The Threats to the Catalan Wallet
Effects of Negative Economic News in the Catalan Independence Debate
Oscar Claramunt Aguayo 11592753
Master’s Thesis
Graduate School of Communication – University of Amsterdam Master’s Program Communication Science
Supervisor: Ruud Wouters 29th June 2018 Word Count: 7317
Abstract
Negative newspaper headlines are argued to drive media selection. Exposure to negative economic news also seems to influence voting intentions. This paper conducted a double
experiment to test these assumptions on Catalan citizens (N=192) in the context of the Catalan independence debate. The results of the first experiment showed that positive newspaper headlines were more likely to be selected than negative headlines in regards to the economic situation of Catalonia. Contrary to expectations, individuals with extreme opinions on the Catalan independence issue did not select negative headlines to a larger extent than moderates. The second experiment investigated the effect of negative articles stressing economic
consequences of either leaving or remaining part of Spain. However, none of these conditions were found to have a significant effect on voting intentions in a referendum. This could be explained by the assumption that participants in such a polarized environment had strong predispositions. Having a shared national identity did not moderate this relationship. These findings partly align with theories on backlash effect, which question negativity bias on media selection. In addition, the study enriches the existing literature on negativity and political behavior in the Catalan context, where limited research on negativity has been conducted. Practical implications for both journalism and the political realm are also discussed.
Introduction
Negativity draws attention. Amongst a variety of scholars and academic domains there is an apparent consensus on this notion. Academics in evolutionary psychology already suggest that we are biologically programmed to pay attention to negativity in order to effectively react to any threat and be able to survive (Fuller, 2010; Shoemaker, 1996; Soroka, 2014). It is also commonly argued that people tend to be attracted to negative news to a greater extent than they are to positive news (Soroka &
McAdams, 2015) and that is one of the reasons why negativity is claimed to be so prevalent in the media (Shoemaker, 1996). In fact, several studies have indicated that negative headlines in the newspapers are selected more often than positive ones by their readers (Donsbach, 1991; Meffert, Chung, Joiner, Walks & Garst 2006; Zillman, Chen, Knobloch & Callison, 2004). However, negativity doesn’t only seem to affect media selection. The exposure to negative messages can also shape citizens’ political attitudes and their voting decisions. Specifically, exposure to negative economic information has been revealed to affect one’s political attitudes and behavior, especially when there is a perceived risk on the personal financial situation (Wahlberg & Sjoberg, 2000). This means that the mere exposure to negative news can influence, if perceived as a real risk, people’s voting decisions.
The effects of negativity in the news have been largely studied by communication scholars both in the US and some countries in Europe like Germany. However, none of them has studied selection and voting intention effects in particularly polarized political contexts like the independence debate in Catalonia. This is precisely the purpose of this research. The following pages aim to bring some light on the topic by filling this research gap. In order to do that, two different studies were conducted. The first one is oriented to clarify the attractiveness of headlines stressing negative economic consequences in comparison to positive ones. This study partly replicates headline selection studies developed in the US and Europe, specifically in Germany, applying a similar methodology in the Catalan case. The second one tries to identify the effects that exposure to news stressing economic consequences may have on the voting intentions in a hypothetical referendum of independence in Catalonia. More precisely, the two experimental studies conducted in this article aim to answer the following research questions:
RQ Study 1: To what extent does a focus on negative economic consequences drive citizens’
news media selection on the Catalan independence issue?
RQ Study 2: Under which conditions can news that stress negative economic consequences have
an effect on citizens’ voting intentions in an independence referendum of Catalonia?
By trying to respond to the previous research questions, these studies show a high academic relevance since they contribute to explain the effects of negative economic news simultaneously on headline selection and voting intentions in the Catalan context. In fact, this empirical research is the first one to investigate both effects in an identic population with a distinct political movement in Europe as a specific context. The Catalan case is a particularly polarized topic where two opposed models (remaining a Spanish region or becoming an independent state) compete to gain support both in the media and among the population. In these conditions, negative economic consequences of either remaining or leaving Spain have become particularly prevalent in the media. Thus, these two studies aim to measure the effects of negative news in a different context than previously examined, which could yield different results.
This research can also be of service to the journalistic industry and the communication
professionals overall, by providing them with specific insights about the behavior of the Catalan citizens in terms of media consumption and voting behavior. More precisely, the conclusions of these studies may assist them in their goal of remaining competitive in a fierce market, threatened by the boom of a fake news industry. Online journalism has particularly empowered media consumers, allowing them to choose from a particularly extensive variety of media outlets and topics. Only by understanding how citizens select and process digital news, journalists and media outlets in general may be able to attract readers and maintain their reputation as reliable sources. On the other hand, the results of the two studies can also be of great relevance for the political realm and political communication professionals in Catalonia, who may better grasp how citizens react to certain types of information in the media and can, thus, design specific communication strategies accordingly.
The following pages will further elaborate on the topic by following the next suggested structure. First, the specific Catalan political context will be discussed. The next section will include the theoretical
background on negative news and media selection as well as negative economic news and voting
intentions. Then, the methodology conducted in the study will be detailed and the results obtained will be explained. Last, the discussion with the conclusions and the limitations of the study will be presented. The Catalan political context
Since the Spanish region of Catalonia is the focal point of this research, some facts about the modern history of the region, as well as some details concerning its recent political events, need to be presented for a better comprehension of the nature and relevance of this work.
With the restoration of democracy after the death of the Spanish dictator Francisco Franco, Spain developed a domestic administrative division that organized the country in 17 autonomous communities (or regions) and 2 autonomous cities (Melilla and Ceuta). Among all of them, only three were considered “historical communities” because of having already
approved their own Statute of Autonomy in the past, during the Second Spanish Republic (1931-1939). Catalonia is one of these regions, together with the Basque Country and Galicia. Located in the northeast of Spain along the Mediterranean Sea, Catalonia regained some local autonomy in the last decades by being in charge of some cultural, educational, health and security affairs. However, it was only during the last decade that a strong pro-independence movement arose in the region and the Catalan government organized an independence referendum. The plebiscite was claimed illegal by the Spanish Government and never counted with its approval. As a consequence, political polarization in the region certainly increased and the political conflict reached its peak. Today, the Catalan former president Carles Puigdemont has fled the country to avoid imprisonment and a new nationalist government has been appointed, after months of a Spanish direct control over the Catalan institutions.
