• No results found

Do immigration and the economy weigh the same? : the effect of topic and framing on attitudes towards EU deepening

N/A
N/A
Protected

Academic year: 2021

Share "Do immigration and the economy weigh the same? : the effect of topic and framing on attitudes towards EU deepening"

Copied!
44
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

DO IMMIGRATION AND THE ECONOMY WEIGH THE SAME? The Effect of Topic and Framing on Attitudes Towards EU Deepening.

Nelleke Keuper 10917179 Master’s Thesis

Graduate School of Communication

Research Master’s Programme Communication Science

Supervisor: Dr. Bert N. Bakker Date: 03-02-2017

(2)

Abstract

As the European Union (EU) is facing some severe challenges, there is an increased incentive to closely investigate the predictors of attitudes towards the EU. Particularly towards EU deepening (i.e. further integration of EU policies), as this seems to have been overlooked to a great extent. In this study, the topics that were salient in the UK during the Brexit campaign (i.e., immigration and economy) are examined in order to determine why the Leave campaign was so successful. In particular, the effect of risk and gain frames on EU deepening was explored. In an experiment (N = 224), risk and gain frames were combined with different topics to test whether there is a stronger effect of risk frames when it comes to the ‘hard’ topic of the economy in predicting attitudes towards EU deepening. The results show no effect of the frames on EU deepening. This does not differ over the different topics, nor is it influenced by a person’s risk acceptance or political knowledge. This study indicates the importance of studying EU deepening attitudes in addition to the abundance of EU widening studies that already exist. In addition, it does not support the assumption that attitudes can be influenced in a one-instance exposure to a frame, contradicting previous research.

(3)

Do Immigration and the Economy Weigh the Same? The Effect of Topic and Framing on Attitudes Towards EU Deepening.

Many European countries are facing increased public debate over the European Union (EU). There are pressing issues that member states are trying to solve in unison, like the refugee crisis. The Brexit referendum was the first of a few controversial political events in 2016, surprising the world and polling agencies when a majority of Great Britain’s citizens voted to leave the EU. This proves the importance of public opinion to the legitimacy and continuation of the EU. The surprise with which the outcome was met, shows that the antecedents of public opinion are at least in this case not thoroughly understood. With the upcoming elections in the Netherlands, Germany, and France, understanding the influence of news media on attitudes towards the EU is very important. Especially since polling agencies expect an electoral win for Euro-sceptical parties like the Freedom Party (PVV) in the Netherlands, and Front National in France. Therefore, this study aims to explore the main topics, and presentation of these topics in the Brexit campaign and test them in a Dutch context.

Among many other factors, there was a clear distinction between the topics used by the Brexit Remain and Leave campaigns, and the use of risk frames in the presentation of these topics (Hobolt, 2016). Prospect theory (Tversky & Kahneman, 1981) states that people are influenced by the way information is presented. If the information is presented in terms of risks people are likely to become risk averse, but if it is presented in terms of gains people are more likely to take risks. Framing research has shown that public opinion is influenced by frames in the media (e.g. Vliegenthart, Schuck, Boomgaarden, & de Vreese, 2008; de Vreese & Boomgaarden, 2006), and risk frames can lead to more negative attitudes (Schuck & de Vreese, 2006). So, according to prospect theory research, using a risk frame should be the

(4)

more effective means of communication (Tversky & Kahneman, 1981). However, this does not hold true for the Brexit campaign, where the risk framed message on the economy did not seem to raise support for the Remain camp. As there was a difference in topics used in the campaign, it can be questioned whether the effect of framing information as a risk is equal for all topics.

The Leave campaign was focused on the core slogan “Take back control”, a direct reference to taking back sovereignty from the EU towards the national government. This taking back of control was mainly focused on the topic of immigration. The Remain campaign seemed to solely focus on the economic risks associated with leaving the EU (Hobolt, 2016; Menon & Fowler, 2016). Johnston and Wronski (2015) show that opinions on ‘hard’ topics, like the economy, are more likely to be influenced by (party) cues than ‘easy’ topics like immigration. Thus, people are expected to follow the cues by their preferred party more so on a ‘hard’ topic than an ‘easy’ topic. This does not show from the Brexit outcome, as people did not follow the risk framed message of the Remain camp on the economy, but seemed to give priority to information on the cultural ‘soft’ topic.

In this study, I try to gain a better understanding of how risk (and gain) framing is affected by the relative ease or difficulty of the topic. To examine this, I want to look at an important, but largely overlooked aspect of EU attitudes – deepening of the EU; in other words, the support for a more central government and handing over part of the national sovereignty. Framing studies have mainly looked at the effects on EU widening (e.g., de Vreese & Boomgaarden, 2006; Schuck & de Vreese, 2006), and research on support for EU deepening is scant. The research question is: Are attitudes towards deepening the EU

influenced by risk and gain frames, and is the size of this effect dependent on the topic of the frame?

(5)

Prior research has shown that risk frames will be more effective in influencing

attitudes towards EU deepening (de Vreese & Kandyla, 2009). In addition, as people seem to be more willing to take cues on a topic that they are less familiar with (Johnston & Wronski, 2015), it is expected that the effect of the risk frame will differ between a cultural and an economic topic. Not only does this study add to the framing literature and EU attitudes literature, it also sheds more light on message dynamics in the public debate. To date, no study has combined ‘easy’ and ‘hard’ issues with risk and gain framing. As there is a distinction in topics focused on by certain political agents, a stronger effect of a specific combination of risk frame and topic might create an even bigger divide in public opinion. EU Attitudes

EU attitudes are multidimensional and consist of attitudes towards – among other dimensions – widening the EU and deepening the EU (Hobolt, 2014; Karp & Bowler, 2006). The support for widening the EU is related to the expansion of the Union by allowing

additional member states to join, whereas support for EU deepening is related to more intensive cooperation and integration of policies throughout the existing Union. Although Boomgaarden, Schuck, Elenbaas, and de Vreese (2011) found an EU attitudes component of ‘EU strengthening’ which contains items on both deepening and widening the EU, other studies clearly find different constructs (Bakker & de Vreese, 2016; Hobolt, 2014). Individual differences are found for deepening and widening of the EU in terms of socio-economic variables. Highly skilled people are more likely to support deepening the EU, but are unlikely to support EU enlargement (Hobolt, 2014). In addition, there are differences in the personality traits related to support for EU deepening and traits related to support for EU widening

(6)

Karp and Bowler (2006) find that people opposing either deepening or widening of the EU are not ambivalent, but rather think of the EU in ways of instrumental self-interest, i.e., what either widening or deepening can mean for them personally. This stance has provided the EU with legitimacy, but it might also lead people to oppose further integration of the EU as it becomes clear what the consequences could be on a personal level. With the use of referendums regarding EU policies, public support for the EU is vital to the survival and further integration of the EU. As the attitudes people hold towards the EU are subject to variation and are flexible over time (Gabel, 1998), it is important to understand what causes these attitudes to change.

