1
The Effects of Corruption News on the Brazilian Political System
Olivia Dias Mirisola Guariba University of Amsterdam
Faculty of Social and Behavioural Sciences
oliviamirisola@gmail.com Student number: 11351454 Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science: Political Communication Supervisor: Lukas Otto
Abstract
This paper examines how exposure to corruption news related to ‘Operation Car Wash’ has a negative effect on Brazilians’ political trust in their institutions and trust in democracy. I define political trust as trust in the government, Congress, political parties and politicians. Operation Car Wash is the biggest corruption scandal in the country and one of the most famous scandals in the world, which involved politicians and executives from the Brazilian elite. First, I hypothesise that participants exposed to an article with a large amount of information on corruption will have lower evaluations of trust in political institutions and the political system than participants reading an article with less information about corruption. I then hypothesise that reading an article about corruption that involves political actors leads to less trust in political institutions and the political system than reading an article that does not include political actors. Finally, I hypothesise that political trust in the institutions mediates the effects of exposure to corruption news on trust in the political system. Therefore, to test my hypothesis, an online experiment was conducted among Brazilians citizens (N = 460). The findings show no significant relationship between exposure to corruption news, and political trust in the institutions and trust in the political system.
Keywords: political trust, trust in the institutions, trust in the political system, democracy, corruption, corruption news
The Effects of Corruption News on the Brazilian Political System
Corruption in Brazil is endemic. The country has a history of fraud and scandals involving not only the public sector and the politicians, but also the private sector.
According to Transparency International’s Corruption Index 2017, Brazil ranked 96 out of 180 countries. Nevertheless, in the last four years, Brazil has been shaken by a massive corruption scandal, known as ‘Operation Car Wash’. It began as a money laundering investigation into Petrobras, one of the most important oil companies in the world, and then turned into a corruption and bribery scandal involving senators, presidents, congressman and CEOs of constructions companies.
What differentiates this scandal from others is that the investigation has taken a toll on Brazil’s political and economic system. Operation Car Wash has exposed corruption at the highest levels of the government and in the country’s largest companies. The
investigation involved more than 30 business owners, including owners of the ten largest contractors in the country. Furthermore, politicians from 14 political parties were involved in the scandals. What made the Operation so famous, to the point where it received 96% of Brazilians’ approval according to a report from the Ipsos Institute, is that for the first time in Brazil’s history politicians and prominent executives were facing jailtime.
Aside from the remarkable fact that the country is investigating and punishing politicians and executives for their crimes, Operation Car Wash has had several political consequences. One of the most critical is the loss of confidence in the institutions. According to recent research from GfK Verein, the level of confidence of Brazilians in their political institutions is one of the lowest in Latin America. Only 6% of Brazilians trust their politicians.
This loss of confidence could be explained by the constant exposure of the citizens to news related to Operation Car Wash. For almost four years, Brazilians have been exposed
to news about scandals, corruption and economic crisis. They have seen politicians of major political parties involved in the corruption scandal or even worse, engaged in strategies to stop the investigation.
Moreover, they saw the former President Dilma Rousseff suffer impeachment in the middle of her second term, which was the second impeachment of a president in 30 years. Furthermore, the vice president, Michel Temer, who became president following the impeachment, has only 5% approval of his government and is also facing corruption charges. Additionally, Brazil’s Supreme Court judged 50 of the most important politicians in the country, from six different political parties, to have been involved in the corruption scandal of Petrobras.
This constant exposure to corruption news can lead not only to a loss of confidence in the institutions but also to a lack of confidence in the democratic system as a whole, since two out of the three (Executive and Legislative) principle democratic bodies are involved in corruption cases. In fact, according to a report from the Ipsos Institute, 43% of citizens support a military intervention in the country. Brazilian political analysts claim that the citizens are disappointed with their political leaders and believe everyone is corrupted; therefore, only military intervention can ‘clean’ the country.
Brazil is disenchanted with politics in general, and this has serious consequences with the democratic system as a whole. According to Garcia (2018), this disillusion with the political system will be felt in the next election, where, according to polls, the number of abstentions will be the highest in Brazil’s history. This is even more worrying if we consider that voting is compulsory in Brazil.
Although political trust or the lack of confidence in the political institutions and the democratic system are not a new topic of studies, they are a new phenomenon in the actual Brazilian context. In 2010, according to IBOPE - one of the most respected polling
institutes in Brazil, the level of confidence in political institutions was at 80%, one of the highest approval ratings in the country’s history. Nevertheless, eight years later, with four of them under a severe economic and political crisis, the situation could not be much different. Therefore, I decided to investigate the effects of corruption news on the level of confidence in the political institutions and the political system. Thus, the central question in this study is as follows: Does the exposure to corruption news related to Operation Car Wash have an adverse effect on Brazilians’ political trust in their institutions and
confidence in democracy?
In the next section, I synthesise and discuss the importance of political trust and define different forms of trust; I then show how news consumption influences trust in general and finally I explain the effects of corruption information on trust. Second, I test my assumptions and present the results and discussion of the analysis. Lastly, I outline the limitations and implications of this study.
Theoretical Framework
Although political trust is a complex term, it has been exhaustively studied by scholars, since trust in the institutions is an indicator of political legitimacy. Researchers have suggested that there has been a decrease of levels of political trust in the world (Catterberg & Moreno, 2006; Dalton, 2004; Hetherington, 1998), especially in Latin America, where the region has a complicated history in terms of its relationship with democracy and succumbing to populism.
Political trust is generally defined as citizens’ confidence in political institutions, such as trust in the government, the Congress and political parties; or trust in a particular politician or particular institution (Catterberg & Moreno, 2006; Zmerli & Newton, 2011). The confidence in the political institutions is part of the legitimation of the democratic system, since political trust is an essential indicator of a healthy civic and democratic
society (Anderson & Tverdova, 2003). Therefore, for a democracy to function properly, citizens should believe and support their political institutions.
