Gives you wings or not? Exploring the impact of viewers’ responsibility attribution and surprise on their attitude, identification and trust

Abstract

Purpose

The purpose of this paper is to explore the impact of critical, journalistic documentaries on viewers. More precisely, it investigates the effects of responsibility attribution and surprise on stakeholder attitude, trust and identification.

Design/methodology/approach

In a quasi-experimental pre-post setting, 127 participants viewed a documentary about Austrian beverage and marketing company Red Bull. The film inquired into the deaths of six extreme athletes sponsored by the company. As a critical, investigative piece, the documentary was designed to give viewers the impression that Red Bull was, at least partially, responsible for the athletes’ deaths.

Findings

Results show that responsibility attribution, the feeling of being surprised and being in a state of negative affect, had a significant impact on viewers’ attitude and trust toward, and identification with Red Bull.

Originality/value

The study adds insights on surprise as a factor in viewers’ assessment of responsibility. The study is original in terms of methodology by using real-time rating to ascertain which sequences trigger changes in responsibility attribution among viewers. Furthermore, implications of the study’s findings with regard to inoculation theory are discussed.

Citation

Seiffert-Brockmann, J., Ruppel, C. and Einwiller, S. (2020), "Gives you wings or not? Exploring the impact of viewers’ responsibility attribution and surprise on their attitude, identification and trust", Corporate Communications: An International Journal, Vol. 25 No. 1, pp. 113-127. https://doi.org/10.1108/CCIJ-07-2019-0087

Introduction

In today’s competitive markets, trust, a positive attitude toward brands and stakeholder-company identification are invaluable assets for companies. Trust, attitude and identification can play a mitigating role by softening the impact of a crisis on a company. Therefore, companies are eager to safeguard their public perception when they come under public scrutiny due to (alleged) corporate misconduct. Filmmaker Michael Moore was among the first to popularize the genre of critical, corporate-bashing documentaries, which due to their “pervasive negative portrayals are consistent with [an] operational definition of crisis” (Pompper and Higgins, 2007, p. 429). As organizational stakeholders see these documentaries, they might attribute greater crisis responsibility toward an organization, the effect being “that increased attributions of crisis responsibility by stakeholders produce lower reputational scores among those same stakeholders” (Coombs, 2007, p. 166). Films can create impressions among viewers that the portrayed company bears responsibility for certain events. This can, in turn, threaten the company’s reputation and brand(s).

Austrian beverage producer and marketing company Red Bull is a noted example. In 2013, Red Bull came under public scrutiny after a series of deaths of athletes sponsored by Red Bull. On April 29, 2013, the German television network ARD aired a documentary titled “The Dark Side of Red Bull” (Büchel, 2013), in which the deaths of six extreme athletes who were sponsored by Red Bull were portrayed. The documentary voiced the hypothesis that Red Bull was partially to blame for the athletes’ lethal accidents, as a result of the company’s branded entertainment marketing strategy which relies on ever-more-spectacular images and dangerous stunts. Red Bull, apart from its public condolence statements, was portrayed as staying silent in the wake of the incidents. For viewers, this seemingly created the impression of a preventable crisis (Coombs, 2007), with Red Bull bearing partial responsibility.

Using the example of Red Bull, the purpose of this study is to research how viewers’ perception of crisis responsibility, while watching a critical documentary, leads to negative outcomes for the organization. Thus, the violation of existing expectations is a key in this situation. For viewers to change their minds about a company, they should be surprised to learn something about the company they did not know before. For that change to be negative, they should also be in a state of negative affect.

To that end, research was guided by the following two research questions:

RQ1.

What is the impact of viewers’ attribution of responsibility on trust and attitude toward, and identification with, Red Bull?

RQ2.

What is the impact of viewers’ surprise and negative affect on responsibility attribution, trust, attitude and identification?

Documentaries are complex pieces that are produced to provide viewers with an argument. Assuming that the filmmakers wanted to convince their audience to change their stance toward Red Bull, it is interesting to establish which parts of the film made them attribute responsibility and to whom. Therefore, the third and fourth research questions were:

RQ3.

What sequences made viewers attribute responsibility toward Red Bull?

RQ4.

What sequences made viewers attribute responsibility toward the individual athletes?