Risk and Gain Frames

One of the antecedents of attitudes towards the EU is media influence. Many studies have looked at the presence of frames in media (Lecheler, de Vreese, & Slothuus, 2009; Scheufele, 2000; Shah, Kwak, Schmierbach, & Zubric, 2004) and the effects of these frames on public opinion (de Vreese & Boomgaarden, 2006; de Vreese, van der Brug, & Hobolt, 2012; Lecheler, Keer, Schuck, & Hänggli, 2015; Schuck & de Vreese, 2006). In their

overview, Cacciatore, Scheufele and Iyengar (2016) differentiate between conceptualisations of framing in research. One area of framing research was inspired by Tversky and

Kahneman’s (1981) prospect theory, specifying that frames are different but logically

equivalent ways of presenting information, which activate cognitive knowledge structures and determine how information is evaluated. Goffman (1974), on the other hand, inspired

sociological theories of framing, focusing on the organising structure that gives meaning by selecting some arguments over others. The current study follows the latter, with a definition of framing as “an emphasis in salience of certain aspects of a topic” (de Vreese, 2002, p. 27).

(7)

Vliegenthart, Schuck, Boomgaarden, and de Vreese (2008) find that people are influenced by frames in the media. A frame presenting benefits lead to higher public support for European integration and a frame presenting conflict lead to a decrease in public support. In addition, Semetko, van der Burg, and Valkenburg (2003) find that people were more supportive of the supranational-idealistic ideals of the EU after extensive news coverage of the EU summit taking place in Amsterdam. Zaller (1992) proposed that for people’s attitudes to be influenced by the media, there should be a tendency to report one-sided news, which is supported by findings of de Vreese and Boomgaarden (2006). Although de Vreese and Boomgaarden (2006) find that the media in the Netherlands provide a balanced account overall, there is a difference between news on paper and news on television. When only consuming television news, there is a risk of a one-sided account and subsequent negative influence (de Vreese & Boomgaarden, 2006). These studies indicate that a focus in the media on a certain aspect of an issue can influence public opinion accordingly.

There is a body of research on how the use of frames affects attitudes towards EU widening. Consistent negative information in the news has a negative effect on EU widening attitudes (e.g., de Vreese & Boomgaarden, 2006; de Vreese, van der Brug, & Hobolt, 2012). Risk framing theory suggests that presenting a risk frame should lead to more negative attitudes, and an opportunity frame to more positive attitudes (Cacciatore, Scheufele, & Iyengar, 2016). Indeed, risk frames have a negative effect on EU widening attitudes whereas gain frames have a positive effect on these attitudes (e.g. Schuck & de Vreese, 2006).

However, little research has focused on the use of frames with regards to attitudes towards EU deepening.

De Vreese and Kandyla (2009) have found similar results for EU deepening as studies examining the effects of frames on EU widening (i.e., Schuck & de Vreese, 2006). Risk

(8)

frames are likely to be negatively related to support for a common EU foreign and security policy (i.e., EU deepening), whereas positive relations are found between opportunity frames and support for this policy (de Vreese & Kandyla, 2009). Previous studies have used a more diffuse measure of EU integration, often making use of Eurobarometer standardised

questions. In this study, I want to explore the external antecedents of EU deepening alone, and test whether there is an effect of frames on attitudes towards EU deepening, using a broader measure of deepening than studies have done so far. Risk and gain frames are used to explore their effect on EU deepening and to compare with findings of EU widening research (Schuck & de Vreese, 2006).

In this study risk frames are conceptualised as focusing on the disadvantageous outcomes of a certain policy, and opportunity frames as focusing on the advantageous outcomes of the same policy (Schuck and de Vreese, 2006; Rothman & Salovey, 1997). People are expected to be more positive towards deepening the EU when they have read a text concerned with the positive outcomes, and expected to be more negative after reading a text concerning the risks of deepening the EU. As de Vreese and Kandyla (2009) find an effect of framing on a deepening policy, similar results are expected for a broader measure of EU deepening. The first hypothesis is thus: H1 Risk frames will lead to more negative attitudes (a) and opportunity frames to more positive attitudes (b) towards deepening the EU. Risk Acceptance

Several individual difference variables have been shown to influence the framing effect; personality traits (Levin, Gaeth, & Schreiber, 2002), fear of globalisation (de Vreese & Kandyla, 2009), risk acceptance (Kam & Simas, 2010), issue involvement (De Graaf, de Bruijn, & Van den Putte, 2015), and even neurological differences (e.g. De Martino, Kumaran, Seymour, & Dolan, 2006). Ehrlich and Maestas (2010) find that risk orientation

(9)

affects the influence of frames, by serving as a filter. People may encounter different frames in different instances, but what becomes relevant to them depends on their individual

characteristics, like their risk orientation. People who are exposed to the risk of loss will become more likely to avoid risk, an effect which is stronger when they are low on risk acceptance. On the other hand, people who are exposed to the possibility of gains will be likely to accept risks, even more so when they score high on risk acceptance.

More risk accepting individuals are thus expected to show less negative influence of a risk frame on their attitudes towards EU deepening. As they are more comfortable with a risky option, their attitudes are less likely to be affected when information is presented in terms of risk. The hypothesis holds: H2 The negative effect of risk frames on EU deepening will be stronger for individuals scoring low on risk acceptance.

Hard and Easy Issues

People respond differently to information that is presented in terms of risk than information presented as an opportunity. The previously presented studies have looked at different topics to test the effects of framing, but the effect of the topic has not been taken into account. It might be that there is a greater tendency to discuss one topic in terms of risks than another. In the recent Brexit referendum, the Remain camp used the economy as their main message, framed in terms of risks (Hobolt, 2016; Menon & Fowler, 2016). The Leave camp however, had a focus on immigration issues related to the EU. It might be that the framing of risks is more effective for a certain topic. In the Brexit campaign, it seems like the framing of the economy in terms of risks was not as effective as using the topic of immigration, although there was not a clear emphasis on risk framing for the immigration topic.

This raises the question if there is a difference in the effect of frames on ‘hard’ (i.e. economic), or ‘easy’ (i.e. cultural) topics on attitudes towards the EU (de Vreese,

(10)

Boomgaarden, & Semetko, 2008). ‘Easy’ topics are topics people are familiar with and most likely have already formed opinions on. These opinions are more likely to be based on gut feelings (Carmines & Stimson, 1980). ‘Hard’ topics are more abstract and require more knowledge and motivation to form an opinion on (Carmines & Stimson, 1980). These topics are less likely to be influenced by gut feelings or emotions, and therefore thought to be influenced more by party cues: people look to someone they trust for their decision (Johnston & Wronski, 2015). If framing of information in terms of risks is more likely to occur for a certain topic (e.g. economy), following framing theory, attitudes are more likely to be influenced by this information. However, there might be a different susceptibility to external influences for different topics (Valkenburg & Peter, 2013). Indeed, Johnston and Wronski (2015) show that people are more likely to take cues on a ‘hard’ economic issue than on an ‘easy’ cultural issue.

De Vreese, van der Brug, and Hobolt (2012) find an effect for economic information in accordance with the frame (i.e. positive information leads to more positive attitudes and negative information to more negative attitudes on support for Turkish accession to the EU). They expected to find a stronger main effect for the cultural information as research has shown the importance of ‘easy’ over ‘hard’ issues in relation to EU enlargement, but this was not found (de Vreese, Boomgaarden, & Semetko, 2008). However, as the frames were

constructed in terms of expert advice, one could argue that the effect of the frame depended on the likelihood of people taking cues on a certain topic. This may have caused the lack of effect of the cultural frame, as this is an ‘easy’ topic and people are less likely to take cues on such a topic.

This idea of susceptibility to external cues can be applied to the information context, with people being less influenced by information when it is related to immigration (an ‘easy’

(11)

issue) than information related to the economy (a ‘hard’ issue). To test this: H3a The negative effect of risk frame on EU deepening will be stronger for ‘hard’ issues than for ‘easy’ issues.

