1 CROSS NATIONAL CONFLICT SHIFTING: AN EXAMINATION OF THE DEEPWATER HORIZON OIL SPILL By SARAH GRAVINA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DE GREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2011
2 2011 Sarah Gravina
3 To my parents and grandparents for t heir constant support, encouragement, and inspiration
4 ACKNOWLEDGMENTS First I must offer my deepe st thanks to my adviser, Dr. Juan Carlos Molleda, because without his patience, guidance, and knowledge, this thesis wo uld have never been completed. I would also like to thank my committee members, Dr. Ka thleen Kelly and Dr. Sora Kim. I appreciate their e xpertise and encouragement throughout this process. I must thank my family for they have offered me unconditional love and support throughout this long journey. My parents and grandparents have served as a constant source of inspiration to me throughout th e years and have instilled within me the desire to achieve and the strength to overcome obstacles. I also appreciate the kindness, love and support offered to me by the Crittenden family over the years. Finally I must offer gratitude to my friends. I appr eciate the support provided by my fellow classmates in the department, specifically Tim Deardourff, Tamekia Massaline, Andreea Savu, Eri c Schumacher, and Weiting Tao. They have played a role in shaping my experience in the program as well as in broadening my perspectives on public relations.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 7 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 International Public Relations ................................ ................................ ................. 10 Cross National Conflict Shifting Theory ................................ ............................ 10 Situational Crisis Communication Theory ................................ ......................... 12 Background ................................ ................................ ................................ ............. 12 Relevance of St udy ................................ ................................ ................................ 14 Purpose of Study ................................ ................................ ................................ .... 14 2 LITERATURE REVIEW ................................ ................................ .......................... 17 Cross National Conflict Shifting ................................ ................................ .............. 17 Application of Propos itions to the Deepwater Horizon Oil Spill ............................... 20 Corporate Social Performance Issue ................................ ................................ 20 TNC Alerted the Audiences ................................ ................................ .............. 21 Human Interest Focus ................................ ................................ ...................... 22 Direct Involvement of Transnation al Corporation in Crisis ................................ 22 Boycottable, Tangible Products ................................ ................................ ........ 22 Numbers of Actors Involved in Crisis ................................ ................................ 23 Transnational Corporation Headquarter Location ................................ ............. 24 Situational Crisis Communication Theory ................................ ............................... 25 Corporate Apologia ................................ ................................ .......................... 26 Corporate Impress ion Management ................................ ................................ 26 Image Restoration Theory ................................ ................................ ................ 27 Situational Crisis Communication Theory ................................ ......................... 28 Research Ques tions and Hypotheses ................................ ................................ ..... 31 3 METHODOLOGY ................................ ................................ ................................ ... 34 Quantitative Content Analysis of Media Coverage ................................ .................. 34 Quantitative Content Analysis on News Releases ................................ .................. 36 Coding Sheet Construction ................................ ................................ ............... 36 Intercoder Reliability ................................ ................................ ......................... 37
6 4 FINDINGS ................................ ................................ ................................ ............... 39 Case Study -The Deepwater Horizon Oil Spill ................................ ........................ 39 Statistical Findings ................................ ................................ ................................ .. 42 Hypothesis One ................................ ................................ ................................ 42 Hypothesis Two ................................ ................................ ................................ 43 Hypothesis Three ................................ ................................ ............................. 45 Hypothesis Four ................................ ................................ ............................... 46 Hypothesis Five ................................ ................................ ................................ 48 Hypothesis Six ................................ ................................ ................................ .. 49 Hyp othesis Seven ................................ ................................ ............................ 50 5 DISCUSSION ................................ ................................ ................................ ......... 52 Summary and Interpretation of Findings ................................ ................................ 52 Implications for Public Relations Practice ................................ ............................... 55 Theoretical Implications ................................ ................................ .......................... 56 Limitations and Further Research ................................ ................................ ........... 57 APPENDIX A ARTICLE CODING SHEET ................................ ................................ .................... 59 B NEWS RELEASE CODING SHEET ................................ ................................ ....... 62 LIST OF REFERENCES ................................ ................................ ............................... 63 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 67
7 LIST OF FIGURES Figure page 4 1 Mean plot of word count and news source of article ................................ ........... 43 4 2 Frequency of actor in headlines ................................ ................................ .......... 44 4 3 Frequency of actor in lead paragraphs ................................ ............................... 45 4 4 Frequency of article focus ................................ ................................ .................. 47 4 5 Frequency of crisis response strategies in media coverage ............................... 49 4 6 Frequency of crisis response strategies in news releases ................................ .. 51
8 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requi rements for the Degree of Master of Mass Communication CROSS NATIONAL CONFLICT SHIFTING: AN EXAMINATION OF THE DEEPWATER HORIZON OIL SPILL By Sarah Gravina December 2011 Chair: Juan Carlos Molleda Major: Mass Communication This study used the cross national conflict shifting theory a s well as the situational crisis communication t heory to analyze the Deepwater Horizon Oil Spill The literature review included a thorough explanation of the cross national conflict shifting and situational crisis c ommunication theories, describing their foundations and e volution. The research included a case study and quantitative content analysis. Research questions were answered by a case study. Seven hypotheses were de veloped based on the literature review. Statistical testing was used to test these h ypotheses. Five of the seven hypotheses rec eived full or partial support. Two of the hypotheses received no support. The findings suggested that news sour ces in different countries cover ed a crisis with varying len gth, frequency and source use. The differ ence in length, frequency, and source use determine d the depth of media coverage by each country. Difference in article foci between the two news sources was present In this particular case, the crisis response messaging was consistent, but it was determi ned that the media may not cover the intended corporate response in its entirety.
9 This study contributed to the growing body of knowledge in global p u blic relat ions and crisis communication. The study had implications for theory building and international public relations practice.
