Overcoming the Power of the Voice: addressing the #MeToo issue in media

Master Thesis

Graduate School of Communication Corporate Communication

Overcoming the Power of the Voice: addressing the #MeToo issue in media

Name: Yana Piven ID: 12017914 Date: 28-06-2019

Abstract

High visibility, enabled by social media, allows groups, that used to have little to no opportunity to stand for their rights, to spark major social issues, like the #MeToo campaign. The #MeToo issue is focused around victims of sexual harassment and assault at the working place. Crisis related to the topic like this, often creates uncertainty for the corporate

communications specialists around the approaches to address this issue, while still preserving the most favorable outcome for the company’s reputation. This study is aimed to examine the effects of how the chosen crisis response strategy, combined with the previous crisis history, impacts the media framing of the issue. The media coverage on 14 organizations, in which key executives were accused of sexual harassment, was analyzed with a test using the content analysis method. The research produced a sufficient basis for practical guidelines, which can be used to approach the unfolding crisis and get the most favorable coverage in media. The main findings of the study show that the denial crisis response strategy can outperform the rebuild and diminish strategy in the media framing of the crisis.

Introduction

The growth and all-pervasiveness of social media lead the word of today to the

visibility and transparency, higher than ever before, challenging all the areas of human life as well as the operations of organizations (Treem & Leonardi, 2013). These new affordances create paradoxes and misbalance between the positive and negative effects they bring in, this way creating a new call for communications specialists (Birkinshaw & Cable, 2017). Social media affordances uncover the new dimensions of the social life, melting the work and private domains (Van Zoonen, Verhoeven & Vliegenthart, 2016), providing the opportunity and platform to expose the topics that remained hidden before, this way allowing the groups that used to have a little to no opportunity to stand for their rights.

This new feature of social reality is capable to spark major social issues, like #MeToo, challenging not only the society to address the sensitive topics raised by those who can finally be heard, but also exposing the internal and often unspoken processes of the

organizations. Caputi, Nobles and Ayers (2019) confirmed the concern of the corporate world with the #MeToo issue in their research, stating that the level of concern and attention to addressing and preventing the sexual harassment and assault has substantially grown. The evidence of a search for the methods and ways to address this topic by the means of communications confirms the lack of empirical background and the need to be addressed from the communications perspective (Caputi, Nobles & Ayers, 2019).

This paper is focused on the #MeToo issue, the significance of which can barely be overestimated. The infamous tweet by the Hollywood actress Alyssa Milano, who claimed herself a sexual harassment survivor, jump-started the issue. Her short message, encouraging the other victims to use the hashtag #MeToo and join the movement has gone viral in no time. Since the tweet was published in October 2017, it was retweeted more than 23,000 times and got over 52,000 likes (Milano A., 2017), inspired more than 500,000 other tweets

using the hashtag “#MeToo” (“More than 12M "Me Too" Facebook posts, comments, reactions in 24 hours”, 2017). The escalation of the #MeToo issue has reached an

overwhelming speed, peaking in December 2017, when traditional Times report brought all the silence breakers to the traditional “Person of the Year 2017” spotlight (Zacharek,

Dockterman, & Sweetland Edwards, 2017). The massive media and public attention, that was drawn to the sexual harassment at the workplace could not leave the current state of things unchanged, creating a new challenge for the corporate world: “now what”? (“Now What?”, 2019).

Apart from the social significance, the especially severe cases of the #MeToo issue were proven to have a devastating effect on the corporate domain. The most representative case - the Weinstein Company Inc. was forced into declaring bankruptcy after the severe reputational backfire, proving that #MeToo can harm not only the reputation of the individuals but the company itself, leading to the total loss of its legitimacy (“Company Overview of The Weinstein Company Inc.”, n.d.). The boost of the attention brought to the topic highlights the important aspect of modern society: sexual misconduct at the working environment is lacking the clarity on how to address and prevent it (Caputi, Nobles & Ayers, 2019), the issue is under-communicated and can be extremely harmful for the reputation of the companies involved if handled incorrectly.

Main concepts

The initial idea behind the #MeToo hashtag has outgrown the borders of just a Twitter conversation, gaining force in the offline reality, inducing the new wave of problems for the corporate world from the communications perspective. Organizations, affected by #MeToo-related scandals are facing the complicated dilemma of taking the decision on the strategic approach to address this problem. Scandals involving sexual misconduct can be studied through the prism of the situational crisis communication theory (SCCT) proposed by

Coombs (2007), which implies the urge of organizations to adopt an overarching crisis response strategy (Coombs, 2007) and create the overall tone and direction of the further communications with its stakeholders. Provided the high sensitivity of problems, that often become the base for social issues, these types of crises are usually getting quickly into the media spotlight and as discussed above are proven to be especially harmful to the valuable corporate assets, like the reputation (Coombs, 2007).

Moderation effect

In addition, given that these types of scandals usually involve the high degree of emotions, it's not surprising that the individuals are being “trialed by media” (Zarkov & Davis, 2018). This process means the media posing the certain attitude and providing the moral judgement for their audience, creating social discontent and urge to punish the accused person, which often extrapolates the negative attributions of the audience towards the

company. This way, one of the vital fronts during the potentially damageable events for the organization is taking control over the way media frames the organizations’ messages. The heat can potentially rise even more if the organization ever had similar cases in the past (Coombs, 2004). The repetitive issue within the organization can strongly affect, how the situation is portrayed in the media and how it is framed, drawing even more attention to the fact that the organization did not undertake any actions to take the situation under control. This will mean the company can suffer even bigger reputational losses in case of the unfolding crisis. Taking this into consideration, the ultimate strategy to address the crisis in order to achieve the maximum plausible outcome in media resources is still yet to be defined and is a current and concerning problem for the practitioners and academics to address.

This leads us to the principal research question as follows: how the prior crisis history affects the media framing applied to the different types of corporate crisis response strategies on the viral societal issues?

The lion's share of the current empiric papers does not investigate the #MeToo or other viral issues from the communications perspective and more precisely through the prism of the corporate crisis response strategies, revealing the gap in the academic research. The lack of academic research in this area is posing the challenge for practitioners to build

communications strategies for organizations, as the only tools they have are their own assumptions and little to no research data. Presented paper to elaborate the framework, based on the existing communications theories, available research on #MeToo from the adjacent practices and test it with the quantitative analysis techniques. Content analysis approach to be employed in order to investigate what the crisis response strategy mediated by presence or absence of the crisis history will lead to the most favorable coverage by the media. An analysis is based on the available media pieces covering the organizations, whose top managers were involved in the #MeToo issue, covering the most recent outlook on the problem.