This new government was appointed six months after the 2017 Catalan elections, which resulted in the most polarized parliament in the recent history of the region. Ciutadans, known as
one of the main unionist parties, became the first party in votes and seats (36 out of 135), being the first non-nationalistic party to win an autonomic election in the Catalan history. The two main pro-independence parties, Junts per Catalunya and Esquerra Republicana de Catalunya came second and third, with 34 and 32 seats respectively. Nevertheless, these results allowed pro-independence parties combined to obtain the majority of seats in parliament (although not of votes). A continuous dramatic fall of the Partit dels Socialistes de Catalunya (PSC) in the ballots, framed in the crisis of social democratic parties in Europe and reinforced by an ambivalent position in Catalonia’s self-determination rights debate, allowed Ciutadans to overpass this party in most of PSC’s traditional strongholds. However, an alternative explanation could be taken into account. Ciutadans had a strong position against the independence movement in Catalonia during the election campaign and its leader, Inés Arrimadas, often referred to the potential negative economic consequences of a hypothetical independence of the region. Moreover, the
pro-independence block often made reference to the possible economic gains or losses for Catalonia, depending on whether the region would become independent or remain in Spain. In an election context with an extremely high participation, where new voters that previously tended to abstain from voting did participate, the economic gains and loses approach may have played an
important role, if not crucial, for many citizens’ voting decisions. These potential economic consequences of the permanence or the independence of Catalonia were also often portrayed in the Spanish and Catalan media outlets, especially during electoral campaign. However, it is yet to investigate how Catalan voters may perceive negative economic information regarding the issue, in comparison to positive facts and how this, in turn, affects their media selection and political behavior.
Theoretical Background Negative News and Media Selection
The progressive fragmentation the media landscape has experienced during the last decades (Prior, 2007) has empowered citizens by allowing them to choose the media they consume every day. With the boom of digital journalism, more and more citizens tend to select their news from an online environment and among them, the majority does so by reading news texts rather than watching videos (Newman, Fletcher, Kalogeropoulos & Nielsen, 2017). It is in this digital context that extensive amounts of information and disinformation coexist and have the opportunity to reach the general public. For this reason, understanding what specific news attributes may affect our selection of information becomes more relevant than ever. Among others, one of the characteristics that seem to catch our attention and influence our media selection is precisely the presence of negativity in the news.
So far, many scholars have pointed out that we are naturally attracted to negative
information because it is perceived as more salient and informative (Fiske, 1980; Pratto & John, 1991; Steiner, 1979). There are mainly two different theoretical approaches that have aimed to explain the underlying mechanisms of our attraction to negativity. These are the expectancy violations theory and the automatic evaluative categorization.
The first one, developed by Burgoon (1993) and further supported by Meffert et al. (2006) and Soroka (2014), is based on the assumption that individuals tend to pay attention to events that violate their own expectations. More precisely, since most of people’s day-to-day lives are primarily positive, negative information stands out and becomes salient.
On the other hand, the second theory suggests that we tend to prioritize negative information because it is the automatic psychological response that happens in our brain when
evaluating any message (Carretié, Albert, López-Martín, & Tapia, 2009; Ito, Larsen, Smith & Cacioppo, 1998; Pratto & John, 1991). Our brain processes negativity quickly in order to detect any potentially dangerous situations or threats and react in time to avoid any possible harm. As Mutz (2007) puts it, we immediately have a “gut-level reaction” to any negative content.
More specifically, concerning the media environment, it is commonly argued that negative content is particularly present in the news because humans are more interested in or reactive to negative information (Shoemaker, 1996). This means that not only news consumers but also journalists themselves are attracted to negativity. In fact there is a body of scholars that have developed an extensive literature on the gatekeeping role of journalists and their tendency to publish negative information because of its apparent higher newsworthiness (Shoemaker, 1991; Shoemaker & Vos, 2009). Actually, the mere reason d’être of the media has historically been related to stressing out the negative events in the world. Carlyle (1841), for instance, already defined the media as the “fourth estate” whose aim is precisely to identify problems and making them salient while denouncing the abuses of the state.
In regards to media consumption and news selection, negativity seems to attract
newspapers’ readers, in particular. According to a study conducted by Donsbach (1991) among readers of several German newspapers, individuals tended to select negative headlines about politicians more often than positive ones. In another study, Zillmann et al. (2004) also discovered that participants tended to choose negative news content, in comparison to factual news,
particularly when conflict or agony was present in the news. Meffert et al. (2006) also found support for this negative bias in media selection, especially among individuals with strong initial preferences or extreme viewpoints. In fact, these participants tended to select and spend more time reading negative news about their preferred candidate, even if they ended up having an
overall more positive evaluation of the candidate. Thus, extreme political views also seem to be a predictor of the negativity bias. According to a study developed by Keene, Shoenberger, Berke, & Bolls (2017) extreme conservatives tend to have a higher defense system activation (DSA), a correlate of the negativity bias of the aversive system, in comparison to more moderate
individuals. Taking into account the findings mentioned above, it is expected that:
H1: Headlines that stress negative economic consequences regarding the Catalan independence debate are more likely to be selected than those ones focusing on positive economic consequences.
H2: Participants with extreme views on the topic are more likely to select headlines that stress negative economic consequences regarding the Catalan independence debate than those with more moderate views.
Figure 1: Conceptual Model for H1 and H2
Negative Economic News and Voting Intentions
Negative news can be found throughout the entire media spectrum and amongst a vast variety of sections and topics. Among all of them, economic news have been claimed to be particularly prevalent in the media and seem to have a special capacity to influence people’s behavior. As Harrington (1989) concluded, most of American broadcasters networks give
Negative Economic
Consequences Headline Article Selection
Extremism of Issue Stance
substantially more coverage to negative economic trends than to positive ones. This tendency in the media can be partly explained by the conclusions from the research developed on loss aversion theory, which suggests that individuals tend to care more about a loss than about a gain of the same importance (Kahneman & Tversky, 1979, Tversky & Kahnmean, 1991). In fact, these findings have been constantly observed across lab and real world environments (Kahneman & Thaler, 1991; Shoemaker and Kunreuther, 1979; Tversky, Slovic & Kahneman, 1990) as well as at an aggregate-level, in macroeconomic dynamics. Bowman, Minehart & Rabin (1999) for instance, explain that consumption tends to drop to higher levels when the economy contracts, than it improves when the economy expands.