10 CHAPTER 1 INTRODUCTION International Public Relations The increasing trend of globalization among various types of organizations has created several opportunities and challenges. The expansion of operations across the globe ha s had a significant impact on the communicative appr oaches taken by organizations. Geographic borders no longer provide distinct communication boundaries for globalized organizations. This phenomenon has allowed the flexibility and pace of communications b etween organizations and their publics to increase. Wang (2005) identified the key players controlling com munication across borders as non governm ental organizations (NGOs), governments, and global media. In order to gain an understanding of the impact th at globalization has had on the public relations industry; one must examine the operations of various types of global organizations. The theories of cross national conflict shifting and situatio nal crisis communication are b riefly discussed in Chapter 1 wi th a more thorough description provided in the literature review. The relevance and purpose of this study is also discussed in Chapter 1 Cross National Conflict Shifting Theory The theory of cross national conflict shifting was developed to interpret the shift of crises involving the t ransnational corporations from the country in which the crisis occurs to the country or countries where the transnational corporation is headquartered and operates (Molleda, 2011). This theory provides the opportunity t o exam ine the operations of transnational corporations around the globe and assess the principles of the cross national confl ict shift including: characteris tics of the conflict shift, the manner
11 in which the conflict rea ches local stakeholders, and the key publ ics involved in the conflict (Molleda, 2011). Molleda and Quinn (2004) developed 10 propositions that serve as the foundation of this theory and through several studies conducted by Wang (2005); Molleda, Connolly Ahern, and Quinn (2005); Kim and Molleda (2 005); Molleda (2011); and Molleda, Bravo, Davila, and Botero (2011) the propositions have been t ested and the theory advanced. The core elements of a cross national conflict shift, indicated by the origin al 10 propositions, include: the incident involves a corporat e social performance issue, the origination of the conflict, number of involved part ies, commercial, tangibl e and boycottable products, corpor ation headquarter location, direct involvement of corporation, party that alerts t he audience of conflict and conflict has human interest focus (Molleda, Connolly Ahern, & Quinn, 2005). With the influx of crises involving transnational corporations in recent years, the dra matically from what was initially a dismal body of literature (Taylor, 2000;Wang, 2005; Harlow, Brantley & Harlow, 2011; Kim, Avery, & Lariscy, 2009). Theor ies from the crisis response literature have been used to complement studies incorporating the cross national conflict shifting theory (Wang, 2005; Molleda, et al., 2011). In the study conducted by Wang (2005), the author used the cross national conflict shifting principles in conjunction with the typology of crisis responses developed by Coombs (2007) to examine the DuPont Teflon crisis. Molleda et al. (2011) conducted a case study and a content analysis to test the cross national conflict shifting propositions
12 and determine the crisis response strategies used by the Chiquita Brand in the transnationa l crisis it faced in C olombia and the United States. In this study, Molleda et al. (2011) discussed the application of the cross national conflict shifting propositions to the Chiquita Brand case and emphasized the importance of understanding the crisis re sponse strategies used by a transnational corporation to further the understanding of global public relations. Situational Crisis Communication Theory The situational crisis communication theory conceptualized by Coombs (1995) provides an option for deter mining the crisis responses disseminated by a transnational corporation. The core of the situational crisis communication theory is to connect the crisis type with the appropriate crisis response strategy (Coombs & Holladay 2003). Coombs (2007) identified fou r groups of response postures including: denial, diminishment, rebuilding, and bolstering. Within these postures, lie several response strategies that can be used by a transnational corporation when facing a cross national conflict shift. The purpose o f this study is to examine the Deepwater Horizon oil spill through the combination of the cross conflict shifting theory and situational cr isis communication literature. A case study will be conducted to provide a thorough description of the case within th e context of the cross nat ional conflict shifting theory. A quantitative content analysis will be conducted to test hypotheses concerning the cross national conflict shifting theory and the situational crisis communication theory. Background On April 20, 2 010, the British Petroleum (BP) oil rig, Deepwater Horizon operating in the Gulf of Mexico, exploded. The initial explosion killed 11 workers, injured 17, and
13 commenced one of the largest environmental crises i n history. Throughout the summer of 2010 larg e amounts of oil leaked from the damaged well into the Gulf while workers attempted to mend the leak with several failed attempts, including several types of containment domes as well as a method known as a top kill. According to the Deepwater Horizon Inci dent Joint Information Center (2010), more than five million barrels of oil left the damaged well and entered the Gulf throughout the summer months. During the crisis, BP was criticized for its crisis response and the criticisms flooded the ne ws media in s everal countries. The news coverage varied in tone and content across the globe. The attribution of blame for the crisis was placed not only on the organization as a whole but also on several key actors, including the CEO at the time, Tony Hayward. Through out the summer of 2010, the shift of the crisis from the United States to the United Kingdom became apparent, as the United States was no longer the only country feeling the effects from the incident. Although the direct physical effects from the crisis, s pecifically damage to the environment and several industries, influenced the United States, issues began to arise in the United Kingdom and other world locations where BP has significant operations and business interests. Not only was the transnational cor poration facing financial and reputational consequences in both countries, but also the social influence in the United States began to specia lly affect the United Kingdom. Protesting in the United States began to not only attack the organization, but also contained anti British sent iment (Leonard, 2010). The increase in opposition toward the British concerned U.K. government officials. Their
14 concern centered on the potential damage that could be done to the U.K. economy if the negative focus toward BP cont inued (Reuben, 2010) Relevance of Study This study holds relevance particularly due to the timely nature of the crisis and Horizon spill to date and it is a crisis that will h ave implications for public relations practitioners and educators for a long while. Grant (2010) described the presence that topic of crisis communications and cas e studies in public relations textbooks for years to 13). Purpose of Study The purpose of the stu dy is to incorporate the cross national conflict shifting literature along with research from the crisis communication perspective to examine the tra nsnational crisis faced by BP, known as the De epwater Horizon oil spill. As previously mentioned, the BP crisis originated in the United States as primarily an environmental and financial crisis involving a corpo rate social performance issue. Growing dissa tisfaction with the corporation allowed the conflict to shift to the United Kingdom as more of a financial, political, and reputational crisis. The case study will illustrate the application of the original cross national conflict shiftin g propositions to the BP case. This case study will also explain other components of the cross national conflict s hifting theory in relation to the case including the identification of key actors involved in the crisis. A quantitative content an alysis will be conducted to examine three aspects of the BP crisis including : components of the cross nationa l conflict shifting theory, the crisis
15 response disseminated by th e TNC in news releases, and the crisis response strategies covered by newswire agencies in the countries sele cted for study. The countries selected for study include the United States, the host country of the corporation and the location in which the crisis originated, as well as the United Kingdom, the headquarters of th e transnational corporation. The component s of the cross national confl ict shifting theory that will be analyzed in the content analysis include the actors identified in the headlines and lead paragraphs of the print media coverage, the key terms used in the headline and lead paragraph of the cove rage, and the amount of space devote d to articles The crisis r esponses disseminated BP and covered in the media will be determined based on the typology conceptualiz ed by Coombs (1995) within the s it uational crisis c ommunication theory. Through the analys is of the BP case, the goal of this study is to further the understanding of global public relations and pro vide additional guidance to transnational corporations facing conflicts or cris es in several world locations. Referring to the research on cross nat world, this knowledge is essential for TNOs to learn how to face cross national conflict eda et al., 2011, p. 1). I Following Chapter 1 the literature review will provide a thorough description of the studies c onducted by researchers concerning the cross national conflict shifting and situational crisis commu nication theories. Research questions and hypotheses will also be presented in the Chapter 2 Chapter 3 will follow the literature review and will
16 describe the two methodologies selected for study: case study and quantitative content analysis. Chapter 4 wi ll present a case study of the Deepwater Horizon oil spill and statistical finding s used to test the hypotheses. Chapter 5 of the thesis will disc uss and interpret the findings. In Chapter 5 limitations of the study and suggestions for further research wi ll also be made.