Political Knowledge

Previous studies do not show a straightforward effect of political knowledge on the influence of frames (see Chong and Druckman [2007] for an overview). Schuck and de

Vreese (2006) find that people are more likely to be influenced by frames when they were less knowledgeable, and more so by the risk frame than the gain frame. Johnston and Wronski (2015) find that the effect of information on a ‘hard’ topic is very much dependent on exposure to political discourse. People are more likely to be able to translate their personal needs into policy preferences when they are more exposed to political news. It could thus be argued that people higher on political knowledge have a better understanding of what their policy preferences are, and will therefore be less influenced by external information. Thus, it is expected that people with high political knowledge will show less of a framing effect, because a ‘hard’ issue might not be hard for them at all: H3b The difference in the negative effect of risk frames on EU deepening between ‘hard’ and ‘easy’ topics will be smaller for individuals scoring high on political knowledge.

Lastly, the final hypotheses apply to the combination of the framing and topic conditions. The effect of the risk frame is expected to be bigger than the effect of the gain frame (Schuck & de Vreese, 2006), especially in the ‘hard’ topic: H4a The negative effect of the risk frame on EU deepening in the ‘hard’ topic is expected to be larger than the positive effect of the gain frame on EU deepening in the ‘hard’ topic. In addition, people high in political knowledge are expected to be influenced less, therefore the difference between the risk and gain frame is expected to be smaller: H4b The difference in the strength of the effect will be smaller for people high in political knowledge.

(12)

Method

Participants

Participants were recruited using a snowball sampling method. Conscious efforts were made to include people of various age groups and socio-economic backgrounds. This was done by recruiting participants at the University of Amsterdam, but also through family and friends from various age groups and backgrounds. The survey took just over ten minutes to complete, no incentives were offered for participation. It was distributed on social networking sites and using e-mail in the period between 30 November and 20 December 2016. This resulted in a final sample of N = 224 (Mage = 40.05, SDage = 16.53, 53.4 % Female). A

One-sample t-test showed that the mean age was significantly different to the national mean (M = 41.50; CBS), t (223) = -4.93, p < .001, and especially the distribution of age in the sample and the population was distinctly different (graphs can be found in Appendix A). The age group of 20 to 30 year olds is overrepresented in the sample, where the age group of 30 to 50 year olds is underrepresented. However, as random allocation to one of six conditions was assured, this should not influence the results.

Design

The study used a 3 x 2 experimental design of frame (risk/gain/neutral) and topic (the economy/immigration). In addition, the moderating role of political knowledge and risk acceptance was tested. A pre-test was conducted to test the manipulation materials. Stimulus Material

Six short articles were written for this study: three with the topic of immigration, including a risk frame, gain frame, and neutral frame article. In addition, three articles concerned the economy, again with a risk, gain, and neutral frame. The neutral conditions

(13)

were slightly shorter than the treatment conditions, and formed the introductory paragraph. For the risk conditions, the article went on to name a number of risks involved with deepening the EU (forming an EU constitution) for either the Dutch economy, or the impact of

immigration on Dutch culture. The gain condition went on to name the same number of advantages of deepening the EU for either the Dutch economy, or for refugees as well as Dutch society. A complete version of the articles can be found in Appendix B.

A pre-test was conducted to test the manipulation (N = 24). Participants were

randomly allocated to read one out of six articles after which they answered the manipulation check items (e.g. whether the article was about risks, gains, or neither [neutral]; whether the article was about the economy or immigration). If people indicated the article was about risks or gains, estimation items were posed (“How big would you say the risks/advantages are according to the article?” [1 very small – 7 very large]). This provided an opportunity to measure the perceived risks and gains between topics and adjust accordingly.

The analyses showed that there was a marginally significant difference between the framing conditions in terms of reported risk (i.e. people reading a risk frame reporting risks; χ2

(4) = 8.84, p = .065) and a significant difference in terms of reported advantages, χ2 (2) = 7.77, p = .021. This indicates that the risk and gain frames are differentiating between groups, as they should, although for risk it does not reach significance in this small sample. Looking at the percentages of answers, the problem seems to be in the gain condition, in which 55,6% of the participants reported perceiving risks. Splitting the results between topics shows that the articles on economy are correctly identified by 75.0% of participants in the gain and neutral conditions, and 66.7% of participants in the risk condition. Articles on immigration were correctly identified by 80.0% in the neutral and 100% in the risk condition, but 80.0% of

(14)

participants in the gain condition perceived the article to contain risks. This indicates that the economy-risk condition and the immigration-gain condition should be improved.

Another chi square difference test was conducted to see whether the topics were correctly identified. Both the economic article (χ2

(2) = 13.55, p = .001) and the immigration article (χ2

(2) = 17.18, p < .001) show a significant difference between groups. Exploration of the percentages show that for the economy topic everyone answered correctly, for

immigration 76.9% correctly identified the article being about immigration. A One-way ANOVA was conducted with the estimation items (on the relative strength of

risks/advantages) and topic, which was not significant, indicating no difference in the strength of frames between topics (Risk: Mdifference = 1.50, F (1) = 2.46, p = .152; Gain: Mdifference = 0.57, F (1) = 0.35, p = .571). This indicates that the frames in the different topics are comparable in strength, which was the aim.

Overall, the manipulation check shows that for the gain frame article there is a tendency to perceive risks as well, especially in the immigration topic. The text in this condition was adjusted to eliminate any arguments that could be perceived as a risk, and opportunities were emphasised even more. In the economy condition, risks were emphasised more as the percentage recognising risks in the risk frame condition showed room for

improvement. As the other articles were perceived in the way they were intended, and did not differ in strength over the two topics, these were deemed fit for the study.

Measures

EU Deepening. A total of five items on EU deepening were collected from Bakker and de Vreese (2016) and Boomgaarden et al. (2011) which fit the context and purpose of the current study (e.g. ‘I am in favour of efforts being made to unify Europe’). The items were answered on a 7-point Likert scale ranging from 1 (totally disagree) to 7 (totally agree). This

(15)

resulted in a reliable scale (EV = 3.29, Cronbach’s α = .86, R2 = .66, M = 3.66, SD = 1.33), the complete set of items can be found in Appendix C.

For validation purposes, some additional analyses were run. Correlations between EU deepening and the control variables, and an OLS-regression analysis of the control variables on EU deepening are presented in Table 1 and Table 2 in Appendix D. A comparison with the measure used by Boomgaarden et al. (2011) indicates that the pattern of predictors is slightly different. Boomgaarden et al. (2011) found that government approval and anti-immigration attitudes were the biggest predictors of EU strengthening. In addition, age and left and right ideology were significant predictors in their model. For the current measure, gender,

education, immigration attitudes and national identity were significant predictors of EU deepening. Government approval was not tested in this study and ideology was measured as a continuous variable, which was not significant. There seems to be a distinct difference in the strengthening measure from Boomgaarden et al. (2011) and the current EU deepening measure, as was expected.