To answer these questions, a quasi-experimental study, embedded in a pre/post-survey design was conducted. Using questionnaires, brand trust, brand attitude and stakeholder-company identification were measured before and, to include surprise, after reception of the documentary. Applying real-time rating (RTR), responsibility attribution was measured to evaluate what sequences preceded viewers’ responsibility attribution toward either Red Bull or the athletes. The results provide insights on how responsibility attribution and violations of expectations impact the aforementioned outcome variables. The paper covers new ground in showing that surprise and negative affect experienced due to the perception of new information has an impact on viewers’ trust and attitude. The consequences of these findings with regard to potential inoculation strategies are discussed.

Conceptual framework

Crisis responsibility and trust are inextricably linked. Many see trust as essential for organizations to maintain their license to operate (Coombs and Holladay, 2014). The impression that an organization could be responsible for a crisis may entail stakeholders’ losing trust, changing their attitude toward and/or identifying less with that organization. Essentially, losing trust, changing attitude and diminished identity are rooted in the violation of expectations (Luhmann, 2000). Stakeholders expect organizations to behave in a certain way and are disappointed if they do not. Disappointment, i.e. being surprised that behavior is different than expected, and crisis responsibility attribution are key variables to understand the outcomes of the perception of critical documentaries among viewers, i.e. changes in attitude, trust and identification. In combining the aspects of responsibility attribution during a crisis, trust, attitude and identification toward and with a company and negative affect and surprise, this study provides insights on how such critical documentaries affect viewers and thus stakeholders.

Attribution of responsibility in crises

Crisis communication is one of the most prominent fields of research in public relations due to its high practical relevance (Verhoeven et al., 2014), with situational crisis communication theory (SCCT) being the primary theoretical framework (Avery et al., 2010). Its core idea is the application of distinct crisis response strategies aligned to the level of perceived crisis responsibility to mitigate or repair damages to an organization’s reputation (Coombs, 2007). If the reputational threat during a crisis is not addressed, organizational-public interactions may be influenced negatively (Ma and Zhan, 2016) as a result.

According to Coombs (2007), “crisis responsibility is a function of stakeholder attributions of personal control for the crisis by the organization – how much stakeholders believe organizational actions caused the crisis” (p. 166). Therefore, the higher the attribution of crisis responsibility, the higher the reputational threat to the organization (Coombs, 1998; Ma and Zhan, 2016). Lee (2005) suggested that “consumers’ causal attribution would influence their judgement of the organization’s responsibility in a crisis, which in turn would affect their cognitive, affective, and behavioral responses toward the organization” (p. 367). In her study on consumer reactions to a crisis, Lee was able to demonstrate that the judgement of crisis responsibility entails a diminished impression of the organization as well as a decrease in trust, which, in turn, influenced purchasing intentions (Lee, 2005). An and Gower (2009) have shown that responsibility attribution is the frame most likely applied by media during crisis coverage, especially when covering a preventable crisis. According to the authors, “news coverage was more likely to emphasize the organizational level of responsibility than the individual level” (An and Gower, 2009, p. 111). Therefore, viewers of critical documentaries should be more inclined to associate responsibility with the organization in question than with the persons representing it on screen. Furthermore, negative publicity in the wake of a crisis has the potential to tarnish and damage corporate image (Dean, 2004). Exposure to critical coverage makes viewers more likely to change their responsibility attribution according to the content of the coverage – with, potentially, negative outcomes for the organization.

Trust, attitude and identification in crises

While SCCT does not explicitly mention trust as a variable, it nevertheless is essential as an antecedent to behavioral intentions. After all, decisions to trust or not trust are based on a trustees’ reputation. For the purpose of this study, the definition of Sirdeshmukh et al. (2002) is used; they define “consumer trust as the expectations held by the consumer that the service provider is dependable and can be relied on to deliver on its promises” (p. 17). In a broader sense, as Luhmann (2000) argued, trust is the expectation that things will unfold in the future in a specific way. However, there is always the risk involved that expectation “may lapse into disappointment” (Luhmann, 2000, p. 98).

Hence, dealing with crises is, to a certain extent, dealing with violated expectations among stakeholders to maintain or regain their trust (Kim et al., 2009). When looking at the relationship between response strategy and trust, many studies suggest that the active attempt to regain trust by means of crisis communication has an overall positive impact on trust and the endangered brand. Park (2017) was able to demonstrate that strategic silence – the opposite approach – results in a negative influence on consumer trust, attitude, reputation and behavioral intentions toward the company in question. This supports Kim et al.’s (2009) rationale that silence is no optimal strategy for trust repair.

