Fake News: Does Correcting it Matter in a Corporate Context?
Examining the impact of the misinformation source and how its correction
influences perceptions of organizational reputation and credibility
Chloe Madigan-Ouellette | 11985518 | Master’s Thesis | Graduate School of Communication | Corporate Communication | Supervisor Toni van der Meer | June 28 2019
Abstract
Misinformation has proven to be an increasingly relevant topic in our current society, affecting multiple industries. From a communication perspective, it is crucial to understand when
misinformation is most damaging, depending on where it comes from and how to correct it. Through an online experiment (N = 227), this study examines the role of the source that spreads the misinformation as well as if and how it is corrected in a corporate context. To understand what can influence how this information is processed by its receiver, media literacy and media scepticism are concepts incorporated in this study. Findings demonstrated how misinformation can be refuted with the use of a corrective statement, regardless of the level of detail, yet the source of this misinformation does not make a difference. This study situates itself within the current research on misinformation by shedding light on the importance of timely corrective information when organizations are confronted with misinformation being spread about them. Also, by highlighting the little importance allocated to the source of misinformation, it
emphasizes how organizations should monitor various sources of known issue arenas. Keywords: corporate misinformation, misinformation source, correction type
Introduction
During the 2016 elections in the United States of America, there was a rumor circulating about a human trafficking ring situated in a pizza restaurant called Comet Ping Pong in
Washington, DC. This rumour unraveled into something bigger than presumably intended and is forever known as “Pizzagate”, a nod to the Watergate crisis of the 1970s (Kroll, 2018).
Throughout the duration of this spreading of misinformation, the owners of the restaurant experienced many negative repercussions which included receiving death threats online and in person (Wade, 2019). This example illustrates the far-reaching consequences of misinformation on different aspects of society, including corporations.
Although most prevalent in the political sphere where truth is constantly being disputed in policy making, corrective efforts are increasing and partisanship plays a large role in how individuals process information or believe it to be true (Thorson, 2016), the concept of
misinformation is undeniably present in the corporate realm as well. “Pizzagate” demonstrates the power of the spread of misinformation and how it has the potential to negatively impact a business, which can lead to an organizational crisis. Misinformation in the form of rumours or conspiracies can impact crucial aspects of an organization such as its reputation and credibility because the public may perceive it to be real and change their perception of the organization accordingly. Misinformation is not exclusive to one sector as it is something that is wide-spread across societies and industries (Figueira & Oliveira, 2017). Thus, this study explores the
communicative role of misinformation in organizational practices. Examining how
communication processes relate to misinformation will lead to a better understanding of when inaccurate information is most harmful to an organization.
That being said, the place from which the misinformation originates can have impact on the way the information is processed. When it comes to the media, the public might interpret news pieces differently depending on the reliability of the media outlet that is disseminating it (Lewandowsky, Ecker, Seifert, Schwarz, & Cook, 2012). There is a difference in information processing between reading an article by a traditional and reputable media outlet versus a less credible one because perceptions change depending on the information source (Shin, Lian, Driscoll & Bar, 2018).
One strategy that could help organizations deal with the dissemination of misinformation is through corrective information. Depending on the severity of the crisis, the organization can release a statement de-bunking the misinformed facts made about them (Coombs, 2007).
Basically, these corrective actions “reassure stakeholders that they are safe thereby reducing their psychological stress” (Coombs, 2007, p. 165) and contingent to the type of crisis, can eliminate rumours or avoid reputational damage (Coombs, 2007). That being said, the way in which the corrective information is formulated is an important factor to consider. The amount of detail included and omitted as well as the type of information presented will have a different impact (Chan, Jones, Jamieson, & Albarracin, 2017). To counteract persisting misinformation, various strategies exist for organizations to use.
Through an experiment, the following study will examine the interplay between the misinformation source and the type of corrective information, as independent variables, in regards to the spreading of misinformation in a corporate context. The overarching research question of this study will be: How does the misinformation source and corrective information after-the-fact influence the public’s perceptions of an organization’s reputation and credibility?
Media literacy and media scepticism will be taken into consideration as possible characteristics in the receiver that can influence the processing of this information.
Theoretical Framework Defining Misinformation
The concept of misinformation is at the core of this research, specifically, within the framework of a corporate crisis. Here, fake news, the more widely-used term, and
misinformation will be synonymous with one another as this can be found in much of the existing literature on the topic (Tandoc, Wei Lim & Ling, 2018). The term misinformation is often described as being the driving force of the post-truth era we are currently living in (Waisbord, 2018). That being said, this concept appears to be difficult to define within
communication and media literature. Many authors propose fairly broad definitions that point to sharing information or a news story that is false but claims to be true (Shin et al., 2018). These authors found different types of fake news: satire, parody, fabrication, photo manipulation, advertising, and propaganda with the common thread being the attempt to pass as being legitimate. If news is constructed by journalists, fake news is co-constructed by the audience because it lies in the extent to which they perceive it to be real (Tandoc et al., 2018). In times of crisis in an age of information abundance, people are often presented with messages that appear to be true but later argued to be inaccurate in nature (van der Meer & Jin, 2019). Jang and Kim (2018) found that misinformation is more easily disseminated when there is high demand for the issue at hand, like in the context of the 2016 US presidential election. These authors studied fake news as misleading information on social media that exists “to deceive the public for ideological and/or financial gain” (Jang & Kim, 2018, p. 295). Misinformation is an important topic to
explore because this type of information is more easily retained and the media leans towards disseminating it as oppose to preventing it (Southwell & Thorson, 2015).
The challenge of correcting misinformation
In the context of misinformation, it is important to address the act of correcting it and more specifically, why this can be challenging. A reason for this is because retracting
misinformation does not necessarily change the way the audience perceives the piece of
information (Lewandowsky et al., 2012). This can be applied to various industries which means it is undeniable that corrective information can be an important tool for organizations when done properly. Correcting misinformation can be expensive for the organization due to investing resources in creating an effective message and ensuring sufficient exposure (Southwell & Thorson, 2015). Del Vicario, Bessi, Zollo, Petroni, Antonio, Caldarelli, Stanley and Quattrociocchi (2016) found that when it comes to spreading information online, the main motivation for users is social homogeneity. In the digital space, people tend to share stories that reflect their views which then reinforces their ideologies. This is a challenge to overcome because if the information presented does not reflect a certain narrative, people may ignore it (Del Vicario et al., 2016).