Even though political confidence in the institutions and support for the political system almost always seem to be complementary concepts, Easton (1975) made a
distinction between them. According to the author, specific support is given to authorities and institutions such as politicians, members of Congress, political parties and the courts. Thus, citizens can oppose the governmental authorities and if they are dissatisfied with their work as representatives they can vote to throw the incumbents out of the office. On the other hand, there is the concept of diffuse support, that is directed towards the political regime. That means that citizens can lose their identification with the democratic system, and therefore, lose their confidence in the system and the political community as a whole.
In this paper, I argue that Brazil is losing confidence in its political institutions and in the democratic system; however, the loss of trust in the political system is a symptom of the loss of confidence in the authorities.
However, is this loss of confidence in the political institutions beneficial or prejudicial to the democratic system? Scholars seems to disagree on this subject. For instance, Citrin (1974) argued that the decline of trust is not a serious problem, because it reflects the dissatisfaction of citizens with politicians, not the system. Thus, citizens can change the politician, and the democratic system will be safe. Nevertheless, some scholars suggest that this decline of trust in the institutions represent a desire for change in the democratic outcome (Catterberg & Moreno, 2006; Norris, 1999). However, Easton (1975) contended that political discontent with authorities is not always a signal for political outcome changes. Nonetheless, Miller (2006) found that distrust in the political institutions can affect the political system directly. These results are in line with Norris (1999), Moisés
(2008) and Morlino (2004), who argued that low levels of political trust represent a risk to the democratic system and affect the quality of the political system.
Exposure to News Articles
Although the literature highlights that there is no consensus among scholars about the loss of confidence in the political institutions and their effects on the democratic system, there is another factor we need to consider: the role of the media. Hetherington (1998) argued that the level of political trust in institutions or in the system depends on the distribution of negative or positive information that citizens receive about the government. It is known that after the Watergate case, the media assumed the role of gatekeeper, acting as the Fourth Estate in checking and investigating the government and politicians and reporting abuse of power (Thompson, 2000). Therefore, the media plays an important role when we analyse the levels of confidence in the political institutions and democracy. According to Thompson, (2000) political scandals are unthinkable without the mass media, since not only do they cover the case, but they often begin the investigation.
Bowler and Karp (2014) claimed that scandals involving politicians can have a negative influence on citizens, leading to a decrease of trust in the political institutions and the political system. On the other hand, Moy and Scheufele (2000) argued that media use affects only social trust, not institutional trust. Therefore, the media only affects the level of confidence among citizens, not confidence in authorities or institutions. Nevertheless, there is a belief among scholars that news media in general is associated with the problems of the democratic system.
According to Avery, (2009) not only can the news media influence the level of confidence in the political institutions and confidence in democracy, but the type of media can also affect the level of trust. Citizens with a high level of political confidence become more trusting when exposed to newspaper news and less trusting when exposed to
television news. Nevertheless, Avery argued that citizens with low levels of political trust do not become more or less trusting following the news, regardless of the news source.
Therefore, the literature suggests that greater exposure to political news and news media can lead to distrust in the government and the democratic system (Avery, 2009). This could help explain the low levels of political confidence in Brazil, since in the last four years citizens have been constantly exposed to news about bribery involving government representatives and companies.
Moreover, the research suggests that not only can exposure to news media decrease the level of confidence in the political institutions and democracy, but that corruption or exposure to corruption can also lead to a decrease of trust in the institutions and the political system (Wroe, Allen, & Birch, 2013). More than simply affecting the levels of confidence, corruption is a threat to the consolidation of democracy because bribery harms the democratic principles of transparency, accountability and fairness (Anderson &
Tverdova, 2003).
Exposure to Corruption News
Therefore, the theories presented and discussed in this paper are in line with the actual Brazilian political scenario. ‘Operation Car Wash’ began with a money laundering investigation into the most important company in the country (Petrobras); nonetheless, the case turned into a political scandal when the investigation revealed that politicians were receiving money from Petrobras to buy political support and finance political campaigns.
Although corruption has always been part of Brazil’s political history (Geddes & Neto, 1992); Operation Car Wash gave new meaning to political scandal, since the investigation has reached the highest levels of the Brazilian government and corporate elites.
The investigation had a significant impact on Brazilian society. According to a report of LatinoBarometro 2017, it was the first time in 22 years of research that corruption has appeared as the main concern of a country. Moreover, the survey reveals even more threatening data: 97% of Brazilians believe that the government does not govern them, but instead considers other interests that have nothing to do with the demands of society. These results are in line with the Edelman Trust Barometer 2017, revealing that the level of confidence of citizens plunged in the last year to the worst level since 2010. Only 24% of the Brazilians believe in the government, and 62% of them say that the national system as a whole has collapsed and is no longer able to meet the demands of the society. Moreover, the report blames the corruption as the cause of this drastic decrease of trust in the
institutions.
Despite the involvement of the Brazilian elite business in the corruption scandals, the level of trust in companies and executives remains relatively high when compared to the level of trust in the political institutions and democracy. The same Edelman Trust
Barometer 2017 affirmed that 61% of citizens trust executives and companies. According to Yakoff Sarkovas, CEO of Edelman Trust, although Brazilians surveyed are fearful of corruption, people seem to separate the image of the companies involved from the image of executives. Based on Stapenhurst (2000), one of the main reasons for this contradiction is because corruption in the public sphere harms the confidence of the citizens in the institutions, since the politicians are using federal funds for private gain. Thus, corruption in the private sector harms the investors in a company and clients of a company; however, in the public sector, it harms the citizens and the confidence they put in the politicians to govern in the interests of society.