The presented outcomes to become a means to produce the recommendations for the practitioners on possible approaches to mitigating the risks of the unfolding crisis related to the viral societal issue by choosing the most suitable combination of the response strategy and crisis history, that will lead to the most favorable coverage by the media. The research outcomes are expected to be of further use for both – practitioners and academics, providing more insights and possibility to create a new outlook on how to choose the most effective strategy to address the scandals or the hot social topics to minimize the reputational threats posed by the framing of the media coverage. Empirical base, build up as the result of the study to provide a solid base for the further analysis and research, provided academic literature is scarce on the viral societal issues, due to its relatively new and emergent nature.

The viral societal issues, induced by the active involvement of the social activists and new affordances of the social media, such as high visibility, transparency and high speed of the information dissemination, and the #MeToo issue, in particular, are quite a recent topic in the communications science area (Van Zoonen, Verhoeven & Vliegenthart, 2016). The current paper is to present the new approach, created on the basis of the existing research in the crisis communications domain, provided the nature of the #MeToo issue, which is concerned with the scandals based on the sexual misconduct at work, and the research on #MeToo issues itself from the connected practices.

The first concepts, to be employed in the presented study, are based on the Coombs’ (2007) research within the situational crisis communication theory (SCCT) - the crisis itself and the crisis response strategy. The crisis was defined by Coombs as the “perception of an unpredictable event that threatens important expectancies of stakeholders and can seriously impact an organization’s performance and generate negative outcomes” (Youngblood, 2010, p.174). In his crisis communications research, Coombs already pinpointed the importance of the viral societal issues due to the rise of the social activism and significant progress in the communication technology (Youngblood, 2010), which at that point were in their very early stage, starting to gain power and velocity on the social arena.

The viral societal issues can be described as ones disseminated via social media,

induced by the scandal or social hot topic that become viral and have an impact on the offline reality (Poell, 2015). Viral societal issues differ from the issues promoted in the news, as this is not the agenda promoted by media players any more, but the issues that are meaningful for society, this way they can be characterized by a higher degree of unpredictability. The organizational crisis happened in relation to the viral societal issues is harder to control and predict how it would unfold. #MeToo issue is an accurate depiction of how a hot topic supported by social activists and influencers, boosted by the speed and visibility afforded by

new means of communication technology can catalyze the corporate crises, harm the reputation and impact the stakeholder’s attitudes. This rationale allows #MeToo issue to be approached using the SCCT theory.

The SCCT framework, developed by Coombs (2007), provides the grounded approach for the understanding of the organizational crisis and communications around it from the theoretical perspective and provides clear guidelines and recommendations to build the post-crisis communications. The core of the SCCT proposition is that the post-crisis communications approach can be developed, based on several characteristics of the crisis and can be used as a strategy to maximize reputational protection (Coombs, 2007). According to SCCT - in order to protect the reputation and stakeholder’s positive attitude company’s communications have to be cohesive, telling the unite story, in line with the chosen crisis response strategy.

The concept of the crisis response strategy brings the most value for the current study, allowing us to start the framework with the general approach, that the organization have chosen for communications and enable to examine what strategy can be more effective in communicating the #MeToo related crisis specifically to the media. Coombs (2007) divides the crisis response strategy into the primary and secondary. In the current research, only the primary crisis response strategy to be examined provided the focus on the immediate communications of the organization and the media coverage of it.

The primary crisis response strategies can be divided into denial, diminish and rebuilding. Denial strategy includes the two directions of communications: attacking or confronting the accuser, denying the fact that crisis happened, and scapegoating, accusing the third party of anything that happened in the organization (Coombs, 2007). Crisis response within the diminish strategy can be looking for an excuse or justification for the actions of the organization (Coombs, 2007). Finally, the rebuilding strategy implies the compensation to the

victims and apology to the stakeholder, taking all the responsibility for the crisis onto the organization (Coombs, 2007).

Crisis response strategy concept, proposed by Coombs (2007) was criticized for not providing the exhaustive list of strategies that practitioners actually use. For instance, Kim, Avery and Lariscy (2009) argued, that the practitioners don’t use the theory along with the guidelines in the real-life setup. According to their research, more than 90% of the studied sample researchers determined the usage of the mixed crisis response strategies (Kim et al., 2009). The research also revealed that in the cases where the crisis response strategy was used in isolation from other ones it was proven to be ineffective. However, the outcomes of the research also highlight, that many of the organizations are still unable to appropriately choose and combine response strategies, which reveals an existing gap between the academic research and practical implementation. For the current research purposes, we adopt the

methodology proposed by Coombs (2007), provided it represents the overarching approach to the crisis response strategies and the alternative empirically proven its efficiency approach, was not revealed in the presented academic research basis.

The effectiveness of the crisis response strategy applied to the #MeToo crisis can be examined by looking at how media is depicting the crisis, happened within the organization. In order to examine which of the crisis response strategies perform best, we implement the media framing concept into the study as the dependent variable. Framing is the ability to stress on certain events or objects (Scheufele, 2004). Media framing is the ability to influence the audience and cause certain reactions of the society by drawing attention to certain

information while moving the other information to the shadow. Entman (1993) elaborated on the framing concept around the ability of the media to choose certain aspects of the perceived reality and raise or lower its salience along with providing the preferable problem definition, interpretation and moral evaluation with the further suggestions.

This way, frames perform the range of functions, like defining a problem, diagnose causes, make moral judgements and suggest remedies (Entman, 1993). Each of them might be or not be present in the media piece. More importantly, media framing can propose the moral judgements, measure the morality of the actors and the consequences of their actions (Entman, 1993), which can seriously damage the corporate assets or vice versa - recover them. Media framing can make the scandal related to #MeToo more salient by the means of the written text, promoting and emphasizing the problem occurred within the organization to their audience. Moreover, provided the ability of the media framing to influence the audience and propose certain attitude towards the organization which directly impacts corporate reputation, this concept fits well in the study as the indicator of the effectiveness of the crisis response strategy to address the #MeToo related crisis.