17 CHAPTER 2 LITERATURE REVIEW The following literature review focuses on two main theories that will guide this study (a) cross national conflict shifting and (b) situational crisis communication theory. The literature review on cross n ational conflict shifting will include the conceptualization of the theory, the propositions that guide it, and the prior studies that have been conducted to examine it. The literature reviewed on situational crisis communication theory centers on the theo ries that were utilized to develop the specific theory including corporate apologia, corporate impression manage ment, and image repair theory. The discussion on situational crisi s communication theory focuses on the core of theory lying in attribution theo ry, the typology of crisis responses, and the studies conducted that contributed to the evolution of the theory. Cross National Conflict Shifting The cross national conflict shifting theory, conceptualized by Molleda and Connolly Ahern (2002), was first in troduced at the 2002 Association for Education in Journalism and Mass Communication a nnual conference in Miami, Fla. The authors derived research from the international management perspective, specifically from German authors Welge, Holtbrugge, and Berg, t o serve as the foundation for the theory. Molleda (2011) defined the process of a cross national conflict shift: Cross national conflict shifting is the transfer of incidents or crises faced by a transnational corporation (TNC) from the country where the situation originates to another or multiple world countries where the TNC is headquartered or executes major operations. (p. 50) The study conducted by Wang (2005) demonstrated a reverse cross national conflict shift in that the conflict occurred in the ho me country of the corporation and transferred
18 to the host country. The home country of a corporation is the country in which the organization is headquartered and/or holds a main office; while the host country or countries serve as the multiple locations i n which the organization executes operations on a smaller scale (Molleda, 2011). The study conducted by Taylor (2000) revealed the effect of culture on the involving the Coca Cola cris is that originated in Belgium. The author used the cultural dimensions o response to the cri sis that spread across Europe. Molleda (2011) identified the study conducted by Taylor (2000) as a key part of the initial conceptualization of the cross nati onal conflict shifting theory. The study produced by Taylor (2000) served as a central role in the theory development due to its emphasis on the examination of communication processes across borders (Molleda, 2011). To advance the development of the cross national shifting theory Molleda and Quinn (2004) identified three main characteristics of the cross national shift process : the c haracteristics of the issue, the ways a national conflict reaches transnational audiences, and the parties involve d or affect ed by the conflict. These characteristics led to the development of 10 initial propositions by Molleda and Quinn (2004) and re constructed by Molleda (2010). Seven of the original 10 propositions will be applied to the Deepwater Horizon oil spill later in t his section to serve as justification for using the cross national conflict shifting theory to guide this study. Molleda, Connolly Ahern, and Quinn (2005) advanced the research on the cross national conflict shifting theory through a study centered on the examination of media coverage regarding the bribery s candal in Lesotho, Africa. The study supported the
19 initial 10 propositions of th e cross national conflict shift. The study found that governmental sources were prominent in news coverage and a large port ion of the media coverage originated from the home country due to th e local impact of the conflict. In this particular study, the home country was Canada and the host location was Africa. Kim and Molleda (2005) developed three additional propositions to in corporate into the cross national conflict shifting and crisis management theory. Kim and Molleda (2005) utilized the cross national conflict shifting theory and the crisis communicati on response typology ian The se additional propositions will not be used in this study but are important points in the discussion of the evolution of t he cross national conflict shifting theory. As previously mentioned, the study conducted by Wang (2005) demonstrated a reversed cross natio nal conflict shift occurrence. Wang (2005) incorporated theories in crisis management along with the cross national c onflict shifting theory to analyze the the resulting media coverage. The findings suggested that when a reverse cross national conflict shift case is present it can be interpreted through three angles in cluding: crisis management performance, level of media interest on the issues, and the social and cultural context of the involved country. Lim and Molleda (2009) deviated from using the typical method of studying the cross national conflict shifting propo sitions, content analysis, by applying an experimental research method to examine the attitude and behavior of host customers
20 when influenced by the cross national conflict shift. Lim and Molleda (2009) found that a crisis from a transnational corporation would result in the formation of negative attitudes and behaviors when audience segments were exposed to media coverage resulting from the cross national conflict shift. The results of the study conducted by Lim and Molleda (2009) also demonstrated the imp ortance of the implementation of appropriate crisis response strategies by a transnational corporation in order to maintain a favorable reputation among key publics. Molleda and Laskin (2010) identified the connection between the literature on the coordina tion and control of a transnational corporation and a cross national conflict shift. with the cross national conflict shifts depend on and are largely determined by the c The connection between these two research areas highlights the importance of underst anding transnational processes. Application of Propositions to the Deepwater Horizon Oil Sp ill To serve as the justification of the Deepwater Horizon oil spill as a cross national conflict shift as well the use of the theory to guide this study, seven of the original 10 propositions of the cross nation al conflict shift theory are applied to the specific case. Three propositions have been excluded due to irrelevance to the case. Corporate Social Performance Issue Molleda (2010) highlighted the issue of corporate social performance as being a significant point in a cross national conflict shift. M olleda and Quinn (2004) further competing voices appear to be related to labor, the environment, human rights,
21 employee treatment and workplace safety issues, custome r satisfaction, and product The Deepwater Horizon oil spill is considered to be one of the worst envir onmental disasters in history. According to the Deepwater Horizon Incident Joint Information Center (2010), more than 6,000 birds, 600 t urtles, and 100 mammals were found deceased near the oil spill location. The disaster had considerable influence on the fishing and restaurant industries with nearly 88,522 square miles of fisheries closed in the Gulf of Mexico at the peak of the s The preventable crisis nature and the extensive damage it has caused to the environment, wildlife, related industries, and economy identifies it as a significant corporate social performance issue. TNC Alerted the Audiences The organ izational actors involved in the cross national conflict shift alerted the media and other audience segments regarding the initial explosion of the Deepwater Horizon oil spill prior t o the onset of media coverage. Transocean, owner of the Deepwater Horizon oilrig, issued the first news release within 24 hours of the trigger ev ent (Transocean, 2010). British Petroleum then followed suit with a news release confirming the validity of the initial stateme nt by Transocean (British Petroleum, 2010). Molleda (20 11) identified one of the advantages of the transnational corporation being the first to issue an announcement as being a proactive measure in protecting the Unfortunately in this case, the organization at the focus of the medi a coverage was BP and they did little to alert audience segments beyond confirming the announcement of Transocean Perhaps a more proactive approach from BP of the crisis.
22 Human In terest Focus Due to the enormity of the consequences of the oil spill and the industries affected, a large portion of the attention on the crisis encompa ssed the human interest angle. The human interest focus of a cross national conflict shifting garners m ore attention form audience segments (Mol leda & Quinn, 2004). Protestors across the United States brought negative attention to the response efforts of BP Employees in several industries including tourism and fishing lost their jobs and failed t o rebuild their organizations. The estimated cost to the fishing industry, calculated in September 201 0, was estimated to be around $ 2 billion (Jarvis, 2010). Direct Involvement of Transnational Corporation in Crisis As operators of the Deepwater H orizon oilrig, BP was directly involved in and partially to blame for the oil spill crisis. The organization not only faced a criminal investigation from the United States government but also became the focus of several lawsuits from those parties affected by the spill. In August 2010, BP reported that claims alone had cost the organiza tion $400 million (British Petroleum, 2010). Boycottable, Tangible Products The boycottable characteristic of the product s produced by BP allowed for more intensity when it c ame to protests and boycotting. Products that are high profile and originate from an industry considered controversial will contribute to a shift in conflict (M olleda & Quinn, 2004). As previously mentioned, protests occurred throughout the United States during the months following the spill. The protests not only had negative implications for the potential economic revenu e generated by BP in the United States, but also for the oil company in the United Kingdom (Leonard, 2010).