Therefore, since Operation Car Wash seems to have had a significant impact on the Brazilian political scenario, it is interesting to investigate the effects of corruptions news
on trust in the political institutions and democratic system in Brazil. I thus present the following hypotheses:
H1: Participants exposed to an article with a large amount of information on corruption
will have lower evaluations of trust in political institutions and the political system than participants reading an article with less information about corruption.
H2: Reading an article about corruption that involves political actors leads to less trust in
political institutions and the political system than reading an article that does not include political actors.
H3: Political trust in the institutions will mediate the effects of exposure to corruption
news on trust in the political system.
Methods Design
In order to test the hypotheses, an online experiment was employed with a 2 (exposure to corruption news) x 2 (actor involved) between-subjects factorial design (Table 1). Exposure to corruption news was the first experimental factor with two levels: low and high. Actors involved was the second experimental factor with two levels: executives and politicians. Therefore, there were four conditions and a control group. Table 1
2 (Actors involved: Executive x Politician) 2 (Exposure to corruption news: Low x High) between-subjects factorial design with four conditions.
Actors Involved Corruption News
Low High
Executives 1 3
Politicians 2 4
Furthermore, participants were randomly allocated to one of the four conditions and read four different articles. After reading the text, participants answered a question about political trust in the institutions and trust in the democratic system. Additionally,
participants also responded to a post-test questionnaire, with a manipulation check and some socio-demographic questions.
Sample
Participants were reached through a convenience sample. The link to the online experiment was distributed through social media websites (Twitter, Facebook and
WhatsApp) and sent only to Brazilian citizens. The link for the research was online from 4 May to 9 May 2018. In total, 461 participants agreed to participate in the experiment; nevertheless, one participant had to be excluded because he did not meet the criterion of age. Therefore, the analysis was conducted over a final sample of 460 participants. Furthermore, 84 participants were assigned to condition one; 94 to condition two; 88 to condition three; and finally, 92 to the last condition. Additionally, 104 participants were assigned to the control group. Moreover, 245 of the participants were female and 214 were male. The mean age of the group was M = 47 (SD = 15.77).
Stimulus Material
Regarding stimulus, four articles related to ‘Operation Car Wash’ were constructed. The articles were retrieved from a real Brazilian newspaper, and the modifications were done according to the conditions. The advantage of using existing content for the stimulus is the low level of artificiality present in the content; nevertheless, this can prejudice the level of comparability across the conditions. However, since ‘Operation Car Wash’ is a well-known topic in Brazil, creating new content about this subject would be perceived as unreal by the participants. Each article presented to the participants contained three
paragraphs: the introduction, main body and the conclusion. The manipulation is mainly found in the first and last paragraph.
Therefore, for this research I manipulated both independent variables - Exposure to corruption news and Actors involved; thus, participants read four types of stories.
In the first condition, participants read an article about Operation Car Wash, which contains only one case of a corruption scandal and a name of a famous executive linked to the scandal. Additionally, participants in the second condition read a piece of news which included mention of a politician but mentioned only one case of corruption. Moreover, in the third condition participants received an article that contained three cases of corruption and mentioned an executive involved in corruption; and finally, participants in the fourth condition read a story that included three examples of corruption scandals and a politician involved in them. In addition, participants in a control group did not read any news. The stimulus material is found in Appendix A.
To verify whether the participants perceived exposure to corruption news and actors involved accurately, a manipulation check was conducted. Therefore, participants were asked to answer how many types of corruption cases the article mentioned (one or more than one) and if the actor involved in the news articles works in the private or public sector.
Procedure
The online survey experiment was created using Qualtrics. First, the participants began the experiment by reading a consent letter. In the consent letter, participants were informed of the goal of the study and the expected time to complete it. Moreover,
participants were told the study had an academic purpose only, and that the anonymity of the data was guaranteed. Furthermore, participants had the right to withdraw consent and stop, and if they had any doubts about the study, they could contact the researcher.
Therefore, if participants agreed to the terms and conditions, they were directed to the questionnaire.
The questionnaire began with demographics questions, where participants indicated their age, gender and educational level. Therefore, participants answered questions about media consumption, level of interest in politics and political ideology.
Participants were subsequently randomly assigned to either one of the four
conditions and asked to read the news article below carefully. Thus, after finishing reading the article, a manipulation check was carried out to determine whether they followed the stimulus material.
Participants then answered questions about trust in the political institutions, such as trust in the president and government; and questions to assess the level of confidence for democracy. Finally, participants were debriefed about the study and informed that the articles were modified for the academic purpose. The entire questionnaire is found in Appendix B.
Measurement Instruments
Political trust in the institutions. Participants’ degree of trust in the political
institutions was assessed using the six items from the definition given by Easton (1975), that measured the level of confidence in authorities and authorities’ institutions, meaning Politicians, Political Parties, Congress, President, Police and Courts. The questions are measured on a ten-point scale from 1 (‘No trust at all’) to 10 (‘Complete trust’). An exploratory factor analysis indicated that the scale was dimensional, with factor 1 explaining 45.95% of the variance and factor 2 explaining 67.83% of the variance in the six original items. The scale also proved reliable, as indicated by a Cronbach´s Alpha of .71, (M = 3.46; SD =1.26).
Trust in the political system. Participants’ degree of trust in the political system
was assessed using four items from de Latinobarometro 2017. The questions were adapted to the Brazilian context. Using a 4-point scale, the items asked to what extent people agreed or disagreed with the statements: (1) ‘Democracy is preferable to any other kind of government’; (2) ‘For people like me, it doesn’t matter whether we have a democratic or non-democratic regime’; (3)‘Under some circumstances, an authoritarian government can be preferable to a democratic one’ and (4) ‘I am satisfied with the level of democracy in the country’. The items were recoded, since some of them were formulated in a negative way. Furthermore, an exploratory factor analysis indicated that the scale was
unidimensional (only one component with eigenvalue above 1.00), explaining 63.88% of the variance in the four original items. The scale also proved reliable, as indicated by a Cronbach´s Alpha of .70 (M = 3.41; SD =.53).