Certainly risk has been repeatedly found to be a powerful mobilizing factor and media exposure can influence the formation of an individual’s perception of it, according to Wahlberg and Sjoberg (2000). For instance, when studying media economic information Garz (2012) discovered that individual’s perceptions of job security decreases with a higher exposure to information regarding the labor market. In addition, exposure to negative economic news seems to be connected with a decreased willingness to take risks. As Tausch & Zumbuehl (2016) explain, negativity in economic news can trigger a perceived increase of general risks in job security, economic development or the general economic situation.
Other authors have claimed that negative economic coverage not only has a direct impact on the perceived risks of job security, economic development and general economic situation of the country but it can also have an effect on voting behavior (MacKuen, Erikson, and Stimson 1992; Wahlberg & Sjoberg, 2000). However, as Mueller (1970) stated, an issue must have certain saliency in order to influence the vote of the electorate. This salience, he claimed, is often achieved when economic news is negative, since negative information is often perceived as more
important and relevant than positive news. Thus, there are indications that voters may take negative economic information into consideration when they approach the ballot boxes. In fact, there is a general consensus among scholars on the belief that support towards a government decreases to a higher extent when the economy is deteriorating than it increases when it is improving (Bloom & Price, 1975; Nannestad & Paldam, 1997). More specifically, there is evidence that US presidents have been historically more penalized electorally for negative economic trends than for positive ones (Soroka & McAdams, 2015). Some scholars like Mueller (1970), have even been able to explain the popularity of a given US president simply by looking at the unemployment rate.
In addition, economic perceptions have also been suggested to affect support for more specific political causes. For instance, when studying the effect that economic expectations had in the referendum of the euro in Denmark, Semetko and Vreese (2004) found that those who assessed their own and the country’s economic expectations as positive were more likely to vote yes. In contrast, people with lower personal and national financial expectations were more inclined to vote no. The influence of economic expectations has also been studied specifically in relation to the independence movement in Catalonia. According to Muñoz and Tormos (2015), economic expectations can predict the increase or decrease of support for independence in Catalonia, but only for those that have a dual national identity to a higher or lower extent. Thus, we can assume that the economic factor may have a potential impact on the voting behavior in a hypothetical referendum of independence in Catalonia for those who identify themselves as both, Catalan and Spanish. This means these citizens may be particularly affected by the potential economic risks they see on either staying in Spain or becoming an independent state when deciding their vote in a plebiscite. On the contrary, those citizens feeling exclusively Catalan or
Spanish may not even consider the economy as a relevant factor. As a result their voting intentions may not vary (at least to the same extent) depending on the economic risks involved. Taking the previously discussed findings into account, the following hypotheses are proposed:
H3: Participants are more likely to vote YES in an independence referendum after being exposed to a pro-independence article stressing the negative economic consequences of staying in Spain than one that stresses the positive economic consequences of the Catalan independence.
H3b: This effect is stronger for those who share a double national identity.
H4: Participants are more likely to vote NO in an independence referendum after being exposed to an against-independence article stressing the negative economic consequences of the independence of Catalonia than one that stresses the positive economic consequences of staying in Spain.
H4b: This effect is stronger for those who share a double national identity.
Figure 2: Conceptual Model for H3, H3b, H4 and H4b
Methodology
The data used to test the proposed hypotheses for Study 1 and study 2 was collected through a survey distributed online only among adult Catalan citizens. In order to engage with the participants, the communication departments of the Catalan parties represented in the
Negative Economic
Consequences Article Referendum Voting Intentions
parliament were contacted, as well as several pro and against independence organizations. The online questionnaire was designed in Catalan and translated into Spanish so that participants could choose the language they were more comfortable with. Although the vast majority of Catalans understand both languages, a survey design in either one or the other could have
potentially lead to biases in the participant’s perception of the objectivity or purpose of the study. For that reason it was considered necessary to elaborate the survey in both languages. A total of 192 Catalans participated in the study, which was estimated to take 10 minutes of the
respondent’s time only. As a first step of the questionnaire, the participants were asked to indicate their age to guarantee only individuals above 18 years old took part in the research. Later, a battery of questions regarding national identity, ideology, party voting intentions and issue stance were asked in order to discern the predispositions of each participant. Following, participants were exposed to the questions concerning Study 1 and Study 2.
Sample
192 Catalan citizens participated in the study, 100 of which were female (52.1%), 86 were male (44.8%) and 6 (3.1%) identified as “other”. Adults aged between 25 and 34 were prevalent comprising of 46.6% of the total sample. Following, were young adults (aged 18-24) at 21.9%, middle age adults (aged 35 to 49) at 17.7%, and seniors (aged 50 to 64) at 13%. The remainder (1% of the sample) was 65 years old or above. On an ideological scale from 0 to 7 (0 being extreme left and 7 being extreme right), the average participant was 2.66. Regarding their support for the pro-independence movement, the average participant rated her/his support as 3.71 (from 0 being no support at all to 7 extremely support) and gave a 2.92 rate to her/his support to unionism. A 51% completed the questionnaire in Catalan while a 49% of respondents preferred
to do it in Spanish. However, when asked in which language they preferred to read the article included in Study 2 only a 32.8% chose to do it in Spanish, while a 67.2% did it in Catalan. Study 1: Negativity Bias on Headline Selection
In order to test the potential effect of negative economic consequences on news selection, participants were randomly assigned to one out of the four possible conditions. Each condition included two different headlines presented in a random order from which participants had to choose the one they preferred to read. A stable headline was provided across the four conditions, emphasizing the support rates for independence in the last national public opinion survey. A second headline, which was manipulated, stressed either the positive or negative consequences of the independence or the permanence of Catalonia in Spain. More specifically the five headlines used (translated into English) were the following:
Stable Headline
Support for independence stops being the preferred option among Catalans for the first time since 2012.
