The role of news media and social media in a case of
corporate misinformation:
Do corrective and/or supportive messages affect corporate
credibility?
Jeanne Viergever | 11388730 | Master’s Thesis | Graduate School of Communication | Corporate Communication | Toni van der Meer| 28th of June
2
Abstract
The rise of misinformation today, such as the false belief that climate change is not caused by humans, is a matter of societal concern. This study takes a closer look at
misinformation in the corporate context. The importance of corporate communication lies in maintaining a favorable reputation and a strong relationship with stakeholders. In a case of corporate misinformation it is possible that an organization loses its favorable position and the trust of its stakeholders. This study examines whether exposure to corporate misinformation regarding climate change affects the corporate credibility of a large fossil fuel company, as an outcome variable to understand and assess how people respond to and interpret
misinformation after they are exposed to different sources who are correcting and/or supporting this misinformation. News media and social media are important sources in the process of correcting and supporting misinformation. This is the first study that takes a closer look at corporate misinformation. This study used a 1 shot experiment with 4 conditions. Findings suggest that conflicting messages after initial exposure to misinformation
significantly resulted in a decrease of corporate credibility. Supporting messages significantly resulted in an increase of corporate credibility. Furthermore, in case of conflicting media coverage, high levels of trust in a specific source could result in an (unjustified) increase in corporate credibility. The recommendations for news media institutions and online publics is to keep correcting and debating (mis)information, as denouncing false information has effect on an organization’s image.
Keywords: corporate misinformation, corrections and support; corporate credibility; news media; social media
3
Introduction
During the 2016 US presidential elections the term ‘Fake News’ became a hot item. The spread of false information prior to the 2016 US presidential election may have resulted in the election of US President Donald Trump and the concept of ‘Fake News’ got a lot of media coverage. However, fake news is not a new concept. It actually existed for a long time under the name of misinformation. Misinformation refers to the presence of, or belief in, objectively incorrect information (Bode & Vraga, 2015). When people are exposed to factually incorrect information, this could have serious consequences for society. When people are misinformed, they will make decisions for themselves and others which are not in their best interests, based on misperceptions (Lewandowsky, Ecker, Seifert, Schwarz, & Cook, 2012). For example, people will not vaccinate their children anymore as a result of a false information flow about the possible link between vaccination and autism (Lewandowsky et al., 2012).
With the continuing increase of new technologies, leading to the rise of social media, the increased availability of misinformation is a growing concern. Social media make it possible to rapidly spread information around the world. With social media and all its opportunities it is easier to find credible evidence-based information existing alongside personal opinion and poor quality data (Ennals, Byler, Agosta & Rosario, 2010). This is a result of the current high-choice media environment where people are exposed to a lot of information (Van Aelst, Strömbäck, Aalberg et al., 2017). Due to the information overload it is very hard for people to know whether the information they find on the internet is true or false.
The concept of misinformation has been mainly studied in political communication, especially in a polarized political context. These studies have shown the potential negative consequences of misinformation (Fridkin, Kenney & Wintersieck, 2015; Shin & Thorson, 2017; Thorson, 2016; Van Aelst et al., 2017 and Werner, 2016). Especially when people already hold misperceptions, the consequences of misinformation strengthened the pre-held misperception (Thorson, 2016). This study takes a closer look at misinformation in the corporate context. In corporate communication, misinformation is also an important issue, though this has not been given attention in scientific literature. Consequently, this is the first study which takes a closer look at corporate misinformation.
Unfortunately, large corporations also engage in spreading misinformation. For example, several large corporations within the fossil fuel industry are significantly funding
4
attempts to mislead the general public about the trustworthiness of climate science (Antilla, 2005). Furthermore, large corporations have a long history of influencing public debate, for example by means of lobbying. The information available to the public could easily influence public debate and therefore change political policies which would impose a regulatory burden on the fossil fuel industry (Jacques, Dunlap & Freeman, 2008). For corporations within the fossil fuel industry, spreading misinformation could lead to more lenient regulations for them to adhere to. In addition, corporate wrongdoing attracts the attention of the media and
watchdog organizations, triggering questions about why companies engage in Corporate Social Responsibility (CSR) and how they contribute to social well-being (Skarmeas, Leonidou & Saridakis, 2014). Spreading doubts by referring to the uncertainty of scientific conclusions – whether about smoking, climate change, or GMOs – is a very popular strategy for misinforming the population (Oreskes & Conway, 2010).
From a media perspective, the influence of (mis)information has to do with the function of the current high-choice media environment. There is an overflow of information and it is easy to access multiple sources for information, though this leads to access to a lot of false information as well(Lewandowsky et al., 2012). For the audience it is easy to get lost in the information overload. Multiple sources, with their correction or support of information, play an important role in the confusion as it gets harder for audiences to evaluate the
credibility of accessible information. When it comes to corporate misinformation, news media and the online public are important actors in combating misinformation. Both sources have the ability to correct for corporate misinformation. Corrections appear to have an effect in debunking misperceptions (Bode & Vraga, 2015). Therefore, it is an effective way to combat misinformation. On the contrary, supportive information will conserve the misperceptions which could eventually result in decision making not in societies best interest (Lewandowsky, Ecker, Seifert, Schwarz & Cook, 2012).
Online publics sometimes support corporate misinformation due to their
misperceptions. In the Internet age, we are all publishers; but few of us rigorously verify what we read online (Besterman, 2013). Therefore, it is possible for false or misleading information to be spread through social media. Alternatively, there are still examples were social media expose accurate information which combats misinformation (Bode & Vraga, 2015). The information flows on social network sites may succeed in exposing people to more diverse political and social orientations than the users seek out themselves (Bode & Vraga, 2015). In
5
addition, users could come into contact with verified sources, via shared links and suggested messages, which could correct misinformation.
Furthermore, news media could shape public opinion by framing an issue or problem in a certain way (Mahon & Wartick, 2003). However, news media often copy organizational press releases as they lack resources and time (Schafraad, van Zoonen & Verhoeven, 2016). Unconsciously, this could result in news content containing incorrect information. On the other hand, news media, in their watchdog function, have to protect the interests of the public and have to be relatively objective. By fact-checking organizational statements, they directly address the accuracy of a statement (Werner, 2016). In this manner, they will show that they are a trustworthy institution and secure the public interest.
The corrections and/or support of corporate misinformation could affect the attitude of citizens about the trustworthiness of a company and therefore the perceived corporate
credibility. The aim of this study is to find out whether exposure to corporate misinformation affects the attitude of citizens towards a large fossil fuel company after they are exposed to different sources correcting and/or supporting this corporate misinformation.
The fossil industry is claimed to be the largest influencer of climate change. Organizations within this industry have to engage in strong stakeholder management. Maintaining a favorable reputation is one of the most important things of corporate communications. However, large fossil fuel organizations have different interests (i.e. lobbying, make a profit, stakeholder management and CSR engagement) which make it complicated to be really transparent and open about their motives. These different interests could result in skepticism towards the company and for the fossil industry, especially in the field of CSR activities. Skepticism often reflects doubt about the truth concerning certain statements and can play a role in understanding the effect of misinformation (Skarmeas, Leonidou & Saridakis, 2014).