Control variables. For this study, three control variables were selected: age,
gender and education level. Age was assessed in years (M = 47 years; SD = 15.77); gender was measured in female/male (53.4% females); and education level (M = 4.46; SD = 1.07) was measured in a seven-point scale where 1 is ‘fundamental school’ and 7 is ‘PhD’. Manipulation check
In order to test if the manipulation check for exposure to corruption news was successful, I conducted a Chi-square analysis. For this investigation, I used Condition (Actors Involved and Exposure to corruption news) as the independent variable and
Manipulation check for corruption level as the dependent variable. I found a significant but weak association between the variables, χ2 (3) = 10.61, p < .001, τ = .03. When taking into account the condition that the participants were assigned to, the level of corruption was improved by 3%. Accordingly, this indicates that the participants perceived the differences in the conditions and the manipulation check was successful.
Furthermore, to test if the manipulation check for actors involved in corruption news was successful, I conducted a Chi-square analysis, where Condition (Actors Involved and Exposure to corruption news) was the independent variable and Manipulation check for Actors Involved was the dependent variable. I found a significant and strong association between the variables, χ2 (3) = 184.92, p < .000, τ = .51. Therefore, when taking into account the condition that the participants were assigned, the recognition of the actors involved in the news was improved by 51%. Thus, this indicates that the participants perceived the differences in the conditions and the manipulation check was successful. Randomization Check
To check if the randomization of participants to the conditions was successful, a Chi-square analysis with Conditions (Actors involved and Exposure to corruption news) as the independent variable and Gender as the dependent variable was conducted. The Chi-square revealed no significant difference between gender in the four groups. Thus, participants’ gender in low levels of exposure to corruption news in the private sector (executives) was no different from participants’ gender in low levels of exposure to corruption news in the public sector (politicians) or high levels of exposure to corruption news in the private sector, or high levels of exposure to corruption news in the public sector, χ2 (6) = 4.96, p = .54. Therefore, the results indicate that no differences existed between conditions on gender; I can therefore conclude the randomization was successful.
Secondly, in order to test whether the randomization of participants to the control group and experimental group was successful, I conducted a One-way ANOVA. This analysis used Condition (Actors Involved and Exposure to corruption news) as the independent variable and Age as the dependent variable. The results show no significant difference between the mean age in the four groups. Therefore, participants’ mean age in low levels of exposure to corruption news in the private sector condition (M = 47.49; SD =
16.03) was no different from participants’ mean age in the low levels of exposure to corruption news in the public sector condition (M = 47.30; SD = 15.25); the mean age in high levels of exposure of corruption news in the private sector condition (M = 46.86; SD = 16.16); or the mean age in high levels of exposure to corruption news in the public sector condition, (M = 48.58; SD = 15.81), F (3, 465) =.304, p = .823, η² = 00. Moreover, based on these analyses, I can assume that the randomization was successful since there is no significant difference in age across conditions.
Results
Mean Differences for Political trust in the Institutions and Trust in the Political
System on Exposure to Corruption News
In order to test the first hypothesis, a MANOVA analysis was conducted to investigate if participants’ exposure to an article with large amounts of information on corruption will lead to less trust in the political institutions and political system compared to participants’ exposure to an article with less information about corruption. Therefore, for this analysis, I used exposure to corruption news as the independent variable and trust in the political institutions and trust in the political system as the dependent variables.
A non-significant Box´s M test (p =.161) indicates the homogeneity of covariance matrices of the dependents variables across exposure to corruption level. A MANOVA on political trust in the institutions and trust in the political system revealed no significant multivariate effect on exposure to corruption news, F (2, 355) = .818, p = .442; Wilk's Λ = .995, partial η2 = .00.
The multivariate effect shows no significant difference between the low (M = 3.40; SD = 1.21) exposure to corruption news and high (M = 3.52; SD = 1.27) exposure to corruption news on political trust in the institutions. Moreover, there is no difference between the low (M = 3.44; SD = .53) exposure to corruption news and high (M = 3.39; SD
= .53) exposure to corruption news on trust in the political system. Therefore, I reject the first hypothesis.
Mean Differences when Political Actors are Involved in the Scandal
To test the second hypothesis, two ANOVAS were conducted. I aimed to
investigate whether reading an article about corruption that involves political actors leads to less trust in political institutions and trust in the democracy system than reading an article that does not include political actors. For the first analysis, a One-Way ANOVA was conducted with the condition (Exposure to corruption news and Actors Involved) as the independent variable and political trust in the institutions as the dependent variable. The analysis of variance showed no significant effect of exposure to corruption news with actors involved in the scandal on trust in the political institutions, F (3,456) =.31, p =.812, η² = .00. A Bonferroni posthoc test revealed no significant differences between political trust in the institutions and the other groups (all p > .05; see Table 2 for means and standard deviations). Therefore, I reject the second hypothesis.
Table 2
Trust in the political institutions
Variable N M SD
Low Exposure to corruption news with executives Low Exposure to corruption news with a politician High Exposure to corruption news with executives High Exposure to corruption news with a politician
84 94 88 92 3.46 3.35 3.48 3.50 1.19 1.23 1.22 1.33
To test the second analysis, a One-Way ANOVA was conducted with condition (Exposure to corruption news and Actors Involved) as the independent variable and
political trust in the system (trust in the democracy) as the dependent variable. The analysis of variance showed no significant effect of exposure to corruption news with actors
A Bonferroni posthoc test revealed no significant differences between trust in the political system and the other groups (all p > .05; see Table 3 for means and standard deviations). Thus, I reject the second hypothesis.