Moreover, studies by several authors (Claeys et al., 2010; Dean, 2004; Kim et al., 2004) lend support to the hypothesis that apology strategies and rebuild strategies yield more positive outcomes compared to a no-comment strategy. Concurrently, Chung and Lee (2017) reported apologies oriented toward responsibility softened public anger, distrust and negative impressions. As Leonidou et al. (2013) suggest, “consumer trust is partly rooted in ethical considerations pertaining to the firm’s marketing activities” (p. 526), i.e. the perception of a company’s conduct from a moral point of view should have an impact on how stakeholders place trust in that company.

When viewers watch a documentary on a critical issue, they rely on a working model of organizational crisis response strategies. This model embodies their expectations of how the organization should behave. A credible, critical narrative in a documentary might challenge these expectations and, thus, cast doubt on whether organizational action during the crisis was appropriate. If viewers find themselves surprised by what they see, a negative impact should be expected.

In the same way as trust, attitude affects behavioral intentions (McDonald et al., 2010). Attitude is shaped by crisis information available (Lu and Huang, 2018) and the way companies act during a crisis (Shim and Yang, 2016). Following Spears and Singh (2004), attitude is defined as “a relatively enduring, unidimensional summary evaluation of the brand that presumably energizes behavior” (p. 55). This definition can be applied in the present case since Red Bull has a monolithic brand structure, i.e. it is a company and a consumer brand at the same time.

Trust and attitude fuel stakeholder–organization identification. Identification means individuals feel a sense of connectedness to an entity, like a brand or a company, and the degree to which aspects of its perceived identity are self-defining (Einwiller et al., 2006). For consumers, the “identification process has a significant impact on individual consumer behavior” (Tuškej et al., 2013, p. 53). This includes, among others, an impact on their decisions (Ahearne et al., 2005), loyalty (Bhattacharya et al., 1995) and commitment (Casaló et al., 2008).

By pointing out inconsistencies between the purported identity and a company’s conduct, journalists, watchdogs and/or critical stakeholders can attack companies and their brands. The attack on Red Bull is best captured by the full title of the documentary: “The Dark Side of Red Bull – When Red Bull does not give you Wings” – the title being the direct negation of Red Bull’s slogan, “Gives you Wings.” It is thus justified to assume that the documentary represented the deliberate attempt to sway viewers to reconsider their stance toward Red Bull and change their behavior.

Changes in behavior may include a more critical stance toward the company, a switch to other brands, or even an active, public boycott of the company. For the purpose of this study, the intention of spreading negative word-of-mouth (NWOM, e.g. Einwiller et al., 2017) was chosen as an indicator of viewers’ willingness to change their behavior. However, to achieve changes in viewers’ trust, attitude, identification and behavior, their expectations need to be violated by what they see to trigger a re-evaluation process.

Violations of expectations: the role of affect and surprise in crises

In a crisis situation, organizations need to fulfill public expectations to secure general public support (Hwang and Cameron, 2008). It is suggested that emotions are an important factor in this process. So far, emotions play a minor role in the crisis communication literature (e.g. Coombs and Holladay, 2005; Kim and Cameron, 2011; Kim and Niederdeppe, 2013). Learning about the tragic deaths of athletes might come as a surprise to viewers, affect them emotionally, and might thus trigger a re-evaluation of their stance toward Red Bull.

Regarding emotions in general, Lu and Huang (2018) assert, “crisis communication scholars tend to ignore the possibility that emotion plays as strong a role as rationality” (p. 98) in the assessment of crisis situations. Cho and Gower (2006) found individuals’ emotional response to a crisis to be a predictor of responsibility and blame. Jin (2009) reported emotions to be influential “on publics’ coping strategy preference” (p. 310). From an organizational point of view, Van der Meer and Verhoeven (2014) demonstrated that emotional signals embedded in corporate messages had a positive effect on organizational reputation. Botha (2014) reported emotions, particularly their intensity, to be a key influencing factor on whether content is liked or shared. In a similar fashion Jin et al. (2016) showed that emotions are significant predictors of communication behavior, such as information seeking and sharing. Hence, negative affect is one way to energize viewers.