Variety of approaches to study the framing can be applied: the major part of the researches aimed to study the framing is using the unique frames – those that are uniquely tailored for the needs of the particular study. This diversity in the framing is getting closer to the point of academical disagreement, being criticized for the lack of integrity, which makes the analysis more complicated and comparison – unreliable (Borah, 2011). The expansion of the frames in the communications studies are not bringing any advancements for the framing theory in general (Borah, 2011).

For the current study, researchers have chosen to not dive into the issue-specific frames and go in line with the already existing approach to media framing. In order to examine how the media framing applies to the different types of corporate crisis response strategies in regards to the viral societal issues, precisely #MeToo, we would use the approach to framing proposed by Kepplinger, Geiss and Siebert (2012), provided it fits well within the crisis communications domain and was tested and proved its efficiency. First, they break down crisis depiction in the media to the focal aspects: crisis is portrayed as “either (a) small or

large, (b) caused or not caused by human behavior, (c) by people following selfish goals or altruistic/common goals, (d) by people who were or were not aware of the consequences of their decisions, and (e) by people who could have acted differently or who were constrained to act the way they did” (Kepplinger et al, 2012, p.661).

This approach, proposed by Kepplinger (2012), can also be critically assessed for reducing the dimension of the frames, as one of the main critical points to the majority of the attempts to put frames into the analysis as proposed by Scheufele (2004). Scheufele (2004) also argues that the most adequate approach to measuring the media framing in the case of studying scandals and controversial issues is value framing. In the case of working with the scandals, it is important to understand how media is positioning the value and importance of the occurred events. However, such an approach has a range of risks within the content analysis approach – it can become too subjective and impact the validity of the results. Provided the complexity of the concept and the restrictions, posed by the chosen method of analysis, the media framing concept in the presented work was based on the characteristics proposed by Kepplinger et al, 2012. The researchers to attribute the frame to the media piece: guilt frame is present if media portrays the conflict through blaming the person who is responsible for the misdeed in line with all five aspects or frame components, at the same time, the excuse frame consists of all the opposite aspects (Kepplinger et al, 2012). In order to add a little more dimension to the framing concept, we would consider complete guilt frame when all five aspects will imply accusing the responsible person and fragmentary guilt frame when only a few of the aspects are present. The same approach was employed for the excuse frame: we would consider complete excuse frame when all five aspects imply justifying the actions of the responsible person and proposing an excuse for the person’s actions to the audience and the fragmentary excuse frame when only a few of the aspects are doing so. The no frame to be included in the study, as the neutral option for the

media coverage, that does not contain any frame but reports the situation happened in the maximum possible objective matter. Having that said, the following hypotheses to be tested:

H1: Denial (H1a) and diminish (H1b) crisis response strategies will lead towards higher usage of the guilt frame in the media coverage.

H2: Rebuilding crisis response strategy will lead towards higher usage of the excuse frame in the media coverage.

The mediator, chosen for this study is prior crisis history, this concept concerns whether the organization was involved in a similar crisis before (Coombs, 2007). Prior crisis history to be operationalized as “present” or “absent”. For the needs of the current research only the crises cases, connected to the harassment on the working place to be included in the analysis. Presumably, the negative prior history can enhance the damage on the reputation of the organization from media perspective which will lead to the more negative judgement, whereas the absence of prior history can be beneficial (Coombs & Holladay, 2011). This leads us to formulate the hypothesis as follows:

H3: Companies with the present and absent prior crisis history will lead to the significantly different media framing of the organizational crisis related to #MeToo.

And the hypothesis to test the interaction effect:

H4: There is a significant difference in the media frame applied to the #MeToo related crisis, dependent on the chosen crisis response strategy and prior crisis history.

The formulated hypotheses along with the key concepts are visually presented in the conceptual framework in Figure 1.

Figure 1. Conceptual framework

Methods

The presented research is focused on the #MeToo issue, as an example of the most recent, representative and impactful viral issues, induced by the acute societal problem. This issue was specifically chosen for the research in order to understand how prior crisis history is affecting the relationships between the crisis communications strategy, crisis history and the media framing. These relationships are most apparently reflected in the media coverage mentioning the occurred crisis, therefore the quantitative content analysis was chosen as the method for the study.

At the moment of conducting the research, the total 263 cases of accusation in the sexual misconduct of the celebrities, politicians, CEOs and others can be referred to the #MeToo issue (North, 2019). Provided the focus of the current research on corporate communications, 14 companies from the business and technology domain were chosen for further analysis. The non-probability purposive sampling technique to collect the articles was employed. On the one hand, this sampling approach has its pitfalls, such as the researcher’s bias, provided the companies and media pieces were chosen to a high extent based on the

researcher’s own judgement. Therefore, there is a high chance of the risk, that some categories might prevail over others, which definitely imply risks to the generalizability of the research. On the other hand, the chosen sampling method includes companies that operate in different sectors of business, which helps to mitigate the generalizability threats. The proposed technique also allowed us to collect the materials that are open for the public domain, eliminating the risks posed by ethical considerations while collecting and processing the data and, at the same time, making the research feasible.

Research unit, employed into the analysis is the media piece, reporting the #MeToo related scandal covering the individual and the corresponding to him/her organization, that was chosen for the research. The sample of the media pieces was systematically retrieved using the NexisUni news search tool, as the trustworthy source (Krippendorff, 2004), using the following model for the search terms: “Company name AND Name Last Name of the accused person”. The search for every organization to be executed separately and sorted by relevance. The research units to be chosen within the timeframe from October 2017 until April 2019, covering the entire lifecycle of the #MeToo issue. Hits were narrowed down to the English media only, newspapers, web-based publications, magazines and journals were chosen in order to eliminate other categories that can be irrelevant for the analysis, provided the corporate communications focus of the research, and the five top results were drawn for analysis. In case the first two media pieces had the same text - the researcher skipped to the next article in the list.

