication channels. John Howard, for example, showed to be successful in promoting fear through speeches (De Castella et al., 2009). Therefore, it is interesting to look further into the use and effects of different
communication channels. Politicians communicate with citizens in different ways, like through speeches, interviews, debates, and press conferences (Norris, 2005). Wilders, for example, does not only use Twitter, but he also gives speeches in the Second Chamber, speeches for anti-Islam organizations like Pegida and he writes columns for his personal website (GeertWilders.nl, 2014, 2015). Twitter might just be a small part of his
communication strategy in order to spread fearful messages and to get political support, while other ways of communicating with citizens can also contain threats and fear appeals. A lot of these communication channels, however, have not been studied before. Future research could therefore take these suggestions into account.
In June 2015 it became clear that right-wing populism remains to be an important topic for future research. Right-wing populist parties in Europe, including the PVV and the Front National, have formed a fraction in the European Parliament, which allows them fight together against immigration, the EU and the Islam (NRC, 2015c). A better understanding of the use and effects of fear by right-wing populism is therefore more important than ever, while these parties claim to be “the voice of resistance” (RTL Nieuws, 2015) and fear is used “at a time of high political tension” (De Castella et al., 2009, p. 22). Therefore, fear can become more prominent in today’s society. These recent developments demonstrate that the topic of this study will keep its societal relevance for people around the world, as everyone has to deal with these politics in one way or another.
References
AD. (2015). “Spreek moslims niet aan op misdaden van een ander.” Retrieved March 21, 2015, from
http://www.ad.nl/ad/nl/1012/Nederland/article/detail/3831025/2015/01/16/Spreek-moslims-niet-aan-op-misdaden-van-een-ander.dhtml
Akkerman, T. (2005). Anti-immigration parties and the defence of liberal values: The exceptional case of the List Pim Fortuyn. Journal of Political Ideologies, 10(March 2015), 337–354. doi:10.1080/13569310500244354
Alharbi, A. M. (2014). Risk communication discourse: A content analysis of some Australian media coverage of cyclones in Queensland, Australia in 2011. International Journal of
Human Sciences / Uluslararası İnsan Bilimleri Dergisi, 11(1), 1019–1036.
doi:10.14687/ijhs.v11i1.2803
Altemeyer, B. (1998). The other “authoritarian personality.” In Advances in Experimental
Social Psychology (p. 402). Academic Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=1s7Q6e2NF9EC&pgis=1
Altheide, D. L. (1997). the News Media , the Problem Frame , and the Production of Fear.
The Sociological Quarterly, 38(4), 647–668. doi:10.1111/j.1533-8525.1997.tb00758.x
Altheide, D. L. (2003). Notes Towards A Politics Of Fear, 1(1), 37–54.
Blanquart, G., & Cook, D. M. (2013). Twitter Influence and Cumulative Perceptions of Extremist Support: A Case Study of Geert Wilders. Australian Counter Terrorism
Conference, 1–11.
Borland, R., Yong, H.-H., Wilson, N., Fong, G. T., Hammond, D., Cummings, K. M., … McNeill, A. (2009). How reactions to cigarette packet health warnings influence quitting: findings from the ITC Four-Country survey. Addiction (Abingdon, England),
Brader, T. (2005). Striking a Responsive Chord: How Political Ads Motivate and Persuade Voters by Appealing to Emotions. American Journal of Political Science, 49(2), 388– 405. Retrieved from
http://www.jstor.org/stable/3647684?seq=1#page_scan_tab_contents
Broersma, M., den Herder, B., & Schohaus, B. (2013). A Question of Power. Journalism
Practice, 7(4), 388–395. doi:10.1080/17512786.2013.802474
Broersma, M., & Graham, T. (2013). Twitter As a News Source. Journalism Practice,
7(March 2015), 446–464. doi:10.1080/17512786.2013.802481
Brown, S., & Locker, E. (2009). Defensive responses to an emotive anti-alcohol message.
Psychology & Health, 24(5), 517–28. doi:10.1080/08870440801911130
Coon, D., & Mitterer, J. (2012). Introduction to Psychology: Gateways to Mind and Behavior
with Concept Maps and Reviews. Retrieved from
https://books.google.com/books?hl=nl&lr=&id=jecJAAAAQBAJ&pgis=1 De Castella, K., McGarty, C., & Musgrove, L. (2009). Fear Appeals in Political Rhetoric
about Terrorism: An Analysis of Speeches by Australian Prime Minister Howard.
Political Psychology, 30(1), 1–26. doi:10.1111/j.1467-9221.2008.00678.x
De Hoog, N., Stroebe, W., & de Wit, J. B. F. (2005). The impact of fear appeals on processing and acceptance of action recommendations. Personality & Social
Psychology Bulletin, 31(1), 24–33. doi:10.1177/0146167204271321
De Villiers, E. N. (2008). The effect of the level of fear appeals on attitude towards
advertising and behavioural intention.
Duckitt, J., & Sibley, C. G. (2009). A Dual-Process Motivational Model of Ideology, Politics, and Prejudice. Psychological Inquiry, 20(2-3), 98–109.