However, for viewers to change their mind, another factor needs to be considered: surprise. While a video can elicit, in this case, negative, emotions, this does not necessarily entail a change of opinion. Red Bull is a well-known brand/company, and most people have a well-founded opinion about Red Bull. To reassess one’s stance while and after watching a documentary, it seems that the violation of expectations plays an important role as well, i.e. viewers need to be surprised by what they see.

Surprise has hardly been a feature of SCCT research, in the sense that crises are usually surprising events (Lyon and Cameron, 2004). In contrast, surprise as a variable on the receiving end has only rarely been considered so far. Choi and Lin (2009) found surprise to be associated with crisis responsibility, and Kim and Cameron (2011) pointed out the importance of emotions in crises situations, with surprise being the dominant emotion in their experimental study. In that view, negative affect is the fuel for a viewer’s reassessment process, and surprise serves as a trigger. Critical documentaries can take viewers aback, even throw them into a state of shock. That makes these films powerful in public discourse and a threat to trust and attitude toward, and identification with, companies in times of crises.

Methodology

Procedure and participants

To answer the research questions, a pre-post study was conducted among students of an Austrian university. In the study, an 18-min-long cut version (uncut 45 min) of the original documentary “The Dark Side of Red Bull” (Büchel, 2013) served as the stimulus prior to the post-measurement. As the strongest and best-known Austrian brand (Hrebicek, 2018) Red Bull represented an interesting and appropriate case to study among Austria-based participants. Furthermore, using a student sample was justified since young consumers are the main target group for Red Bull (Albrecht, 2018).

Two weeks before watching the stimulus, participants were surveyed on the central variables of attitude, trust and identification. This pre-measurement was a part of a general study on opinions of beverage manufacturers in the Austrian market (including Coca-Cola and Makava, a producer of organic beverages in Austria) to avoid potential priming effects. Students were asked to participate in a follow-up study at the end of the survey. Bonus points were awarded for participation without any preconditions. For the second part, participants were invited to a screening of the documentary. Before each session, viewers were advised on the disturbing content of the film, which contained, among others, images of people dying. Thus, participants were asked to give their informed consent. Aborting the study would have had no impact on the incentive; however, all participants agreed to proceed. While watching the documentary, viewers were instructed to attribute responsibility for the deaths of the featured athletes by using a RTR device (Leiner and Fahr, 2016).

With regard to RQ3 and RQ4, a content analysis of the stimulus was conducted to ascertain which sections of the film caused changes in viewers’ responsibility attribution. The stimulus contained 63 sequences belonging to six distinguishable content categories:

• Cat1 – Athletes: all sequences that portrayed the deceased athletes (18 sequences; combined length: 345 s).

• Cat2 – Expert interviews: all sequences that contained statements given by experts on either the incidents or Red Bull’s overall marketing strategy (9 seq; 207 s).

• Cat3 – Red Bull imagery: all sequences where Red Bull products were shown during the documentary (6 seq; 106 s).

• Cat4 – Friends and relatives: all sequences where friends or relatives of the deceased were interviewed (13 seq; 205 s).

• Cat5 – Accidents: all sequences that showed accidents and/or the circumstances leading to them (11 seq; 152 s).

• Cat6 – Organizational communication: all sequences that contained reactions or statements regarding the events or reflected Red Bull’s communication and marketing strategy (5 seq; 69 s).

Immediately after watching the stimulus, the post-survey was conducted. In total, 127 students participated in the study. None had seen the documentary before. The sample was predominantly female (88.2 percent), and the average age was 22.57 (SD=3.21). In the pre-study, participants showed, on average, a positive attitude toward Red Bull (M=4.67, SD=1.35). Similarly, trust in Red Bull was also reported to be positive (M=4.97, SD=1.13). However, identification was very low (M=2.42, SD=1.21).