The chosen methodology has some restrictions concerning the number of variables, that can be successfully measured with the content analysis approach. Along with that, provided that the content analysis approach is based on the analysis of the meanings, it traditionally suffers from the higher degree of ambiguity, than other research methods. This pitfall of the methodology to be mediated by the reduction of the ambiguities with the maximum

exhaustiveness of the codebook instructions, that includes some real examples for higher certainty. The research will benefit from adding another dimension: conducting an

experiment of what strategy can perform better, comparing the different combinations of the available crisis response strategies with history. This approach will add up some depth to the understanding of viral societal issues. However, due to the time and resources limit, this is the subject for further research on this matter.

The research conceptual framework consists of the independent variable, dependent variable and the mediation effect and is graphically presented in Figure 1. It is expected that the independent and the dependent variable have the association which is moderated by the prior crisis history. The independent variable of the study is the crisis response strategy, that is operationalized as “denial”, “diminish” and “rebuild” (Coombs, 2007). For the cases where the crisis response strategy is unclear - the option “not applicable” is provided - these options will be excluded from the analysis. The crisis response strategy to be chosen by the

researcher for every particular organization based on media coverage.

The dependent variable is the media framing, operationalized on the five-point scale, representing the spectrum from the “guilt frame” to the “excuse frame” (Kepplinger et al, 2012), applied towards the company. The frame to be determined for each analyzed media piece. The mediator variable is the crisis history, which is represented in the binary mode “present” or “absent” (Coombs, 2007), determined by the researcher on the basis of each particular article - whether this aspect is mentioned in the article or not.

The sample of the media pieces (n = 67), drawn for analysis was coded in accordance with the codebook (Appendix 1). Two experts in communications coded 25% (n = 18) of the sample to test the inter-coder reliability. The inter-coder reliability test, after two rounds of coder training, revealed satisfactory results for the key variables: crisis response strategy (Krippendorff’s Alpha: .89, percentage of agreement: 25%; κ = .9, p < .05), media frame

(Krippendorff’s Alpha: .75, percentage of agreement: 23%; κ = .75, p < .05) and crisis history (Krippendorff’s Alpha: .64, percentage of agreement: 22%; κ = .64, p < .05), which allowed the researcher to proceed with the analysis.

Results

The collected sample articles resulted in 87 observations, provided some of the publications covered more than one organization involved in the study going through the corresponding crisis. Collected sample mostly includes coverage for United States (n = 70), other cases (n = 16) is the coverage in British (n = 5), Indian (n = 3), Swiss (n = 1) or other global (n = 7) media resources. In the sample, most of the companies have chosen the rebuild (n = 44, 51.2%) crisis response strategy (n = 86, Mode = 3), denial (n = 25, 29.1%) was the second most popular strategy and finally, diminish (n = 9, 10.5%) was the least applicable strategy. 67.4% of the media pieces did not mention previous crisis history (n = 58) in the companies that were covered and for 32.6% articles mentioned prior crisis related to harassment on the workplace (n = 28).

Media frames were distributed as follows: guilt frame was applied to 50% of the cases reported (n = 43), fragmentary guilt frame was used for 31.4% cases (n = 27), fragmentary excuse frame was used for 12.8% cases (n = 11) and the full excuse frame was used the least, only for 5.8% cases (n = 5). A major part of the resources depicted crises related to sexual harassment as the large-scale crisis (n = 58, 67.4%) and the rest of the publications tried to depict the crisis as insignificant occurrence (n = 28, 32.6%). After the exploration of the descriptive characteristics of the sample the hypotheses testing was conducted in SPSS, the syntax is available in Appendix 2.

To test the first and second hypotheses the cross-tabulation analysis was conducted based on the level of the dependent and independent variables which were both nominal. Results reflect that the denial (H1a) and diminish crisis response strategy (H1b) were more

often are reported in the media using the guilt frame. However, the organizations who used the denial strategy were more distributed on the spectrum, as can be seen in Table 1. The results revealed that one of the assumptions for both conducted Chi-square was violated and the expected count was below 5. Further analysis of the results revealed a significant

association between the denial crisis response strategy and how the company is framed in media, Chi-squared (3, N = 86) = 18.56, p < .05. Based on the cross-tabulation analysis hypothesis 1a can be partially accepted. The analysis also revealed a non-significant

association between the diminish crisis response strategy and how the company is framed in media, Chi-squared (3, N = 86) = 2.44, p = .49. Based on the cross-tabulation analysis hypothesis 1b can be rejected.

The dependent and independent variables for the second hypothesis were both measured on the nominal level. Cross-tabulation results revealed that the rebuild crisis response strategy was reported in the media using the full guilt frame. The results revealed a non-significant association between the rebuild crisis response strategy and the media framing, Chi-squared (3, N = 86) = 6.22, p = .10. Based on the cross-tabulation analysis hypothesis 2 can be rejected.

Table 1. Denial, diminish and rebuild crisis response strategy and the media frames applied

Guilt frame Fragmentary guilt frame

Fragmentary

excuse frame Excuse frame Total

Denial 32.0% 24.0% 24.0% 20.0% 100%

Diminish 66.7% 33.3% 0.0% 0.0% 100%

Rebuild 56.8% 31.8% 11.4% 0.0% 100%

To test the third hypothesis an independent samples t-test was conducted, examining whether companies with prior similar crisis history scored differently on the media frame

spectrum than the companies without. Levene’s test for equality of variances was

insignificant, which meets the assumptions for the T-Test and confirms the heterogeneity of the results, F = .00, p = .960. Results revealed that companies with the present prior crisis history score lower on the media frame scale, which means they are portrayed more often using the guilt frame (M = 1.68, SD = 1.16) than those who had no prior crisis history (M = 2.05, SD = 1.28), t(84) = -2.64, p = .194, 95% CI [-0.94, 1.94], however, the test proved the difference insignificant. Hereby, we reject the third hypothesis.

To test the fourth hypothesis the factorial analysis of variance (ANOVA) was carried out. We cannot assume equal variances, F (5, 72) = 20.05, p < .05, hereby we proceed with the notion that the test was conducted with the violation of one of the assumptions. Denial strategy more often led to applying the excuse frame (M = 2.76, SD = 1.62) than the rebuild (M= 1.66, SD = .96) and diminish (M= 1.33, SD = .5). A post-hoc test indicated that the only significant difference found was that between the rebuild and denial crisis response strategies (Mdifference = 1.10, p <.001). There was found no significant main effect of the crisis response strategy on the media frame applied, F(2, 72) = .50, p = .607 and no significant main effect of the prior crisis history on the media frame applied, F(1, 72) = 4.85, p = .031. However, the significant interaction effect was revealed of the crisis response strategy and the crisis history on the media frame applied, F(2, 72) = 6.71, p = .002. The graphic depiction of the

interaction can be observed in Figure 2.