Headline Condition 1 (Negative economic consequences of becoming independent): Half of Catalonia’s GDP, at risk because of companies escaping to other Spanish regions due to the independence movement process.
Headline Condition 2 (Negative economic consequences of remaining in Spain): Half of Catalonia’s GDP, at risk because of companies escaping to other Spanish regions due to the threats of the Spanish government.
Headline Condition 3 (Positive economic consequences of remaining in Spain): Catalonia creates more jobs since Spain took control of the Catalan institutions.
Headline Condition 4 (Positive economic consequences of becoming independent): Catalonia creates more jobs than before the pro-independence movement process started. Study 2: Economic Consequences on Voting Intentions
In addition, a second experiment following a 2x2 between subjects design aimed to explain the participants’ voting intentions in a hypothetical referendum of independence of Catalonia depending on news exposure. More specifically, this experiment intended to compare the effects that negative and positive economic consequences (of the secession or the
permanence in Spain) could have on voting affirmatively or negatively in a referendum on the issue. Participants were, thus, assigned to read one article out of four conditions and were then asked a question regarding their voting intentions in a hypothetical referendum of independence of Catalonia.
Two factors were manipulated across the four conditions. These were the negative or positive economic consequences factor, and the permanence in Spain or becoming independent factor. More specifically, each of the four conditions stressed the following approaches: negative economic consequences of becoming independent, positive economic consequences of becoming independent, negative economic consequences of remaining in Spain and positive economic consequences of remaining in Spain. Two new variables were computed in order to separate the pro-independence articles from the against-independence articles and be able to test H3, H3b, H4 and H4b. The same process was conducted in order to separate positive consequences’ articles from negative consequences’ articles. The full articles to which participants were exposed are translated into English in the Appendix A. For the original layouts see Appendix B (in Catalan) and Appendix C (in Spanish).
Table 1: Conditions of Study 2, with a 2x2 factorial design
Type of political event
Remaining as a region in Spain Becoming an independent state
Positive Consequences
“Catalonia will enjoy prosperous economic consequences if it remains in Spain” Article stressing the potential positive
consequences for the Catalan economy if the region stays in Spain
“Catalonia will enjoy prosperous economic consequences if it splits from Spain”
Article stressing the potential positive consequences for the Catalan economy if the region becomes independent from Spain.
Negative Consequences
“Catalonia will suffer catastrophic economic consequences if it remains in Spain” Article stressing the potential negative
consequences for the Catalan economy if the region stays in Spain
“Catalonia will suffer catastrophic economic consequences if it splits from Spain”
Article stressing the potential negative consequences for the Catalan economy if the region becomes independent from Spain.
In order to ascertain the realism of the manipulation in Study 2, participants were asked to what extent they thought the article could have been published in an online newspaper (on a scale from 0 to 7). The article was perceived as moderately realistic (M = 4.50, SD = 2.11) which indicates that participants could react similarly in a real-world scenario.
Manipulation Check
In addition, two manipulation checks were conducted for each of the conditions in Study 2 by including two questions regarding the negativity or positivity of the consequences
mentioned in the article, as well as the general portrayed standpoint on the Catalan
independence. For both questions, a scale from 0 to 7 was used, being 0 extremely negative and 7 extremely positive. All manipulation checks were successful for both factors. The articles that stressed positive economic consequences were perceived as more positive (M = 4.76, SD = 2.09) than those stressing negative economic consequences (M = 1.45, SD = 1.64); t(176.63) = -12.15,
p < .001. On the other hand, the articles that presented a pro-independence standpoint on the issue were rated as more favorable to the independence (M = 5.33, SD = 2.33) than those that supported Catalonia remaining in Spain (M = 1.38, SD = 1.72); t(176.71) = -13.38, p < .001. Randomization Check
Study 1
When conducting a randomization check, the result showed that gender (χ2(6) = 10.14, p = .119) and national identity (χ2(12) = 18.07, p = .114) were randomly distributed among the selected headline conditions (manipulated headlines). When running an ANOVA for ideology F(3,89) = 0.40, p = .989 and for age F(3,89) = 0.19, p= .900, it was proved that randomization was also successful for both variables across conditions. The participants that selected the manipulated headline were also randomly distributed in terms of support for independence F (3,89) = 1.65, p = .184 as well as support for unionism F(3, 89) = 0.51, p = .677.
Study 2
When conducting a randomization check in Study 2, the result showed that gender (χ2(6) = 3.62, p = .728) and national identity (χ2(12) = 9.73, p = .640) were randomly distributed among the four article conditions. When running an ANOVA for ideology F(3,188) = 0.26, p = .858 and for age F(3,188) = 0.52, p= .667, it was stated that randomization was also successful for both variables across conditions. The participants that were exposed to one of the four article
conditions were also randomly distributed in terms of support for independence F (3,188) = 1.01, p = .388 as well as support for unionism F(3, 188) = 0.26, p = .851.
Variable Measurements
Headline Selection (Dependent Variable in Study 1)
In order to check whether the negative or the positive headlines were selected more often, the four headline groups were recoded into two new variables. The first variable was called Negative Selection and was computed by combining the two possible negative headline choices. The second one was computed by a combination of the two existing positive headlines and it was called Positive Selection. The stable headline choice across the four groups was treated as a missing value.
Extremism of Issue Stance (Moderator in Study 1)
In order to operationalize extremism of issue stance, two items were included. The first item asked participants about their support towards the independence movement in Catalonia, on a scale from 0 (no support at all) to 7 (extremely support). The second item, asked about their support in relation to the unionist movement, again on a scale from 0 (no support at all) to 7 (extremely support), which was reverse-coded to align it with the previous item. A factor analysis was then conducted resulting in the items loading on one factor, with an Eigenvalue of 1.67 and explaining an 83.33% of the variance with a Cronbach’s alpha = .80. A new variable was then computed by calculating the mean of both items (M = 3.90, SD = 2.49). Following, the participants were split into two groups, depending on their extremism on the issue. Those respondents that answered 0-1 and 6-7 were labeled as extremists (N = 93) while the rest were considered to be moderate (N = 99).