Correcting misinformation in a corporate crisis
Within the corporate context, there are several challenges that can arise when attempting to correct misinformation. It has the potential to spiral into a crisis by threatening the
organization’s reputation. Not only can the act of correcting misinformation be difficult for the organization itself, there are other factors that come into play. Carvalho, Klagge and Moench (2011) explored the notion of a false news shock where separate pieces of information cancel one another like when United Airlines stock plummeted after a report from years prior resurfaced
about their parent company filing for bankruptcy. The media corrected it but it took some time for the United Airlines stock to normalize. The correction of misinformation can be treated as a crisis response strategy and in a crisis scenario, a response from the organization involved can be highly effective (Coombs, 2007).
Corporate reputation and credibility
Misinformation poses direct threats to corporate reputation and credibility. Within the framework of a corporate crisis and the spread of misinformation about an organization,
corporate reputation has proven to be a powerful determinant for the way stakeholders perceive an organization (Matuleviciene & Stravinskiene, 2015). Many authors in the communication field confound it with corporate image or as a tool of corporate marketing (Gotsi & Wilson, 2001). In this study it will be referred to as the overall evaluation of the extent to which a firm is substantially good or bad (Tat Keh & Xie, 2009). The importance of a positively perceived reputation is undeniable. Highly perceived reputations can lead to organizational sustainability as well as benefit the company by enabling a competitive advantage compared to companies with a lower perceived reputation (Tat Keh & Xie, 2009). Equally, organizations with a positive
reputation tend to have more leverage in times of crisis where stakeholders will likely be more lenient if the organization makes a mistake (Coombs, 2007). These are the reasons corporate reputation is at the centre of this research.
Although corporate credibility is intrinsically related to corporate reputation, they are two separate entities. Credibility will be referred to here as “the perceived expertise and
trustworthiness or truthfulness of a firm” (Newell & Goldsmith, 2001, p.237). As with corporate reputation, stakeholders may be more lenient in their perceptions of the organization in times of crisis towards those with high corporate credibility (Coombs, 2007). Credibility is a concept that
is comprised of perceptions held by an individual or group about an organization (Findley Musgrove, Choi & Cox, 2018). Notably, companies with high corporate credibility usually have consumers trusting and liking them. Credibility is an element worth exploring seeing as it has the power to shape public perceptions of an organization.
Misinformation source
An element that will be explored more specifically in relation to misinformation is the source. The source that is spreading the information can have an impact on how it is processed by the public (Tandoc, 2019). In this context, the media are a crucial player. They have the capacity to shape public opinion with the information they share and prove to be large
contributors in the dissemination of misinformation (Lewandowsky, et al., 2012). Plus, they are a key source in times of crisis (van der Meer, 2018). Journalists tend to play a part in
daily-sensemaking processes as well as in the way truth is constructed (Waisbord, 2018). Although unlike traditional journalism, the Internet does not have gatekeepers that filter the information which is why online platforms are more susceptible to the spread of misinformation
(Lewandowsky et al., 2012). Generally, false rumours are driven by non-traditional news sources but social media platforms such as Twitter are also where the sharing of misinformation is fairly common (Shin et al., 2018). These non-traditional news websites make it so that the notion of truth is being constantly disputed as they reconstruct the definition of a newsworthy topic (Waisbord, 2018). Source credibility and news credibility work hand-in-hand because the media are a heuristic cue for the public to evaluate the credibility of the information and judge the extent to which it can be trusted (Tandoc, 2019). Also, when highly motivated to analyze the information, individuals found news articles more credible when shared by an established media outlet than by other sources (Tandoc, 2019). This study expects, that when inaccurate
information about an organization comes from a more credible media outlet it is more likely to damage reputation and credibility than when it comes from a less credible news website. H1: Those exposed to misinformation by a less credible news website will have a more positive perception of the organization’s reputation (H1a) and credibility (H1b) than those exposed to misinformation by a more credible news source.
Corrective information type
The mere presence of a corrective statement by the organization can be crucial when it comes to a misinformation crisis (Coombs, 2014). Equally, the type of corrective information released by an organization is essential in understanding how to act in a misinformation crisis to minimize reputational and credibility damage. It is comparable to a denial response strategy in a crisis scenario which can be effective when circumstances permit (Coombs, 2014). When it comes to correcting misinformation, the way a statement is constructed can make a difference to the audience. The ones that are more detailed are proven to be more effective, especially when offering new information because merely counterarguing the previous false claims may reinforce them (Chan et al., 2017). Although there is doubt within this field about the effectiveness of corrective information by an organization because misinformation can be memorable and audiences hard to reach (Lewandowsky et al., 2012), there is a general consensus that if done properly, the correction can be well received by the intended audience (Southwell & Thorson, 2015). Equally, the fact of having this corrective information present in the first place should have an influence on the audience in terms of how they perceive the organization (van der Meer & Jin, 2019). This study presumes that by presenting new facts rather than simply contradicting the initial misinformation, the audience should react more positively to the statement, and therewith, the organization (Lewandowsky et al., 2012).
H2: Those exposed to a detailed corrective statement by the organization after exposure to the misinformation will have a more positive perception of the organization’s reputation (H2a) and credibility (H2b) than those exposed to the less detailed or no corrective statement by the organization.
It is worth exploring the relationship between the two concepts mentioned above. In many studies they are intrinsic seeing as the debunking of misinformation can be examined in relation to the diffusion of this misinformation online and from where it originates (Shin et al., 2018) but also how in reality, misinformation cannot exist without a truthful counterpart. The lowest threat, which in this case is when the misinformation originates from a non-credible
source, followed by a detailed corrective statement would be the least harmful to an organization. H3: Those exposed to misinformation by a less credible news website as well as the detailed corrective statement will have a higher perception of the organization’s reputation (H3a) and credibility (H3b) than those exposed to misinformation by a more credible news source and concise or no corrective statement by the organization.