Measurements

In the pre- and post-measurement, participants’ general attitude toward the company was gauged using four seven-point semantic differentials taken from Spears and Singh (2004): bad (1)–good (7), unpleasant–pleasant, unappealing–appealing, unlikable–likable. The four items were averaged to form an attitude index (pre: M=4.67, SD=1.35, α=0.89; post: M=3.50, SD=1.43, α=0.72). Trust toward the company was measured with five seven-point semantic differentials from Sirdeshmukh et al. (2002): undependable (1)–dependable (7), incompetent–competent, of low integrity–of high integrity, unresponsive to customers–responsive to customers, not trustworthy–trustworthy. The five items were averaged to form a trust index (pre: M=4.97, SD=1.13, α=0.85; post: M=3.67, SD=1.15, α=0.80). Participants indicated their identification with the company on an eight-item scale, taken from Einwiller et al. (2006), including the following statements:

• “I am somewhat associated with Red Bull”;

• “I have a sense of connection with Red Bull”;

• “I consider myself as belonging to the group of people who are in favor of Red Bull”;

• “Customers of Red Bull are probably similar to me”;

• “Employees of Red Bull are probably similar to me”;

• “Red Bull shares my values”;

• “Being a customer of Red Bull is part of my sense of who I am”; and

• “Purchasing Red Bull’s mutual funds would help me express my identity.”

Responses to each statement were measured on a seven-point scale (1=strongly disagree; 7=strongly agree) and averaged to obtain an index for customer–company identification (pre: M=2.42, SD=1.21, α=0.91; post: M=2.13, SD=0.91, α=0.85). Participants’ negative affect was estimated using six items taken from the German version of the PANAS (Janke and Glöckner-Rist, 2012): hostile, irritated, upset, distressed, scared and afraid. For each item, participants indicated to what extent they experienced the stated condition after seeing the documentary, using a seven-point scale ranging from “not at all” (1) to “extremely” (7). The six items were averaged to form a negative affect index (M=4.37, SD=1.04, α=0.73). To evaluate the participants’ level of surprise, they were asked to answer the question, “How much were you surprised to learn about the death of the athletes?” (1=not surprised at all; 7=very surprised, M=3.99, SD=2.10).

Furthermore, participants’ intention to spread NWOM was measured with one item taken from Wolter et al. (2016): “I will say negative things about Red Bull to other people” (1=highly unlikely; 7=most likely, M=3.65, SD=1.77).

Finally, responsibility attribution was measured during the reception of the stimulus using a mobile RTR (Seiffert-Brockmann and Jarolimek, 2017) application for tablets (Leiner and Fahr, 2016). Participants were advised to use the device as follows:

While watching, please evaluate continuously who, in your opinion, is responsible for the death of the respective athlete. The more you think Red Bull is responsible, the more you move the switch towards 100. The more you think the individual athlete is responsible, the more you move the switch towards 0.

(0 – sole responsibility athlete, 100 – sole responsibility Red Bull).

To determine which sequences were succeeded by changes in responsibility attribution, the mean attribution scores of all participants were averaged for every sequence. Then, the change in the mean attribution score was calculated using the formula for gradients:

m = Δ y Δ x = y2 − y1 x2 − x1 .

With x being the timestamp in seconds at the end of a sequence and y the mean responsibility attribution score. A positive value indicated a shift in responsibility attribution toward Red Bull; a negative value indicated a shift toward the individual athletes. The five biggest swings toward each end of the scale were analyzed to determine what content preceded changes in responsibility attribution among viewers.

Results

To test the effects of the stimulus on participants’ attitude, trust and identification toward Red Bull, three mixed repeated-measures ANOVAs were performed. The analyses included as within-subjects variable the pre-post measurements of attitude or trust toward or identification with the company and two between-subjects factors as moderators: participants’ attribution of responsibility and participants’ perceived surprise concerning the portrayed deaths. For the first factor, participants were categorized based upon their averaged RTR responsibility attributions – either into participants attributing responsibility stronger to the companies (>50, n=72) or participants attributing responsibility stronger to the athletes (n=55). For the second factor, participants were categorized based on a median split (median=4.00) into surprised (n=63) and not surprised (n=64).

As was to be expected based on the average attitude in the pre- (M=4.67) and post-measurement (M=3.50), ANOVA results showed a significant main effect of the film (F(1, 123)=80.10, p<0.001). This negative effect on attitude toward the company was stronger for participants attributing responsibility more to the company compared to participants attributing responsibility to the athletes (see Figure 1(a)), as indicated by a significant interaction effect F(1, 123)=15.05, p<0.001. The interaction effect between participants’ surprise and the repeated attitude measurement failed to reach significance, F(1, 123)=2.54, p=0.113 (see Figure 1(b)).