The effects of misinformation could be stronger for people who are skeptical towards CSR strategies than for people that are not. They might be more focused on and aware of potential misinformation regarding climate change. They might also think that the company only engages in CSR activities to improve its image (Elving, 2013). Therefore, skeptical people could be more aware of possible attempts of manipulation. Likewise, people who are media skeptical will be more focused on and aware of potential mistakes of news institutions. They are skeptical about the way mainstream news institutions function in society (Tsfati,
6
2003). Consequently, CSR skepticism and media skepticism could affect the relation between exposure to different messages and the perceived corporate credibility in a negative way.
This research has to be interpreted in the realm of corporate communication. This research examines the effect of corporate misinformation and the attempts of news media and the online public to correct and/or support for this misinformation. It addresses the gap between the consequences corporate misinformation has on the attitude towards a large fossil fuel industry company and which role news media and social media play in this corrective role. Addressing this gap is of importance because the new possibilities of social media allow misinformation to be spread more easily. Ultimately, this study examines if different
messages from news media and social media, which either support and/or correct the misinformation, will affect corporate credibility and reputation. Therefore, the following research question has been designed: How can news media and online publics’ messages, with correction and support of misinformation, after exposure to corporate misinformation
regarding climate change, affect the attitude of people regarding corporate credibility? Theoretical Framework
Misinformation and corporate communication
The consequences of misinformation can be tremendous for society. According to Lewandowsky et al. (2012) misinformation may form the basis for political and societal decisions that are directly oppose society’s best interest. This means that if individuals are misinformed, they may likely make decisions for themselves and their families that are not in their best interest and can have serious consequences, for example parents who believe the claims of a vaccination-autism link which could result in their decision to not vaccinate children (Lewandowsky et al., 2012). Therefore, holding misperceptions towards any societal topic could have serious consequences.
Research on misinformation finds that people who are exposed to misinformation, and afterwards are provided with corrective information through social media, experience a decrease in misperceptions about the particular topic (Bode & Vraga, 2015). It is thus
effective to correct misinformation online. However, much research on misinformation effects is directed at various factors that can increase or decrease an individual’s susceptibility to misinformation (Loftus & Palmer, 1974; Greene, Flynn & Loftus, 1982; Tousignant, Hall & Loftus, 1986). Notable is that corrective information is not always helpful in overcoming misperceptions. This is especially the case when people already hold misperceptions strongly
7
related to their ideology. In addition, when information agrees with prior-held beliefs, people find this information more credible and reliable than disconfirming information (Jerit & Barabas, 2012).
This research examines what the effect would be if a large corporation in the fossil fuel industry misinforms its stakeholders about climate change. The importance of corporate communication in fact lies in maintaining a favorable reputation and a good relationship with the stakeholders of companies. Without the trust of an organization’s stakeholders, there is no legitimacy to operate (Patriotta, Gond, & Schultz, 2011). Furthermore, when an organization violates moral standards of honesty and integrity, this could have detrimental effects on trust (Werner, 2016). Therefore, the reputation of a company could be influenced.
Fombrun (1996) explicitly incorporates corporate credibility as one important aspect of corporate reputation.Corporate credibility is the degree to which consumers, investors and other stakeholders believe in the trustworthiness and expertise (Fombrun, 1996). Corporate credibility is especially important in the context of misinformation, as the spread of corporate misinformation could have a potential negative effect on corporate reputation. Reputation is formed not only over time, but also over time as a function of complex interrelationships and exchanges between and among stakeholders and the organization in different contexts (Mahon & Wartick, 2003). A good relationship is necessary if organizations want to maintain their reputation. Furthermore, within political science it is proven that the violation of moral norms effects trust in politicians (Werner, 2016). This could also be the case in corporate
communication. The effect could be the same for spokespersons of organizations who are misinforming the public.
Lewandowsky et al. (2012) mention different sources which can provide
misinformation. One of these sources is corporate interests. Corporate interests have a long history of influencing public debate by proclaiming incorrect information (Lewandowsky et al., 2012). Especially when it comes to issues such as the environment. This issue has the potential to motivate policies that would impose a regulatory burden on certain industries, e.g. the fossil industry (Jacques, Dunlap & Freeman, 2008). In the fossil fuel industry there is legal and scientific evidence for the process of the willful manufacture of mistaken beliefs which is described as ‘agnogenesis’ (Bedford, 2010). Misinformation, or false information, in this research is focused on the scientific consensus there is about climate change.
8
High-choice media environment
Due to the high-choice media environment and the overload it provides, it is easy to get access to (mis)information on the internet and often this information is contradictory or overlaps. Until recently, mass media were the main source of information people needed to be free and self-governing (Kovach & Rosenstiel, 2007). However, since social media became of more importance the media environments have changed fundamentally (Van Aelst et al., 2017). By the growing use of social media networks the media-choice environment becomes more complex and it may foster the dissemination of misinformation (Lewandowsky et al., 2012). Regarding new media, the internet has paced immense quantities of information but at the same time it has also contributed to the spread of misinformation (Lewandowsky et al., 2012). Consequently, the general public could be confused about the correctness and trustworthiness of accessible information and sources. At this point, social media and traditional media are of interest.
Firstly, social media sites are network sites where people can share their thoughts and feelings. In this research, social media are described as ‘the public’s voice’. The macro level of social media, described in Bruns and Moe (2014), which constitutes hashtagged exchanges, is of importance. The inclusion of a topical hashtag in a tweet (or on Facebook) means that the message has the potential to reach well beyond the user’s existing number of followers or friends (Bruns & Moe, 2014). Using a hashtag signals a wish to take part in a large
communicative process, potentially with anyone interested in the same topic (Bruns & Moe, 2014). By taking part in a large communicative process, it is easy for incorrect information to slip into these conversations. Hashtags are often used with ad hoc issues and in response to breaking news or other sudden developments. It is possible that a hashtag shows the capacity to destabilize the initial organizational identity (Albu et al., 2016). This example shows how much power social media actually have. For instance, by hijacking an organizational hashtag people could respond to corporate misinformation. Furthermore, social media are accessible for all kinds of publics and they can share anything they believe. That would include people who hold misperceptions towards any societal topics (e.g. climate change). Since there is an overflow of information in online environments, which is often contradictory, it is possible that the general public can be affected by their online peers.
Secondly, it is important for news media institutions to play the role of a watchdog. As a news media institution you have the duty to protect the interests of the public. News media play an important role in the misinformation process. The literature shows that when
9
environmental issues occur, the media adopt protester/antagonist frames (Ihlen & Nitz, 2008; García, 2011). Journalists use frames to craft interesting and appealing news reports (Nisbet, 2009). However, there is also evidence that journalists follow their own discursive strategies when environmental issues occur (Verhoeven, 2016). This means that the news media do not always frame their message in a certain way and do not always have the public interest at heart. By largely copying organizational press releases due to the lack of resources and time it is possible that incorrect information unconsciously slips into news items. Ideally news media have a watchdog function, and more explicitly they have the duty to protect the public interest by, when encountering misinformation, giving corrective information. A recent phenomenon to encounter misinformation is fact-checking. According to Werner (2016) fact-checks operate at another level than average news articles because they present a direct judgement of the accuracy of a statement. Fact-checks are hard to counter-argue because the corrective information is impossible to overlook and dismiss (Werner, 2016).