23 Numbers of Actors Involved in Crisis The act ors involved in the crisis are important considerations when discussing a cross national conflict shift because it can determine the direction of the crisis shift, the communication methods needed from the transnational corporation, and predict the sources used in the media cov erage (Molleda & Quinn, 2004). According to Molleda principal p art At the base level of this crisis there were three organizations involved, BP Transocean, and Halliburton. As previously mentioned, BP was the operator of the Deepwater Horizon oilr ig leasing it from Transocean. Halliburto responsibility in cementing the oilrig in place. According to the National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling (2011), Halliburton was to blame for using a cement mixture to secure the oilrig that was considered unstable and Once the trigger event occurred, a variety of actors became involved in the crisis. From a governmental perspective, the United States government was centra l to th e investigation into the spill. Politicians from United Kingdom were also highly visible and involved in the aftermath of the spill including British Prime Minister David Cameron, London mayor Boris Johnson, and former Trade Minister Lord Norman Teb bit, (Assinder, Several conservationist groups were involved in the cleanup of the spill as well as in the public response to the crisis including the National Oceanic Atmospheric Administration, S ierra Club, among many other s. Media agencies worldwide, including
24 BBC Reuters, Associated Press, and several others were highly involved in the crisis and generated an enormous amount of coverage regarding all facets of the crisis. The media covered the key actors in the crisis inc luding BP the governments of the countries involved and the response from those publics affected by the spill. Several prominent celebrities including film director James Cameron and actor Kevin Costner contributed to the recovery efforts and represented another audience segment generating public response (G abbatt, 2010; Hinckley, 2010). Transnational Corporation Headquarter Location According to Molleda (2010), when the transnational corporation in the cross national conflict shift is headquartered in a developed nation the organization is more likely to garner attention from several audience segments including global media, NGOs, and national governments. British Petroleum, as the transnational corporation that was most highly involved in the crisis, is headquartered in the United Kingdom and produced a crisis in the ho st country of the organization. The home country of a transnational corporation is the locations of its headquarters; the host country is a country in which th e corporation holds operations With both the United States and United Kingdom being home to prominent international newswire agencies, Associated Press and Reuters respectively, it was clear that the attention from medi a alone would be considerable. The active nature of conservationis t groups as well as governmental figures in both countries allowed for the global media coverage to increase throughout the months following the trigger eve nt. With the application of the majority of the cross propositions to t he Deepwater Horizon crisis, it is clear that this case is characterized as
25 a cross national conflict shift This justification establishes relevance for this study and serves as the reasoning for selecting t his theory to guide the study. Situational Crisi s Communication Theory This section of the literature review focuses on the research conducted on the topic of crisis respo nse. This portion of the review also holds relevance to the topic of study and is a s ignificant part of this study. The focus of this section centers on the development and evolution of situational crisis communication theory, conceptualized by Coomb s (1995). The examination of the research on this theory must first begin with the identification of its origin and foundation. Coombs (200 7) defined the term crisis as, ate According to Coombs (2007), the fo undation of situational crisis communication theory lies within the theories of corporation apologia, corporate impression man agement, and image restoration. Corporate apologia and corporate impression management will be briefly discussed later in Chapter 2 with a more thorough description regarding image restoration theory and its connection to situational crisis communication theory following. S ituational crisis communication theory is linked to attribution theory, its derivation from social psychology, in that the attribution of responsibility to the organization involved in the crisis is the core of situationa l crisis communication theory. Coombs (1995) identified three dimensions of attribution that publics assess before attributing responsibility to a n organization in a crisis including: locus, s tability, and controllability. The locus of the crisis reflects the location of the cause of the crisis either upon the
26 actor inv olved or the situation itself. The stability of the crisis acknowledges the previ ous history of the organization and whether a crisis has happ ened prior to the current one. The controllability of the crisis refers to whether the cause of the crisis is controllable or not (Coombs & Holladay, 2003). Corporate Apologia Hearit (1996) defin contain one), but a defense that seeks to present a compelling, counter description of org This response is often used when an organization is facing some sort of et hic al offense during a crisis. Several authors have contributed to the research on apologia, defined their stances and have expanded its core concepts (Benoit, 1997; Hearit, 1999). Three stances of corporate apologia have been developed and these stances have contributed to the development of the response typology in situational crisis communication theory (Hearit & Brown, 2004). Corporate Impression Management Corporate impression management is an other theory that served as the foundation for situationa l cris is communication theory. Conceptualized by Allen and Caill ouet (1994), this theory focused on the connection between crises and legitimacy and incorporated the concept of perception in managing impressions of an organization. Allen and Caillouet identifie d six impression management strategies to guide the process of corporate impression management including: excuse, justification, ingratiation, intimidation, denouncement, and factual distortion. Several of these strategies have been incorporated in the res ponse typology of situational crisis communication theory, including excuse, justification, and ingratiation.