Characteristics that influence the processing of misinformation
Lastly, there are certain characteristics that may influence the processing of information by the receiver in today’s media environment. It is beyond doubt that with the rise of social media and the evolution of the Internet, society online has become more fast paced which means the media have developed in relation to these unspoken rules (Waisbord, 2018). This is one of the reasons why media literacy has become increasingly important. There are many varied definitions of media literacy throughout literature, especially with the increasing presence of mediated messages that make the foundation of online culture (Koltay, 2011). Here media literacy will be referred to as an individual being able access, analyze, evaluate and create
messages (Livingstone, 2004). Those with high levels of media literacy tend to process information from the media differently than others (Koltay, 2011) which means they may be more likely to recognize certain elements that enhance the credibility of a story, for instance. Self-perceived media literacy is a crucial element to consider in this study due to the fact that people tend to believe that others are more susceptible to media influence than they are when processing news information (Jang & Kim, 2018). There also tends to be an out-group and in-group effect with misinformation. Particularly within politics, authors Jang & Kim (2018) found that individuals believed those within their group who, by definition share their same views, tend to be less vulnerable to misinformation than those who exist in an outside group. In sum, this study infers that media literacy in participants will play an important role on how a news article containing misinformation is processed in relation to how they perceive the reputation and credibility of the organization in question.
H4: High media literacy skills will lead to a more positive perception of the organization’s reputation and credibility when exposed to misinformation by a less credible news website (H4a, H4b) and a detailed corrective statement (H4c, H4d) as oppose to misinformation on a more credible news source and concise or no corrective statement by the organization.
Notably, similar to media literacy, media scepticism is a concept worth considering in the context of misinformation because it is also a characteristic that can influence the processing of information (Maksl, Ashley, & Craft, 2015). Interestingly, trust in the media has been in decline over the years (Tsfati, 2003). Although there is not a plethora of studies diving into media scepticism, there are certain studies that look at scepticism in relation to agenda setting or online news sources as well as media scepticism in the context of a hostile media phenomenon (Tsfati, 2003; Tsfati, 2010; Choi, Yang & Chang, 2009). In literature, this term is commonly defined as
the “subjective feeling of alienation and mistrust toward the mainstream news media” (Tsfati, 2003, p.160). Many studies incorporate terms such as likeability and trustworthiness when referring to media scepticism (Tsfati, 2003; Tsfati, 2010; Maksl et al., 2015). Tandoc (2019) studied how individuals perceived their peers to be more credible than a news outlet when it comes to sharing information on social media. In general, people who are sceptical tend to be less affected by the salience of a news item (Tsfati, 2003). A high level of media scepticism implies that receivers mistrust the media, which in turn can influence their perceptions of the information being disseminated (Maksl et al., 2015). Tsfati (2010) found that exposure to alternative online news was directly related to media scepticism. This should mean that they are less trusting in various media outlets in general, regardless of whether the outlet is known for being more fact-based or not and should be less influenced by the credibility of the media source. Thus, those with higher levels of media scepticism are more careful to believe the news that is disseminated by the media, and will use this information to a lesser extent when evaluating the organization.
H5: Those with high media scepticism will have a more positive perception of the organization’s reputation and credibility when exposed to misinformation by a more credible news source (H5a, H5b) and a detailed corrective statement (H5c, H5d) as compared to those exposed to misinformation by a less credible news website and concise or no corrective statement by the organization.
Figure 1. Conceptual Model of Hypotheses
Methods Sample
A total of 227 participants were included in this study, this being once 92 participants were excluded due to non-completion. Through the context of an experiment, a questionnaire was circulated within the researcher’s network and completed online. Participants were asked to share their gender, age, country of origin and education level as background characteristics. The age range of participants was 19 to 80 with the mean age being 32 (M = 32.20, SD = 13.78). Participants were predominantly from Canada (42.6%) with the rest being from countries such as The Netherlands (22%) and Germany (6.6%). Overall, the sample consisted of 27.3% males and 72.7% females with 49.8% of participants having completed a bachelor degree.
A randomization check was conducted to see if the characteristics mentioned above were equally distributed among the conditions. A one-way ANOVA was done for age (p = .680) and a
chi-square test was done for both gender (2 = 2.38, p = .794) and education (2 = 21.47, p = .370) which all demonstrated that these characteristics did not significantly differ per condition. Design
For this research, a 2x3 in-between subjects factorial design was used which included six conditions. The two referring to the media conditions i.e., the more credible news outlet and the less credible news website and the three referring to the corrective information by the
organization i.e., the detailed statement, the concise statement or the absence of a statement. Procedure and manipulation
Upon receiving the questionnaire and opening the link, participants are instructed to read general information about the study as well as to consent to the use of their data. They are then asked to fill in questions related to their background characteristics which included gender, age, country of origin and level of education. Equally, they shared their levels of media literacy and media scepticism in this section. They were also asked whether they have children or not and how frequently they research children related topics online because this could influence the way they react to the crisis scenario which involves children. Following these background variables, participants were randomly assigned to one of the six conditions.
All participants were introduced to the organization Daisy & Co. with a short description of the organization’s background. Daisy & Co. is a fictional organization that sells baby products for parents. The organization and crisis scenario in this experiment were completely fabricated to eliminate any biases participants may have had towards an organization they know or a crisis they are familiar with when responding to questions in the survey. The scenario involves
allegations that one of their products, a pacifier toy, was the cause of lead poisoning in children. This was inspired by a similar misinformed article on asbestos being found in Johnson &
Johnson baby powder (Wischhover, 2018). The choice to write an article on children’s health is also meant to be reflective of the relationship between the anti-vaccine movement and fake news in current society (Young, 2018). This makes the crisis topic relevant, top of mind and realistic.
Afterwards, participants were introduced to one of two news outlets, either Infowars or Reuters depending on the condition they were assigned to. The choice to use British media outlet Reuters and American news website Infowars as the news sources in this experiment enhanced external validity and because participants could have been familiar with the reputations of the respective media outlets seeing as the source of misinformation was a variable being studied. Reuters was ranked as one of the more trusted news sources in the United States and Infowars as one of the least (Ruddick, 2017). Following this, they read the article written about Daisy & Co. for at least one minute (see Appendix A1 and A2 for misinformation news articles). The message was consistent among both sources. The article was meant to emulate fake news which included testimonials from mothers for whom their children were allegedly poisoned. This being without extensive information on the organization and without concrete proof or input from an outside party. On the one hand, the article itself was written in a way that is most commonly associated with fake news articles. On the other hand, it was written in a way that was still representative of a real crisis because participants were meant to interpret it as a realistic scenario for the sake of this experiment.