Referendum Voting Intentions (Dependent Variable in Study 2)
After the exposure to one of the four conditions of the Study 2, participants were presented with a hypothetical scenario. They were asked to consider whether they would most
likely vote NO or YES, on a scale from 0 (for sure no), to 7 (for sure yes), if a referendum about the independence of Catalonia was to take place on the following day (M = 3.91, SD = 2.88).
National Identity (Moderator in Study 2)
National identity was measured using the same scale that is used by the Centre d’Estudis d’Opinió (CEO), the National Public Opinion Center in Catalonia. The item was, thus, operationalized by asking the participants whether they considered themselves to be “Only Catalan”, “More Catalan than Spanish”, “As Spanish as Catalan”, “More Spanish than Catalan” or “only Spanish”. The majority of participants (a 35.9%) identified themselves as “as Catalan as Spanish”, followed by a 25% of respondents being “More Catalan than Spanish” and a 24.5% feeling “Only Catalan”. Exclusively a 9.8% of the respondents identified themselves as “more Spanish than Catalan” and a scarce 6.8% as “only Spanish”. A new variable was then computed with only those participants that had both Spanish and Catalan identities to some extent under the variable name of Shared Identity.
Results Study 1
A chi square test was conducted in order to test whether negative headlines regarding the economic consequences of staying in Spain or becoming independent were picked more
frequently than positive headlines on the issue (H1). Contrary to the expectations, the positive headlines were selected more often by participants (N=54) compared to the negative ones (N=39) when presented next to the same stable headline. The difference between both types of headlines (negative and positive) was significant χ2 (1) = 93.00, p < .001.
Figure 3: Results of Headline Selection Study 1 (H1)
A chi square test was also conducted in order to test the potential effect of extremism of issue stance on choosing a negative headline depending on exposure. The results of the test do not show a significant difference between extremists and moderates regarding negative selection χ2
(1) = 1.71, p = .191. However, extremists selected more positive headlines (N=52) than negative headlines (N=41). On the contrary, moderates chose more negative headlines (N= 53) than positive headlines (N=46). As a result, H2 has to be rejected.
Study 2
In order to examine whether negative economic consequences’ articles had a greater effect on referendum voting intentions than the positive economic consequences’ articles, two separate t-tests were conducted using pro and against-independence articles respectively. The purpose of splitting pro and against-independence articles was to detect any potential differences between them in terms of effects. The first t-test took pro-independence articles (positive
economic consequences of becoming independent and negative economic consequences of 39
54
0 10 20 30 40 50 60
Negative Headline Positive Headline
Positive vs. Negative Headline Selection
remaining in Spain) only in consideration. The Levene test for the equality of variance was not found significant, F = 1.53, p = 0.22 so equal variances were assumed. The result of the test showed that participants exposed to negative economic consequences of remaining in Spain were not more willing to vote YES in a referendum (M = 3.66, SD = 3.02) than those who were
exposed to positive economic consequences of becoming independent (M = 3.83, SD = 2.80). In addition the difference between both groups was not significant t (95) = -2.29, p = .775. As a result, H3 should be refuted.
A two factor-analysis of variance was conducted in order to assess the influence of exposure to negative pro-independence messages in conjunction with the effect of a shared national identity (Spanish and Catalan). However, the results showed no interaction effect between shared national identity and the exposure to a negative or positive pro-independence article for predicting referendum voting intention F(1,92) = 0.97, p = .327. This means that H3b cannot be confirmed.
In order to examine whether negative against-independence articles could have a greater impact on referendum voting intentions than positive against independence articles (H4), a second t-test was developed. The Levene test for the equality of variance was not found
significant, F = .04, p = .842. Therefore, homogeneity of variance could be assumed. The results showed that participants were not significantly more likely to vote NO in a referendum when exposed to negative economic consequences of becoming independent (M = 4.48, SD = 2.86) than they were when they read positive economic consequences of remaining in Spain (M = 3.68, SD = 2.84). Moreover, the relationship turned not to be significant as t(93) = 1.36, p = .176. As a consequence we can state that H4 needs to be rejected too.
To examine whether a shared national identity could have an impact on the relationship between negative or positive against-independence articles and referendum voting intentions a two-factor analysis of variance was also conducted. Contrary to expectations, the results showed no interaction effect between shared national identity and the exposure to a negative/positive against independence article for predicting a negative voting intention, F (2,183) = 0.71, p = .494. This means H4b cannot be confirmed either.
Since none of the moderation effects could be confirmed, potential direct effects that were not hypothesized in advance were also investigated. A shared national identity was found to have a direct effect on voting intentions when exposed to pro-independence articles F (1, 92) = 24.20, p < .001, probably due to a disproportionate difference between only Spanish (N=13) and only Catalan (N=47) participants. As it could be expected, participants that had a shared identity were more likely to vote NO in a referendum (M = 2.94, SD = 2.66) than those participants who had only one identity (M = 5.53, SD = 2.65). On the other hand, when exposed to against-independence articles, a shared national identity also had a direct effect on voting intentions article F (1,183) = 24.73, p < .001. Again, those participants with a shared national identity scored significantly lower in the voting intentions scale (M = 3.20, SD = 2.74), than those with only one national identity (M = 5.47, SD = 2.57). Young age also had a direct effect on voting intentions on those participants exposed to pro-independence articles F (1,91) = 14.26, p < .001 and against-independence articles as F (1,180) = 5.88, p = .016. More precisely, participants that were younger than 34 and viewed a pro-independence article scored significantly higher in the voting intentions scale (M = 4.32, SD = 2.68) than participants that were 35 or older (M = 2.52, SD = 3.01). On the other hand, those participants that saw the against-independence article and were younger than 34 were also significantly more likely to vote YES (M = 4.34, SD = 2.74)
than those older than 35 (M = 2.98, SD = 2.97). Extremism of issue stance was only found to have a direct effect on voting intentions for those participants who were exposed to pro
independence articles F (1,91) = 4.59, p = .035. In this case, participants who had extreme views and were exposed to a pro-independence article scored significantly lower in the voting
intentions scale (M = 3.59, SD = 3.40) than those with more moderate views (M = 3.91, SD = 2.25). Last, as it could be expected, the language chosen in the questionnaire also had a direct effect on voting intentions for both, participants exposed to pro-independence articles F (1, 91) = 26.41, p < .001 and respondents that saw the against-independence articles F (1, 177) = 25.33, p < .001. In the first case, participants that chose Spanish as their preference were significantly more likely to vote NO (M = 2.18 SD = 2.68) than those who chose Catalan (M = 5.33, SD = 2.17). Similarly, in the second case those that chose Spanish as their preference were also significantly more likely to vote NO (M = 2.57 SD = 2.77) than those who chose Catalan (M = 5.19, SD = 2.36).