Misinformation and its correction or support
Communication can become really complex by the overload of information. By correcting misinformation, there can be an attitude change in believing misinformation or people will even support it (Bode & Vraga, 2015). When corrective information is given in relation to climate change it is likely that people will change their beliefs about the company in a more negative way. When the corrective information is not given at all, or if news articles and the online public support the misinformation even more, it is likely that people will still hold the misperception. The evaluation of the company will be more positive because people have not come into contact with corrective facts. However, corrective information could also have a boomerang effect because misperceptions are generally hard to counter (Ecker, Hogan & Lewandowsky, 2017). In practice, providing the public with accurate information on both sides of the story becomes more difficult when individuals share attitude-consistent content with their social networks (Shin & Thorson, 2017). This relates to the high-choice media environment we currently have access to. Furthermore, corrections often fail because of the continued influence effect and reactance theory (Ecker et al., 2017).
When this is related to corporate reputation it is likely that correcting the
misinformation will relate to a decrease in the credibility and the reputation of a company. Corporate credibility, or the extent to which consumers, investors, and other constituents believe in a company's trustworthiness – which is an evaluation of the reputation as well as a company’s communication effort, in this case misinformation – and expertise, makes up a
10
portion of a corporation's image (Fombrun, 1996). The reputation can be harmed when it is suspected of producing dangerous products, when it lies to consumers and others or when it is reported to be in violation of legal and/or ethical norms (Fombrun, 1996). In addition, when traditional and social media adopt protester frames (e.g. frames that challenge the truthfulness of claims by criticizing an organization), it would be argumentative that the company’s credibility will be affected negatively. Claims that are made by the company which aren’t correct are challenged by protester frames. Moreover, the company won’t have a trustworthy image if the claims are corrected. In result, people are more likely not believing the
information anymore after they are exposed to corrective information. This will also work the other way around. Meaning that, when frames without critique are adopted, the credibility of the company will be affected positively. Transferring these arguments and findings to
statements of misinformation of a large fossil industry company, it could be assumed that negative effects of misinformation should be largest for messages that hold corrective information towards this misinformation and that there will be positive effects of
misinformation for messages that have supportive information. Therefore, the following hypothesis is conducted:
H1a: Corrective information after corporate misinformation, as compared to supporting information and mixed information, will relate to a decrease in the perceived corporate credibility.
Contradicting messages
Within the framing literature, it becomes clear that social media and traditional media frame their issues differently (Moody-Ramirez, Lewis & Murray, 2015; Egbunike and
Olorunnisola, 2015; Liu & Kim, 2011; Van der Meer & Verhoeven, 2013; Hamdy & Gomaa, 2012; Wasike, 2013). Since issues can be expected to have different interpretations, frame building often results in a competition of conflicting frames (Chong & DruckLoftusman, 2007b). In a misinformation context it reflects a play of power and boundaries of discourse over an issue. Consequently, different actors are in a way competing over what is right. Conflicting or contradicting messages are common today as result of the high-choice media environment.
Research shows that news articles are more persuasive than Twitter (Wasike, 2017) but others show that Twitter is a medium which could persuade people and is more reliable than news media (Mourao et al., 2015). News media used to be the main information source people had access to back in the day, but with the rise of social media a whole new
11
information source could be exploited. Conflicting messages of different sources as a reaction to corporate information could indicate that information is debated and therefore people could get the feeling that there is something wrong. For the general public, it could be a sign that they are dealing with misinformation. It might result in an activation of their more cognitive way of processing info via de central route described in the Elaboration Likelihood Model (ELM). In the central route to persuasion, cognitive processing involves careful and thoughtful consideration of messages for quality of arguments (Amazeen, Thorson, Muddiman & Graves, 2016). It helps the public to carefully think about what is indeed factional. Therefore, it would be very interesting to examine the role of different sources that spread conflicting information. Taken the literature into account, the following hypothesis is conducted:
H1b: Contradictory information of news media and the public after corporate misinformation will relate to a decrease in the perceived corporate credibility. Source Credibility and Skepticism
This research also seeks to find out which source is more credible. Source credibility is the degree of trust in a source (Kohring & Matthes, 2007). When messages are
contradicting regarding misinformation, it would be logical that people trust the source which they believe to be more credible. People tend to believe a correction if it comes from a trusted source, but research has also shown that if people believe a statement, they judge its source to be more credible (Lewandowsky et al., 2012). Furthermore, an explanation for the results of misinformation effect studies is provided by the source monitoring framework by Johnson, Hastroudi and Lindsay (1993). This framework explains that memories are a result of
processes by which mental experiences are attributed to sources (Cann & Katz, 2005). These attributions are not always correct due to the distributions of different features of information from multiple sources. They overlap and other factors such as prior knowledge, beliefs and desires could influence these memory attributions as well.
Furthermore, the effect of corrective and/or supportive information after corporate misinformation, in this study, could be decreased when people hold a certain degree of media skepticism or CSR skepticism. Skepticism is developed when publics continuously experience inconsistency in behaviors and claims (Webb & Mohr, 1998).Tsfati & Peri (2006) found that the more people mistrusted the mainstream media, the more they exhibited a tendency to seek
12
political information outside of mainstream news. Therefore, it is likely that people who have higher levels of media skepticism value the information on Twitter more.
CSR skepticism is defined as publics’ inclination to question, distrust, and have negative feelings toward an organization’s socially responsible actions and claims (Rim & Kim, 2016). One reason for this could be the way the media portray the scientific consensus on climate change (Brüggemann & Engesser, 2016). Another reason for CSR skepticism could be that many people doubt the extent to which companies live up to their professed standards (Skarmeas & Leonidou, 2013). Skarmeas & Leonidou (2013) found that skepticism reduces the incremental value of the retailers' name in consumers' minds. Furthermore, skeptical responses from consumers and other stakeholders could have a negative effect on the reputation of a company (Ashforth & Gibbs, 1990). In addition, CSR commitment of an organization followed by negative news coverage relates to publics’ suspicion of the
organization’s claims of good deeds (Vanhamme & Grobben, 2009). Also people may believe that the company is preoccupied with its own interests instead of CSR (Forehand and Grier, 2003). This could indicate that people who have higher levels CSR Sketicism are already more focused on possible deception regarding CSR initiatives. With the literature discussed above in mind, the following hypotheses are conducted:
H2: Corrective information of news media and supportive information of the public after corporate misinformation, compared to supporting information of news media and corrective information of the public, will relate to an increase in the perceived corporate credibility, but only when media skepticism is weak.
H3: The effect of corrective information of the NOS and supportive information of Twitter on corporate credibility decreases when people have low trust in Twitter, compared to people with high trust in Twitter, but only when media skepticism is high. H4: The effect of corrective and/or supportive information on corporate credibility decreases when the value of CSR skepticism increases and the effect increases when the value of CSR skepticism decreases.