Participants were then in the next phase of the experiment which involved a corrective statement by the organization Daisy & Co. They either read a detailed or concise statement or saw no statement at all. This statement by Daisy & Co. was an attempt to correct the
misinformation participants had just previously read about the organization but seeing as participants could not tell what was true or not, the news article was not by definition
misinformation. Depending on the condition participants were assigned to, this correction was detailed and extensive to introduce new information about the claims made against them or concise and short as a simple rebuttal denying the claims. For instance, the more detailed statement provides concrete examples of how the products are tested in compliancy with the Child Safety Protection Act which includes lifespan testing to avoid possible choking hazards. The more concise statement was in bullet-point form and included the most essential information such as how Daisy & Co. products are tested in three different ways to ensure their safety for children, without diving into details (see Appendix B1 and B2 for corrective statements).
Following this, participants answered questions related to the reputation and credibility of the organization. The survey ended with questions specifically related to the manipulation they were exposed to. They were then debriefed on the experiment and explained that the crisis scenario was fictitious for the purpose of the research.
To optimize the stimulus material, a pilot test was circulated to see if the manipulation of this experiment was effective. Participants were exposed to one of the six conditions, as well as answered questions regarding the dependent variable reputation and the moderator media literacy, followed by questions related to the manipulation checks. Due to the small sample size the results were not significant, but the means were in the right direction of the hypotheses. For instance, participants were asked to evaluate the credibility of the media outlet, where higher scores are more credible and lower scores are less credible, Reuters (M = 5.14, SD = 1.46) and Infowars (M = 4.00, SD = 1.73). The detailed corrective information by the organization (M = 4.40, SD = .89) did not seem to differ from the concise statement (M = 3.75, SD = 1.50) when participants were asked to evaluate the degree of detail in the statement. Therefore, more
information was added to the detailed statement and the concise statement was made to appear more straightforward with the use of fewer words and enlarged bullet points.
Moderation measures
As mentioned above, two of the elements incorporated as background characteristics were media literacy and media scepticism, which were both examined as moderators in this study. Self-perceived media literacy was measured with six items, which included “I am confident in my ability to judge the quality of the news” and “I know how to recognize false or misleading information” (Vraga, Tully, Kotcher, Smithson & Broeckelman-Post, 2015; Jang & Kim, 2018; Maksl, Craft, Ashley & Miller, 2016), (Appendix C). This scale was a combination of items from different studies, some of which were tailored to the acquisition of media literacy in regards to misinformation. One of the six items was reverse coded and ultimately left out of the final testing, with the Eigenvalue being 2.71 (M = 4.19, SD = .91, Cronbach’s = .77). This scale was measured on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree”, 1 being low self-perceived media literacy and 7 being high self-perceived media literacy.
Media scepticism was measured with items including “I think the news media are fair” and “I think the news media are accurate” from a scale developed by Maksl, et al., (2015) (Appendix D). Three items were reserve coded and two items were left out of the final testing phase (M = 4.35, SD = .94, Cronbach’s = .86). These items were measured on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree” where 1 is low scepticism and 7 is high scepticism.
Dependent measures
After being exposed to the stimulus which was randomly assigned, participants were asked to respond to questions that measured the following dependent variables.
To measure corporate reputation, items were used from a scale developed by Puncheva‐ Michelotti and Michelotti (2010) which touched on customer value, ethical responsibility and emotional appeal. Items included “The company provides excellent value to its customers” and “The company is honest and straightforward in its communication” (Appendix E). All items were measured on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree” (M = 4.24, SD = .97, Cronbach’s = .93).
To measure corporate credibility, a scale developed by Featherman, Miyazaki and Sprott (2010) was used. The scale included items such as “The company is reliable” and “The company is likeable” (Appendix F). All items were measured on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree” (M = 4.12, SD = .90, Cronbach’s = .90).
Manipulation checks
Manipulation checks were conducted to see if the manipulation of the independent variables was successful. Participants were asked to identify the media outlet from which the article originated, being either Infowars or Reuters. A chi-square test was conducted to confirm that participants did significantly correctly identify the media source (2 = 190.66, p < .001). Then, they evaluated the credibility of this media outlet. A one-way ANOVA confirmed that the manipulation was successful showing that participants who read the article by Infowars assessed the media outlet as significantly less credible (e.g. M = 3.58, SD = 1.27) than those exposed to the article by Reuters (e.g. M = 4.58, SD = 1.06); F(5) = 12.21, p < 001). According to the Levene’s test, these results should be taken with caution (p = .008). Participants were then asked the degree to which the statement by the organization was detailed. Those who were not exposed to the corrective information did not see this question. A one-way ANOVA confirmed a
SD = .96) did judge it as being more detailed than those in the concise corrective information condition (M = 4.04, SD = 1.23); F(1) = 23.29, p < .001). According to the Levene’s test, these results should be taken with caution (p < .001).
Data analysis
To test the hypotheses of this study, a series of ANOVAs were used. Both independent variables misinformation source and the corrective information type were nominal variables and the dependent variables reputation and credibility were interval level variables. Also, the Hayes macro-PROCESS was used to examine the moderation effects of media literacy and media scepticism based on the hypothesized interaction effect, as they were measured on an interval level. Notably, the variable of having children or not was controlled for in all of these analyses.
Results
Misinformation source
ANOVAs were used to test the hypothesis that those exposed to the more credible news source, the independent variable, would have a more negative perception of both the credibility and reputation, the dependent variables, of the organization than those exposed to the less credible news source.
An analysis of variance showed that perceptions of the organization’s reputation did not significantly differ from those exposed to the less credible news website i.e., Infowars (M = 4.23, SD = 1.01) in relation to those exposed to the more credible news website i.e., Reuters (M = 4.24, SD = .93; F(1) = .04, p = .834). The same could be applied to the perceptions of the organization’s credibility which did not significantly differ from those exposed to the less credible news website (M = 4.17, SD = .91) in relation to those exposed to the more credible
news website (M = 4.06, SD = .90; F(1) = .47, p = .493). Notably, having children did
significantly influence the outcome with a small effect (p = .010, 2 = .03) Hypotheses 1a and 1b were rejected and those exposed to Reuters article did not perceive the reputation and credibility of Daisy & Co. as more negative than those who read the Infowars article.