27 Image Restoration Theory The central concept of image restoration theory, conceptualized by Benoit (1997), is perception. Perception is consider ed more important than reality and this emphasis is continued by Coombs (2007) in situation al crisis communication theory. Benoit (1997) identified five general response strategies for image restoration with each including several respon se types that compl ement them. Similarities can be drawn between the response strategies in image restoration theory and the response postures in situational crisis communication theory. image repai r strategies during her speech followin g the death of Princess Diana. The Queen used four image repair strategies throughout the speech including: denial, bolstering, defeasibility, and transcendence (Benoit & Brinson, 1999). Benoit and Henson (2009) ana lyzed the image repair discourse strategies used by President Bush during a speech in the aftermath of Hurricane Katrina. The authors found that the speech was considered ineffective due to the improper or minimal use of the image repair strategies (Benoit & Henson, 2009). One of the more recent studies on image repair theory, conducted by Harlow, Brantley and Harlow (2011), analyzed the image repair strategies used during the initial aftermath of the BP Deepwater Horizon oil spill. The findings of the stud y revealed that the organization used the image repair strategies of compensation and corrective action most frequently in the news releases disseminated. The authors identified that one of the main limitations of the study that not all of the news release s fit under the image restoration theory. This study provides justification for the use of the situational crisis
28 communication the ory for the analysis of the news releases from the crisis as well as the response strategies covered in the media. Situationa l Crisis Communication Theory As previously mentioned, attribution theory lies at the core of situational crisis communication theory and provides the foundation for the assessment of the perceived responsibility of the organization in the crisis. Once the assessment of perceived situational crisis communication theory then prescribes response strategies based on that potential reputational threat (Coombs, 2007). The main obj ective of situational crisis communication theory is to connect the crisis type with the appropr iate crisis response strategy. The bridge that connects crisis type with crisis response is the amount of responsibility the publics assign to the organization facing the cri sis (Coombs & Holladay, 2003). Coombs (2007) identified three clusters that crisis types are grouped within including: victim cluster, accidental cluster, and the preve ntable or intentional cluster. Each cluster contains varying degrees of or ganizational responsibility. To connect with the various crisis types, Coombs (2007) identified four groups of response postures including: denial, diminishmen t, rebuilding, and bolstering. These four groups of responses seek to reduce the attributions ass ociated with organizational responsibility and to reduce the negative feeling generated by those attributi ons (Coombs & Holladay, 2003). The denial category of response intends to completely separate the organization from t he crisis and resulting blame. Th e strategies included in the denial posture are attack the accuser, denial, a nd scapegoating (Coombs, 2007). According to Coombs
29 clai The deni al strategy simply denies that a crisis exists. The scapegoating strategy evades responsibility of the crisis by placing on the blame on an external party (Coombs, 2007). The diminishment group of strategies seeks to reduce attribution of blame to the org anization. Excusing and justification are the communicative strategies including in the dimini shment posture (Coombs, 2007). The main objective of the excusing strategy is to diminish the perceived attribution of organizational responsibility. The purpose of the justification strategy is to minimize the perceived damage of the crisis (Coombs, 2007). reputation and its relationship wit h stakeholders (Coombs, 2007). The response stra tegies included in this posture a re compensation and apology. Compensation strategy simply offers some form of compensatio n to the victims of the crisis. The apology strategy forces the organization to publicly take responsibility for the crisis and ask fo rgiveness (Coombs, 2007). Reminding, ingratiation, and victimage are the response strategies incl uded in the bolstering posture. The reminding strategy, sometimes referred to as bolstering (Kim & Liu, in press), promotes the good deeds that he orga nization has done in the past. The ingratiation strategy compliments the involved publics of th e organization (Coombs, 2007). The utilization of the victimage strategy allows the organization to take the role of victim in the crisis (Coombs, 2007).
30 Since the conce ptualization of situational crisis communication theory, several authors have conducted studies to analyze the original response typology and expand on it (Jeon g, 2009: Kim & Liu, in press). responses to an oil sp ill in South Korea using situationa l crisis communication theory. The results of the study revealed that when past poor management history was present, people tended to have higher internal attributions, or more intense feelings toward the actors involved in the crisis rather than the organiza tion as a whole (Jeong, 2009). The study provides important implications for practice because not only does it demonstrate the type of information that elicits negative attitudes from stakeholders but it provides relev ant research for this study due to the similarities in case type. influenza pandemic which allowed for expansion of the original response strategies in situationa l crisis comm unication theory. the recommendations made by Coombs (2007), when there is little attribution to the organization facing a crisis the bolstering posture can be used alone with supplementation from other post ures (Kim & Liu, in pres s). The authors identified two response strategies used by organizations during the pandemic crisis that were not previously a part of the situationa l crisis communication theory. These response strategies a re enhancing and transfer ring. The enhancing strategy focuses on the good deeds an organization is current ly doing (Kim & Liu, in press). strategy looks only at the present. The transferring strategy is used when organizations
31 incorporate third party support in hopes of transferring the credibility of the third party on to the organization facing the crisis (Kim & Liu, in press). Research Questions and Hypotheses The goals of this study include: advancing the cross national conflict shifting theory through the examination of the Deepwater Horizon oil spill crisis, identifying the crisis response strategies covered in the media, and the crisis response strategies disseminated by Brit ish Petroleum in news releas es. Based on the review of literature above and borrowed from the hypothesis development by Molleda et al. (2011), the following research questions and hypotheses have been developed to guide this study. An explanation for hypotheses will be provided to fu rther clarify its development. RQ1: What were the main events following the initial trigger event of the explosion that characterized this crisis as a cross national conflict shift? RQ2: Who were the actors/publics involved in the cross national conflict s hift and what were their roles in the crisis? These research questions will guide the case study. Seven hypotheses will be tested through a quantitative content analysis of the news coverage produced by the Associated Press and Reuters as well as of the n ews releases disseminated by British Petroleum. H1: The international newswire agency from the host country of the conflict (the Associated Press) will produce a great number of stories and publish stories in greater length regarding the crisis. Initially, Molleda et al., (2005) found that a cross national conflict shift is more frequently reported by the home country involved in a conflict because of the impact that th e crisis has on that location. Contrary to this finding, Molleda et al. (2011) found that
32 due to the proximity of the crisis and the significant effects it had on the host country, the host country produc ed more stories on the crisis. Based on this finding, hypothesis one was developed. H2: The headline and lead paragraphs of the news coverage will feature corporate and governm ent responses most frequently in the cross national conflict shift. Molleda et al. (2005) found that government officials and transnational ss national confl ict shifting. The study conducted by Molleda (2011) confirmed this finding revealing that the international newswire agencies reporting on the Mattel crisis favored corporate and governmental responses. H3: Associations betwe en countries of origin and new sources cited will be significant which may indicate that the international newswire agencies analyzed favor different voices in coverage. Molleda et al. (2011) found that there was a significant difference in the sources cited by the Colombian media v ersu s the United States media. The host country (Colombia) favored governmental sources, while the home country (United States) favored corporate sources. H4: International newswire agencies analyzed (the Associated Press and Reuters) will make different empha sis on story focus (financial, legal, corporation accusation, recovery, consumer impact, and government). In the examination of the DuPont Teflon crisis within the context of the cross national conflict shifting theory, Wang (2005) found that story focus differed between the media coverage in China versus the United St ates. Molleda (2011) found that in the
33 Mattel crisis, international newswire agencies highlighted particular aspects of the cross national conflict shift. H5: Compensation and apology will be the cri sis response strategies most frequently covered in the media coverage. H6: Association between news sources and crisis response strategies will be significant which may indicate that news sources from different countries favor different types of co rporate response. H7: Compensation and enhancing will be the cri sis response strategies most frequently used in the corporate response from BP As previously mentioned, based on the crisis types provided by Coombs (2007), the Deepwater Horizon oil spill is identified as a preventable crisis. For this crisis type, the prescribed response strategies include: compensati on and apology (Coombs, 2007). As a supplementary response, the author recommends strategies from the bolstering posture to be used including: reminding victimage, and ingratiation. Enhancing is an additional strategy identified by Kim & Liu (in press), which can be included in the bolstering posture.