Conclusion and Discussion

The aim of this study is to examine, how the prior crisis history can affect the media framing applied to the different types of corporate crisis response strategies on the crises induced by the viral societal issues, related to the #MeToo issue. To address this question the content analysis method was applied, in order to review of the existing media coverage on the #MeToo related issues on organizations from the business and technology domain, whose top-managers were accused in the sexual harassment or misconduct. The total of 14 companies was included in the analysis, covering all the organizations from the corporate sector, whose top management got involved in the scandal connected to the #MeToo issue.

The reviewed literature, chosen for the provided research revealed, that the #MeToo issue itself wasn’t yet in-depth studied through the prism of the communications science. However major problems sparked by this issue are in the communications domain or can be mitigated by the successful usage of communications (Caputi, Nobles & Ayers, 2019). The

current study is the pioneer in approaching #MeToo through crisis communications concepts. In order to do so, the crisis response strategy, crisis history and media framing concepts were adopted to study the viral social issue. A plethora of academic literature analyzed these concepts, which helped to formulate the hypotheses for the study. The main expectations of the study were that the rebuild strategy will outperform the other strategies and will be strongly impacted by the absence or presence of the prior crisis history in the organization.

Conducted research produced results, that can help practitioners to overcome the existing gap in the research by shedding light on how the different crisis response strategy in combination with the previous crisis history was reflected in the media and what media frame was applied. Results revealed contradictory outcomes to the initially hypothesized, however, they allow us to draw the tangible conclusions and formulate the few guidelines on how to approach the unfolding crisis.

The hypothesis testing confirmed, that the denial and diminish crisis response strategies lead to the less favorable depiction of the organization in press. However, it is important to note, that the companies, that have used the denial strategy performed in an unexpected manner, being evenly covered using each of the possible media frames. At the same time, the rebuilding strategy did not perform as well, as it was expected. Most of the articles, about the companies who employed the rebuild strategy, promised to take active actions on fixing the situation, namely get the accused person fired, undertake preventive actions, like implement the new policies related to the sexual harassment on the working place, support the victims, invest in special training for the employees, explaining what consequences they might suffer along with pay close attention to investigating the cases of reported sexual harassment or assault. However, in most cases these organizations still got the unfavorable image in the articles, depicting organizations as ones with toxic culture and employing a complete guilt frame towards the accused individual.

This leads us to the first implication of the results: rebuilding crisis response strategy is not necessarily the best strategy to approach the imminent backlash from the public, also taken into consideration its high price for execution and potential immense risks for reputation (Fuolia, van de Weijerb & Paradis, 2017).

The results on the crisis response history can’t be considered significant, however, as we expected we can see the pattern of media framing the crisis in the least favorable way if the company has a prior crisis history. Furthermore, the interaction testing revealed that the absence of crisis history makes the coverage a little more favorable for the range of crisis response strategies.

This leads us to the next implication of the findings: companies communications specialists have to be precisely cautious with their media activities if the organization has the present similar crisis history in the past.

Interaction test also revealed that the denial crisis response strategy performed better than all the other strategies in terms of the frames applied by media. This leads us to the last implication: denial crisis response strategy is worth considering as the early stage of reacting to the unfolding crisis. Honesty and the strive to rebuild the image as opposed to the initially hypothesised, can be harmful in the long run versus defending the company’s interests (Fuolia, van de Weijerb & Paradis, 2017).

To conclude, the results did not appear to be in line with what was initially expected, however, they did reveal some useful further implications for the practical implementation. First, the denial crisis response strategy does need more attention of both, practitioners and academics. Second, implying the denial strategy might be useful for the organization, experiencing the scandal unfolding, as it is giving the public the benefit of the doubt about the accused entity, instead of the straightforward acknowledgement of the crisis that has happened and attempts to take further responsibility for it. In line with the previous research

findings by van der Meer (2014), the denial frame at some point can be capable to reduce the public’s skepticism towards the organization and impact the wide adoption of the crisis response, proposed by the organization. These processes can be explained by the frames alignment concept discussed by Gerken, Van der Land, and van der Meer (2016), as the life cycle of the crisis ends with the strive of reducing the ambivalence and aligning the variety of proposed frames by the various sides. However, it is important to note, that this theoretical proposition needs further research, provided it wasn’t confirmed by the case studied by Gerken et al. (2016).

Presented research has the number of the limitations, important to be considered for further research in the crisis communications domain, focused on the viral societal issues. First, the limitations concerned the chosen method of the analysis, which, along with its indisputable benefits, has its disadvantages. Content analysis is subjected to the increased error, provided its subjectivity and is prone to reducing the dimensions of the concepts put into analysis. Second important limitation to be noted - the generalizability of the results was somewhat affected by the chosen sampling strategy. Even thought, the sample of the articles was drawn using the software, one can argue that the applied technique is a non-probability purposive sampling technique and it is judgmental and selective. This type of sample has a high chance of obtaining homogeneous sampling, which was expected and later it reflected in the presented analysis. Further, more broad investigation is needed to analyze the more heterogenous sample.

Next, it is important to note, that the examined articles were mostly covering the USA region, which was expected, provided that most of the organizations have their headquarters in the USA and American media would expectedly react on the issue faster, creating the most recent and relevant coverage. This location focus also limits the generalizability of the

organizations used the rebuild crisis response strategy, as was expected in the theoretical part. However, this skewness still allowed us to produce some results, the enlargement of the sample, by for instance, diversifying of the organizations taken for the analysis. Finally, majority of the organizations had no prior crisis history reported, presumably because the type of the crisis is relatively new and was reportedly under-communicated in the past. This way, a longitudinal study of the #MeToo issue can help to provide some tangible results and help overcome these limitations.

Further, the presented study is focused on the very specific and in some ways unique social issue, which is a direct threat to the generalizability of the research results. Moreover, the research included only a very particular range of media frames from the guilt to excuse. For the more complex understanding of the crisis communications, it could be beneficial to include other concepts, like skepticism and valence, enlarging the spectrum of the studied variables to create more opportunities for the complexity of the analysis, which was impossible to do using the content analysis approach.