Corrective information type
ANOVAs were conducted to test the hypothesized effect that participants exposed to the more detailed statement, the independent variable, will have a more positive perception of the organization’s reputation and credibility, the dependent variables, than those exposed to the concise statement or no statement at all.
A One-way ANOVA was conducted to see if the difference between seeing detailed, concise or no correction by the organization had an impact on the perception of the reputation. Results showed that participants did significantly perceive the organization’s reputation, with a large effect size, as more negative with the absence of corrective information (M = 3.57, SD = .76) as oppose to the presence of detailed (M = 4.66, SD = .98) or concise information (M = 4.50, SD = .76; F(2) = 36.42, p < .001, 2 = .25). A Bonferroni post-hoc comparison demonstrated that the absence of corrective information did significantly differ for those exposed to the detailed (p < .001) or the concise (p < .001) corrective information. That being said, there is no significant difference between the two types of corrective information in regards to the perception of the organization’s reputation (p = .756). Similar results were found in relation to organizational credibility. The analysis of variance showed that there is a significant effect of the corrective information, with a medium effect, on how the organization’s credibility is perceived (F(2)= 17.03, p < .001, 2 = .13). A Bonferroni post-hoc comparison demonstrated that the absence of corrective information (M = .37, SD = .79) did significantly differ from the detailed information
(M = 4.44, SD = .88, p < .001) and the concise information (M = 4.26, SD = .87, p < .001). Although there is no significant difference between the two types of corrective information in regards to the perception of the organization’s credibility (p = .621). The fact of having children did have a significant but small effect on the way credibility was perceived (p = .010, 2 = .03). This means that Hypotheses 2a and 2b were only partially supported.
The relationship between misinformation source and corrective information type
A Two-way ANOVA was used to test the hypothesis that those exposed to both a less credible news website and detailed corrective information will have a more positive perception of the organization’s reputation and credibility. This as oppose to participants who were exposed to a more credible news website and either concise or absent corrective information.
The analysis of variance showed that there is no interaction effect of both misinformation sources, the less credible news website and correction detailed (M = 4.69, SD = .98), concise (M = 4.54, SD = .85) and absent (M = 3.63, SD = .77) and the more credible news source and corrective information detailed (M = 4.63, SD = 1.00), concise (M = 4.47 , SD = .71) and absent (M = 3.63, SD = .77 ) on reputation (F(2) = .39, p = .675). The same can be applied to the dependent variable credibility. The analysis of variance showed that there is no significant effect of both misinformation sources, the less credible news website and correction detailed (M = 4.58, SD = .71), concise (M = 4.31, SD = .92) and absent (M = 3.67, SD = .85) and the more credible news source and corrective information detailed (M = 4.31, SD = 1.00), concise (M = 4.22 , SD = .82) and absent (M = 3.67, SD = .74 ) on credibility (F(2) = .67, p = .512). Hence, Hypotheses 3a and 3b were rejected. Notably, the fact of having children did significantly influence the results albeit with a small effect size (p = .010, 2 = .03).
The Hayes Process was used to test the moderating effects of media literacy and media scepticism on the misinformation source and corrective information type, the independent variables, in relation to organizational reputation and credibility, the dependent variables. Analyses were conducted with both moderators integrated to properly analyze the two independent variables and the two dependent variables.
H4(a)(b) and H5(a)(b): It is hypothesized that those with high levels of media literacy will have a more positive perception of the organizational reputation and credibility when exposed to the less credible news website. Also, those with high media scepticism will have a more positive perception of the organization’s reputation and credibility when exposed to the more credible news source. Results showed that there is no significant effect of the moderators media literacy (SE = .93, t = -.04, p = .972, 95% CI [-.21; .20]) and media scepticism (SE = .11, t = .50, p = .618, 95% CI [-16; .27]) on the outcome variable reputation. The interaction effect is also insignificant for either media literacy (SE = .14, t = 50, p = .618, 95% CI [-.21;.36]) and media scepticism (SE = .14, t = -.23, p = .824, 95% CI [-.31; .25]) on the effect of the
misinformation source. Meaning that the level of media literacy and media scepticism of the participant did not differ in their perceptions of the organization’s reputation if they read the article by the less credible news website or the more credible media outlet. For the outcome variable credibility, the process showed that there is no significant effect of the moderators media literacy (SE = .09, t = .31, p = .754, 95% CI [-.16; .22]) and media scepticism (SE = .10, t = .55, p = .580, 95% CI [-.14; .26]). Also, the interaction effect is not significant for either media literacy (SE = .13, t = .22, p = .828, 95% CI [-.23; .29]) or media scepticism (SE = .13, t = -.35, p = .727, 95% CI [-.31; .21]) on the misinformation source. This means that the level of media literacy and media scepticism of the participant did not differ in their perceptions of the
organization’s credibility if they read the article by the less credible news website or the more credible media outlet. Hence, Hypotheses 4a, 4b, 5a and 5b were not supported.
H4(c)(d) and H5(c)(d): It is hypothesized that those with high media literacy and high media scepticism who are exposed to the corrective information will have a more positive perception of the organization’s reputation and credibility as oppose to those who did not see the corrective information. To complete this analysis, the independent variable of corrective
information was recoded into two categories, one with the presence of the correction and one with the absence of it seeing as the amount of detail did not have an effect. Results show no significant effect of the moderators media literacy (SE = .19, t = -.07, p = .947, 95% CI [-.38; .36]) and media scepticism (SE = .18, t = -.28, p = .784, 95% CI [-.41; .31]) on the outcome variable reputation. Also, the interaction effect is not significant for either media literacy (SE = .13, t = .56, p = .576, 95% CI [-.19; .34]) or media scepticism (SE = .12, t = -.52, p = .601, 95% CI [-.18; .31]) on the presence or absence of the corrective statement. Ultimately, the presence of corrective information coupled with media scepticism and media literacy in the participant, did not influence the way they perceive the organization’s reputation and credibility. Also, results show that there is no significant effect of the moderators media literacy (SE = .19, t = .47, p = .642, 95% CI [-.28; .45]) and media scepticism (SE = .18, t = -.58, p = .566, 95% CI [-.47; .26]) on the outcome variable credibility. The interaction effect is not significant for either media literacy (SE = .13, t = -.05, p = .957, 95% CI [-.27; .25]) or media scepticism (SE = .13, t = -.79, p = .433, 95% CI [-.14; .34]) on the presence or absence of the corrective statement. Notably, having children does have a significant effect on how the credibility of the organization is perceived in relation to the levels of media literacy and media scepticism of a participant after
being exposed to the corrective information by the organization (p = .012). Thus, Hypotheses 4c, 4d, 5c and 5d were not supported.