34 CHAPTER 3 METHODOLOGY This study includes a qualitative case study of the Deepwater Horizon oil spill, a quantitative content analysis of the new coverage generated by the international newswire agencies in the home and host country of the crisis, and a quantitative content analysis of the news releases disseminated by Britis h Petroleum during th e crisis. The case study methodology was selected to provide a thorough description of the crisis as well as to answer research questions regarding the case within the context of the cross nat ional conflict shifting theory. According to Wimmer and Dominick studies are conducted when a researcher needs to understand or explain a Quantitative Content Analysis of Media Coverage To test the first six hypotheses a quant itative content analysis is performed on the media cover age following the Deepwater Horizon oil spill in the United Kingdom and the United States. The content analysis methodology was selected to describe the content of the media coverage. The content analysis methodology is defined by Riffe, Lacy, and Fico (20 05) as: Quantitative content analysis is the systematic and replicable examination of symbols of communication, which have been assigned numeric values according to valid measurement rules and analysis of relationships involving those values using statisti cal methods, to describe the communication draw inferences about its meaning, or infer from communication to is context, both of production and consumption. (p. 25) Due to the global scope of the crisis, international newswire agencies were selected as the outlets that c overage would be obtained from. Based on the countries involved in the conflict, the international newswire agencies selected were the Associated Press and Reuters. These particular agencies were selected base d on the justification that the
35 Associated Press is headquartered i n the United States and Reuters in the United Kingdom, allowing representation from each country involved in the conflict. The Lexis Nexis and Factiva databases were used to collect stor ies from each newswire agency. On b The time frame used in both searches was from Apri l 20, 2010 to October 1, 2010. This time frame was selected based on the first date marking the trigger event of the cris is, the initial rig e xplosion. The last date marked the significant organization structural change in the CEO position with the resignation of Tony Hayward and the ascension of Bob Dudley. Additional criterion for selecting the sample involved tagging the latest edition of eac h article as well as excluding thos e that were deemed irrelevant. An example of an involvement in the Lockerbie controversy. Following the final exclusions, the search from both the Associated Press and Reuters resulte d in a sample of 123 articles. The sample from the Associated Press accounted for 75 articles, while there were 48 articles from Reuters. The sta tistical procedures that were used to test th e first six hypotheses inc lude descriptive statistic methods: frequency, one way ANOVA, cross tabu lations, and chi square tests. The one way ANOVA test is used to test the relationship between variables measured at the nominal and ratio levels. An example of that specific type of r elationship as it applies to this study is the relations hip between the word count o f article and the news source. Cross tabulation methods can be used to examine the relationship between variables measured at the nominal level. Examples of that type of
36 re lationship within the context of this study are the relationship between news source and sources cited in headline. Quantitative Content Analysis on News Releases To test the final hypothesis, a quant itative content analysis was performed on the news relea ses diss eminated by British Petroleum. This content analysis focused on crisis response strateg ies used by British Petroleum. The news releases were obtained from the British Petroleum press website and lf of Mexico respons The theme selections were provided by British Petroleum. The news releases were further filtered through the time frame of Apri l 21 2010 to October 1, 2010. This time frame coincided with the time frame i n the media coverage sampling and was selected for the same reasons. The sampling process resulted in a total of 80 news re leases collected for analysis. Coding Sheet Construction The coding sheet was constructed to include the variables selected for study base d on the hypotheses developed. The unit of analysis for each news story included the headline, the lead paragraph, and the source in headline, and the source s in the lead paragraph s Overall, the coding sheet included nine variables: date published, l ength (word count), country of origin, international newswire agency, focus of story, type of source cited in headline, type of source cited in lead paragraph, type of corporate response in story, and the type of corpor ate response in news releases. The va riables selected for the coding sheet were based on the hypotheses developed and the literature review.
37 The story focus was defined by six coding options including : financial, legal, British Petroleum accusation, recovery, consumer impact, and government i nvolvement. These options were coded through a quantification system of numbers one to seven, with each number identifying a certain t ype of focus. The coding options for sources cited in both the headline and lead paragraph included: transnational corpor ation ( BP ), U.S. government, U.S government official, U.K. government, U.K. government official, Non governmental organization (NGO), industry member, citizen, investor, and expert. ng For the type of corporate response, the typology of crisis response strategies in situational crisis communication theory, defin ed by Coombs (2007), was used. The coding options included: attack the accuser, denial, scapegoating excusing, justification, compensation, apology, reminding, enhancing, ingratiation victimage, and transferring. Intercoder Reliability Intercoder reliability was conducted to test the internal validity of the coding she et constructed for this study. The intercoder reliability tests were divided into two sections, a reliability test was done on the media coverage and separate one co mpleted for the news releases. Te n percent of the news articles (N = 12) were randomly sele cted for the reliability test. The reliability sample was coded independently by two coders, the author and a public relations professional. To calculate the reliability coefficient, the formula con ceptualized by Holsti (1969) was used. The formula is 2M divided by N1+N2 with M equivalent to the number of
38 coding decisions agreed upon by the two coders and N1+N2 accounting for the total nu mber of coding decisions made. Based on this formula, the reli ability coefficient was 0.94 with 438 coding decisions in agreem ent out of 468 total decisions. According to minimum reliability coefficient of about 90% or above when usin g Ho 169). Based on this assumption, the coefficient of 0.94 is a suitable indicator of the reliability of this test and allow s the research to move forward. To conduct the reliability test on the news releases, 10 % of the news releases we re collected for testing (N = 8). This reliability sample was independently coded by the same two coders. The formula conceptualized by Holsti (1969) was again used to t Based on that formula, there were 97 coding decisions ag reed upon out of a total of 104 coding decisions, resulting in a r eliability coefficient of 0.93. This coefficient was within the appropriate range outlined by Wimmer and Dominick (2006) and serves as a suitable indicato r of reliability for this test.
39 CH APTER 4 FINDINGS The case of the Deepwater Horizon oil spill is presented to provide a description of the main events and actors of the transnational crisis as they relate to the research questions developed. This case study places the crisis in context o f the cross nati onal conflict shifting theory. Following the case study, the statisti cal findings are presented. News articles collected for the content analysis were used for the case study, as well as additional stories from The Guardian The Telegraph BBC News and Bloomberg Businessweek Case Study -The Deepwater Horizon Oil Spill On April 20, 2010, the oilrig Deepwater Horizon, licensed by BP exploded killing 11 employees and injuring 17. In the days that followed, the rig began to sink into the Gulf of Mexico, continuing to spew oil into the ocean, resulting in one of the worst envi ronmental disasters on record. On April 30, BP CEO Tony Hayward acknowledged the responsibility of the organization in the crisis and promised compensation to those affect ed (Cutler, 2010) In the first few weeks of May 2010, the enormity of the consequences created by t he spill began to be realized. The first major developments included the Coast Guard announcement that oil would reach the shores of the Gulf of Mexico and t he fishing industry began to be affected by the initial restrictions imposed on fishing areas in the The efforts to mitigate the leak during this month included several types of containment domes as well as a method known as a top ki ll. All strategies f ailed to control the oil leaking into the ocean.