Risk acceptance. Risk acceptance was measured using the Risk Acceptance Index developed by Kam and Simas (2010), which was translated to Dutch using close translation (Harkness, 2008). The scale consists of seven items, put together from pre-existing scales (see Kam & Simas, 2010), like ‘I like new and exciting things, even if this means I have to break the rules’. In the original index, most of the items are answered on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree), but to enhance consistency throughout the survey, these were answered on a 7-point Likert scale using the same anchors. As the items were converted to a score between 0 and 1 and summed together; this does not affect the performance of the scale. The sum of the Risk Acceptance Index was then rescaled ranging from risk averse (0) to risk acceptant (1). A full list of items can be found in

(16)

Appendix C. One item (marked with an asterisk in the appendix) was deleted after the

Principal Component Analysis showed that the factor loading was very low (.32; Tabachnick & Fidell, 2001). Analyses were run with two versions of the scale, one including and one excluding the item. As there were no substantial differences in outcome, results are reported using the scale without the item. Reliability of the final scale was good (EVscale = 2.81,

Cronbach’s α = .75, R2

= .47, M = 0.57, SD = 0.14).

Political Knowledge. A total of five factual multiple-choice items regarding EU politics are used to measure political knowledge (e.g. ‘What is the name of the president of the European Commission?’). The items are adapted from Boomgaarden, Schuck, Elenbaas, and de Vreese (2011) and were coded as correct (1) or incorrect/‘don’t know’ (0). The coded answers are then added up to create a scale of political knowledge ranging from 0 (no correct answer) to 5 (all answers correct; EV = 1.50, R2 = .32, Kuder-Richardson 20 = .35). The scale does not seem to perform well in terms of internal consistency, and overall political

knowledge scores were low (M = 1.65, SD = 1.13).

Control variables. Several control variables are considered to examine the effect of frame and topic above and beyond control variables, and to see how these variable relate to the EU deepening measure. Controls included socio-economic variables often used in studies towards EU widening (age, gender, education, religion, income) and political ideology (Boomgaarden, Schuck, Elenbaas, & de Vreese, 2011). Immigration attitudes (Azrout, van Spanje, & de Vreese, 2011; de Vreese, Boomgaarden, & Semetko, 2008), economic evaluation (Azrout, van Spanje, & de Vreese, 2011), and national identity (Azrout, van Spanje, & de Vreese, 2012) were also added as control variables to be able to examine the framing effect above and beyond existing attitudes. All items were answered on a 7-point

(17)

Likert scale ranging from 1 (totally disagree) to 7 (totally agree), and can be found in Appendix C.

Results

Manipulation Check

Chi-square tests were conducted to check whether the manipulation of both topic and frame was effective. In the immigration condition, 82.0% of the participants correctly stated the article was about immigration, and 93.8% of participants in the economy condition correctly identified the article was about the economy. The difference between groups was significant (χ2

(2) = 148.77, p < .001), which indicates that the manipulation of the topic was successful. The manipulation of frames was also checked. A chi-square test showed a

significant difference between the neutral frame and the other framing conditions (χ2 (3) = 40.31, p < .001). In the neutral frame (n = 75), 54.7% of participants correctly stated that the article was neither about advantages nor about risk of deepening the EU, 28.0% indicated that the article was about gains, and 14.7% indicated the article was about risks. For the gain frame (n = 73), 72.6% of participants correctly identified the article being about advantages of EU deepening, which again was significantly different from the other conditions (χ2

(3) = 74.53, p < .001). In the risk frame condition (n = 74) this was very similar, 73.7% of participants correctly indicated the article to be about risks of deepening the EU (χ2 (3) = 101.48, p < .001). This indicates that for each framing condition there was a significant difference with the other conditions, although for the neutral article it was less clear whether risks or gains were mentioned or not.

(18)

To examine whether the allocation of participants across conditions was random, a randomisation check was performed using One-way ANOVA’s with the control variables (age, education, income, religion, and ideology), and the independent variables Frame and Topic. None of the control variables showed a significant difference between frames or topics, indicating that participants were allocated randomly. For gender this was checked with a chi-square test, which was not significant for frame or topic, also indicating random allocation. An overview of the results can be found in Table 4 of Appendix D.

Hypothesis Testing

To examine the effect of risk and gain frames on EU deepening, an ANOVA was conducted. The effect of frame on EU deepening was not significant (F (2, 221) = 2.68, p = .071, η2

= .02), post hoc tests did not show significant differences between the neutral (M = 3.77, SD = 1.39), gain (M = 3.84, SD = 1.26), and risk (M = 3.37, SD = 1.32) frames1. These results show no effect for the frames on the outcome variable of EU deepening, and H1 is rejected. However, these are the overall results for the different frames, the rest of the

hypotheses will be discussed to see whether there is a moderating effect, or whether there are different results for the different topics as expected.

To explore the moderating effect of risk acceptance on the effect of framing on EU deepening, an OLS regression analysis was conducted. The analysis included the risk and gain frame dummies, risk acceptance, and the interaction terms. Hypothesis 2 presumes a more negative effect of risk frames for participants with lower risk acceptance. The model was significant (F (5, 218) = 3.50, p = .005, R2 = .07), but the interaction terms with risk frame (b* = 0.24, p= .446) and gain frame (b* = 0.40, p= .229) were not significant, showing no

influence of the level of risk acceptance on the framing effect. Thus, the evidence does not support H2, and it is therefore rejected.

(19)

Although no effects are shown of risk and gain frames overall, differences were expected between the effects of the frames on EU deepening in the different topics. To test this, another OLS regression was conducted with frame, topic, and an interaction term

between the two. The model was not significant (F (5, 218) = 1.18, p = .319, R2 = .03) and did not show a significant result for the interaction of topic with gain frame (b* = 0.06, p = .654) or risk frame (b* = -0.01, p = .954), indicating no difference between the framing conditions in the two topics. Thus, also when looking at the different topics there is no effect of framing found in this study, H3a is rejected.

Hypothesis 3b expected the difference in the effect of risk frames between the two topics to be moderated by political knowledge, with a smaller effect of the risk frame for people high in political knowledge. Therefore, a three-way interaction model with topic, political knowledge, the risk frame dummy and their interaction terms was tested. The model was not significant (F (5, 218) = 1.45, p = .209, R2 = .03), nor was the interaction term between the risk frame, topic and political knowledge, b* = -0.09, p = .331. It must be noted that political knowledge was generally low (M = 1.65, SD = 1.13 out of a possible 5), so people high in political knowledge were virtually non-existent in this sample.

To see whether the negative effect of the risk frame is larger than the positive effect of the gain frame in the ‘hard’ topic, participants in the economy condition were selected and an OLS regression analysis was conducted with risk frame and gain frame dummy variables. The model was not significant (F (2, 110) = 2.01, p = .139, R2 = .04), showing no effect of framing in the ‘hard’ condition. To test the final hypothesis, political knowledge was added to the model to check H4b. The model was not significant (F (5, 107) = 1.35, p = .249, R2 = .06) nor was the interaction term for gain frame and political knowledge (b* = -0.21, p = .266) or risk frame and political knowledge (b* = -0.30, p = .120). This indicates that there is no effect of

(20)

the frames on EU deepening, and political knowledge is not influencing this effect. Thus, H4a and H4b are rejected.

Discussion

This study developed on existing risk and gain frame literature and investigated the differences in effect of these frames when different topics were used. It was expected that there would be a larger effect of information (and particularly risk frames) if the topic was ‘hard’ compared to ‘easy’ (Johnston & Wronski, 2015). This study cannot show an influence of risk and gain framed information on attitudes towards EU deepening, and no effect of the topic of these frames was found. Risk acceptance nor political knowledge moderated the effect of the frames.