Having said that, the presented research can be adopted as a solid base for further work on the proposed topic. For future directions, one way of addressing this issue can be proceeding with the research using the different methodologies, utilizing the presented

analysis as a base. On the one hand, experimental design of further study can be beneficial for the deeper understanding of the crisis response strategy and how it impacts the reader, on the other hand survey can also be a good fit for the deeper dive into the complexity of employed concepts. As stated before, longitudinal study can bring even more depth and clarity to the issue and the way it can be addressed. Furthermore, provided the nature of the crisis -

studying the emotional aspects of the reader’s reactions can be highly informative and useful for practitioners, despite the convolution of such research. Next, in line with what was stated by Scheufele (2004) in the overview of approaches to studying framing – approaching it via

the valence framing would be beneficial for the better understanding of crisis and social issues depiction. Finally, as was also noted by Schultz, Utz and Göritz (2011) it is important to include the different types of media into the analysis. Current research was focused on the newspapers, magazines and web articles; however, the further research can benefit from including the variety of different resources.

The effects of the denial crisis response strategy need to be studied with more depth and precision. Traditionally, rebuild frame is the most often recommended for the

organizations, that experience crisis, as even though this strategy requires more resource investment, it leads an organization to a new avenue of generating the additional assets for reputation improvement, precisely by showing the will to improve and bring value to the society (Coombs, 2007). However, as both, current and previous researches revealed, the denial strategy also has a lot of potential which can seem irrational, however, works in practice. Furthermore, the durable effects of the crisis response strategy used by the company and the public, media and company’s frames alignment should be studied. In the long-term perspective, the chosen strategy can result in a variety of consequences, which was not researched before.

References

Birkinshaw, J., & Cable, D. (2017, February). The dark side of transparency. Retrieved June 26, 2019, from https://www.mckinsey.com/business-functions/organization/our-insights/the-dark-side-of-transparency

Borah, P. (2011). Conceptual Issues in Framing Theory: A Systematic Examination of a Decade’s Literature. Journal of Communication, 61(2), 246–263.

Caputi, T., Nobles, A., & Ayers, J. (2018). Internet Searches for Sexual Harassment and Assault, Reporting, and Training since the #MeToo Movement. JAMA Internal Medicine, 179(2), 258–259. https://doi.org/10.1001/jamainternmed.2018.5094 Company Overview of The Weinstein Company Inc. (n.d.). Retrieved from

https://www.bloomberg.com/research/stocks/private/snapshot.asp?privcapId=21943957 Coombs, W. (2004). Impact of Past Crises on Current Crisis Communication: Insights From

Situational Crisis Communication Theory. Journal of Business Communication, 41(3), 265–289. https://doi.org/10.1177/0021943604265607

Coombs, W. (2007). Protecting Organization Reputations During a Crisis: The Development and Application of Situational Crisis Communication Theory. Corporate Reputation Review, 10(3), 163–176. https://doi.org/10.1057/palgrave.crr.1550049

Coombs, W., & Holladay, S. (2010). The handbook of crisis communication. Malden, MA: Wiley-Blackwell

Entman, R. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of Communication, 43(4), 51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x Fuoli, M., van de Weijer, J., & Paradis, C. (2017). Denial outperforms apology in repairing

organizational trust despite strong evidence of guilt. Public Relations Review, 43(4), 645–660. https://doi.org/10.1016/j.pubrev.2017.07.007

Gerken, F., Van Der Land, S., & van Der Meer, T. (2016). Crisis in the air: An investigation of AirAsia’s crisis-response effectiveness based on frame alignment. Public Relations Review, 42(5), 879–892. https://doi.org/10.1016/j.pubrev.2016.09.002

Keib, K., Himelboim, I., & Han, J. (2018). Important tweets matter: Predicting retweets in the #BlackLivesMatter talk on twitter. Computers in Human Behavior, 85, 106–115. https://doi.org/10.1016/j.chb.2018.03.025

Kepplinger, H., Geiss, S., & Siebert, S. (2012). Framing Scandals: Cognitive and Emotional Media Effects. Journal of Communication, 62(4), 659–681.

https://doi.org/10.1111/j.1460-2466.2012.01653.x

Kim, S., Avery, E., & Lariscy, R. (2009). Are crisis communicators practicing what we preach?: An evaluation of crisis response strategy analyzed in public relations research from 1991 to 2009. Public Relations Review, 35(4), 446–448.

https://doi.org/10.1016/j.pubrev.2009.08.002

Milano, A. (2017, October 15). If you've been sexually harassed or assaulted write 'me too' as a reply to this tweet. pic.twitter.com/k2oeCiUf9n. Retrieved June 24, 2019, from https://twitter.com/Alyssa_Milano/status/919659438700670976

More than 12M "Me Too" Facebook posts, comments, reactions in 24 hours. (2017, October 17). Retrieved June 24, 2019, from https://www.cbsnews.com/news/metoo-more-than-12-million-facebook-posts-comments-reactions-24-hours/

North, A. (2019, January 09). More than 250 powerful people have been accused of sexual misconduct in the #MeToo era. Here's a running list. Retrieved June 27, 2019, from https://www.vox.com/a/sexual-harassment-assault-allegations-list

Now What? (2019, January 03). Retrieved June 27, 2019, from https://hbr.org/cover-story/2018/01/now-what

Poell, T. (2015). Social media and activist communication. Routledge, London

Scheufele, B. (2004). Framing-effects approach: A theoretical and methodological critique. Communications, 29(4), 401–428. https://doi.org/10.1515/comm.2004.29.4.401 Shelley, M., & Krippendorff, K. (1984). Content Analysis: An Introduction to its

Methodology. Journal of the American Statistical Association, 79(385). https://doi.org/10.2307/2288384

Treem, J. W., & Leonardi, P. M. (2013). Social Media Use in Organizations: Exploring the Affordances of Visibility, Editability, Persistence, and Association. Annals of the International Communication Association, 36(1), 143-189.

https://doi.org/10.1080/23808985.2013.11679130

Utz, S., Schultz, F., & Glocka, S. (2013). Crisis communication online: How medium, crisis type and emotions affected public reactions in the Fukushima Daiichi nuclear disaster. Public Relations Review(39), 40-46.