40 The month of May 2010 not only included the initial attempts to reduce the damage from the oil spill, but also was the first month in which CEO Tony Hayward made several responses regard ing the organization and its involvement in the spill. On May 6, 2010, Hayward made a statement that attempted to shift the blame of the crisis from BP to Tra nsocean, the owner of the rig. Later in May 2010, Hayward made two infamous statements sparking t he public relations controversy that surrounded the CEO and the orga nization throughout the summer. The statements included one comment regarding the small size of the spill in comparison to the size of the Gulf of Mexico, and in the second Hayward express The first day of June 2010, marked the commencement of a criminal investigation into the spill co nducted by the U.S. government. On June 3, 2010 BP launched an advertising campaign showcasing the and feat uring CEO Hayward prominently. As part of the continuing effort to reduce the effects of the spill, another containment cap was placed on the well but failed. On June, 8, 2010, the U.S. National Oceanic and Atmospheric Administration made its first confirmation of undersea oil coming from the damaged well Hayward vehemently denied the presence of plumes coming from t he well (Baltimore, 2010). The month of June 2010 witnessed more public relation s stumbles by the CEO. In hotographed attending a ya cht race on the Isle of Wight. His attendance at the race furthered th e outrage against him.
41 During June 2010, the first signs of the conflict shift bec ame evident in media coverage. Protests in the United States over the spill be gan to not only include the organization, but also expressed anti British sentimen products. corporate and anti oil but occasionally nationality of the corporation slips o rd said (2010, 3). Because of the significant involvement of government in the crisis, the potential conflict shift to the United Kin gdom had serious implications. what they feel are attempts to label Britain as t he villain, the blame game risks turning ). July 2010 proved to be a significant month during the aftermath of the spill for several reasons. The leaking containment cap was removed and replaced and on July 1 5, 2010, the oil flow was stopped completely for the first time since the initial explosion. On July 27, 2010, BP announced the resignation of CEO Tony Hayward and the appointment of h is successor, U.S. Bob Dudley. In the fall and winter months of 2010, s everal significant steps were made in efforts to permanently stop the oil flow and minimize th e damage created by the spill. BP issued a report in September 2010 claiming a portion of the responsibility in the spill while also identifying the organizations that shared t he remaining portion of blame. On September 19, 2010, the d amaged well was declared dead. The company saw a shift in leadership on October 1, 2010, when Bob Dudle y officially took over as CEO. In November 2010, the former CEO Hayward admitted that the crisis response from BP ted in Macalister, 2010, 2).
42 Although the repercussions of the oil spill will be f elt for years to come, the initial crisis finally came to an end in January 2011 when the White House oil commission announced that the three actors to blame for the spill were BP Transocean, and Halliburton. Statistical Findings A quantitative content an alysis was conducted to test the following hypotheses. The sample included articles from two international newswire agencies, the Associated Press and Reuters. The total sample size was 123 articles. The Associated Press published 75 articles and Reuters p ublished 48 articles. The following findings presented by each hypothesis tested. Hypothesis One It was predicted that the international newswire agency from the host country of the cross national conflict shift, the Associated Press, would produce a grea ter number of stories and produce stories greater in length based on word count. This hypot hesis was partially supported. The As sociated Press published more stories regarding the crisis ( N = 75) than Reuters (N = 48). However, based on the word count, th e articles published by Reuters were longer than those publ ished by the Associated Press. The mean word count of articles produced by Reuters (M = 542.98, SD = 239.62) was larger than that of the Associated P ress (M = 186.89, SD = 137.09). The difference i n mean length was statistically significant ( df = 1, F = 109.80, p = .000). The crisis was given more coverage, based on number of stories, in the host country of t he conflict, the United States. The crisis was given more coverage, based on word length of stories, in the home country of the conflic t, the United Kingdom. Figure 4 1 illustrates the difference in mean plots.
43 Figure 4 1. Mean plot of word count and news source of a rticle Hypothesis Two It was predicted that the headline and lead paragraphs of the media coverage would feature corporate and government act ors most frequently in the crisis. This hypothesis was supported. In the total headlines, the corporate actors were most frequently m entioned, accounting for 58 % of headlines (N = 71) followe d by the U.S. govern ment with 17 % (N = 21). U.S. government officials were used as sources in 11 % of headlines (N = 13), followed by BP executives (7%, N = 9). Sources from NGOs were used in 4 % of headlines (N = 5) as well as oil industry members (4%, N = 5). The United Kingdom was used as a source in 2 % of headlines (N = 3) equaling the source frequency of U.S. citizens (2%, N = 3). The least sources mentioned in headlines were investors and
44 experts, each accou nting for 2 % (N = 2). Figure 4 2 illustra tes the frequency of actors in headlines below. Figure 4 2. Frequency of actor in headlines In the total leads, corporate response was the most frequent of mentions accounting for 76 % (N = 94), followed by the U.S. government officials accou nting for 2 2 % (N = 27). The source of the U.S. government was used in 20 % of leads (N = 24). British Petroleum executives were mentioned in 7 % of the leads (N = 9). NGO sources as well as sources from the oil industry were used equally, each in 5 % of the leads (N = 6). The U.K. government was used in 3 % of the leads (N = 4), while U.S. citizens were used as a sour ce the same amount (3%, N = 4). Experts were mentioned in 2 % of the leads (N = 3). United Kingdom officials were mentioned in 2 % of the leads (N = 2), equaling the source frequency of investors (2%, N = 2). Figure 4 3 illustrates the frequency of actor in lead paragraphs below.
45 Figure 4 3. Frequency of actor in lead paragraphs Hypothesis Three It was predicted that the association between news source and source used would be statistically significant, indicating that news sources would favor different voices in headlines and leads. This hypothesis was not supported. Although, cross tabulation showed some differences in sources used in headlines and le ads, chi square tests deemed these associations non significant. For the stories published by the Associated Press, the dominant sources used in headlines were BP in 37 % of headlines (N = 46), the U.S. government (11%, N = 14), and U.S. gove rnment officia ls (8%, N = 10). British Petroleum executives were used as sources in 7 % of headlines (N = 8). NGO sources as well as members of the oil industry were used equally, each mentioned in 2 % of headlines (N = 2). The source of U.S. citizen s was mentioned in 0 .8 % of headlines (N = 1). Three categories of sources
46 were not mentioned in the headlines produced by the Associated Press including: the U.K. government, U.K. government officials, and experts. For the stories published by Reuters, BP was mentioned most frequently in headlines accounting for 20 % of headlines (N = 25), followed by th e U.S. government (6%, N = 7). The U.K. government was used as a source in 2 % of headlines (N = 3), equally the frequency of the use of U.S. gov ernment officials (2%, N = 3). Sources from NGO s and the oil industry members were used equally in headlines, each mentioned in 2 % of headlines (N = 3). The source of U.K. government official s was used in 2 % of headlines (N = 2). U.S. citizens and experts were each also used in the s ame am ount of headlines (2%, N = 2). The least frequently used sources in headlines by Reuters were British Petroleum executives and investors, each in 0.8 % of headlines (N = 1). The Associated Press and Reuters were similar in their use of BP the U.S. g overnment, and U.S. government officials in headlines. One main difference was the lack of mention by the Associated Press of the U.K. government or officials. Also, the Associated Press used BP executive s more frequently than Reuters. Sources from NGOs, the oil industry, or experts were used more often in headlines produced by Reuters. Hypothesis Four It was predicted that the international news agencies analyzed would make diff erent emphasis on story focus. This hypot hesis was partially supported. For t he total articles in the sample, the financial story focus was the most frequent (N = 43), followed by government (N = 24), recovery, (N = 22), British Petroleum accusation (N = 19),
47 consumer impact (N = 11), and legal (N = 4). Figure 4 4, shown below, ill ustrates the frequency of article focus among the sample. Figure 4 4. Frequency of article focus For stories published by the Associated Press, the financial focus was used in 20 % of the total articles (N = 25), followed by the recovery and BP accusati on foci, each in 11 % of total articles (N = 14). The government focus was used in 9 % of articles (N = 11). Consumer impact was the focus in 6 % of articles (N = 7). The least frequent article focus used by the Associated Press was legal (3%, N = 4). For articles published by Reuters, the financial focus was again most dominant, used in 15 % of stories (N = 18). The government focus was used in 11 % of articles (N = 13), followed by t he recovery focus (7%, N = 8). The least frequently used story foci in a rticles published by Reuters was consumer impact (3%, N = 4) and legal, which was used in zero articles. The rankings of story foci between the two new s sources were rather similar. The main dif ference lied in the heavier use of th e government focus by R euters.