ANOVA results also show a significant main effect for the decrease in trust between the pre- (M=4.97) and the post-measurement (M=3.67; F(1, 123)=153.54, p<0.001). As indicated by significant interaction effects, the more the participants attributed responsibility to Red Bull (F(1, 123)=16.53, p<0.001), and the more they reported being surprised (F(1, 123)=13.27, p<0.001), the stronger was their reported decline in trust (see Figure 2(a) and (b)).

Although not as distinct as for attitude and trust toward the company, the ANOVA results also show a significant main effect for the decrease in customer–company identification between the pre-measurement (M=2.42) and the post-measurement (M=2.13; F(1, 123)=14.03, p<0.001). This effect was again stronger for those who attributed responsibility to the company as for those who attributed responsibility to the athletes, as indicated by a significant interaction effect between the repeated identification measurement and participants’ responsibility attributions (F(1, 123)=5.26, p<0.05). Participants’ surprise, however, did not have an impact on the negative effect of the critical coverage on customer–company identification (F(1, 123)=0.68, p=0.411).

Concerning viewers’ negative affect after the reception, both their responsibility attributions and their surprise about the portrayed deaths of the athletes sponsored by the company had a significant effect. Participants who attributed responsibility to the company (M=4.74) showed more negative affect compared to participants who attributed responsibility to the athletes (M=3.89; F(1, 125)=24.99, p<0.001). Surprised participants (M=4.75) reported significantly stronger negative emotions compared to those who were not surprised (M=4.00; F(1, 125)=18.77, p<0.001).

Concerning participants’ intention to spread NWOM after viewing the documentary, again both their responsibility attributions and reported surprise had a significant effect. Participants who attributed responsibility to the company (M=4.24) were more eager to speak negatively about the company than vice versa (M=2.87; F(1, 125)=21.62, p<0.001). Participants who were surprised (M=4.03) showed significantly stronger intention to spread NWOM compared to those who reported no surprise (M=3.27; F(1, 125)=6.22, p<0.05).

To answer research questions three and four, the changes in responsibility attribution were analyzed sequence by sequence (see Figure 3; 22 shifts to athletes, M_m=−0.158; 23 to Red Bull, M_m=0.164; 18 sequences with no shift, m=0).

Category six (organizational communication) shows the biggest average shift to Red Bull per sequence (M_m=0.22), while sequences in category two (expert interviews) account for the biggest shift toward athletes (M_m=−0.07; see Table I).

The five biggest shifts toward the athletes contained two sequences of accidents and the surrounding circumstances (Cat5), two sequences with expert statements on an accident (Cat2), and one sequence that portrayed a close friend of the deceased athlete (Cat4). The two accident sequences were a part of a larger portrayal of proximity flyer Eli Thompson, who died during a fly-by in Switzerland. The first sequence (A4, m=−0.3, t=10 s) described the circumstances of his death in detail. The second sequence (A5, m=−0.375, t=16 s) showed lethal proximity flights not related to Red Bull, but was embedded in the larger portrayal of the deceased athlete. The preceding and succeeding sequences to A4 and A5 showed no particularly strong shift toward athlete responsibility, but it is noteworthy that they provided context that might have been influential in viewers’ decision to attribute responsibility. Sequence A4 was embedded in an expert interview with the first responder at Thompson’s crash site, who described the event. Sequence A5 was embedded in an interview with the athlete’s widow, whom the narrator reports was eight months pregnant when the accident happened.

The other three sequences show BASE jumper Ueli Gegenschatz, who was fatally injured during a jump from a high-rise building in Zurich in 2009. Two of these sequences (A1, m=−0.25, t=4 s, at 16 s; A2, m=−0.33, t=3 s, at 25 s) feature an expert who witnessed the accident, and the third shows a close friend of the deceased who mourns his death (A3, m=−0.375, t=8 s, at 37 s). All three sequences are embedded in footage of the fatal BASE jump.

Interestingly though, these three sequences of responsibility attribution to the athlete are succeeded by two sequences that account for the biggest and second biggest shift toward Red Bull. The first sequence (RB1, m=0.45, t=11 s, at 45 s) shows footage of the preparations before the stunt. The narrator says the jump was a promotional event organized by Red Bull (Cat6). The sequence is followed by a short expert soundbite (m=0.167, t=6 s; not among the top five sequences in terms of attribution change), decrying the “perversion of event marketing,” followed by the title of the documentary (RB2, m=0.25, t=12 s, at 1 min 2 s) against the background of footage of the accident (Cat5). According to the real-time data, the participants’ responsibility attribution started to change the moment they became aware that the BASE jump was a marketing event staged by Red Bull. Without the framing, viewers attributed responsibility to the individual athlete. With the frame applied, they seemed to focus more on the organization than the individual.