Van Zoonen, W., Verhoeven, J. W., & Vliegenthart, R. (2016). Social media’s dark side: inducing boundary conflicts. Journal of Managerial Psychology, 31(8), 1297-1311. https://doi.org/10.1108/jmp-10-2015-0388

Youngblood, S. (2010). Ongoing Crisis Communication: Planning, Managing, and Responding, 2nd Edition (Coombs, W. T.) and Handbook of Risk and Crisis Communication (Heath, R. L. and O’Hair, H. D., Eds.) [Book reviews. IEEE Transactions on Professional Communication, 53(2), 174–178.

https://doi.org/10.1109/TPC.2010.2046099

Zacharek, S., Dockterman, E., & Sweetland Edwards, H. (2017, December). TIME Person of the Year 2017: The Silence Breakers. Retrieved June 26, 2019, from

https://time.com/time-person-of-the-year-2017-silence-breakers/

Zarkov, D., & Davis, K. (2018). Ambiguities and dilemmas around #MeToo: #ForHow Long and #WhereTo? European Journal of Women’s Studies, 25(1), 3–9.

https://doi.org/10.1177/1350506817749436 Appendix 1

Codebook for Content Analysis

The goal of the study is to examine how media portrays the corporate crisis related to #MeToo and how is it affected by the chosen crisis response strategy to address the issue publicly and the company's previous crisis history. For the purpose of the research 14 companies, whose C-suit employees were reportedly involved in the scandals related to the sexual harassment on the workplace, were chosen.

Instructions

The coding should be performed through the dedicated excel spreadsheet. Please read the provided articles carefully and fill out the response number into the corresponding tab in the excel document, following the instructions below.

Q1 Article_ID

Enter the identity code for each article into the corresponding cell as follows: year/month/day/newspaper abbreviation/time(s) coded (e.g. 20181005TNYT1). For the newspaper abbreviation the first letter(s) of the name of the newspaper should be used. The date refers to the date on which the article was published. If the article is coded for the first time – please indicate 1 in the end, if the article was coded more times - please indicate the corresponding number.

________________________________________ Q2 Date

Add the date on which the article was published: day-month-year. ________________________________________

Q3 Country

Please specify the country in which the article was published. Check the media resource’s location if unsure. Please use one of the following countries, or type in the country name if it is different:

2. United Kingdom 3. Canada

4. Other, namely _________________

Q4 Crisis, related to the viral societal issues

Please identify if the article mentions the crisis in the dedicated company, related to the #MeToo issue: sexual harassment, misconduct or assault on the working place.

1. Yes 2. No

Instructions

If you answered no (2) to the Q5 – this is the end of the survey. Q5 Company mention

Does the article mention a company from the list of the companies, included for the research: Google, Barnes & Noble, National Beverage, Dallas Mavericks, Wynn Resorts, Code for Science & Society, Morgan Stanley, OrbiMed, Sherpa Capital, Soros Fund Management, SEIU, Transformation Group, SEIU, 500 Startups, Binary Capital, Uber.

1. Yes

2. No

Instructions

If you answered no (2) to the Q6 – this is the end of the survey. Q6 Company

Indicate what individual and company is mentioned in the article for the current coding round. If more than one company is mentioned, the article is to be coded separately for each company. Therefore, note to pick the company of which you are currently coding.

2. Demos Parneros, Barnes & Noble 3. Nick Caporella, National Beverage 4. Terdema Ussery, Dallas Mavericks 5. Steve Wynn, Wynn Resorts

6. Harold Ford Jr., Morgan Stanley 7. Sam Isaly, OrbiMed

8. Shervin Pishevar, Sherpa Capital 9. Howie Rubin, Soros Fund Management 10. Robert Scoble, Transformation Group 11. Scott Courtney, SEIU

12. Dave McClure, 500 Startups 13. Justin Caldbeck, Binary Capital 14. Travis Kalanick, Uber

Instructions

If the article mentions more than one of the companies selected for this research, the article is to be coded separately for each company.

For the following questions, only code the paragraphs that are related to the current crisis and past crises of the corresponding company. If the text covers other non-related to the crisis company activities (i.e. CSR activities not related to the sexual harassment prevention, stock information, product information, etc.) – this part should not be employed in the analysis.

Q7 Crisis response strategy

Crisis response strategy is the general approach, that the organization have chosen for communications and enable to examine what strategy can be more effective in communicating the #MeToo related crisis specifically to the media. The crisis response

strategies can be divided into denial, diminish and rebuilding. Look for the lines, that describe company’s official statements by the company’s spokespeople, or lines in the text, that mention the company’s position.

Denial strategy includes the two directions of communications: attacking or confronting the accuser, denying the fact that crisis happened, and scapegoating, accusing the third party of anything that happened in the organization. Look for the statements like: “This never happened…”, “False claims…”, “Company/Accused individual declined the accusations…”, etc.

Diminish strategy can be looking for an excuse or justification for the actions of the organization or diminishing the seriousness of the situation by the non-sufficient actions towards the accused. Look for the statements like: “The accused person acted in mutual consent...”, “There are no evidence that the misconduct happened”, “The accused person was not fired”, “The accused person received compensation package”, etc.

Rebuilding strategy implies the compensation to the victims and apology to the stakeholder, taking all the responsibility for the crisis onto the organization. Choose this strategy, if the organization fired the accused person after the allegations. Look for the statements like: “The accused person’s conduct is inconsistent with the corporate values...”, “Was fired for the sexual misconduct…”, “Compensation to the victims…”, etc.

Please choose not applicable only if there is no mention of how organization reacted to the crisis.

1. Denial 2. Diminish 3. Rebuild 4. Not applicable

Q8 Media framing components

Media frame is the ability of the media to choose certain aspects of the perceived reality and raise or lower its salience along with providing the preferable problem definition, interpretation and moral evaluation with the further suggestions

In order to facilitate your choice of the frame please first define the media framing components filling in the answers for the questions below in the excel sheet:

Q8a: How was the scale of the crisis related to the sexual harassment and misconduct depicted in the article?

Look for the adjectives that are used to describe the crisis and that directly highlight the scale of the crisis: “rampant”, “great”, “immense”, “dead serious”, etc for the large crisis. No color added on crisis or inclusion of the words like “insignificant”, “small”, “minor”, etc to be considered as the small scale of crisis.