Table 3
Trust in the political system
Variable N M SD
Low Exposure to corruption news with executives Low Exposure to corruption news with a politician High Exposure to corruption news with executives High Exposure to corruption news with a politician
84 94 88 92 3.46 3.42 3.38 3.40 .53 .54 .53 .53
Effects of Corruption News on Trust in the Political system Mediated by Political trust
in the Institutions
In order to test the third and final hypothesis, a mediation analysis was conducted using a PROCESS macro command and its Model 4 (Hayes, 2013) for the dependent variable (trust in the political system). Therefore, for this analysis, I used exposure to corruption news as the independent variable and political trust in the institutions as the mediator. Moreover, for the study, a bootstrapping analysis with a 5000 sample was used, and a regression equation was computed.
The findings reveal that no mediation was established on trust in the political system through trust in the political institutions. This means the indirect effect from trust in the political system by trust in the political institutions was no different from zero (-.00, SE = .00, 95% CI [ -.01, .00]). When taking into account the indirect effect, the direct effect of trust in the political system remains insignificant (B = -.02, t = -1.00, p = .31). Furthermore, the second interaction of exposure to corruption news on trust in the political system is also insignificant in this model of mediation (B = -.05, t = -.91, p = .36).
trust in the political institutions, it was concluded that trust in the political institutions did not mediate the relationship between the level of corruption and trust in the political system. Thus, the third hypothesis is not supported.
Figure 1
Mediation model for the effects of corruption news on trust in the political system mediated by political trust in the institutions.
Conclusion
The purpose of the present study was to examine whether the effects of corruption news related to ‘Operation Car Wash’ had a negative impact on Brazilians’ trust in their political institutions and trust in the democratic system. First, I predicted that participants exposed to an article with large amounts of information on corruption will have lower evaluations of trust in political institutions and the political system than participants
reading an article with less information about corruption. Secondly, I predicted that reading an article about corruption that involves political actors leads to less trust in political institutions and political system than reading an article that does not include political actors. Finally, I predicted that political trust in the institutions mediates the effects of exposure to corruption news on confidence in the democratic system. Nevertheless, no
Exposure of Corruption news Trust in the Political System Political Trust in the Institutions -.05 -.02 .12
support was found for the predictions, which contrasts with previous work discussed in the theoretical background section.
The literature suggests that exposure to corruption news or exposure to corruption scandals can decrease the level of trust in the political institutions and the democratic system (Wroe et al., 2013). Moreover, the literature also suggests that constant exposure to political news and news media can lead to distrust in government and the democratic system (Avery, 2009). Therefore, with this in mind, I would expect that my results would be significant. Nevertheless, there is another important variable that I did not consider in this analysis: the level of education.
Recent studies claim that the effects of corruption on political trust depend on educational level. For instance, Hakhverdian & Mayne (2012) found that citizens with the lowest level of education are indifferent to the effects of corruption. Nonetheless, for higher educated citizens, corruption presents a corrosive effect on political trust. According to the authors, education not only facilitates the gaining and processing of information, but it also fundamentally informs how citizens react to certain types of information. Moreover, this theory is in line with a report presented by the OECD on educational level. Brazil is 34 out of 38 countries, only ahead of South Africa and Mexico. Therefore, future research should consider using education level as an independent variable since it can influence exposure to corruption news and trust in the political institutions and the political system.
Furthermore, another point that deserves to be discussed is the definition of political trust used in this paper. I choose to define political trust as trust in the political institutions or trust in the government, the Congress and political parties and politicians in general (Catterberg & Moreno, 2006; Zmerli & Newton, 2011). However, political trust is a complicated term, as Easton (1975) defined. According to Easton, there are two different types of political trust or support: specific support and diffuse support. The first is a feeling
shaped by the action of the officeholders and political elites and can change faster according to the political actions, and the second is a feeling towards the political
institutions and the political system. When analysing the Brazilian scenario it is difficult to make this distinction between the terms, because the lack of trust seems to be general.
This leads to one of the most significant limitations of this study: the measures. Even though there is much research and study of political trust in the institutions and the democratic system, most of the studies focus on Western countries or analyse Latin
America. There are fewer studies focusing on Brazil, and the number is even lower when I examine studies published in English. Brazil is a unique country with a particular history and singular political system. Therefore, the measures used in this paper to analyse political trust in the institutions are not accurate for representing the Brazilian context, because they do not capture the nuances the country is experiencing in this moment. Moreover, the questionnaire used in this paper was first formulated in English; however, since Brazilian citizens formed the sample, the questionnaire was translated into
Portuguese. Nevertheless, I am aware that translating the questionnaire could harm the validity of the questions, which could impact the results of the study; nonetheless, according to a report from EF Education First, only 3% of the Brazilian population is fluent in English.
Another limitation of this study is related to the measure chosen to analyse
exposure to corruption news. Initially, my first idea was to measure exposure to corruption news versus no exposure to corruption news; however, since I measured a specific
phenomenon (Car Wash Operation), I realised that it would be almost impossible to discuss the operation and not mention a corruption scandal. Therefore, I changed the measure to high and low exposure to corruption news; nevertheless, during the pre-test, I comprehended that participants were having difficulty distinguishing between the levels.
Thus, I needed to make a more obvious manipulation check, which harms the quality of the research. Furthermore, I recognised that the use of high/ low exposure is perhaps more suitable in a survey and not in an experiment study.
Although this paper has some limitations, future research should further investigate the topic and include new variables to the study to enrich the research. For instance, it would be interesting to include political interest as a moderator of the relationship between exposure to corruption news and trust in the political institutions and political system. Moreover, as mentioned before, I would advise including level of education as one of the independent variables since this is an essential variable to help us understand the
relationship between corruption and political trust. Furthermore, taking the Brazilian case as an example, it would be interesting to investigate the relationship between corruption, lack of confidence in the political institutions and preference for authoritarian regimes.