13
Conceptual Model
Methods Design
In order to test the hypotheses of this study, an experiment embedded in a web-based survey was performed (see Appendix I). This study used a 1 shot experiment with 4
conditions: 1) both NOS and Twitter correcting corporate misinformation, 2) both NOS and Twitter supporting corporate misinformation, 3) NOS correcting and Twitter supporting corporate misinformation and 4) NOS supporting and Twitter correcting corporate misinformation.
Sample
A convenience sample of close relationships and networks has been used. The sample for this study included 160 participants of which 38,8% was male (N = 62) and 61,3% was female (N = 98). The participants were between 20 and 61 years old with an average of 27,01 years old (SD = 9,08). Each condition contains 40 participants who were randomly assigned to the condition. The experiment was online from the 3th of May till the 22nd of May 2017.
Procedure
The experiment was distributed online via Facebook, LinkedIn, and E-mail. After participants clicked on the link, it led to the openings page of the experiment where the participants were informed about the subject of the questionnaire, the ethical code and their anonymity. After reading this informed consent the process of the experiment was explained by showing what the participants could expect and which elements were more to follow. The experiment used a case study which was focused on the large fossil fuel company Shell. After
14
that, participants completed a pre-test questionnaire, which included questions about their attitude towards Shell along with issue attitudes regarding media skepticism. Participants were than told they would see an interview about oil and gas extraction with the CEO of Shell to obscure the purpose of this study. In all conditions, participants had to read an interview that was based on an existing interview with Shell’s CEO Van Beurden. The original
interview was focused on the consequences for the world after the ‘2016 Climate Agreement of Paris’ and Shell’s next steps and actions (Nieuwsuur, 2016). In the manipulated interview, the CEO of Shell confirmed the misperception that climate change is NOT caused by humans. The overall consensus is that humans cause global warming (Cook et al., 2016). For example, CEO van Beurden states in the interview: “It turns out that climate change is not so bad, since there is no evidence that people are causing climate change”. Hereby he denies the overall consensus that humans are the cause of climate change. After all participants have read the interview, they were exposed to the experimental manipulation, where participants were randomly assigned to one of four conditions. In these conditions, participants were exposed to a news article from the NOS, which is a popular traditional news media source in the
Netherlands, and a Twitter feed. These sources contained the experimental manipulation: a) both with corrective information, b) both with supportive information, c) a mixed condition, where the NOS article contained corrective information towards Shell and the Twitter feed contained supportive information towards Shell and d) a mixed condition, where the NOS article contained supportive information towards Shell and the Twitter feed contained corrective information towards Shell. After the manipulation participants had to answer the questions regarding source credibility, connection with Twitter and CSR skepticism. The survey ended with questions regarding demographics and a debriefing.
Stimuli
Both news articles are based on real news items from the NOS and uses the general format employed by NOS.nl. For a translation of the news articles and a print screen of the original in Dutch, see Appendix III. Both news articles used the same photograph but used different headlines. Furthermore, in the conditions with corrective information the article contained four paragraphs and in the conditions with supportive information the article contained three paragraphs. The news article with corrective information mainly critiques Shell’s actions. The news article with supportive information mainly approves Shell’s actions. In the conditions with supportive information, the misperception that humans do NOT cause climate change is added. In the conditions with corrective information, the misperception that
15
humans don’t cause climate change is corrected by this sentence: “(…) the CEO, in fact, trivializes climate policy and is denying climate change”.
The Twitter feed that is manipulated uses the general format employed by Twitter. For a translation of the Twitter feed and a print screen of the original in Dutch, see Appendix III. The Twitter reactions are based on real Twitter statements but for the purpose of the research, the data are manipulated and some of the statements in the conditions with
corrective information are exaggerated. The type of Twitter users that are used are people who are interested in environmental issues and Shell. Since corrective information was easier to find on Twitter, people were exposed to six tweets in the conditions with corrective
information, compared to the conditions with supportive information in which people were exposed to five tweets. The content of the tweets was addressed to Shell and its CEO. In the conditions with corrective information the tweets mainly give critique towards Shell’s actions. For example: “Why is van Beurden denying climate change?”. In the conditions with
supportive information the tweets mainly approve Shell’s actions. For example: “If Shell stops pumping, we directly boom into World War 4. Then nobody nags about 2 degrees”.
Operationalization
Media Skepticism - Media skepticism is a moderator variable. A set of five items measured participants’ mistrust in news media which in this research is called media
skepticism. The scale was constructed by three items of Gaziano and McGrath’s (1986) News Credibility Scale items (e.g. fair, accurate, tell the whole story), one item that was used by the National Election Studies since 1996 (e.g. the confidence in people running the institutions of the press) and one item that was constructed by the researcher herself (e.g. I am skeptical about the accuracy of the news).
The participants could answer the questions on a seven point Likert-scale, ranked from 1 = totally disagree to 7 = totally agree. The higher the participants scored on this scale, the less skeptical they are towards news media. In an exploratory factor analysis, all five items loaded with an explanatory variance of 65,68%. The five items had a component load higher than .631. A reliability analysis showed that Cronbach’s alpha for these five items was (α = .865) (M = 3.66, SD = 1.09. In addition, this scale was split into a dichotomous variable. Participants who initially scored 1 to 3 were assigned to the 0 = media skeptical condition and participants who initially scored from 5 to 7 were assigned to the 1 = non-media-skeptical condition. Participants who initially scored 4 were left out because these people were neutral.
16
CSR Skepticism - CSR skepticism is a moderator variable. A set of three items measured participants’ mistrust in Shell’s CSR communication, hereby called CSR
skepticism. The scale was constructed by three questions; ‘I am skeptical about the accuracy of Shell’s message about sustainable policies’, ‘I believe Shell’s message about sustainable policies’ and ‘I think that Shell has got a hidden agenda in promoting CSR’.
The participants could answer the questions on a seven point Likert-scale, ranked from 1 = totally disagree to 7 = totally agree. The higher the participants scored on this scale, the more skeptical they are towards Shell’s CSR message. In an exploratory factor analysis, all five items loaded with an explanatory variance of 61,80 %. The five items had a component load higher than .753. A reliability analysis showed that Cronbach’s alpha for these five items was (α = .691) (M = 4.57, SD = 1.20. In addition, this scale was split into a dichotomous variable. Participants who initially scored 1 to 3 were assigned to the 0 = non-CSR-skeptical condition and participants who initially scored from 5 to 7 were assigned to the 1 = CSR skeptical condition. Participants who initially scored 4 were left out because these people were neutral.
Source Credibility - The main dependent variable Source Credibility is measured by a set of eight items. The scale was constructed by three items of the Meyer (1988) Credibility Scale (e.g. trustworthy, fair, biased) and five items used by Newell & Goldsmith (2001) (e.g. expertise regarding climate issues, realistic, credible, objective, innovative). Both scales were found reliable, α = .868, α = .920. The sources to which these questions relate are Shell, Shell’s CEO, the NOS and Twitter. The questions about expertise and innovativeness were left out for the NOS and Twitter. For Shell it is a dependent variable, for the NOS and Twitter it is an independent variable.