These findings are in conflict with de Vreese and Kandyla (2009), which could be related to the extended measure of EU deepening used in this study. For this study, a scale was created focusing on attitudes towards further deepening the EU, whereas de Vreese and Kandyla (2009) use a single policy measure (the support for a Common Foreign and Security Policy). The chosen items for the EU deepening scale were previously used and showed to be conceptually different from EU widening (Bakker & de Vreese, 2016; Boomgaarden et al., 2011). As there were little overlapping control variables for de Vreese and Kandyla (2009) and the current study, the measures could not be compared as such. However, there is an arguable difference in asking about support of a single policy, and a measure focusing on deepening the EU more broadly, which may explain the difference in results.

However, the lack of results may also be related to the manipulation that was used. As Druckman (2001) indicated, there are prerequisites for framing to succeed, an important one being credibility of the source. It was not measured whether people found the source to be credible, but to ensure personal preferences could not confound the effect of the information,

(21)

the name of the source was fictitious. This lack of familiarity could have lead people to question the source’s credibility, thereby limiting or eliminating the framing effect. However, de Vreese and Kandyla (2009) also used a fictitious character in their articles manipulating risk and opportunity frames, which did produce a framing effect. Therefore, familiarity is not expected to be of such influence that it would be able to eliminate an existing effect of framing altogether.

So, although familiarity of the source is not likely to explain the lack of results, the nature of the source might be able to do so. Previous research has shown that people are more likely to take cues on ‘hard’ topics compared to ‘easy’ topics (Johnston & Wronski, 2015) but to test this, party cues were used. However, the party system is more diffuse in the

Netherlands (with 13 parties currently in parliament), so it was not feasible to include party cues in the manipulations to avoid influence of party preferences and a splintered sample. Instead, an expert spokesperson was used. As this study did not show a difference between the two topics when it comes to the effect of frames, it might be that the effect of cues is indeed limited to cues from a specific party.

De Vreese, Boomgaarden, and Semetko (2008) found that cultural factors had

substantially more explanatory power in EU enlargement support than economic factors. This would suggest that, regardless of the inability of the economic manipulation to influence people without a party cue, for the cultural topic there should have been an effect of risk and gain frames. This indicates that hypotheses cannot be based on results of EU widening studies when looking at EU deepening, and there is indeed a profound difference in these attitudes as previous studies suggest (Boomgaarden et al., 2011; Hobolt & Brouard, 2010; Ruiz Jiménez & Torreblanca, 2008). More research is needed to explore the effects of information and

(22)

frames on EU deepening attitudes in depth, and see how they relate to findings on EU enlargement attitudes.

It is important to investigate how people understand deepening issues and how important these issues are to their stance towards the European Union. EU deepening is a complex aspect of attitudes towards the EU, but people might feel strongly about it because there is a direct link to national sovereignty. It could be questioned whether EU deepening itself is a ‘hard’ issue as it is very complex, or an ‘easy’ issue as for many people handing over national sovereignty is a very symbolic and cultural matter. More research is needed to determine what level of elaboration EU deepening attitudes are based on, and how this affects the factors influencing these attitudes.

Limitations

Political knowledge was very low in the current sample, which might be influenced by the sampling method. As a snowball sample was used, chances are that the survey was

distributed among people with similar levels of political knowledge, and therefore overall knowledge remained low. However, it might also be due to the questions that were posed to measure political knowledge. The items were adapted from Boomgaarden et al. (2011) but it might be that the selected questions were too difficult, or the multiple-choice answer options harder for people to choose from. It would have been helpful to see the answer options from Boomgaarden et al. (2011) as the difficulty of multiple choice questions is also determined by possible answers. In their sample the mean is a lot higher (M = 4.33 out of a possible 8), and the scale performed a lot better in terms of consistency (Kuder-Richardson 20 = .81 versus .35). As this study only concerns attitudes towards the EU, the questions were selected on their relevance towards the dependent variable. This may have influenced the results as the questions on national politics may have been easier to answer for many people.

(23)

As Johnston and Wronski (2015) found political sophistication to be an important moderator especially in the financial domain when no party cues were present, the low score on political knowledge may have impacted the possibility to find an effect of framing in the ‘hard’ topic. It might be impossible to detect a difference in topics when party cues are not provided. However, there still should be an effect of frames for the ‘easy’ topic, as for Johnston and Wronski (2015) the effect in the social domain was strong even in the absence of party cues. Political knowledge can therefore not explain the lack of effects of framing in the ‘easy’ topic conditions, but the topic of the manipulation might.

It can be debated if the ‘easy’ topic that was chosen was truly easy. An ‘easy’ topic is described as a cultural, highly symbolic topic (Carmines & Stimson, 1980). Immigration has a strong cultural component, and in the article only cultural risks/opportunities were presented. However, as immigration is a very hot topic, the mere mentioning of immigration is likely to trigger other aspects, among which economic concerns. It might therefore be that the ‘easy’ topic as it was presented was not ‘easy’ at all. It is possible people have been predominantly thinking about the economic consequences of immigration, instead of just the cultural aspects that were explicitly mentioned. The same limitations then apply as to the ‘hard’ topic; the absence of party cues might have decreased the ability to find an effect of the risk and gain frames.

Conclusion and Implications

The aim of this study was to explore the influence of topic on the effect of risk and opportunity frames on attitudes towards EU deepening. We must conclude that attitudes towards deepening the EU are not influenced by the risk and gain frames used in this study, and the size of this effect does not seem to be dependent on the topic of the frames. These results have some possible implications. On a positive note, it may indicate that people are not

(24)

influenced by a short article they have just read. Although some other studies do find these effects (e.g. de Vreese & Kandyla, 2009; Lecheler, de Vreese, & Slothuus, 2009), it is quite reassuring to think that people are not that easily influenced by the information they are exposed to. Even if there is a divide in the topics certain political actors choose to talk about and these topics are likely to be presented in terms of risk, this does not seem to have

profound effects on attitudes people hold. This explanation may be welcomed by news media professionals, as recently people have become more critical about the role of the media in steering public debate.

However, a more worrying explanation could be that people no longer believe in the possibility of deepening the EU and building a strong Union. They may have dismissed the information in the article because they just do not believe that deepening the EU could have the described effects. This might not influence their attitudes towards deepening the EU, because they could still prefer more deepening despite not believing in its feasibility. The implications could be quite grim: if people no longer believe the EU can become a strong and decisive Union, the EU will lose legitimacy. In the months to come, it is up to politicians to determine their stance towards the EU and be explicit about their aspirations, creating the first steps towards the future of the Union.

Acknowledgements

First and foremost, I want to thank Dr. Bert N. Bakker for his feedback and the insights provided throughout the process. Special thanks go out to Marion Keuper, who has been a great help in recruiting participants, as well as Zeph M. C. van Berlo for his calm advice, and Keren Kurlander for her motivational speeches. Last but not least, I want to thank Maurice Lapré, for his relentless support in whatever endeavour, making it possible for me to follow my dreams, helping in whichever way possible.

(25)

Notes

1. The presented models are without control variables. All analyses were run with and without the control variables, adding the controls did not lead to different conclusions. For clarity, the analyses without controls are presented.