To conclude, the findings presented in this study do not allow us to claim that exposure to corruption news related to Operation Car Wash has a negative impact on Brazilians’ trust in the political institutions and trust in the democratic system.
Nevertheless, Operation Car Wash and its subsequent developments is an important topic and has been influencing the Brazilian political scene for the past four years; therefore, it is essential to continue to investigate the impact of the Operation in the country.
References
American National Election Studies, University of Michigan, and Stanford University. ANES 2016 Time Series Study (2016). Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2017-09-19.
https://doi.org/10.3886/ICPSR36824.v2
Anderson, C., & Tverdova, Y. (2003). Corruption, Political Allegiances, and Attitudes Toward Government in Contemporary Democracies. American Journal of Political Science, 47(1), 91-109. doi:10.1111/1540-5907.00007
Avery, J. (2009). Videomalaise or Virtuous Circle?: The Influence of the News Media on Political Trust. The International Journal of Press/Politics, 14(4), 410-433.
https://doi.org/10.1177/1940161209336224
Bonin, R. (2010, December 16). Popularidade de Lula bate recorde e chega a 87%, diz Ibope. G1. Retrieved from
http://g1.globo.com/politica/noticia/2010/12/popularidade-de-lula-bate-recorde-e-chega-87-diz-ibope.html
Bowler, S., & Karp, J. (2004). Politicians, Scandals, and Trust in Government. Political Behavior, 26(3), 271-287.
Carmo, M. (2017, October 27). Brasileiros são os que menos confiam em democracia na América Latina, diz pesquisa. BBC Brasil. Retrieved
from http://www.bbc.com/portuguese/brasil-41780226
Catterberg, G., & Moreno, A. (2006). The Individual Bases of Political Trust: Trends in New and Established Democracies. International Journal of Public Opinion Research,18(1), 31-48. https://doi.org/10.1093/ijpor/edh081
Chade, J. (2016, May 11). Brasileiro é quem menos confia em político, diz pesquisa mundial. O Estado de S. Paulo. Retrieved from
https://politica.estadao.com.br/noticias/geral,brasileiro-e-quem-menos-confia-em-politico--diz-pesquisa-mundial,10000050380
Citrin, J. (1974). Comment: The political relevance of trust in government. American Political Science Review, 68(3), 973-988. https://doi.org/10.2307/1959141P
Corporación Latinobarómetro (2017), Informe 2017. Buenos Aires: Latinobarómetro. Retrieved from http://www.latinobarometro.org/latNewsShow.jsp
Dalton, R. (2004). Democratic challenges, democratic choices: The erosion of political support in advanced industrial democracies (Comparative politics). Oxford [etc.: Oxford University Press. doi:10.1093/acprof:oso/9780199268436.001.0001
Easton, D. (1975). A Re-assessment of the Concept of Political Support. British Journal of Political Science, 5(4), 435-457. https://doi.org/10.1017/S0007123400008309
Edelman Trust Barometer 2017 (2017). Edelman. Retrieved from https://edelman.com.br/propriedades/trust-barometer-2017/
Em quatro anos, Lava-Jato já alcançou 14 partidos. (2018, April 08). O Globo. Retrieved from https://oglobo.globo.com/brasil/em-quatro-anos-lava-jato-ja-alcancou-14-partidos-22569538
Figueiredo. J. (2017, October 27). Latinobarômetro: Corrupção aparece pela 1ª vez como principal preocupação para Brasil, diz pesquisa. O Globo. Retrieved from
https://oglobo.globo.com/brasil/latinobarometro-corrupcao-aparece-pela-1-vez-como-principal-preocupacao-para-brasil-diz-pesquisa-21999964
Garcia, R. (2018, April 21). Os riscos das eleições de 2018. Jornal do Brasil. Retrieved from http://www.jb.com.br/artigo/noticias/2018/04/21/os-riscos-das-eleicoes-de-2018/
Geddes, B., & Neto, A. (1992). Institutional Sources of Corruption in Brazil. Third World Quarterly, 13(4), 641-661. https://doi.org/10.1080/01436599208420302
Hakhverdian, A., & Mayne, Q. (2012). Institutional Trust, Education, and Corruption: A Micro-Macro Interactive Approach. The Journal of Politics, 74(3), 739-750. https://www.journals.uchicago.edu/doi/10.1017/S0022381612000412
Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach (Methodology in the social sciences.). Guilford Publications. doi: 10.1111/jedm.12050
Hetherington, M. (1998). The Political Relevance of Political Trust. The American Political Science Review, 92(4), 791-808.
https://doi.org/10.2307/2586304
Ipsos: 96% acreditam que Lava Jato deve seguir, mesmo que haja instabilidade. (2017, January 10). Ipsos. Retrieved from https://www.ipsos.com/pt-br/ipsos-96-acreditam-que-lava-jato-deve-seguir-mesmo-que-haja-instabilidade
Krakovics, F. (2018, March 29). Para 49%, combate à corrupção deve ser prioridade do próximo governo. O Globo. Retrieved from https://oglobo.globo.com/brasil/para-49-combate-corrupcao-deve-ser-prioridade-do-proximo-governo-22535490
Lagos, M. (2000). A máscara sorridente da América Latina. Opinião Pública, 6(1), 1-16. http://dx.doi.org/10.1590/S0104-62762000000100001
Miller, A. (2006). Arthur H. Miller. 1974. Political Issues and Trust in Government: 1964-1970. The American Political Science Review, 100(4), 681-682.
https://doi.org/10.1017/S0003055406352562
Morris, S., & Klesner, J. (2010). Corruption and Trust: Theoretical Considerations and Evidence from Mexico. Comparative Political Studies, 43(10), 1258-1285. https://doi.org/10.1177/0010414010369072
Moy, P., & Scheufele, D. A. (2000). Media effects on political and social trust. Journalism & Mass Communication Quarterly, 77(4), 744-759.
doi:10.1177/107769900007700403
Norris, P. (Ed.). (1999). Critical citizens: Global support for democratic government. OUP Oxford. doi:10.1093/0198295685.001.0001
OECD (2017). Better Life Index Retrieved from
http://www.oecdbetterlifeindex.org/countries/brazil/
Rodrigues, A. (2017, April 18). Por que ainda não somos fluentes em inglês? Revista EXAME. Retrieved from https://exame.abril.com.br/carreira/por-que-ainda-nao-somos-fluentes-em-ingles/
Seligson, M. (2002). The Impact of Corruption on Regime Legitimacy: A Comparative Study of Four Latin American Countries. The Journal of Politics, 64(2), 408-433. https://doi.org/10.1111/1468-2508.00132
Stapenhurst, R. (2000). The media's role in curbing corruption. Washington, DC: World Bank Institute.