The third biggest shift toward Red Bull occurred more than 7 min into the stimulus and was content related to sequences RB1 and RB2. At that point, the documentary portrayed Ueli Gegenschatz’s fate in detail and interviewed other BASE jumpers who knew Gegenschatz professionally. The increased responsibility attribution occurred when fellow BASE jumper Douglas McDougall said Gegenschatz had jumped in bad weather conditions, and there was talk about pressure by his sponsor to jump (RB4, m=0.4, t=25 s, at 7 min 43 s). The sequence ends with McDougall stating that Gegenschatz was a very good and safe jumper (Cat2).

The other two sequences are related to the death of extreme skier Shane McConkey, who died in an attempted ski-wingsuit jump in the Dolomites in 2009. Red Bull produced a documentary about McConkey in 2013 titled McConkey. You have one life, live it. The second strongest shift toward Red Bull occurred (RB3, m=0.43, t=7 s, at 3 min 12 s) when the documentary filmmakers asked Red Bull Media House’s Head of Cinema and International Theatrical Sales, Sophokles Tasioulis, to comment on the film. The filmmaker asks Tasioulis if he is okay making a marketing movie about a deceased athlete. Tasioulis then declines any comment and brushes off the camera (Cat6). The second of the two sequences (and fourth biggest shift in responsibility attribution; RB5, m=0.35, t=17 s, at 15 min 29 s) features Shane McConkey’s father, himself a former extreme skier, speaking, hypothetically, about the pressures to perform during a film shooting (Cat2). He is thereby alleging that, without the cameras rolling, there might not have been an accident.

To sum it up: The two biggest shifts in responsibility attribution occur while viewers learn that a fatal BASE jump was a part of a Red Bull marketing event and in a scene where a Red Bull representative declines to comment on a postmortem biopic of an athlete sponsored by the company. The third and fourth biggest shifts occur when experts allege that the sponsor might have exerted pressure on the athletes.

Discussion and limitations

There are three major limitations to this study. First, even though young people in the age group between 18 and 30 are the most important target group for Red Bull, the sample had two disadvantages: women were overrepresented in the sample, and no groups other than college students were surveyed. Thus, the convenience sample has only limited validity, and results need to be interpreted with care. Second, the measurement of identification proved to be difficult. Only a few participants actually identified themselves with Red Bull. Pre-stimulus identification was very low to begin with (M=2.42). Even though the reduction in identification in the post-survey (M=2.13) proved to be significant, a more robust finding would have likely been with a higher number of identified participants. With 13 people (10 percent) having an identification index score of four or higher in the pre-study, the level of identification among surveyed samples is in line with prior research, having found between 12 percent (Wolter et al., 2016) and 14 percent (Bhattacharya and Elsbach, 2002) of respondents as being identified with a brand. Therefore, future studies on the impact of critical coverage on stakeholder–organization identification should use larger samples of 1,000 respondents and more to include a sufficient number of self-identified persons. However, having the setting of the study in mind, especially using RTR, rendered a larger sample size unrealistic. Finally, the length and complexity of the stimulus posed an analytical challenge. While RTR provided second-by-second data to estimate at what point responsibility attribution shifted, it is still unclear what a part of the content caused it. As viewers watch films, they constantly form and re-evaluate their opinions. Therefore, the RTR data likely just show the tipping point of impression formation, not necessarily its cause. Future studies need to focus on more discrete units of media content to test different content categories against each other to determine what content influences opinion in what way.

Having these caveats in mind, the study nevertheless revealed interesting findings. First, the results are in line with prior research on the SCCT. Data show a significant decrease in the levels of trust and attitude toward, and identification with, Red Bull in the follow-up to the documentary. Hence, critical coverage clearly has an effect on relevant variables that influence the stakeholder–organization relationship. Second, the finding that surprise influenced trust and attitude suggests that no particular crisis response strategy would have protected Red Bull. Being created with a critical mindset toward Red Bull, the documentary left little wiggle room for arguments in favor of Red Bull. However, inoculation theory may provide an alternative approach. Compton et al. (2016) suggested that individuals could be inoculated against persuasive messages by conveying challenging views beforehand. In the words of Pfau et al. (1997), receivers of inoculation “strive to strengthen attitudes, employing content provided through refutational preemption in addition to other knowledge. In this manner, inoculation provides a broad umbrella of protection” (p. 188).