The participants could answer the questions on a seven point Likert-scale, ranked from 1 = totally disagree to 7 = totally agree. The higher participants scored on this scale, the more credible Shell or oil producers in general were evaluated. In addition, the same items were asked in the posttest for Shell’s Credibility. A change score was computed by subtracting posttest attitude (M = 3.53, SD = .99) and pre-test attitude (M = 3.65, SD = .99), with a positive score indicating an increase in corporate credibility and a negative score indicating a decrease in corporate credibility (M = -.12, SD = .72).
Pre-test
A pre-test among a sample of 16 people checked if the manipulation was successful. A Chi-square test with condition as independent variable and ‘critique on Twitter’ as dependent
17
variable was conducted and showed that the manipulation was successful, Χ2 (3) = 16.00, p <.001. In addition, a chi-square test with condition as independent variable and ‘critique of NOS’ as dependent variable was conducted. The manipulation was not successful, Χ2 (3) = 3.05, p = .384, indicating that there had to be some minor adjustments of the NOS articles. There were already noticeable differences between groups in the change in credibility (M = .68, M = 0.41, M = 0.60, M = -.28), but no main effect was found, F (3, 12) = .37, p = .777. Furthermore, the flow of the survey worked. For the whole pre-test questionnaire, see Appendix II.
Manipulation Check
To see if the manipulation for the NOS news articles was successful, a Chi-square tests was conducted. A significant difference between condition on manipulation of the NOS article was found, Χ2 (3) = 44.80, p <.001. Thus, manipulation of the NOS news articles was successful. To see if the manipulation for the Twitter feed was successful, a Chi-square test was conducted. A significant difference between condition on manipulation of the Twitter feed was found, X2 (3) = 97.08, p <.001. Therefore, manipulation of the Twitter feed was successful.
Analyses
For all five hypotheses a two-way ANOVA analysis with ‘condition’, ‘Media Skepticism’ (H2/H3), ‘credibility of NOS’ (H3), ‘credibility of Twitter’ (H3) ‘CSR Skepticism’ (H4) and ‘Attitude Certainty of Shell’ (H5) as the independent variables and ‘change in corporate credibility of Shell’ as the dependent variable was conducted. In the results section the dependent variables ‘change in corporate responsibility’ and ‘behavioral support’ that were asked in the original experiment are left out because there were no relevant significant effects for these dependent variables in comparison with ‘change in credibility of Shell’. ‘Change in credibility of Shell’ is in itself enough representative for answering the research question. The two-way ANOVA has been chosen, since this study wants to examine if there is a relation between different independent interval and dichotomous variables.
For additional information about the manipulations, manipulation checks, randomization checks and control variables, see Appendix III.
Results
To test hypothesis 1a, if there is a main effect of supportive and/or corrective
18
independent variable and the change in Shell’s credibility as dependent variable was
conducted. This analysis showed no significant difference between conditions for change in credibility, F (3, 151) = 1.29, p = .282, η2 = .03. This means that either corrective, supportive or mixed information on misinformation has no difference for the evaluation of Shell’s credibility. Therefore, hypothesis 1a could be rejected. To test hypothesis 1b, if contradictory information relates to a decrease in the perceived corporate credibility, the conditions are divided in three groups; 1) supportive, 2) corrective and 3) contradicting. A two-way ANOVA with the three conditions as independent variable was conducted. This analysis showed also no significant difference between conditions for change in credibility, F (2, 152) = .84, p = .434, η2 = .011. Therefore, hypothesis 1b could also be rejected.
However, the mean differences are interpreted to see if the results are at least in line with the expectations. See Figure 1 for the mean differences. The figure shows that the results are in line with the expectations that only corrective information results in a decrease in Shell’s credibility, only supportive information results in an increase in Shell’s credibility and that contradicting messages result in a decrease in Shell’s credibility.
Figure 1.
Mean differences for change in Shell’s credibility between corrective, supportive and contradicting messages.
To test hypothesis 2, if Media Skepticism moderates the effect of corrective information of news media after misinformation and supportive information of the public after misinformation, a two-way ANOVA with condition as independent variable, Media Skepticism as moderator and the change in credibility of Shell as dependent variable was conducted. This analysis showed no significant main effect of condition, F (3, 134) = 1.60, p
-0,18 -0,16 -0,14 -0,12 -0,1 -0,08 -0,06 -0,04 -0,02 0 0,02
19
= .193, η2 = .04, nor a significant main effect of Media Skepticism, F (1, 134) = 1.31, p = .254, η2 = .01. There is also no significant interaction effect between these two factors, F (3, 134) = .25, p = .862, η2 = .01. Therefore, hypotheses 2 could not be supported.
To test hypothesis 3, if Source Credibility and Media Skepticism moderate the effect of corrective information of news media and supportive information of Twitter, an ANCOVA with condition as independent variable, Media Skepticism and Credibility of the NOS and Twitter as moderators and the change in credibility of Shell as dependent variable was conducted. Condition had a main effect on change in Shell’s credibility (F (3, 87) = 3.41, p = .021 η2 = .11). The conditions with only corrective information (M = -.43, SE = .19) and the mixed condition in which the NOS showed supportive information and Twitter showed corrective information (M = -.41, SE = .17) showed a decrease in the credibility. The conditions with only supportive information (M = .23, SE = .16) and the mixed condition in which the NOS showed corrective information and Twitter showed supportive information (M = .11, SE = 17), showed an increase in the credibility. There is also a significant interaction effect between condition and the credibility of Twitter (F (3, 87) = 2.98, p = .036 η2 = .09). A post-hoc test on credibility of Twitter revealed that there was a significant effect within the mixed condition in which the NOS gives corrective information and Twitter gives supportive information (p = .040). It shows a decrease in the evaluation of Shell’s credibility for
participants who do not have trust in Twitter (M = -.23, SE = .16). In contrast, the evaluation of Shell’s credibility increased for participants who do have trust in Twitter (M = .56, SE = .35). See Figure 2 for the differences between groups per condition. It shows that when participants have a high degree of trust in Twitter, it doesn’t matter what news media say. These participants have such a strong attitude towards Twitter that the attempted correction of the NOS about the misperception has no effect on the credibility of Shell. These participants will believe the misperception of climate change due to their strong degree of trust in Twitter. However, there is still no interaction effect between the change in credibility of Twitter and Media Skepticism, F (1, 87) = .02, p = .892, η2 = .00. Therefore, hypotheses 3 is not supported.
20
Figure 2.
Change in Shell’s credibility per condition for participants who have low and high trust in Twitter.
To test hypothesis 4, if CSR Skepticism moderates the effect of message framing for Shell’s credibility, a two-way ANOVA with condition as independent variable, CSR
Skepticism as moderator and the credibility of Shell as dependent variable was conducted. This analysis showed a significant main effect of condition, F (3, 127) = 3.12, p = .028, η2 = .07. Apparently, CSR Skepticism is influential since a main effect is visible which was not found in the analysis for hypothesis 1. However, there is no interaction effect with CSR Skepticism (F (1, 127) = 1.87, p = .174). Therefore, CSR Skepticism can be explained as a ‘control variable’ in the analyses for hypothesis 4.