EenVandaag. (2015). Debat Pechtold vs Wilders. Retrieved March 19, 2015, from http://politiek.eenvandaag.nl/tv-items/57675/debat_pechtold_vs_wilders Enli, G. S., & Skogerbø, E. (2013). Personalized Campaigns in Party-Centred Politics.
Information, Communication & Society, 16(March 2015), 757–774.
doi:10.1080/1369118X.2013.782330
Feldman, S., & Stenner, K. (1997). Perceived Threat and Authoritarianism. Political
Psychology, 18(4), 741–770. doi:10.1111/0162-895X.00077
GeertWilders.nl. (2014). Speech Geert Wilders in Nashville, USA: new International
Freedom Alliance (IFA) to stop the islamisation of the free world. Retrieved April 13, 2015, from http://www.geertwilders.nl/index.php/94-english/1881-speech-geert- wilders-in-nashville-usa-new-international-freedom-alliance-ifa-to-stop-the-islamisation-of-the-free-world
GeertWilders.nl. (2015). Geert Wilders Weblog. Retrieved June 1, 2015, from http://geertwilders.nl/
Glassner, B. (2004). Narrative Techniques of Fear Mongering. Social Research, 71(4), 819– 826. Retrieved from
http://www.jstor.org/discover/10.2307/40971980?uid=3738736&uid=2&uid=4&sid=2 1106441192293
Gleason, B. (2013). #Occupy Wall Street: Exploring Informal Learning About a Social Movement on Twitter. American Behavioral Scientist, 57(7), 966–982.
doi:10.1177/0002764213479372
Google. (2015). Google Trends. Retrieved April 11, 2015, from
http://www.google.nl/trends/explore#q=islamic state%2C islamitische staat&geo=NL&cmpt=q&tz=
Hansen, J., Winzeler, S., & Topolinski, S. (2010). When the death makes you smoke: A terror management perspective on the effectiveness of cigarette on-pack warnings. Journal
of Experimental Social Psychology, 46(1), 226–228. doi:10.1016/j.jesp.2009.09.007
Huddy, L., Feldman, S., Capelos, T., & Provost, C. (2002). The Consequences of Terrorism: Disentangling the Effects of Personal and National Threat. Political Psychology,
23(3), 485–509. doi:10.1111/0162-895X.00295
Huddy, L., Feldman, S., Taber, C., & Lahav, G. (2005). Threat, Anxiety, and Support of Antiterrorism Policies. American Journal of Political Science, 49(3), 593–608. doi:10.1111/j.1540-5907.2005.00144.x
Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129(3), 339–375.
doi:10.1037/0033-2909.129.3.339
Jost, J. T., Napier, J. L., Thorisdottir, H., Gosling, S. D., Palfai, T. P., & Ostafin, B. (2007). Are needs to manage uncertainty and threat associated with political conservatism or ideological extremity? Personality & Social Psychology Bulletin, 33(7), 989–1007. doi:10.1177/0146167207301028
Klein, K. N., & Mattson, M. (2009). Breast Self-Examination Pamphlets : A Content Analysis Grounded in Fear Appeal. Health Communication, 12(1), 1–21.
doi:10.1207/S15327027HC1201
Mathes, R., & Pfetsch, B. (1991). The Role of the Alternative Press in the Agenda-Building Process: Spill-over Effects and Media Opinion Leadership. European Journal of
Communication, 6(1), 33–62. doi:10.1177/0267323191006001003
McMahan, S., Witte, K., & Meyer, J. (1998). The Perception of Risk Messages Regarding Electromagnetic Fields : Extending the Extended Parallel Process Model to an
Unknown Risk. Health Communication, 10(3), 247–259. doi:10.1207/s15327027hc1003
Mythen, G., & Walklate, S. (2008). Terrorism, Risk and International Security: The Perils of Asking “What If?” Security Dialogue, 39(2-3), 221–242.