References

Azrout, R., van Spanje, J., & de Vreese, C. H. (2011). Talking Turkey: Anti-immigrant attitudes and their effect on support for Turkish membership of the EU. European Union Politics, 12(1), 3–19. doi:10.1177/1465116510389498

Azrout, R., van Spanje, J., & de Vreese, C. H. (2012). When News Matters: Media Effects on Public Support for European Union Enlargement in 21 Countries. Journal of Common Market Studies, 50(5), 691–708. doi:10.1111/j.1468-5965.2012.02255.x

Bakker, B. N., & de Vreese, C. H. (2016). Personality and European Union attitudes: Relationships across European Union attitude dimensions. European Union Politics, 17(1), 25–45. doi:10.1177/1465116515595885

Boomgaarden, H. G., Schuck, A. R. T., Elenbaas, M., & de Vreese, C. H. (2011). Mapping EU attitudes: Conceptual and empirical dimensions of Euroscepticism and EU support. European Union Politics, 12(2), 241–266. doi:10.1177/1465116510395411

Cacciatore, M. a., Scheufele, D. a., & Iyengar, S. (2016). The end of framing as we know it … and the future of media effects. Mass Communication and Society, 19(1), 7–23.

doi:10.1080/15205436.2015.1068811

Carmines, E. G., & Stimson, J. A. (1980). The two faces of issue voting. The American Political Science Review, 74(1), 78–91. doi:10.2307/1955648

(26)

De Graaf, A., de Bruijn, G.-J., & Van den Putte, B. (2015). Effects of issue involvement and framing of a responsible drinking message on attitudes, intentions, and behavior. Journal of Health Communication, 20(8), 989–994. doi:10.1080/10810730.2015.1018623

De Martino, B., Kumaran, D., Seymour, B., & Dolan, R. J. (2006). Frames, biases, and rational decision-making in the human brain. Science, 313(5787), 684–687. doi:10.1126/science.1128356

de Vreese, C. H. (2002). Framing Europe. Television news and European integration. Amsterdam: Aksant/Transaction.

de Vreese, C. H., & Boomgaarden, H. G. (2006). Media effects on public opinion about the enlargement of the European Union. Journal of Common Market Studies, 44(2), 419– 436. doi:10.1111/j.1468-5965.2006.00629.x

de Vreese, C. H., Boomgaarden, H. G., & Semetko, H. A. (2008). Hard and soft: Public support for Turkish membership in the EU. European Union Politics, 9(4), 511–530. doi:10.1177/1465116508095149

de Vreese, C. H., & Kandyla, A. (2009). News framing and public support for a common foreign and security policy. Journal of Common Market Studies, 47(3), 453–481. doi:10.1111/j.1468-5965.2009.01812.x

de Vreese, C. H., van der Brug, W., & Hobolt, S. (2012). Turkey in the EU?: How cultural and economic frames affect support for Turkish accession. Comparative European Politics, 10(2), 218–235. doi:10.1057/cep.2011.27

Druckman, J. N. (2001). On the limits of framing effects: Who can frame? The Journal of Politics, 63(4), 1041–1066. doi:10.1111/0022-3816.00100

(27)

case of free trade. Political Psychology, 31(5), 657–684. doi:10.1111/j.1467-9221.2010.00774.x

Gabel, M. (1998). Public support for European integration: An empirical test of five theories. Journal of Politics, 60, 333–354. doi:10.2307/2647912

Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA: Harvard University Press.

Harkness, J. A. (2008). Comparative survey research: Goals and challenges. In E. D. de Leeuw, J. J. Hox, & D. A. Dillman (Eds.), International handbook of survey methodology (pp. 56–77). New York: Taylor & Francis.

Hobolt, S. (2014). Ever closer or ever wider? Public attitudes towards further enlargement and integration in the European Union. Journal of European Public Policy, 21(5), 664–680. doi:10.1080/13501763.2014.897746

Hobolt, S. B. (2016). The Brexit vote: a divided nation, a divided continent. Journal of European Public Policy, 1763(September), 1–19. doi:10.1080/13501763.2016.1225785 Hobolt, S., & Brouard, S. (2010). Contesting the European Union? Why the Dutch and the

French rejected the European constitution. Political Research Quarterly, 64(2), 309–322. doi:10.1177/1065912909355713

Johnston, C. D., & Wronski, J. (2015). Personality dispositions and political preferences across hard and easy issues. Political Psychology, 36(1), 35–53. doi:10.1111/pops.12068 Karp, J. A., & Bowler, S. (2006). Broadening and deepening or broadening versus deepening:

The question of enlargement and Europe’s “hesitant Europeans.” European Journal of Political Research, 45(3), 369–390. doi:10.1111/j.1475-6765.2006.00302.x

(28)

Lecheler, S., De Vreese, C. H., & Slothuus, R. (2009). Issue Importance as a Moderator of Framing Effects. Communication Research, 36(3), 400–425.

doi:10.1177/0093650209333028

Lecheler, S., Keer, M., Schuck, A. R. T., & Hänggli, R. (2015). The effects of repetitive news framing on political opinions over time. Communication Monographs, 7751(April), 1– 20. doi:10.1080/03637751.2014.994646

Levin, I. P., Gaeth, G., & Schreiber, J. (2002). A new look at framing effects: Distribution of effect sizes, individual differences, and independence of types of effects. Organizational Behavior and Human Decision Processes, 88(1), 411–429. doi:10.1006/obhd.2001.2983 Menon, A., & Fowler, B. (2016). Hard or soft? The politics of Brexit. National Institute

Economic Review, (238), 4–12. doi:10.1177/002795011623800110

Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: the role of message framing. Psychological Bulletin, 121(1), 3–19.

doi:10.1037/0033-2909.121.1.3

Ruiz Jiménez, A. M., & Torreblanca, J. I. (2008). Is there a trade-off between deepening and widening? What do Europeans think? EPIN Working papers, 17(April), 1–31, retrieved from http://aei.pitt.edu/9401/.

Scheufele, D. A. (2000). Agenda-Setting, Priming, and Framing Revisited: Another Look at Cognitive Effects of Political Communication. Mass Communication & Society, 3(2–3), 297–316. doi:10.1207/S15327825MCS0323_07

Schuck, A. R. T., & de Vreese, C. H. (2006). Between risk and opportunity: News framing and its effects on public support for EU. European Journal of Communication, 21(1), 5– 32. doi:10.1177/0267323106060987

(29)

Semetko, H., van der Burg, W., & Valkenburg, P. M. (2003). The influence of political events on attitudes towards the European Union. British Journal of Political Science, 33(4), 621–634. doi:10.1017/S0007123403000280

Shah, D. V., Kwak, N., Schmierbach, M., & Zubric, J. (2004). The interplay of news frames on cognitive complexity. Human Communication Research, 30(1), 102–120.

doi:10.1093/hcr/30.1.102

Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics. Boston: Allyn and Bacon.

Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211, 453–458. Retrieved from

http://www.jstor.org.proxy.uba.uva.nl:2048/stable/1685855

Valkenburg, P. M., & Peter, J. (2013). The differential susceptibility to media effects model. Journal of Communication. doi:10.1111/jcom.12024

Vliegenthart, R., Schuck, a. R. T., Boomgaarden, H. G., & De Vreese, C. H. (2008). News coverage and support for European integration, 1990-2006. International Journal of Public Opinion Research, 20(4), 415–416. doi:10.1093/ijpor/edn044

Zaller, J. (1992). The nature and origins of mass opinion. Cambridge: Cambridge University Press.

(30)

Appendix A Figure 1. Distribution of age in the sample.

Figure 2. Distribution of age in the Dutch population.

(31)

Appendix B Manipulation articles

The additions to the neutral article for the gain and risk [in parentheses] frames are underlined.

Immigration

Komt er toch een Europese grondwet?