A post-hoc test on type of information revealed that there was a significant difference between the condition with supportive information and the condition in which the NOS had supportive information and Twitter had corrective information (p = .020). In the condition with supportive information after misinformation (M = .29, SE = .15) the credibility of Shell increased, in contrast to participants in the condition in which the NOS had supportive information and Twitter had corrective information. In this condition the credibility of Shell decreased after exposure to the news item and Twitter feed (M = -.29, SE = .12). See Figure 3 for the means per condition. It seems like Twitter is the more important source here, since the NOS didn’t correct for the misinformation and people on Twitter did. Only when Twitter corrects the misinformation, there is a change in credibility, not when media do, see Figure 3.
-0,8 -0,6 -0,4 -0,2 0 0,2 0,4 0,6 0,8 Corrective Supportive NOS - Corrective / Twitter - Supportive NOS - Supportive / Twitter - Corrective
21
By correcting the misinformation, it could be that people will doubt the NOS more than Twitter. When the correction for misinformation doesn’t happen at all, in case of the
supportive condition, it appears that the credibility even increases. This is what happens if a misperception is conserved.
Figure 3.
Change in Shell’s credibility per condition after exposure to information from the NOS and Twitter.
To take a closer look at the difference between supportive, corrective and contradicting information, the conditions are divided in three groups; 1) supportive, 2) corrective and 3) contradicting. A two-way ANOVA shows also a significant main effect of condition, F (2, 129) = 3.38, p = .037, η2 = .05. The difference is significant for the condition with only supportive information and the condition with contradicting information (p = .033). This result shows that mixed and contradicting messages after exposure to corporate
misinformation could result in a decrease in corporate credibility (M = -.18, SE = .09) in comparison with supportive information (M = .28, SE = .13). See figure 4 for the differences between the types of information participants were exposed to.
-0,4 -0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 Corrective Supportive NOS - Corrective / Twitter - Supportive NOS - Supportive / Twitter - Corrective
22
Figure 4.
Increase/Decrease in credibility per type of information participants were exposed to.
Furthermore, there is no significant main effect of CSR Skepticism, F (1, 127) = 1.87, p = .174, η2 = .02, nor a significant interaction effect between these two factors, F (3, 127) = 2.04, p = .111, η2 = .05. However, there is still no moderation effect of CSR Skepticism. Therefore, hypothesis 4 could not be supported.
Discussion
Theoretical implications
The general aim of the study was to examine the effect of corporate misinformation and the attempts of news media and the online public to correct and/or support for this misinformation. It addressed the gap between the consequences corporate misinformation have on the attitude towards a large fossil fuel industry company.
The results related to the first hypothesis showed that there was no difference between the change in the corporate credibility of Shell after exposure to misinformation and its correction, support or mixed messages. A possible explanation for this finding is that the ideology and personal worldviews of the participants could have played a role in the evaluation of Shell (Lewandowsky et al., 2012). Furthermore, it is possible that people remembered what they answered before the manipulation, and knew that they were manipulated by the different messages resulting in minor differences in the post-test. This could be explained by the Cognitive Consistency Theory.
Moving on to the second hypothesis, this study examined if media skepticism moderates the effect of the mixed messages. The results show that media skepticism didn’t
-0,2 -0,1 0 0,1 0,2 0,3 0,4
23
have a moderating effect. People who are media skeptical and people who aren’t media skeptical evaluated Shell the same, regardless which messages they read. Nevertheless, it is possible that people who are media skeptical trust news media in general more than social media. In addition, if people believe a statement or in this case a misperception, they judge its source to be more credible (Lewandowsky et al., 2012) despite the fact whether the source gives supportive information or corrective information and from which source the information comes from. Believing or disbelieving misinformation could thus have a stronger effect than media skepticism. This is in line with different studies that show that prior knowledge, beliefs and desires could influence memory attributions and therefore the belief or disbelief in
misperceptions (Cann & Katz, 2005; Loftus & Palmer, 1974; Greene, Flynn & Loftus, 1982; Tousignant, Hall & Loftus, 1986; Jerit & Barabas, 2012).
The results related to the third hypothesis showed that there was no moderating effect of media skepticism and the credibility of Twitter. This is in contrast with Tsfati & Peri (2006) who found that the more people mistrusted the mainstream media, the more they seek information outside mainstream media. A possible explanation could be that news media skeptics could also be more skeptical in general towards any information source. However, the analysis showed an interaction effect between condition and the credibility of Twitter. People who have high trust in the credibility of Twitter showed a decrease in the credibility of Shell and people who have low trust in the credibility of Twitter showed an increase in the credibility of Shell. This effect was only visible for the condition in which people saw corrective information of the NOS and supportive information on Twitter. A possible
explanation could be that, if people have high trust in a particular source, in this case Twitter, they would believe anything that is said in this source, even though it is not true. In addition, it is possible that they already believed the misperception before the manipulation and that their attitude again was confirmed by the statements on Twitter.
Furthermore, it is possible that the NOS wasn’t able to correct the misperception if people have big trust in the information on Twitter. It doesn’t matter what news media say because they will always have more trust in Twitter. This finding fits with the notion that these participants have such a strong attitude towards Twitter that the attempted correction of the NOS about the misperception has no effect on the credibility of Shell. Shell’s credibility even increased for these participants after the manipulation. Due to their strong degree of trust in Twitter and therefore the public’s voice, it is possible that Shell’s image remains favorable.
24
On the contrary, the NOS was able to correct the misperception if people don’t have much trust in the information on Twitter. The argument for this explanation is provided by Lewandowsky et al. (2012) who say that people tend to believe a correction if it comes from a trusted source. In addition, the mass media were until recently the main source providing the information people needed (Kovach & Rosenstiel, 2007) and thus could still be seen as a trusted source. Due to the correction, Shell’s image and reputation decreased for people who don’t have much trust in the information on Twitter.
These results show that Twitter is a very important source in the context of
misinformation. A possible explanation could be that social media have become one of the primary means by which people learn about worldwide development and that information on Twitter is rapidly spread. Furthermore, misinformation is within easy reach since freedom of the press exists. Where news media institutions still verify stories and news items, no
overreaching state checks whether the statements of the online public are based on truth.
The results related to the fourth hypothesis showed there was no moderating effect of CSR skepticism. However, the analysis showed that there was a main effect of condition. The participants who only saw supportive information and the participants who saw contradicting information differed significantly. This difference could be explained by the fact that in the supportive information condition people didn’t see corrective information at all. This resulted in an increase in the credibility of Shell. It means that if people don’t come across corrections of misinformation, they’ll never know that Shell provided misinformation and they’ll
(wrongly) judge Shell as credible. The results for this (wrong) judgement are discussed by Lewandowsky et al. (2012) and could have serious consequences for society. For example, people will support the fossil industry because they think this industry is doing well while actually this industry is performing against society’s regulations regarding climate change. On the contrary, Shell’s credibility decreased for people who see mixed messages. This is in line with Frombrum (1996) who says that if a company lies, it could be harmful for their image. Furthermore, Werner (2016) states that it is hard to counter-argue fact-checks because the corrective information is impossible to overlook. Corrective information could in this case have a stronger effect on the attitude towards Shell than the supportive information. It is also possible that the contradicting messages activated the central route of processing information described in Amazeen et al. (2016) which may have resulted in critically thinking about whether information is true or not.