1. Large

2. Small

Q8b: What, according to the article, were the goals of the individual accused in sexual harassment and misconduct?

Look for the attempts in the article to justify the goals of the individual accused in the scandal. All other options imply the selfish goals.

1. Selfish goals

2. Altruistic/common goals

Q8c: What, according to the article, was the awareness of the individual accused in sexual harassment and misconduct?

To identify the positioning of the individual who is not aware of the consequences, look for the hints that can point on the attempts in the article to find excuses for the individual accused in the scandal, or mentions that the individual was sure about the mutual consent. All other cases will identify that the individual was aware of the consequences.

1. Crisis situation caused by people who were aware of the consequences of their

decisions

2. Crisis situation caused by people who were not aware of the consequences of their

decisions

Q8d: To what extent the responsibility is attributed to the individual accused in sexual harassment and misconduct, according to the article?

Cases of the responsibility not attributed to the accused individuals can try to rationalize the behaviour, for example: “Toxic company’s culture boosted the behaviour related to sexual misconduct...”, etc. Other options imply the responsibility of the individual.

1. Crisis situation caused by people who could have acted differently

2. Crisis situation caused by people who were constrained to act the way they did

Instructions

The following questions relate to the frame that is applied by the media to report the crisis related to sexual harassment or misconduct occurred in the company. Please indicate the frame, based on the set of previously answered questions about media frame component, as follows: when all five aspects are accusing the responsible person of the crisis that happened, the article implies the guilt frame. When only a few of the media frame components are accusing the responsible person – choose the fragmentary guilt frame.

If the article is trying to understate and belittle the individual’s level of responsibility for the crisis occurred in all five media frame component choose the excuse frame. When only a

few of the media frame components are referred to minimizing personal responsibility – choose the fragmentary excuse frame.

If the article does not contain any frame, but reports the situation happened in the maximum possible neutral matter – please choose no frame option.

Q9 Media frame

Please identify the media frame, applied in the article to the company’s crisis response strategy.

1. Guilt frame

2. Fragmentary guilt frame

3. No frame

4. Fragmentary excuse frame

5. Excuse frame

Q10 Prior crisis history

Please identify if the prior crisis history or the similar nature (sexual harassment and misconduct at the working place) of the respective company is mentioned.

1. Present

2. Absent

Instructions

Before you end the questionnaire, please, save the article you have coded as a .doc, .docx or .pdf file naming it with the identity code, as follows: year/month/day/newspaper abbreviation (e.g. 20181005TNYT).

Appendix 2 SPSS Syntax

*Uploading the excel sheet to the SPSS GET DATA /TYPE=XLSX /FILE='C:\Users\12017914\Downloads\dataset_results.xlsx' /SHEET=name 'Sheet1' /CELLRANGE=FULL /READNAMES=ON /DATATYPEMIN PERCENTAGE=95.0 /HIDDEN IGNORE=YES. EXECUTE.

DATASET NAME DataSet2 WINDOW=FRONT.

*Running descriptive analysis for the key variables for analysis itself and for the description of the sample

FREQUENCIES VARIABLES=Q7Strategy

/STATISTICS=STDDEV MEAN MEDIAN MODE SKEWNESS SESKEW /ORDER=ANALYSIS.

FREQUENCIES VARIABLES=Q10Priorcrisishistory

/STATISTICS=STDDEV MEDIAN MODE SKEWNESS SESKEW /ORDER=ANALYSIS.

FREQUENCIES VARIABLES=Q9Mediaframe

/STATISTICS=STDDEV MEDIAN MODE SKEWNESS SESKEW /ORDER=ANALYSIS.

FREQUENCIES VARIABLES=Q8aScaleofthecrisis

/STATISTICS=STDDEV MEDIAN MODE SKEWNESS SESKEW /ORDER=ANALYSIS.

*Creating dummy variables for the further analysis

RECODE Q7Strategy (1=1) (2=0) (3=0) (4=0) INTO denial_dummy. VARIABLE LABELS denial_dummy.

EXECUTE.

RECODE Q7Strategy (1=0) (2=1) (3=0) (4=0) INTO diminish_dummy. VARIABLE LABELS diminish_dummy.

EXECUTE.

RECODE Q7Strategy (1=0) (2=0) (3=1) (4=0) INTO rebuilding_dummy. VARIABLE LABELS rebuilding_dummy.

EXECUTE.

RECODE Q9Mediaframe (1=1) (2=1) (3=0) (4=0) (5=0) INTO guilt_dummy. VARIABLE LABELS guilt_dummy.

EXECUTE.

*Hypothesis 1 and 2 testing CROSSTABS

/TABLES=denial_dummy BY Q9Mediaframe /FORMAT=AVALUE TABLES

/STATISTICS=CHISQ /CELLS=ROW

/COUNT ROUND CELL. CROSSTABS

/TABLES=diminish_dummy BY Q9Mediaframe /FORMAT=AVALUE TABLES

/STATISTICS=CHISQ /CELLS=ROW

/COUNT ROUND CELL. CROSSTABS

/TABLES=rebuilding_dummy BY Q9Mediaframe /FORMAT=AVALUE TABLES

/STATISTICS=CHISQ /CELLS=ROW

/COUNT ROUND CELL. *Hypothesis 3 Testing T-TEST GROUPS=Q10Priorcrisishistory(1 2) /MISSING=ANALYSIS /VARIABLES=Q9Mediaframe /CRITERIA=CI(.95). FREQUENCIES VARIABLES=Q9Mediaframe

/STATISTICS=STDDEV VARIANCE MEAN MEDIAN MODE SKEWNESS SESKEW KURTOSIS SEKURT

/ORDER=ANALYSIS. *Hypothesis 4 Testing

UNIANOVA Q9Mediaframe BY Q10Priorcrisishistory Q7Strategy /METHOD=SSTYPE(3)

/INTERCEPT=INCLUDE

/POSTHOC=Q7Strategy(BONFERRONI)

/PLOT=PROFILE(Q10Priorcrisishistory*Q7Strategy) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO YAXIS=AUTO

/PRINT DESCRIPTIVE HOMOGENEITY /CRITERIA=ALPHA(.05)