BRUSSEL In 2005 stond een Europese grondwet al hoog op de politieke agenda. Er werd toen in nationale referenda, onder andere in Nederland, tegen de grondwet gestemd. Door de huidige vluchtelingenproblematiek wordt in Brussel nu opnieuw gesproken over diepere samenwerking binnen de EU. De vluchtelingen crisis loopt steeds hoger op en een breed gedragen oplossing is nog lang niet in zicht. Het plan voor een Europese grondwet werd ingebracht door de Zweedse eurocommissaris Jansson en moet ervoor gaan zorgen dat de daadkracht van de EU wordt vergroot en beslissingen sneller kunnen worden genomen. Socioloog Marnix Revel ziet enorme voordelen in verdere integratie van de Europese Unie: “Asielzoekers kunnen beter worden verdeeld over alle lidstaten, waardoor er minder

asielzoekers naar Nederland komen en ze dus beter kunnen worden opgevangen.” Daarnaast kunnen door een diepere samenwerking beslissingen sneller worden genomen, omdat de Raad van de Europese Unie deze goedkeurt en niet alle nationale parlementen in afzonderlijke lidstaten. Het grote voordeel van deze slagvaardigheid van de EU is dat vluchtelingen minder lang in onzekerheid verkeren en eerder kunnen beginnen met hun integratie in de Nederlandse cultuur. “Snellere integratie is heel belangrijk voor het behoud van de Nederlandse identiteit, omdat de nieuwe Nederlanders hierdoor beter inburgeren en geen kans krijgen om te

radicaliseren”, aldus Revel. Hij pleit er dan ook voor dat politieke partijen in Den Haag vóór een Europese grondwet te stemmen. (Socioloog Marnix Revel wijst op de enorme risico’s van verdere integratie van de Europese Unie: “Nederland zal minder te zeggen hebben over de hoeveelheid asielzoekers die opgevangen moeten worden. Er kunnen geen strengere grenscontroles worden ingesteld of grenzen helemaal worden gesloten.” Daarnaast hebben lidstaten minder inspraak omdat beslissingen niet meer door de nationale parlementen worden goedgekeurd, maar door de Raad van de Europese Unie. “Door de invoering van de grondwet hebben we niet langer zelf controle over onze nationale veiligheid. De versnelde procedure geeft bovendien te weinig tijd voor een goede inburgering, wat het risico op radicalisering van

(32)

immigranten sterk vergroot. De grote culturele verschillen zorgen er bovendien voor dat onze Nederlandse identiteit steeds meer verloren gaat”, aldus Revel. Hij pleit er dan ook voor dat politieke partijen in Den Haag tegen een Europese grondwet te stemmen.)

Economy

Komt er toch een Europese grondwet?

BRUSSEL In 2005 stond de Europese grondwet al hoog op de politieke agenda. Er werd toen in nationale referenda, onder andere in Nederland, tegen de grondwet gestemd. Door de huidige economische situatie wordt in Brussel nu opnieuw gesproken over diepere

samenwerking binnen de EU. De economische crisis is nog niet voorbij, nog steeds worden er extra maatregelen getroffen om de economische groei te stimuleren. Het plan voor een

Europese grondwet werd ingebracht door de Zweedse eurocommissaris Jansson en moet ervoor gaan zorgen dat de daadkracht van de EU wordt vergroot en beslissingen sneller kunnen worden genomen. Econoom Marnix Revel ziet enorme voordelen van verdere integratie van de Europese Unie: “De concurrentiekracht van Europa zal toenemen op de wereldmarkt. Hierdoor zal ook de groei van de Nederlandse economie verder worden gestimuleerd.” In het afgelopen jaar is de Nederlandse economie met 2,1% gegroeid, Revel voorspelt dat dit kan oplopen tot 3% bij verdere integratie van de EU. Deze groei heeft een positief effect op de werkloosheid, die sinds 2014 al met 1,5% is gedaald. “Verdere Europese integratie leidt tot een sterkere en snellere daling van de werkloosheid van nog eens 1,5%. Daarnaast biedt meer eenheid in de regelgeving een duidelijk wettelijk kader, wat de

internationale handel versoepelt”, aldus Revel. Hij pleit er dan ook voor dat politieke partijen in Den Haag vóór een Europese grondwet te stemmen. (Econoom Marnix Revel waarschuwt voor de enorme risico’s van verdere integratie van de Europese Unie voor de Nederlandse economie: “De Europese economie wordt instabieler, en de concurrentiekracht van Europa op de wereldmarkt zal afnemen. Dit heeft directe gevolgen voor de handelspositie van

Nederland, en geeft een reëel risico op een nieuwe recessie.” In het afgelopen jaar is de Nederlandse economie met 2,1% gegroeid, Revel voorspelt dat dit zal teruglopen tot 1,4%. Deze daling heeft een negatief effect op de werkloosheid, die sinds de dalende trend van 2014 weer met 1 tot 1,5% zal toenemen. “Verdere Europese integratie beperkt onze controle over de Nederlandse economie en legt onze welvaart in handen van Europa”, aldus Revel. Hij pleit er dan ook voor dat politieke partijen in Den Haag tegen een Europese grondwet te stemmen.)

(33)

Appendix C National Identity Helemaal mee oneens Mee oneens Beetje mee oneens Niet mee eens/niet mee oneens Beetje mee eens Mee eens Helemaal mee eens

Nederlander zijn betekent

veel voor me.       

Ik ben er trots op een

Nederlands burger te zijn.       

Ik voel me verbonden met

andere Nederlanders.       

Ik voel me meer Nederlands

dan Europees.       

Economic Evaluation

Slechter 1 2 3 4 5 6 Beter 7

Hoe denkt u dat de economische situatie in Nederland zich de komende twaalf maanden zal

ontwikkelen?

      

Hoe denkt u dat de economische situatie in de Europese Unie zich de

komende twaalf maanden zal ontwikkelen?

      

Hoe denkt u dat uw persoonlijke financiële situatie zich de komende

twaalf maanden zal ontwikkelen?

(34)

Immigration Attitudes Helem aal mee oneens Mee oneen s Beetje mee oneens Niet mee eens/niet mee oneens Beetje mee eens Mee eens Helemaal mee eens

Immigranten maken misbruik

van sociale voorzieningen.       

Immigranten zijn een bedreiging voor de veiligheid in

Nederland.

      

Immigranten krijgen een nadelige behandeling van de

overheid.

      

Immigranten zijn een belangrijke oorzaak voor criminaliteit in Nederland.

      

Immigranten zijn een verrijking

voor de Nederlandse cultuur.       

De religieuze praktijken van immigranten vormen een bedreiging voor de Nederlandse

levensstijl en tradities.

(35)

Risk Acceptance

1. Sommige mensen vinden dat je voorzichtig moet zijn met grote veranderingen in je leven, stel dat zij op 1 staan. Anderen zeggen dat je nooit veel zult bereiken in het leven zonder risico’s te nemen, stel dat deze mensen op 7 staan. Weer anderen zitten daar tussen in. Waar zou u zichzelf plaatsen op die schaal van 1 tot 7?

o 1 o 2 o 3 o 4 o 5 o 6 o 7

2. Stelt u zich voor dat u op paarden aan het wedden bent en u heeft veel gewonnen in de derde of vierde race. Zou u blijven spelen of zou u uw winst innen? *

o 1 Zeker blijven spelen

o 2 Waarschijnlijk blijven spelen o 3 Weet niet

o 4 Waarschijnlijk mijn winst innen o 5 Zeker mijn winst innen

3. Over het algemeen, hoe makkelijk of moeilijk is het voor u om het nemen van risico’s te accepteren?

o 1 heel makkelijk o 2 best wel makkelijk o 3 redelijk moeilijk o 4 heel moeilijk

(36)

4 - 7. Geef aan welk antwoord het beste bij u past. Helemaal mee oneens Mee oneens Beetje mee oneens Niet mee eens/niet mee oneens Beetje mee eens Mee eens Helemaal mee eens Ik wil vreemde plekken graag verkennen.       