The Global Competitiveness Report 2017-2018. (2017). World Economic Forum.
Retrieved from https://www.weforum.org/reports/the-global-competitiveness-report-2017-2018
Thompson, J. B. (2010). Political scandal: Power and visibility in the media age. Cambridge: Polity Press.
Tumber, H., & Waisbord, S. R. (2004). Introduction: Political scandals and media across democracies, volume I. American Behavioral Scientist, 47(8), 1031-1039.
http://dx.doi.org/10.1177/0002764203262275
Wroe, A., Allen, N., & Birch, S. (2013). The role of political trust in conditioning perceptions of corruption. European Political Science Review, 5(2), 175-195. doi:10.1017/S1755773912000094
Zmerli, S., & Hooghe, M. (2011). Political trust: Why context matters. Colchester, UK: ECPR Press.
Appendix Appendix A
Stimulus Material
Condition 1: Low level of exposure to corruption news with an executive involved in the scandal
02/05/2018 07h07 - Atualizado em 02/05/2018 13h27
PF cumpre mandados da 33ª fase da Lava Jato
Operação investiga esquema bilionário de lavagem de dinheiro
Da Redação
A Polícia Federal realizou nesta terça-feira, 2, a 33 ª fase da Operação Lava Jato, batizada de Operação Resta Um. A investigação mira a construtora Queiroz Galvão, suspeita de pagar propina para evitar investigações de uma CPI no Senado.
Foi preso no Rio de Janeiro o ex-presidente da construtora Ildefonso Colares Filho. Ele foi alvo de um pedido de prisão temporária de cinco dias, que pode ser revertido para uma prisão preventiva, na qual não há prazo para soltura.
As investigações indicam que a Queiroz Galvão teria pago 10 milhões de reais em propina para evitar que as apurações da Comissão Parlamentar de Inquérito (CPI) da Petrobras, em 2009 tivessem sucesso em descobrir os crimes que já haviam sido praticados até então.
Condition 2: Low level of exposure to corruption news with a politician involved in the scandal
02/05/2018 07h07 - Atualizado em 02/05/2018 13h27
PF cumpre mandados da 33ª fase da Lava Jato
Operação investiga esquema bilionário de lavagem de dinheiro
Da Redação
A Polícia Federal realizou nesta terça-feira, 2, a 33 ª fase da Operação Lava Jato, batizada de Operação Resta Um. A investigação mira o ex-senador Gim Argello (PTB-DF), suspeito de receber propina para evitar investigações de uma CPI no Senado.
Foi preso no Rio de Janeiro o ex-senador Gim Argello (PTB-DF). Ele foi alvo de um pedido de prisão temporária de cinco dias, que pode ser revertido para uma prisão preventiva, na qual não há prazo para soltura.
As investigações indicam que o ex-senador teria recebido 10 milhões de reais em propina para evitar que as apurações da Comissão Parlamentar de Inquérito (CPI) da Petrobras, em 2014 tivessem sucesso em descobrir os crimes que já haviam sido praticados até então.
Condition 3: High level of exposure to corruption news with an executive involved in the scandal
02/05/2018 07h07 - Atualizado em 02/05/2018 13h27
PF cumpre mandados da 33ª fase da Lava Jato
Operação investiga esquema bilionário de lavagem de dinheiro
Da Redação
A Polícia Federal realizou nesta terça-feira, 2, a 33 ª fase da Operação Lava Jato, batizada de Operação Resta Um. A investigação mira a construtora Queiroz Galvão, suspeita de fraudar licitações da Petrobras; formação de cartel e de pagar propina para evitar investigações de uma CPI no Senado.
Foi preso no Rio de Janeiro o ex-presidente da construtora Ildefonso Colares Filho. Ele foi alvo de um pedido de prisão temporária de cinco dias, que pode ser revertido para uma prisão preventiva, na qual não há prazo para soltura.
As investigações indicam que a Queiroz Galvão teria pago 10 milhões de reais em propina para evitar que as apurações da Comissão Parlamentar de Inquérito (CPI) da Petrobras, em 2009 tivessem sucesso em descobrir os crimes que já haviam sido praticados até então. Segundo o Ministério Público Federal a Queiroz Galvão formou, com outras empresas, um cartel de empreiteiras que participou ativamente de ajustes para fraudar licitações da Petrobras. Além dos ajustes e fraude a licitações, as evidências colhidas nas investigações revelam que houve corrupção, com o pagamento de propina a funcionários da Petrobras.
Condition 4: High level of exposure to corruption news with a politician involved in the scandal
02/05/2018 07h07 - Atualizado em 02/05/2018 13h27
PF cumpre mandados da 33ª fase da Lava Jato
Operação investiga esquema bilionário de lavagem de dinheiro
Da Redação
A Polícia Federal realizou nesta terça-feira, 2, a 33 ª fase da Operação Lava Jato, batizada de Operação Resta Um. A investigação mira o ex-senador Gim Argello (PTB-DF),
suspeito de receber propina para evitar investigações de uma CPI no Senado, formação de cartel e fraude de licitações.