Discussion
This research aimed to support the extensive body of literature that concludes that negativity drives attention and influences voting intentions in a particularly polarized political context in Europe: the independence debate in Catalonia. However, the results of the studies conducted did not point towards the same direction. In fact, contrary to the expectations, negative headlines were correlated with a lower capability of being selected, in comparison to positive headlines. In addition, those participants with extreme views did not select negative headlines more often than moderates. Last, voting intentions did not appear to be influenced by
negative news to a higher extent than positive news and a shared identity did not increase nor decrease the chances of that influence to happen.
The fact that positive headlines were more often selected than negative ones differs from the conclusions of the theoretical framework detailed previously. Nevertheless, there is also an extensive body of research that claims negativity in the news can produce a backlash effect, especially when a certain level of incivility is present or it is showed next to positive news
(MacKuen, Wolak, Keele, & Marcus, 2010; Muddiman, Pond-Cobb & Matson, 2017). It is worth mentioning that negative headlines used in Study 1 could be considered to have a certain level of incivility as they literally blamed (in the original version in Spanish and Catalan) one of the parts (either the pro-independence or the unionist movements) for the negative economic
consequences, which may have in turn triggered a backlash effect. However, another possibility is that the headlines themselves were perceived as less novel than the positive ones. Although the topics of the headlines stressing both, positive and negative economic consequences had been previously published by the Catalan and Spanish media, negative news may have been better remembered, and, as such, considered less original (Pratto & John, 1991). Furthermore, positive headlines used in the study were considerably shorter in the number of words and letters than the negative ones, which could have influenced selection too. Additional research should replicate this study with totally new information, not previously reported by the media (which would imply weaker predispositions on the topic too) and with a similar number of words across conditions, in order to remove the novelty and length factors from the equation.
The fact that negative economic news in Study 1 failed to attract the participants may also be explained by the same two theories that have defined the negativity bias itself. First, according to the expectations violation theory, we pay attention to negativity because it is less
present in our daily lives (Soroka, 2014). However, negative news is often prevalent in the media, which may allow positive information to stand out over the rest. From this perspective, the results of this study can be useful for journalists (at least in Catalonia). Only by breaking this negativity vicious circle journalists may be able to properly draw attention towards their news, differentiating themselves from other types of journalism and disinformation sources (including fake news) that tend to use a more uncivil and negative tone.
Second, although automatic processing theory argues that negativity bias is attitudinal and evaluative, this does not imply a behavioral reaction (Ito et al., 1998).Muddiman et al. (2017) found that some individuals consciously avoid selecting negative information when being exposed to both, positive and negative headlines. According to Pew (2002) women are
particularly likely to avoid exposure to news due to its high level of negativity. Regardless of gender differences, these findings suggest that even when we pay more attention to negative messages, we can consciously avoid reacting to them; a decision that comes immediately after evaluating the stimulus (Carretié, Albert, López-Martín & Tapia 2009).
Furthermore, the results of Study 1 do not seem to detect any influence of extremism of issue stance on choosing negative information in the media. In fact, those that had an extreme view on the independence of Catalonia were more likely to select positive headlines than negative ones, although this difference was not significant. This may be due to an
overrepresentation of extreme left wing supporters in the sample, who have been suggested to have a particularly lower negativity bias in comparison to extreme right wing voters (Keene et al., 2017).
With regards to Study 2, the results showed no effect for negative economic
economic information has a greater impact on voting behavior than positive information (Soroka & McAdams, 2015), this effect was not found in the study. This may, perhaps, be explained by the probable pre-exposure of participants to information on the topic and their consequent strong formed opinion. The familiarity with the issue and the highly polarized context in which this issue takes place may have limited the capability of the stimulus to shape any opinions or influence voting intentions. Future research should aim to replicate a similar experiment with an exposure to more than one stimulus over time, and with a less familiar topic. By doing so, the effects of the negative economic news on political attitudes and voting behavior may be more likely to happen (Garz, 2012).
Even when taking national identity into consideration, those who considered themselves as both, Spanish and Catalan to a certain degree, did not show a higher disposition to vote YES or NO in the referendum depending on the negativity and viewpoint of the article. This may be explained, once again, by the previous exposure to related articles in the media, and an already shaped pre-disposition in regards to the topic, which limited the capability of the manipulation to have any effect.
Thus, it can be presumed that the high existing polarization in Catalonia and the participants’ strong predispositions on the issue may have obstructed any of the hypothesized effects on voting intentions. This factor should be emphasized as an important conclusion since it means citizens may be becoming less likely to adjust their opinions on the issue depending on the information they receive. Hence, polarization on the issue could even become stronger with time.
On the other hand, this study also brought some light on the parameters that may influence voting intentions in a potential referendum of independence in Catalonia. In fact,
language, age and national identity seem to be strong predictors of the voting intentions in a referendum of independence in Catalonia. Spanish speakers, older than 35 years old with a shared national identity are more likely to vote NO in a referendum, at least after being exposed to an against-independence economic article. On the contrary, Catalan speakers, younger than 34 years old, with only Catalan nationality that are exposed to pro-independence economic articles are more likely to votes YES in a referendum on the issue than the rest of the population. These results become particularly useful for political parties and pro or against independence
organizations, since they contribute to profiling their potential supporters, as well as those citizens that are more likely to be reticent towards their ideas.