Table 1. Moderation effect on reputation in ANOVA
Media Literacy Media Scepticism
Low M (SD) High M (SD) Low M (SD) High M (SD) Corrective information Present 4.52 (.70) 4.67 (1.11) 4.55 (.78) 4.61 (.99) Absent 3.37 (.86) 3.72 (.65) 3.60 (.77) 3.53 (.79) Misinformation source Infowars 4.21 (.98) 4.25 (1.07) 4.25 (.87) 4.21 (1.19) Reuters 4.22 (.83) 4.27 (1.04) 4.17 (.94) 4.32 (.93) Table 2. Moderation effects on credibility in ANOVA
Media Literacy Media Scepticism
Low M (SD) High M (SD) Low M (SD) High M (SD) Corrective information Present 4.29 (.74) 4.44 (1.04) 4.32 (.87) 4.38 (.88) Absent 3.55 (.89) 3.76 (.71) 3.65 (.79) 3.68 (.79) Misinformation source Infowars 4.13 (.83) 4.22 (1.02) 4.15 (.88) 4.20 (.91) Reuters 4.05 (.87) 4.08 (.93) 4.00 (.91) 4.13 (.88)
Discussion
The objective of this study was to investigate if the source of misinformation and if the presence and type of corrective information has an impact on organizational reputation and credibility as perceived by the public. Through experimental research, this study observed the absence of influence of the media source when it comes to misinformation and the impact of corrective information following the spread of this misinformation.
The results of this study demonstrated the lack of importance allocated to the source of misinformation when it comes to public perceptions of an organization’s reputation and credibility. Surprisingly, participants did not respond to the news article differently from the source that is known for being fact-based as opposed to the source that is known for
sensationalizing news content even though participants did rate these sources to be different in terms of their credibility. Although they acknowledged one source to be more credible than the other, it did not influence how they processed the information and in turn, evaluate the
organization in question. This is an interesting finding because it shows that the media source may not have as much of an impact as initially assumed (Waisbord, 2018) and that the credibility of a news source does not have an impact on the way news is consumed. The public blindly consumes news articles even if it is aware that the information source is not credible, which is disconcerting. Although many studies on misinformation do not empirically examine the impact of the media as its source, many studies do look at the construction of misinformation and how it is spread (Figueira & Oliveira, 2017; Shin et al, 2018; Tandoc et al., 2018) which mirrors the finding that the source of misinformation plays a smaller role than assumed. However, as based on previous research, the importance may lie in the source of the correction of misinformation instead (van der Meer & Jin, 2019).
That being said, it is important that the information deemed inaccurate be corrected. These results highlight that the mere presence of this rebuttal does have an impact on the way the public perceives the reputation and the credibility of the organization in question, which is in line with the crisis response strategies theory by Coombs (2007). This research also demonstrates that correcting misinformation means debunking but the amount of detail in the debunking by the organization following a crisis does not necessarily matter to the public. This reflects findings from van der Meer and Jin (2019) that highlighted how the amount of elaboration in a rebuttal did not make a difference in times of a public health crisis on participants’ attitudes. Perhaps this is due to the fact that misinformation can stick to memory more than regular information and the audience lends more weight to the content of the statement rather than to the negation of the misinformation (Lewandowsky et al., 2012). This could also be because the audience allocated more importance to the credibility of the source that is doing the correction rather than the amount of detail in the statement itself (van der Meer & Jin, 2019) and seeing as they were not familiar with the organization, they did not perceive the statement to be credible (Lewandowsky et al., 2012). This highlights the necessity for correcting misinformation but also the timely response from an organization involved in a crisis, to avoid others ‘stealing thunder’ (Coombs, 2014). It could also mean that the crisis may warrant a more elaborate response at a different point in time after the initial response to the misinformation has been released.
Also, the concept of media literacy did not influence the way in which the news article was processed by the receiver. This could be due to the fact that media literacy is still difficult to define in current literature. Although its importance in society is increasing, the way in which it is taught and acquired is still being developed (Koltay, 2011). A study by Waisbord (2018) highlighted how media literacy is not sufficient in the processing of news articles because
journalism proves to be a malleable industry that is constantly evolving. Equally, media
scepticism is a concept that did not influence the way the news article was processed in relation to the perceptions of the reputation and credibility of the organization. This contradicts the finding by Tsfati (2010) that finds media sceptics tend to view information they see online as more credible than those of traditional media outlets. This could be explained by the idea that perceptions of one’s levels of media literacy and scepticism may not reflect behavior in reality. This is in line with the public not evaluating information differently from sources they deem to credible or not. It could also be that media literacy and media scepticism do influence how information is processed but to a lesser extent on the evaluation of the organization. Possibly, individuals overestimate their capacities to consume media in their everyday environment because, for media literacy in particular, it is becoming increasingly important to be able to navigate the current media landscape (Koltay, 2011).
Implications
This research highlights the importance of the organization’s role in a post-crisis situation involving the spread of misinformation. Organizations should note that when misinformation is being spread, the fact that it originates from a sensationalizing news website or a credible media outlet, may not make a big difference to the public. Thus, it is important for organizations to monitor the information being disseminated on all different types of online platforms and media outlets exactly for this reason. Equally, communication practitioners must prepare their
responses to misinformation that impacts them. Even though the degree of detail may not have a large influence on the public, the mere presence of a rebuttal can change the way their reputation and credibility are perceived by stakeholders. Time is also an important factor to consider. If the organization corrects the misinformation in a timely manner, it could potentially influence public
perception in a positive way. This reflects the importance of having predetermined responses to a possible crisis scenario and assessing potential threats online (Marwitz, Maxson, Koch,
Aukerman, Cassidy, & Belongerf, 2008). Limitations & Future Research
It is necessary to acknowledge certain limitations of the present study as well as to offer potential avenues for future research on this topic. There is a challenge in conducting research on the concept of misinformation that is undeniable. Studying fake news is inherently difficult in an experiment, because when creating an article about a fabricated crisis that happened to a
fabricated organization, this can already be considered fake news by definition (Tandoc et al., 2018). It is a challenge to create an experiment about fake news, in hopes that participants react accordingly to maintain internal validity, without disclosing that the article they are reading is fake news. It would be interesting for future studies to dive into if perceptions of the organization could vary if the misinformation and rebuttal originates from other sources in a corporate crisis. Also, the content itself within the fabricated news article would be a notable variable to consider. The present study involved children in a crisis, which was ultimately controlled for, but it would be worth exploring different types of misinformation content to see if they vary in their degrees of impact in the context of a corporate crisis.