Foi preso no Rio de Janeiro o ex-senador Gim Argello (PTB-DF). Ele foi alvo de um pedido de prisão temporária de cinco dias, que pode ser revertido para uma prisão preventiva, na qual não há prazo para soltura.
As investigações indicam que o ex-senador teria recebido 10 milhões de reais em propina para evitar que as apurações da Comissão Parlamentar de Inquérito (CPI) da Petrobras, em 2014 tivessem sucesso em descobrir os crimes que já haviam sido praticados até então. Segundo o Ministério Público Federal Gim Argello é acusado de formação de quadrilha, fraude a licitações e recebimento de propina.
Appendix B
Experiment Questionnaire
Dear Participants,
Welcome and thanks to you for your interest in our study.
We are interested in knowing the mechanisms behind exposure to news related to Car Wash Operation and the trust in the political institutions and system. Therefore, you will read some news articles and answer some questions on the subject. Also, we will ask some questions related to your profile, which will help us interpret the results.
It is important to note that the questionnaire is entirely anonymous. We will not be able to identify the participants so that you can respond according to your opinions and beliefs. That said, altogether anonymous data can be shared with other researchers for scientific purposes only. Participation is purely voluntary.
You may refuse to participate in the survey and may quit at any time without having to give a reason to do so. After completing the questionnaire, you will have up to 10 days to request to exclude the use of your data in the survey. None of these actions will have adverse consequences.
You can contact the principal researcher Olivia Dias Mirisola Guariba
(olivia.diasmirisolaguariba@student.uva.nl) or your supervisor Lukas Otto (lpotto@uva.nl) for any questions, as well as with the Ethics Committee for any claim or questions:
ASCoR Secretariat Ethics Committee University of Amsterdam PO Box 15793 1001 NG Amsterdam Tel: + 31 20-525 3680 Email: ascor-secr-fmg@uva.nl
The questionnaire takes approximately 10 minutes. It is imperative that you carefully read all the questions and information provided.
Please check the box below to indicate that you have read and understood the information provided about the scope of the research, anonymity and the possibility of withdrawing your answers at a later stage.
√ Yes, I have read and understood the declaration of participation, and I agree to participate in the research study.
1-) How old are you? 2-) What is your gender?
Male Female Other
3-) What is the highest level of education you have completed? Elementary School
High school
Technical education
Higher Education or Graduation Postgraduate
Master PhD
4-) During a normal week, how many days do you watch, read, or listen to news on TV, radio, print newspapers on the Internet?
1 2 3 4 5 6 7
5-) How often do you follow news about politics? Always
Often Sometimes Never
6-) To what extent do you interested in politics? Very interested
A little interested Not very interested I'm not interested
7-) When it comes to politics, people talk about "left" and "right". In general, where would you place your political ideology on this scale?
1 (Left) 2 3 4 5 6 7 8 9 20 (Right)
8-) Regarding the text, you just read how many types of corruption was mention? One
Three
9-) Regarding the text you just read the actor mentioned in the article was from the public or private sector?
Private Sector
10-) "How much of the time do you think you can trust the government in Brazil to do what is right?
Just about always Most of the time Only some of the time DN
11-) "Would you say the government is pretty much run by a few big interests looking out for themselves or that it is run for the
benefit of all the people?" Few Big Interests Benefit of All
Don't Know, Depends
12-) "Over the years, how much attention do you feel the government pays to what the people think when it decides what to do?”
A good deal Some Not much Don't Know
13-) Would you say that Brazil is going in the right direction or do you think the country has gone out of the way?
Right direction Got out of the way Do not know
14-) "Do you think that people in the government waste a lot of money we pay in taxes, waste some of it, or don't waste very much of it?"
A Lot Some
Not Very Much Don't Know
15-) "Do you think that people that running the government are corrupt?” A Lot
Some
Not Very Much Don't Know
16-) In your opinion, how reliable do you consider President Michel Temer? Being 1 "Not at all reliable" and 10 "Very trustworthy"
17-) Do you approve, disapprove, or neither approve nor disapprove of the way Michel Temer is handling his job as president?”
Totally Approve Approve
Disapproval
Totally Disapproval
18-) "Do you approve or disapprove of the way the Congress has been handling its job?"
Totally Approve Approve
Disapproval
Totally Disapproval
19-) "Do you approve or disapprove of the way the STF (Justice) has been handling its job?"
Totally Approve Approve
Disapproval
Totally Disapproval
20-) "Do you approve or disapprove of the way the Police has been handling its job?"
Totally Approve Approve
Disapproval
21-) From 1 to 10 how much do you consider the following political institutions to be trustworthy? Being 1 "Not at all reliable" and 10 "Very trustworthy"
Congress Political Parties Politicians STF Police
22-) In general, how satisfied are you with the functioning of democracy in Brazil? On this scale, 1 means "not satisfied " and 10 means "totally satisfied"
23-) To what extent is it important for you to live in a country that is governed in a democratically? On this scale, 1 means "nothing important" and 10 means "totally important"
24-) Under some circumstances, an authoritarian government can be preferable to a democratic one
strongly agree agree
disagree strongly disagree
25-) For people like me, it doesn’t matter whether we have a democratic or non-democratic regime
strongly agree agree
disagree
strongly disagree
Thank you for agreeing to participate in this study!
This research aims to investigate the effects of exposure to corruption news related to Operation Car Wash and how this exposure affects citizen confidence in political
institutions and the democratic system. The articles used in this research are real and taken from renowned communication vehicles; however, they have been modified for academic purposes.
If you have any questions or concerns regarding this study, its purpose or procedures, or if you have any problems related to the research, please contact the researcher, Olivia Dias Mirisola Guariba (olivia.diasmirisolaguariba@student.uva.nl).