25
Practical implications
Looking at the practical insights of this study, the findings of this study could be important for different news institutions and social media publics. It shows that combating corporate misinformation is helpful and could result in attitude change towards Shell’s credibility. For news media institutions it’s a confirmation that the role as watchdog is still applicable when people who don’t have much trust in social media get in contact with their corrective information. However, they have to find a way to penetrate into people who do have much trust in social media.
Furthermore the findings could be important for Shell and its communication
department. When you as an organization decide to say things that aren’t true it is most likely that either news media institutions or the public will criticize the company. This study shows that this could decrease the company’s credibility. Therefore, the recommendation for organizations is to be transparent in your communication and don’t spread false information in case you need to get around regulations and/or laws. In the end this strategy will always backfire.
Limitations
The present study had a number of limitations. The main limitation of this study was that the relation between the independent variables and the dependent variables easily could have been influenced by other factors. The research attempted to control predetermined knowledge, beliefs and desires, though it is impossible to account for all predetermined beliefs.
The research attempted to control predetermined knowledge, beliefs and desires, though it is impossible to completely take all these attitudes out of the equation
Furthermore, the experiment contained a lot of reading material. It is possible for participants that after reading the interview, the article from the NOS and the Twitter feed, they already forgot what they have read. This could be an explanation for the minor differences between the conditions. Likewise, there were a lot of questions that the
participants had to fill in after the manipulation. It is possible that they were too rushed while filling in the questionnaire or that they didn’t have the focus anymore at the end of the
questionnaire.
Thirdly, the present study is limited in its inability to generalize the results for the whole fossil industry. Since this study is a case study the results only count for this particular
26
organization. Conclusions only could be drawn for Shell. Further research is necessary to see if corporate misinformation in other industries could have an effect on the credibility of a company.
Fourthly, the content validity can be questioned. The manipulations didn’t contain even numbers of paragraphs in the news article and even numbers of Twitter reactions. It is also possible that the misinformation wasn’t explicit enough, although it was realistic.
Lastly, some of the questions were too hard to understand or it was hard to form an opinion about it because of the topic of the interview, the news article and the Twitter feed. Some participants made use of the feedback pace in the questionnaire to explain why some questions were too difficult. This could have resulted in answering a lot of questions neutral.
Future research
Future research could address the limitations of this study by doing a similar
experiment in the realm of corporate communication in different industries. In future research it would be good to categorize participants by testing if they already hold the misperception. Furthermore, the inoculation theory could be tested in a corporate context by giving
participants a mild attack before the manipulation takes place and people are exposed to misinformation and its correction, support or mixed approach. There can be varied in factors by giving mild attacks and no attacks towards a company before the manipulation and afterwards there can be tested whether people will support the company. There could also be varied in condition with adding messages that tell both sides of the story and misinformation there could be examined if explicit and more implicit ways of misinformation have different effects. Adding more fact-checks would also be good to examine. Furthermore, how
employees of an organization react to misinformation that is given by the company they work for and how it influences their job satisfaction and attitude towards the company could also be studied. This could be very interesting for employers. What consequences does
misinformation have for your own employees and other important stakeholders? Your own employees are yet the driving force of a company.
Conclusions
Overall this study examined what the effect is of supportive, corrective and
contradicting information of different sources after exposure to misinformation on corporate credibility. This study broadened the field of research on misinformation in the corporate communication context. It appeared as though trust plays an important role in the evaluation
27
of Shell’s credibility when people are confronted with opposing messages from different sources. If misinformation is supported and people have high trust in the source that supports the misinformation, correction doesn’t help anymore. Furthermore, when CSR skepticism is added as a moderator a main effect of condition appears. This result is valuable for news institutions and social media publics who provide information to counter (corporate)
misinformation. It shows that combating misinformation does have an effect on the reputation of a company, even if people are exposed to information that supports the misinformation.
References
Amazeen, M. A., Thorson, E., Muddiman, A., & Graves, L. (2016). Correcting Political and Consumer Misperceptions: The Effectiveness and Effects of Rating Scale Versus Contextual Correction Formats. Journalism & Mass Communication Quarterly, 1077699016678186.
Antilla, L. (2005). Climate of scepticism: US newspaper coverage of the science of climate change. Global environmental change, 15(4), 338-352.
Ashforth, B. E., & Gibbs, B. W. (1990). The double-edge of organizational legitimation. Organization science, 1(2), 177-194.
Bedford, D. (2010). Agnotology as a teaching tool: Learning climate science by studying misinformation. Journal of Geography, 109, 159–165.
Besterman, M. (2013). Someone is Wrong on the Internet.
Bode, L., & Vraga, E. K. (2015). In related news, that was wrong: The correction of misinformation through related stories functionality in social media. Journal of
Communication, 65(4), 619-638.
Brüggemann, M., & Engesser, S. (2017). Beyond false balance: How interpretive journalism shapes media coverage of climate change. Global Environmental Change, 42, 58-67. Bruns, A., & Moe, H. (2014). Structural layers of communication on Twitter. In K. Weller, A.
Bruns, J. Burgess, M. Mahrt, & C. Puschmann (Eds.), Twitter and society (pp. 15–28). New York, NY: Peter Lang
Cann, D. R., & Katz, A. N. (2005). Habitual acceptance of misinformation: Examination of individual differences and source attributions. Memory & cognition, 33(3), 405-417. Chong, D., & Druckman, J. N. (2007b). A theory of framing and opinion formation in
competitive elite environments. Journal of Communication, 57, 99–118.
Cook, J., Oreskes, N., Doran, P. T., Anderegg, W. R., Verheggen, B., Maibach, E. W., & Nuccitelli, D. (2016). Consensus on consensus: a synthesis of consensus estimates on human-caused global warming. Environmental Research Letters, 11(4), 048002 Ecker, U. K., Hogan, J. L., & Lewandowsky, S. (2017). Reminders and Repetition of
Misinformation: Helping or Hindering Its Retraction?. Journal of Applied Research in
28
Egbunike, N., & Olorunnisola, A. (2015). Social media and the #Occupy Nigeria protests: Igniting or damping a Harmattan storm? Journal of African Media Studies, 7(2), 141e164.
Elving, W. J. (2013). Scepticism and corporate social responsibility communications: the influence of fit and reputation. Journal of Marketing Communications, 19(4), 277-292. Ennals, R., Byler, D., Agosta, J. M., & Rosario, B. (2010, April). What is disputed on the
web?. In Proceedings of the 4th workshop on Information credibility (pp. 67-74). ACM.
Fombrun, C. J. (1996), Reputation. Boston, MA: Harvard Business School Press.
Forehand, M. R., & Grier, S. (2003). When is honesty the best policy? The effect of stated company intent on consumer skepticism. Journal of Consumer Psychology, 13, 349– 356. doi:10.1207/S15327663JCP1303_15
Fridkin, K., Kenney, P. J., & Wintersieck, A. (2015). Liar, liar, pants on fire: How fact-checking influences citizens’ reactions to negative advertising. Political
Communication, 32(1), 127-151.