The fact that the moderators in this study media literacy and media scepticism did not have an impact may have been due to self-perception and socially desirable responses. It would be worth trying a different method in measuring these concepts to see if they can influence participants in a misinformation crisis. In a similar vein, media is a broad term that has been widely defined with the potential of being misinterpreted. It is necessary in future research to provide a clear-cut definition of what media means within the context of the research. In
particular when using moderators such as media literacy and media scepticism, it would be interesting to define the type of media that is being referred to, that is either credible or not credible. That being said, characteristics within an individual that shape the way they consume news should be further explored going beyond media literacy and scepticism.
In conclusion, this study sought to shed light on the topic of misinformation within the corporate context. Misinformation about an organization should be treated as a crisis. The type of media as a misinformation source does not change perceptions of information and the timely presence of the correction of misinformation is crucial for an organization to maintain a positive reputation and credibility. Equally, individual characteristics such as media literacy and media scepticism do not have as large of an influence as initially presumed because human behaviour does not always reflect self-perceptions. Communication practitioners must consider these
findings when dealing with misinformation and future research should build upon them to further attempt to understand the phenomenon that is fake news.
References
Arke, E. T., and Brian A. Primack, B. A. (2009). Quantifying media literacy: development, reliability, and validity of a new measure. Educational Media International, 46(1), 53-65. doi: 10.1080/09523980902780958
Carvalho, C., Klagge, N., & Moench, E. (2011). The persistent effects of a false news shock. Journal of Empirical Finance, 18, 597-615. doi: 10.2139/ssrn.1363762
Chan, M. S., Jones, C. R., Jamies, K. H. and Albarracin. D. (2017). Debunking: A Meta-Analysis of the Psychological Efficacy of Messages Countering Misinformation. Psychological Science, 28(11), 1531-1546. doi: 10.1177/0956797617714579
Chen, Y., Conroy, N. J., & Rubin, V. L. (2015). Misleading Online Content: Recognizing Clickbait as “False News”. Proceedings of the 2015 ACM on Workshop on Multimodal Deception Detection, 15-19. doi: 10.1145/2823465.2823467
Choi, J., Yang, M., and Chang, J. JC. (2009). Elaboration of the Hostile Media Phenomenon: The Roles of Involvement, Media Skepticism, Congruency of Perceived Media Influence, and Perceived Opinion Climate. Communication Research, 36(1), 54-75. doi:
10.1177/0093650208326462
Coombs, T. W. (2007). Protecting Organization Reputations During a Crisis: The Development and Application of Situational Crisis Communication Theory. Corporate Reputation Review, 10(3), 163-176. doi: 10.1177/089331802237233
Coombs, W. T. (2014). State of crisis communication: Evidence and the bleeding edge. Institute of Public Relations.
Del Vicario, M., Bessi, A., Zollo, F., Petroni, F., Antonio, S., Caldarelli, G., Stanley, H. E., & Quattrociocchi, W. (2016). The spreading of misinformation online. Proceedings of the National Academy of Sciences of the United States of America, 113(3), 554-559. doi: 10.1073/pnas.1517441113
Ihlen, Ø., & Pallas, J. (2014). Mediatization of corporations. In K. Lundby (Ed.), Handbook on mediatization of communication (pp. 423–442). De Gruyter Mouton.
Jang, S. M., and Kim, J. K. (2018). Third person effects of fake news: Fake news regulation and media literacy interventions. Computers in Human Behavior, 80, 295-302. doi:
10.1016/j.chb.2017.11.034
Featherman, M. S., Miyazaki, A. D., and Sprott, D. E. (2010). Reducing online privacy risk to facilitate e‐service adoption: the influence of perceived ease of use and corporate
credibility. Journal of Services Marketing, 24(3), 219-229. doi: 10.1108/08876041011040622
Figueira, A. and Oliveira, L. (2017). The current state of fake news: challenges and opportunities. Procedia Computer Science, 121, 817-825. doi:
10.1016/j.procs.2017.11.106
Findley Musgrove, C., Choi, P., and Cox, K. C. (2018). Consumer Perceptions of Green Marketing Claims: An Examination of the Relationships with Type of Claim and Corporate Credibility. Services Marketing Quarterly, 39(4), 277-292. doi:
10.1080/15332969.2018.1514794
Gordon, L. and Shapiro, A. (2012). Priming correct information reduces the misinformation effect. Memory & Cognition, 40(5). 717-726. doi: 10.3758/s13421-012-0191-7 Gotsi, M., Wilson, A. M. (2001). Corporate reputation: seeking a definition. Corporate
Communications: An International Journal, 6(1), 24-30. doi: 10.1108/13563280110381189
Kellner, D., and Share, J. (2005). Towards Critical Media Literacy: core concepts, debates, organizations, and policy. Discourse: Studies in the Cultural Politics of Education, 26(3) 369-386. doi: 10.1080/01596300500200169
Koltay, T. (2011). The media and the literacies: media literacy, information literacy, digital literacy. Media, Culture & Society, 33(2), 211-221. doi: 10.1177/0163443710393382 Kroll, A. (2018, December 9). John Podesta is Ready to Talk About Pizzagate. Rolling Stone.
Retrieved from: https://www.rollingstone.com/politics/politics-features/john-podesta-pizzagate-766489/
Lewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz, N., and Cook, J. (2012). Misinformation and Its Correction: Continued Influence and Successful Debiasing. Psychological Science in the Public Interest, 13(3), 106-131. doi:
10.1177/1529100612451018
Livingstone, S. (2004). Media Literacy and the Challenge of New Information and Communication Technologies. The Communication Review, 7(1), 3-14. doi: 10.1080/10714420490280152
Maksl, A., Ashley, S., & Craft, S. (2015). Measuring News Media Literacy. The National Association for Media Literacy Education’s Journal of Media Literacy Education, 6(3), 29-45.