While it is hard to inoculate viewers against allegations of potential misconduct voiced by experts (RB4 and RB5), the case is different when it comes to marketing and communication strategy. Some viewers seemed genuinely surprised and shocked to learn Red Bull relies on dangerous stunts as a part of its strategy and the number of lives it cost. Red Bull is notorious for its no-comment approach to such tragic events. Communicating extreme athletes’ occupational hazards as a part of an inoculation strategy could make a difference – and athletes would even agree.

That this could be a viable strategy is indicated in some of the thoughts provided by participants in the post-survey. Different viewers highlighted that people who take extreme risks should also be held accountable. Participant A046 argued, “I think, for the most part, the athlete is to blame.” In a similar manner, respondent A085 said, “there is no way Red Bull is responsible alone.” Making the question of responsibility salient could thus have an impact if communicated as a part of the response strategy. More recently, Einwiller and Johar (2013) were able to demonstrate that inoculation works among critical and disidentified stakeholders. Thus, lowering the potential for surprise through inoculation could have had an effect in this case.

Third, complementary to the findings of Lee and Chung (2012), the lack of a communicative response of Red Bull regarding the incidents raised negative affect among participants if they were surprised. It seems participants felt deprived of information or a statement issued by Red Bull. This was especially apparent in sequence RB3, which accounted for the second biggest shift in responsibility attribution toward Red Bull. Hence, no-response postures fare bad, inoculation or not. Finally, not only was negative affect stronger among the surprised and those who deemed Red Bull being responsible, the intention to spread negative word-of-mouth was also significantly higher.

Conclusion

This study found support for the central claim of the SCCT, that responsibility attribution is the key in crisis perception among stakeholders. Responsibility attribution toward a company in a crisis situation leads to negative outcomes in relevant dimensions such as attitude, trust, identification, and behavioral intention. Critical media content has an impact on stakeholders, and it does influence relevant behavioral outcomes; hence, documentaries targeting organizations present a viable reputational threat.

However, one should also notice that these indicators for changes in trust, attitude, identification and behavioral intention, should be taken with a grain of salt, at least in the case of Red Bull. Despite many athletes’ deaths over the years, the company remains strong in sales and revenue and has done so steadily over the past years (Hrebicek, 2018). It is no secret that Red Bull uses extreme marketing measures to advocate its brand – and many people who consume Red Bull products know that. Moreover, while many viewers feel sorry for the athletes’ relatives and acknowledge their suffering, they also say, “Nevertheless, I will continue to drink Red Bull” (Participant A002). Thus, future research should try to research the effects of content on different segments of stakeholders, e.g. according to their level of identification with a company. This, and hitherto little-regarded factors like surprise, could shed light on the sometimes-contradictory evidence in the field. Finally, inoculation strategies could be tested in experimental settings to determine whether negative outcomes can be mitigated.

Figures

Figure 1

1a (left) and 1b (right)-Effects of responsibility attribution and surprise on attitude towards the company

Figure 2

2a (left) and 2b (right)-Effects of responsibility attribution and surprise on trust towards the company

Figure 3

Mean overall responsibility attribution (left scale; 0 – athlete solely responsible, 100 – Red Bull solely responsible) and change in responsibility attribution from sequence to sequence (right scale). The five biggest changes in responsibility attribution towards Red Bull and the athletes are indicated in the order of their appearance in the video

Table I

Characteristics of content categories in the stimulus

Category	Mean gradient responsibility attribution	No. of seq.	Average length per seq.	Shifts toward Red Bull	Shifts toward athletes
1 – Athletes	Mm =−0.01	18.	19.17 s	6	7
2 – Experts	Mm =−0.07	9	23.00 s	2	7
3 – Red Bull imagery	Mm =0.05	6	17.67 s	2	0
4 – Friends and relatives	Mm =0.01	13	15.77 s	6	5
5 – Accidents	Mm =−0.04	11	13.82 s	3	3
6 – Organizational communication	Mm =0.22	5	13.80 s	4	0