As a conclusion, we may be able to say negativity draws attention. However attention does not necessarily translate into media selection nor this will automatically have a stronger impact on one’s voting behavior. In fact, it seems the potential economical risks of the independence of Catalonia or of its permanence in Spain for the Catalan wallet do not
particularly attract citizens’ news selection. Moreover, once exposed to them in the media, they do not seem to shape voting intentions among the Catalan citizens; at least, in the current
political context and taking into consideration the abundant information flow regarding the issue that Catalan citizens have been exposed to in the recent years.
References
Bowman, D., Minehart, D., & Rabin, M. (1999). Loss aversion in a consumption-savings model. Journal of Economic Behavior and Organization, 38, 155–178. doi: 10.1016/S0167-2681(99)00004-9
Bloom, H.S. and Price, H.D., 1975, The effects of aggregate economic variables on congressional elections, American Political Science Review 69, 1232–1269. doi: 10.2307/1955283
Burgoon, J. K. (1993). Interpersonal expectations, expectancy violations, and emotional communication. Journal of Language and Social Psychology, 12, 30-48. doi:10.1177/02619 27X93121003
Carlyle, T. (1841). Heroes: Hero-worship and the heroic in history. New York: Charles Scribner and Sons.
Carretié, L., Albert, J., López-Martín, S., & Tapia, M. (2009). Negative brain: an integrative review on the neural processes activated by unpleasant stimuli. International Journal of Psychophysiology,
71(1), 57-63. doi: 10.1016/j.ijpsycho.2008.07.006
Donsbach, W. (1991). Exposure to political content in newspapers: The impact of cognitivedissonance on readers’ selectivity. European Journal of Communication, 6, 155–186. doi:
10.1177/0267323191006002003
Fiske, Susan T. 1980. “Attention and Weight in Person Perception: The Impact of Negative and Extreme Behavior,” Journal of Per-sonality and Social Psychology 38 (6): 889–906. doi: 10.1037/0022-3514.38.6.889
Fuller, J. (2010). What is happening to news: The information explosion and the crisis in journalism. University of Chicago Press.
Garz, M. (2012). Job insecurity perceptions and media coverage of labor market policy. Journal of
Labor Research, 33(4), 528-544. doi: 10.1007/s12122-012-9146-9
Harrington, D. E. (1989). Economic news on television: The determinants of coverage. Public Opinion
Quarterly, 53(1), 17-40. doi: 10.1086/269139
Ito, T. A., Larsen, J. T., Smith, N. K., & Cacioppo, J. T. (1998). Negative information weighs more heavily on the brain: The negativity bias in evaluative categorizations. Journal of personality and
Kahneman, D., & Thaler, R. (1991). Economic-analysis and the psychology of utility: Applications to compensation policy. American Economic Review, 81, 341–346.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–292.
Keene, J., Shoenberger, H., Berke, C., & Bolls, P. (2017). The biological roots of political extremism: Negativity bias, political ideology, and preferences for political news. Politics and the Life
Sciences, 36(2), 37-48. doi:10.1017/pls.2017.16
Newman, N., Fletcher, R., Kalogeropoulos, A., Levy, D. A., & Nielsen, R. K. (2017). Reuters institute digital news report 2017.
MacKuen, M. B., Erikson, R. S., & Stimson, J. A. (1992). Peasants or bankers? The American electorate and the US economy. American Political Science Review, 86(3), 597-611. doi:10.2307/1964124 MacKuen, M., Wolak, J., Keele, L., & Marcus, G. E. (2010). Civic engagements: Resolutepartisanship
or reflective deliberation. American Journal of Political Science, 54, 440-458. doi:10.1111/j.1540-5907.2010.00440.x
Meffert, M. F., Chung, S., Joiner, A. J., Waks, L., & Garst, J. (2006). The effects of negativity and motivated information processing during a political campaign. Journal of Communication, 56, 27-51. doi:10.1111/j.1460-2466.2006.00003.x
Muddiman, A., Pond-Cobb, J., & Matson, J. E. (2017). Negativity bias or backlash: Interaction with civil and uncivil online political news content. Communication Research. doi:
10.1177/0093650216685625
Mueller, J.E., 1970, Presidential popularity from Truman to Johnson, American Political Science Review 64, 18–39. doi:10.2307/1955610
Muñoz, J., & Tormos, R. (2015). Economic expectations and support for secession in Catalonia: between causality and rationalization. European Political Science Review, 7(2), 315-341. doi:10.1017/S1755773914000174
Mutz, D. C. (2007). Effects of “in-your-face” television discourse on perceptions of a legitimate opposition. American Political Science Review, 101(4), 621-635.
doi:10.1017/S000305540707044X
Nannestad, P., & Paldam, M. (1997). The grievance asymmetry revisited: A micro study of economic voting in Denmark, 1986–92. European Journal of Political Economy, 13(1), 81–99.
doi:10.1016/S0176-2680(96)00034-1
The Pew Research Center. (2002, June 9). Public’s News Habits Little Changed by September 11. Retrieved from http://www.people-press.org/files/legacy-pdf/156.pdf
Pratto, F., & John, O. P. (1991). Automatic vigilance: the attention-grabbing power of negative social information. Journal of personality and social psychology, 61(3), 380. doi: 10.1037/0022-3514.61.3.380
Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political
involvement and polarizes elections. Cambridge University Press.
Semetko, H. A., & de Vreese, C. H. (2004). Political campaigning in referendums: Framing the referendum issue. Routledge.
Shoemaker, P. J. (1996). Hardwired for news: Using biological and cultural evolution to explain the surveillance function. Journal of communication, 46(3), 32-47. doi:
Shoemaker, P. J., Danielian, L. H., & Brendlinger, N. (1991). Deviant acts, risky business and US interests: The newsworthiness of world events. Journalism Quarterly, 68(4), 781-795. doi:10.1177/107769909106800419
Shoemaker, P. J., & Kunreuther, H. C. (1979). An experimental study of insurance decisions. Journal of Risk and Insurance, 46, 603–618. doi: 10.2307/252533
Shoemaker, P. J., & Vos, T. P. (2009). Gatekeeping theory. New York, NY: Routledge.