doi:10.1177/0967010608088776
Norris, P. (2005). Political parties and democracy in theoretical and practical perspectives. Retrieved May 14, 2015, from
https://www.ndi.org/files/1950_polpart_norris_110105.pdf
NRC. (2014). Openbaar Ministerie vervolgt Wilders voor discriminatie. Retrieved May 13, 2015, from http://www.nrc.nl/nieuws/2014/12/18/openbaar-ministerie-vervolgt-wilders-voor-discriminatie/
NRC. (2015a). “FBI onderzoekt of schutters Texas banden hadden met terreurgroepen” - nrc.nl. Retrieved June 1, 2015, from http://www.nrc.nl/nieuws/2015/05/04/een-van-de-verdachten-schietpartij-texas-in-beeld-bij-fbi/
NRC. (2015b). Twee mannen gearresteerd die mogelijk dader Kopenhagen hielpen - nrc.nl. Retrieved June 1, 2015, from http://www.nrc.nl/nieuws/2015/02/16/deense-politie-twee-mannen-gearresteerd-die-mogelijk-dader-kopenhagen-hielpen/
NRC. (2015c). Wilders op Twitter: PVV vormt fractie in Europarlement. Retrieved June 18, 2015, from http://www.nrc.nl/nieuws/2015/06/15/wilders-op-twitter-pvv-vormt-fractie-in-europarlement/
Oxford Dictionary. (2015). Threat - definition from the Oxford dictionary. Retrieved April 21, 2015, from http://www.oxforddictionaries.com/definition/english/threat
Parmelee, J. H., & Bichard, S. L. (2011). Politics and the Twitter Revolution: How Tweets
Influence the Relationship between Political Leaders and the Public. Lexington
Peters, G. Y., Ruiter, R. A. C., & Kok, G. (2014). Threatening communication: A qualitative study of fear appeal effectiveness. International Journal of Psychology, 49(2), 71–79. Retrieved from
http://web.b.ebscohost.com/ehost/pdfviewer/pdfviewer?vid=1&sid=cd70cca5-9b54-42c8-b7c9-c787ed9c949c%40sessionmgr114&hid=128
Plutchik, R. (1991). The Emotions. Retrieved from
https://books.google.com/books?hl=nl&lr=&id=JaQauznPoiEC&pgis=1 Roskos‐Ewoldsen, D. R., Yu, J. H., & Rhodes, N. (2004). Fear appeal messages affect
accessibility of attitudes toward the threat and adaptive behaviors. Communication
Monographs, 71(1), 49–69. doi:10.1080/0363452042000228559
RTL Nieuws. (2014). Australië bang voor nieuwe aanslag. Retrieved June 1, 2015, from http://www.rtlnieuws.nl/nieuws/buitenland/australie-bang-voor-nieuwe-aanslag RTL Nieuws. (2015). Nieuw rechts blok EU: “Wij zijn de stem van het verzet.” Retrieved
June 17, 2015, from http://www.rtlnieuws.nl/nieuws/politiek/nieuw-rechts-blok-eu-wij-zijn-de-stem-van-het-verzet
Rydgren, J. (2007). The Sociology of the Radical Right. Annual Review of Sociology, 33(1), 241–262. doi:10.1146/annurev.soc.33.040406.131752
Schmuck, D., & Matthes, J. (2014). How Anti-immigrant Right-wing Populist
Advertisements Affect Young Voters: Symbolic Threats, Economic Threats and the Moderating Role of Education. Journal of Ethnic and Migration Studies, 1–23. doi:10.1080/1369183X.2014.981513
Schuermans, N., & De Maesschalck, F. (2010). Fear of crime as a political weapon: explaining the rise of extreme right politics in the Flemish countryside. Social &
Slone, M. (2000). Responses to Media Coverage of Terrorism. Journal of Conflict Resolution,
44(4), 508–522. doi:10.1177/0022002700044004005
Small, T. (2011). What the Hashtag? Information, Communication & Society, 14(6), 872–895. doi:10.1080/1369118X.2011.554572
Trouw. (2014). Welkom thuis, jihadist. Retrieved March 19, 2015, from
http://www.trouw.nl/tr/nl/4500/Politiek/article/detail/3805377/2014/12/07/Welkom-thuis-jihadist.dhtml
Twitter. (2015a). About Twitter. Retrieved May 21, 2015, from https://about.twitter.com Twitter. (2015b). Geert Wilders (@geertwilderspvv) | Twitter. Retrieved May 21, 2015, from
https://twitter.com/geertwilderspvv?ref_src=twsrc%5Eappleosx%7Ctwcamp%5Esafar i%7Ctwgr%5Eprofile
Volkskrant. (2014). Dit zijn de 100 invloedrijkste twitteraars van Nederland. Retrieved April 13, 2015, from http://www.volkskrant.nl/dossier-top-200-van-de-macht/dit-zijn-de-100-invloedrijkste-twitteraars-van-nederland~a3552978/
Volkskrant. (2015a). Teruglezen liveblog - "Gijzelnemer schoot 4 mensen dood’ | Aanslag op Charlie Hebdo. Retrieved June 1, 2015, from
http://www.volkskrant.nl/dossier-aanslag-op-charlie-hebdo/live-geen-doden-gevallen-zegt-franse-justitie~a3826119/ Volkskrant. (2015b). Wie angst wil zaaien, vindt in een open samenleving altijd een weg |
Commentaar. Retrieved April 22, 2015, from http://www.volkskrant.nl/dossier- commentaar/wie-angst-wil-zaaien-vindt-in-een-open-samenleving-altijd-een-weg~a3825777/
Vossen, K. (2011). Classifying Wilders: The Ideological Development of Geert Wilders and His Party for Freedom. Politics, 31(3), 179–189.
Wilson, G. (2013). The Psychology of Conservatism. Retrieved from
https://books.google.com/books?hl=nl&lr=&id=7rlTAQAAQBAJ&pgis=1
Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model.
Communication Monographs, 59(4), 329–349. doi:10.1080/03637759209376276
Witte, K. (1994). Fear Control And Danger Control: A Test Of The Extended Parallel Process Model (EPPM). Communication Monographs, 61, 113–134.