Ik doe graag dingen

die me bang maken.       

Ik hou van nieuwe en opwindende ervaringen, ook als ik

hiervoor de regels moet overtreden.

      

Ik geef de voorkeur aan vrienden die

spannend en onvoorspelbaar zijn.

      

Manipulation Check

Beantwoord u alstublieft de volgende vragen over het artikel dat u net heeft gelezen. Ging het artikel dat u heeft gelezen over:

o De voordelen van verdere integratie binnen de Europese Unie o De risico’s van verdere integratie binnen de Europese Unie o Geen van beide

o Weet niet

Werd er in het artikel gesproken over: o Immigratie

o Economie o Weet niet

(37)

EU Deepening

Nu volgen er een aantal stellingen over de EU, geef aan in hoeverre u het ermee eens bent. Er zijn geen juiste antwoorden, het gaat om uw mening, dit is vaak het eerste antwoord dat in u opkomt. Helemaal mee oneens Mee oneens Beetje mee oneens Niet mee eens/niet mee oneens Beetje mee eens Mee eens Helemaal mee eens

De Europese Unie zou één land

moeten worden.       

Ik ondersteun meer beslissings-/beleidsbevoegdheid op

EU-niveau.

      

De beslissingsbevoegdheid van de Europese Unie moet worden

vergroot.

      

Ik ben voor inspanningen om

Europa verder te verenigen       

Ik heb persoonlijk voordeel van het Nederlands lidmaatschap

van de EU.

(38)

Political Knowledge

Nu volgen er enkele vragen met betrekking tot uw kennis over de EU. 1. Hoeveel Eurocommissarissen zijn er in de Europese Commissie?

o 15 o 28 o 51

o Weet niet

2. Wat is de naam van de huidige voorzitter van de Europese Commissie? o Jean-Claude Juncker

o Romano Prodi o Frans Timmermans o Weet niet

3. Welk land houdt momenteel het voorzitterschap van de EU? o Nederland

o Malta o Slowakije o Weet niet

4. Wat is de naam van de Nederlandse Eurocommissaris? o Neelie Kroes

o Frans Timmermans o Jeroen Dijsselbloem o Weet niet

5. Hoeveel landen zijn op dit moment in onderhandeling over lidmaatschap van de Europese Unie?

o 6 o 8 o 12 o Weet niet

(39)

Control Variables Wat is uw leeftijd? ……… Wat is uw geslacht? o Man o Vrouw o Anders

Wat is uw hoogste voltooide opleiding (of opleiding die u binnenkort verwacht te voltooien)? o Alleen Basisonderwijs/ LBO/VBO

o VMBO/MAVO o MTS/ MBO

o HAVO/Athenaeum/Gymnasium/VWO

o HBO/ HTS/ HEAO/ WO-bachelor/ kandidaats o WO-master/ doctoraal

Wat is uw gemiddelde maandelijkse inkomen? (Als u geen inkomen heeft omdat u bijvoorbeeld studeert, kiest u dan voor de eerste optie).

o < EUR 1150,–

o EUR 1150,– tot 1800,– o EUR 1800,– tot 2600,– o >EUR 2600,–

Beschouwt u zichzelf als gelovig? o Ja

o Nee o Anders

Rekent u zichzelf tot een bepaalde kerkelijke gezindte of levensbeschouwelijke groepering en zo ja: tot welke?

o Geen kerkelijke gezindte of levensbeschouwelijke groepering o Rooms-Katholiek o Nederlands Hervormd o Gereformeerde kerken o Protestantse Kerk Nederland o Islam

(40)

In de politiek wordt soms gesproken over links en rechts. Wanneer u denkt aan uw eigen politieke opvattingen, waar zou zichzelf dan plaatsen op een lijn van 0 tot en met 10, waarbij 0 staat voor links en 10 voor rechts?

o Links 0 o 1 o 2 o 3 o 4 o 5 o 6 o 7 o 8 o 9 o Rechts 10

(41)

Appendix D

Table 1

Results of the exploratory factor analysis of the EU deepening scale using PAF

Factor 1

De EU zou één land moeten worden. .53

Ik ondersteun meer beslissings-/beleidsbevoegdheid op EU-niveau.

.89

De beslissingsbevoegdheid van de Europese Unie moet worden vergroot.

.92

Ik ben voor inspanningen om Europa verder te verenigen.

.86

Ik heb persoonlijk voordeel van het Nederlands lidmaatschap van de EU.

.56

Eigenvalue 3.29

R2 .66

(42)

Table 2

Pearson correlations of EU deepening with control variables (N = 224).

EU deepening Age -.32 Sex -.03 Education .32 Income -.16 Religion .06 Ideology -.23 National Identity -.28 Immigration attitudes -.33 Economic evaluation .04

(43)

Table 3

OLS regression of control variables on EU deepening.

B SE Age -.003 .007 Gender (female) -.40* .17 Income -.16 .09 Education .23** .08 Ideology -.06 .04 Economic evaluation .06 .09 Immigration attitudes -.19* .09 National identity -.24** .07 F 6.92*** N 224 R2 .25

Note: Cell entries are unstandardized b coefficients and standard errors. *** p<.001, **p<.01, *p<.05

(44)

Table 4

Randomisation check: ANOVA results for control variables.

Frames Topics Age F(2) = 0.53 p = .592 F(1) = 0.00 p = .993 Education F(2) = 0.33 p = .718 F(1) = 0.30 p = .582 Income F(2) = 0.004 p = .996 F(1) = 0.26 p = .611 Religion F(2) = 0.96 p = .386 F(1) = 0.23 p = .635 Ideology F(2) = 0.78 p = .459 F(1) = 0.38 p = .536 Gender* F(2) = 0.42 p = .811 F(2) = 4.35 p = .133 Note: *chi-square test for gender

Referenties

GERELATEERDE DOCUMENTEN

“Positive” = Evaluate where Player X is mentioned negatively related to human capital (this because it will make the transfer more likely): Assess this option if the

Other important supporting industries are online platforms as DMSs, SM, ranking companies and music magazines as DJ MAG, which promote and market artists on a

De factoren die het meeste invloed lijken uit te oefenen op de uitvoering van het rolstoelenbeleid zijn: starheid, flexibiliteit en het beleidsnetwerk van de gemeenten,

Government ministries tend to show higher significance in the impact of its received attention on Twitter (i.e. amount of tweets about a press releases information) on both news

By analyzing 400 images published in Germany, France, Denmark and Portugal during the German election years 2009, 2013, 2017 and the refugee crisis 2015, the study finds that there

Furthermore, the ARPES data demonstrates that the electronic structure of Au(111) is modulated by the molecular network on a macroscopic scale, which indicates the possibility of

The study found a long-run relationship between government spending and job creation in the mining sector but there was no evidence of long-run

Treatment of fibrotic mice with the lower dose of IFNα and GPI reduced the CD31 protein expression (Fig. 3A3, 3A5, 3B), confirming the anti-angiogenic effect of IFNα