Soroka, S. (2014). Negativity in democratic politics: Causes and consequences. Cambridge:Cambridge University Press.
Soroka, S., & McAdams, S. (2015). News, politics, and negativity. Political Communication, 32(1), 1-22. doi: 10.1080/10584609.2014.881942
Steiner, J. (1979). Human facial expressions in response to taste and smell stimulation. InH. W. Reese & L. P. Lipsitt (Eds.), Advances in child development and behavior (Vol. 13,pp. 257–295). New York: Academic Press. doi: 10.1016/S0065-2407(08)60349-3
Tausch, F., & Zumbuehl, M. (2016). Stability of risk attitudes and media coverage of economic news. doi: 10.1016/j.jebo.2018.01.013
Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless choice: A reference dependent model. Quarterly Journal of Economics, 41, 1039–1061. doi: 10.2307/2937956
Tversky, A., Slovic, P., & Kahneman, D. (1990). The causes of preference reversal. The American Economic Review, 80, 204–217. Retrieved from
https://www.jstor.org/stable/2006743?seq=1#page_scan_tab_contents
Wahlberg, A. A., & Sjoberg, L. (2000). Risk perception and the media. Journal of risk research, 3(1), 31-50. doi: 10.1080/136698700376699
Zillmann, D., Chen, L., Knobloch, S., & Callison, C. (2004). Effects of lead framing on selective exposure to Internet news reports. Communication Research, 31(1), 58-81. doi:
Appendices
Appendix A: English translations of articles used in experimental conditions of Study 2
Condition “Negative consequences of becoming independent”
Catalonia will suffer catastrophic economic consequences if it splits from Spain The consequences of a Catalan independence would have far-reaching catastrophic implications for the region’s economy, its businesses and its people, according to some of the most renowned economists in the world.
As the Economics Professor at Cambridge University, Alex Middleton, explains, the establishment of a border would result in a loss of jobs, decreased income and social wealth for all citizens. Those losses would be triggered by the consequent obstacles to trade, by the
spending needs of the new state and by the financial difficulties related to its exit from the EU. Catalonia could see its economy diminished, as today 35.5 percent of Catalan exports are to the Spanish market. The region would also need to increase its expenses in order to build new state structures (embassies, central banks, etc.), which will become the most important part of the government’s budget.
If becoming an independent state, Catalonia would directly exit the European Union. In fact it would only be able to become part of the union again years later with an unanimous “yes” from all EU members, including Spain and its allies, who are unlikely to support its addition to the union.
Condition “Positive economic consequences of becoming independent” Catalonia will enjoy prosperous economic consequences if it splits from Spain
The consequences of a Catalan independence would have far-reaching beneficial implications for the region’s economy, its businesses and its people, according to some of the most renowned economists in the world.
As the Economics Professor at Cambridge University, Alex Middleton, explains, the full sovereignty of Catalonia would result in a gain of jobs, an increased income and social wealth for all citizens. Those gains would be triggered by the constant improvement of foreign trade, by the correction of its fiscal deficit and by the financial benefits of being a small state in the EU.
Catalonia could see its economy boosted, avoiding the current tax detour to Spain, which would have direct consequences on the citizen’s public services (education, health, etc.), which become the most important part of the government’s budget.
If becoming an independent state, Catalonia would become again part of the European Union in a short period of time. This would happen once Catalonia received a unanimous “yes” from all EU members who are soon likely to ascertain its benefits of its permanence.
Condition “Negative economic consequences of remaining in Spain”
Catalonia will suffer catastrophic economic consequences if it remains in Spain The consequences of Catalonia staying in Spain would have far-reaching catastrophic implications for the region’s economy, its businesses and its people, according to some of the most renowned economists in the world.
As the Economics Professor at Cambridge University, Alex Middleton, explains, remaining as an autonomous Spanish region would result in a loss of jobs, decreased income and social wealth for all citizens. Those losses would be triggered by the economic drainage from less wealthy regions in Spain, by the lack of investment from the state and by the financial problems related to the Spanish economy’s crisis.
Catalonia could see its economy diminished, as Catalonia contributes today with a 21 per cent of Spanish taxes and only gets a 16 per cent in return. The region would also need to increase its financial contribution to other regions in Spain if inequality among them keeps arising. Only by doing this, those regions will be able to maintain their public services (education, health, etc.), which become the most important part of their government’s budget.
If staying as an autonomous region, Catalonia would directly be left behind in the European Union. With the struggles of Spain in the financial crisis, Catalonia would feel the rejection from European citizens and businesses around Europe, who are unlikely to keep supporting economically the country for a long time.
Condition “Positive economic consequences of remaining in Spain”
Catalonia will enjoy prosperous economic consequences if it remains in Spain
The consequences of Catalonia staying in Spain would have far-reaching beneficial implications for the region’s economy, its businesses and its people, according to some of the most renowned economists in the world.
As the Economics Professor at Cambridge University, Alex Middleton, explains, staying as an autonomous region Catalonia would result into a gain of jobs, increased income and wealth for all citizens. Those gains would be triggered by the constant improvement of foreign trade, by the correction of the fiscal deficit due to a new financial system, as well as by the economic benefits of being part of the EU.
Catalonia could see its economy boosted, taking into consideration that already a 65.5% of Catalan exports are already to the foreign market. The region would also start benefiting from a new financial system that would correct its fiscal deficit, resulting in a substantial improvement of its public services (education, health, etc.), which become the most important part of the government’s budget.
If remaining as a region in Spain, Catalonia would stay in the European Union, becoming an economic node in it. In fact, Catalonia would most likely receive the recognition from all the members in the Union, included Spain, for the substantial progress that would convert it into the economic hub of Southern Europe.
Appendix B: Stimuli in Catalan
Economic consequences of becoming independent
Positive Negative
Economic consequences of remaining in Spain
Appendix C: Stimuli in Spanish
Economic consequences of becoming independent
Positive Negative
Economic consequences of remaining in Spain