García, M. M. (2011). Perception is truth: how U.S. newspapers framed the Go Green conflict between BP and Greenpeace. Public Relations Review, 37, 57–59.
Gaziano, C., & McGrath, K. (1986). Measuring the concept of credibility. Journalism quarterly, 63(3), 451-462.
Greene, E., Flynn, M. S., & Loftus, E. F. (1982). Inducing resistance to misleading information. Journal of Verbal Learning & Verbal Behavior, 21, 207-219. Hamdy, N., & Gomaa, E. H. (2012). Framing the Egyptian uprising in Arabic language
newspapers and Social media. Journal of Communication, 62(2), 195-211.
Ihlen, & Nitz, M. (2008). Framing contests in environmental disputes: paying attention to media and cultural master frames. International Journal of Strategic Communication, 2, 1–18.
Jacques, P. J., Dunlap, R. E., & Freeman, M. (2008). The organisation of denial: Conservative think tanks and environmental scepticism. Environmental Politics, 17, 349–385. Jerit, J., & Barabas, J. (2012). Partisan perceptual bias and the information environment. The
Journal of Politics, 74(03), 672–684. doi:10.1017/S0022381612000187.
Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological bulletin, 114(1), 3.
Kohring, M., & Matthes, J. (2007). Trust in news media: Development and validation of a multidimensional scale. Communication research, 34(2), 231-252.
Kovach, B., & Rosenstiel, T. (2007). The elements of journalism: What news people should know and the public should expect. New York, NY: Three Rivers Press
29
Lewandowsky, S., Ecker, U. K., Seifert, C. M., Schwarz, N., & Cook, J. (2012). Misinformation and its correction continued influence and successful debiasing. Psychological Science in the Public Interest, 13(3), 106-131.
Liu, B. F., & Kim, S. (2011). How organizations framed the 2009 H1N1 pandemic via social and traditional media: Implications for U.S. health communicators. Public Relations Review, 37, 233e244
Loftus, E. F., & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning & Verbal Behavior, 13, 585- 589.
Mahon, J. F., & Wartick, S. L. (2003). Dealing with stakeholders: How reputation, credibility and framing influence the game. Corporate reputation review, 6(1), 19-35.
Meyer, P. (1988). Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly, 65(3), 567-574.
Mohr, L. A., Eroǧlu, D., & Ellen, P. S. (1998). The development and testing of a measure of skepticism toward environmental claims in marketers' communications. Journal of
consumer affairs, 32(1), 30-55.
Moody-Ramirez, M., Lewis, T., & Murray, B. (2015). The 2013 Steubenville rape case: An examination of framing in newspapers and user-generated content. Southwestern Mass Communication Journal, 30(2), 1e22
Mourao, R. R., Yoo, J., Geise, S., Araiza, J. A., Kilgo, D. K., Chen, V. Y., et al. (2015). Online news, Social media, and European Union attitudes: A multidimensional analysis. International Journal of Communication, 9, 3199e3222
Nieuwsuur (2016, 6 februari). Shelltopman van Beurden, het uitgebreide interview. Accessed at http://nos.nl/nieuwsuur/video/2085259-shelltopman-van-beurden-het-uitgebreide-interview.html
Nisbet, M. C. (2009). Communicating climate change: Why frames matter for public engagement. Environment: Science and Policy for Sustainable Development
Newell, S. J., & Goldsmith, R. E. (2001). The development of a scale to measure perceived corporate credibility. Journal of Business research, 52(3), 235-247.
Oreskes, N., & Conway, E. M. (2010). Defeating the merchants of doubt. Nature, 465(7299), 686-687.
Patriotta, G., Gond, J. P., & Schultz, F. (2011). Maintaining legitimacy: Controversies, orders of worth, and public justifications. Journal of Management Studies, 48(8), 1804-1836. Rim, H., & Kim, S. (2016). Dimensions of corporate social responsibility (CSR) skepticism
and their impacts on public evaluations toward CSR. Journal of Public Relations
Research, 28(5-6), 248-267.
Schafraad, P., van Zoonen, W., & Verhoeven, P. (2016). The news value of Dutch corporate press releases as a predictor of corporate agenda building power. Public Relations Review, 42(3), 451-458.
30
Shin, J., & Thorson, K. (2017). Partisan Selective Sharing: The Biased Diffusion of Fact‐ Checking Messages on Social Media. Journal of Communication.
Skarmeas, D., & Leonidou, C. N. (2013). When consumers doubt, watch out! The role of CSR Skepticism. Journal of Business Research, 66(10), 1831-1838.
Skarmeas, D., Leonidou, C. N., & Saridakis, C. (2014). Examining the role of CSR skepticism using fuzzy-set qualitative comparative analysis. Journal of business
research, 67(9), 1796-1805.
Thorson, E. (2016). Belief echoes: The persistent effects of corrected misinformation. Political Communication, 33(3), 460-480.
Tousignant, J. P., Hall, D., & Loftus, E. F. (1986). Discrepancy detection and vulnerability to misleading postevent information. Memory & Cognition, 14, 329-338.
Tsfati, Y. (2003). Does audience skepticism of the media matter in agenda setting?. Journal of
Broadcasting & Electronic Media, 47(2), 157-176.
Tsfati, Y., & Peri, Y. (2006). Mainstream Media Skepticism and exposure to sectorial and extranational news media: The case of Israel. Mass Communication & Society, 9(2), 165-187.
Van Aelst, P., Strömbäck, J., Aalberg, T., Esser, F., de Vreese, C., Matthes, J., ... & Papathanassopoulos, S. (2017). Political communication in a high-choice media environment: a challenge for democracy?. Annals of the International Communication
Association, 41(1), 3-27.
Van der Meer, T. G., & Verhoeven, P. (2013). Public framing organizational crisis situations: social media versus news media. Public Relations Review, 39(3), 229-231.
Vanhamme, J., & Grobben, B. (2009). “Too good to be true!”. The effectiveness of CSR history in countering negative publicity. Journal of Business Ethics, 85, 273–283. doi:10.1007/s10551-008-9731-2
Verhoeven, P. (2016). The co-production of business news and its effects: The corporate framing mediated-moderation model. Public Relations Review, 42(4), 509-521.
Wang, A. (2007). Priming, framing, and position on corporate social responsibility. Journal of
Public Relations Research, 19(2), 123-145.
Wasike, B. (2013). Framing news in 140 characters: How social media editors frame the news and interact with audiences via Twitter. Global Media Journal e Canadian Edition, 6(1), 5e23.
Wasike, B. (2017). Persuasion in 140 characters: Testing issue framing, persuasion and credibility via Twitter and online news articles in the gun control debate. Computers in
Human Behavior, 66, 179-190.
Webb, D. J., & Mohr, L. A. (1998). A typology of consumer responses to cause-related marketing: From skeptics to socially concerned. Journal of Public Policy and Marketing, 17, 226–238.
31
Werner, H. (2016). Facts are for losers? The effect of fact-checking on trust in politicians and trust in media sources during the US presidential campaign 2016.