Maksl, A., Craft, S., Ashley, S., and Miller, D. (2016). The Usefulness of a News Media Literacy Measure in Evaluating a News Literacy Curriculum. Journalism & Mass Communication Educator, 72(2) 228–24. doi: 10.1177/1077695816651970
Marwitz, S., Maxson, N., Koch, B., Aukerman, T., Cassidy, J., and Belongerf, D. (2008). Corporate crisis management: Managing a major crisis in a chemical facility. Journal of Hazardous Materials, 159(1), 92-104. doi: 10.1016/j.jhazmat.2007.07.126
Matuleviciene, M. and Stravinskiene, J. (2015). The Importance of Stakeholders for Corporate Reputation. Inzinerine Ekonomika-Engineering Economics, 26(1), 75–83. doi:
10.5755/j01.ee.26.1.6921
Newell, S. J. and Goldsmith, R. E. (2001). The development of a scale to measure perceived corporate credibility. Journal of Business Research, 52(3), 235-247. doi: 10.1016/S0148-2963(99)00104-6
Puncheva‐Michelotti, P., and Michelotti, M. (2010). The role of the stakeholder perspective in measuring corporate reputation. Marketing Intelligence & Planning, 28(3), 249-274. doi: 10.1108/02634501011041417
Ruddick, G. (2017, August 9). Four UK news sources among top 10 most trusted in US – survey. The Guardian. Retrieved from: https://www.theguardian.com/media/2017/aug/09/four-uk-news-sources-among-top-10-most-trusted-in-us-survey
Shin, J., Jian, L., Driscoll, K., and Bar, F. (2018). The diffusion of misinformation on social media: Temporal pattern, message, and source. Computers in Human Behavior, 83, 278-287. doi: 10.1016/j.chb.2018.02.008
Tandoc, E. C. (2019). Tell me who your sources are: Perceptions of news credibility on social media. Journalism Practice, 13(2), 178-190.
Tandoc, E. C., Wei Lim, Z., & Ling, R. (2018). Defining "Fake News". Digital Journalism, 6(2), 137-153. doi: /10.1080/21670811.2017.1360143
Tat Keh, H., and Xie, Y. (2009). Corporate reputation and customer behavioral intentions: The roles of trust, identification and commitment. Industrial Marketing Management, 38(7), 732-742. doi: 10.1016/j.indmarman.2008.02.005
Thorson, E. (2016). Belief Echoes: The Persistent Effects of Corrected Misinformation. Political Communication, 33(3), 460-480. doi: 10.1080/10584609.2015.1102187
Thorson, E., and Southwell, B. (2015). The Prevalence, Consequence, and Remedy of
Misinformation in Mass Media Systems. Journal of Communication, 65(4), 589-585. doi: 10.1111/jcom.12168
Tsfati, Y. (2003). Does Audience Skepticism of the Media Matter in Agenda Setting? Journal of Broadcasting & Electronic Media, 47(2), 157-176. doi: 10.1207/s15506878jobem4702_1
Tsfati, Y. (2010). Online News Exposure and Trust in the Mainstream Media: Exploring Possible Associations. American Behavioral Scientist, 54(1), 22-42. doi:
10.1177/0002764210376309
van der Meer, T. & Jin, Y. (2019). Seeking Formula for Misinformation Treatment in Public Health Crises: The Effects of Corrective Information Type and Source, Health Communication. doi: 10.1080/10410236.2019.1573295
Vraga, E., Tully, M., Kotcher, J., Smithson, A., and Broeckelman-Post, M. (2015). A Multi-Dimensional Approach to Measuring News Media Literacy. Journal of Media Literacy Education 7(3), 41 – 53.
Wischhover, C. (2018). Johnson & Johnson accused of hiding asbestos in its baby powder for decades. Vox. Retrieved from:
https://www.vox.com/the- goods/2018/12/14/18141265/johnson-johnson-talc-asbestos-lawsuits-cover-up-stock-price
Wade, P. (2019, January 26). Someone Started a Fire at Comet Ping Pong, the Infamous Pizzagate Restaurant. Rolling Stone. Retrieved from:
https://www.rollingstone.com/politics/politics-news/comet-ping-pong-fire-785020/ Waisbord, S. (2018). Truth is What Happens to News. Journalism Studies, 19(3), 1866-1878.
doi: 10.1080/1461670X.2018.1492881
Young, Z. (2018, November 21). How anti-vax went viral. Politico. Retrieved from: https://www.politico.eu/article/how-anti-vax-went-viral/
Appendices
Appendix A1
Appendix A2
Reuters Article on Daisy & Co. Crisis
Appendix B1
Daisy & Co. Detailed Statement
.2. Daisy & Co.
onAppendix B2
Appendix C
Questions on Media Literacy
On a scale of 1 to 7 ranging from strongly disagree to strongly agree, please indicate the extent to which you agree with the following statements.
1. I am confident in my ability to judge the quality of news 2. I have the skills to interpret news messages
3. I am in control of the information I get from the news media
4. I am often confused about the quality of news and information (reverse coded and excluded)
5. If I pay attention to different source of news, I can avoid being misinformed 6. I know how to recognize false or misleading information in media
Appendix D
Questions on Media Scepticism
On a scale of 1 to 7 ranging from strongly disagree to strongly agree, please indicate the extent to which you agree with the following statements.
1. I think the news media are fair (reverse coded)
2. I think the news media tell the whole story (reverse coded) 3. I think the news media are accurate (reverse coded)
4. I do not think the news media can be trusted
5. I think the news media prioritize being first to report a story (excluded)
6. I think the news media get in the way of society solving its problems (excluded) 7.
Appendix E
Questions on Corporate Reputation
On a scale of 1 to 7 ranging from strongly disagree to strongly agree, please indicate the extent to which you agree with the following statements.
1. The company provides excellent value to its customers (excluded) 2. The company offers high-quality products
3. The company stands behind its products
5. The company cares about its customers
6. The company is honest and straightforward in its communication 7. I have a good feeling about this company
Appendix F
Questions on Corporate Credibility
On a scale of 1 to 7 ranging from strongly disagree to strongly agree, please indicate the extent to which you agree with the following statements.
1. The company is honest 2. The company is reliable
3. The company is insincere (reverse coded) 4. The company is trustworthy