48 Proportionally, Reuters used the British Petroleum accusation l ess than the Associated Press. This could be attributed to the proximity of the crisis bei ng farther away from the home country of British Petroleum, so less blame is placed on the corp oration. Hypothesis Five It was predicted that the compensation and apology crisis response strategies would be the most prominent strategies covered in the media by the international newswire agencies selected for analysis. This hypothesis was supported. In the total articles, the most frequent crisis response strategy covered was enhancing, covered in 33 % of the articles (N = 41). The compensation strategy was covered in 29 % of articles (N = 35). These two strategies were covered much more frequently t han any of the other strategies. The excusing strategy was covered in 5 % of articles (N = 6), followed by the justi fication strategy (4%, N = 5). The scapegoating and transferring strategies were each covered in 2 % of the articles (N = 3). The attack the accuser, denial, and ingratiation strategies were each used in 0.8 % of the articles (N = 1). The reminding strategy was not covered in any of the articles. Figure 4 5, shown below, illustrates the frequency of crisis response strategies covered by the tw o international newswire agencies.
49 Figure 4 5. Frequency of crisis response strategies in media coverage Hypothesis Six I t was predicted that the association between news source and crisis response strategy would be significant, indicating that the news sources preferred to cover differe nt strategies in its coverage. This hypothesis was not supported. Chi square tests deemed the association between news source and crisis response strategy non significant. Although there was no statistically significant a ssociation, there was a slight difference between the crisis response strategies used by the two news sources. For the Associated Press, enhancing strategy was covered most frequently in 24 % of the articles (N = 29), followed by the compen sation strategy (21%, N = 26). The excusing strategy was covered in 4 % of the articles (N = 5), followed by the justi fication strategy (3%, N = 4). The scapegoating (N = 3) and victimage strategies (N = 2) were each covered i n 2 % of the articles. The transferring and in gratiation strategies were
50 each covered in 0. 8 % of articles (N = 1). The attack the accuser, denial, apology, and reminding strategies were not covered by the Associated Press. For Reuters, the enhancing strategy was covered in 10 % of the articles (N = 1 2), followed by the comp ensation strategy (7%, N = 9). The transferring strategy was covered in 2 % of the articles (N = 2). The attack the accuser, denial, excusing, and justification strategies were each covered in 0. 8 % of articles (N = 1). Both the Ass ociated Press and Reuters covered the enhancing and compensat ion strategies most frequently. Reuters covered the attack the accuser, denial, and transferring strategies more freque ntly than the Associated Press. The scapegoating strategy was covered more f requently by the Associated Press rather than Reuters. Hypothesis Seven I t was predicted that BP would most often use the compensation and enhancing strategies in their corporate response to crisis. This hypothesis was supported. The crisis response strat egies used by the corporation were identified through the news releases publ ished on the corporate website. The enhancing strategy was used in 81 % of news releases disseminated by BP (N = 65), followed by the compen sation strategy (65%, N = 52). The ingra tiation strategy was present in 15 % of the news releases (N = 12), followed by the transf erring strategy (13%, N = 10). The excusing strategy and the justification strategy were equally used in the news releases, each being found in 8 % of news releases ( N = 6). The reminding strategy was used in 5 % of the news releases (N = 4), followed by the v ictimage strategy (4%, N = 3). The scapegoating strategy was found in 3 % of articles (N = 2), followed by t he denial strategy (1%, N = 1). The attack the accuser and apology strategies were not used by the organization in its news releases. Figure 4 6,
51 shown below, illustrates the frequency of crisis response strategies used in news releases disseminated by British Petroleum. Figure 4 6. Frequency of crisis res po nse strategies in news releases When the crisis response strategies used by the organization in news releases was compared to the crisis response strategies covered by the media, there were some similarities and differences. The attack the accuser strate gy was covered by the media but was not used by the organization in its corporate The reminding strategy was used by the organization but was not covered in the media. The ingratiation strategy was used by the organization, but was m inimally covered in th e media. The apology strategy was not present in the corporate response or in the media coverage. In Chapter 5 the findings will be discussed and interpreted. The implications of this study to practice as well as to the ory building will be discussed. The limitations of this study and suggestions for further research will also be made in Chapter 5
52 CHAPTER 5 DISCUSSION Summary and Interpretation of Findings The case of the Deepwater Horizon oil spill has been described within the context of the cross nati onal conflict shifting theory and it has been shown that the crisis faced by British Petroleum started in the United States and spread to the United Kingdom. There were similarities and differences in the media coverage disseminated by the two countries th rough the newswire agencies selected for analysis, the Associated Press and Reuters. The Associated Press published a greater number of stories, although Reuters published stories in greater length. ANOVA testing showed significant differences in mean word length between the two news sources. Reuters also used a greater variety of sources in both its headlines and lead paragraphs of its articles. The Associated Press and Reuters both focused its headlines and leads on government and corporate sources. Reute rs featured a variety of source use in its coverage through the heavier incorporation of sources from NGOs, industry members, experts, investors, and U.S. citizens. The greater length of articles and the variety of sources used indicated that Reuters publi shed stories in more detail on the crisis. There was a minimal U.K. government source presence in the headlines and leads of the articles published by both the Associated Press and Reuters. This minimal presence indicated that the government of the host co untry, the United States, garnered more media coverage. The difference in government media presence could be attributed to the proximity of the crisis to the host country and the government intervention into the crisis by the host country. The U.S. governm ent was under great
53 pressure to defend its citizens and industries during this crisis due to the breadth of the Although government and corporate sources were the most widely used by the two news sources there were some differences in so urce use. The differences were made evident by the variety of sources used by Reuters as compared to the sources used by the Associated Press. These differences were congruent with previous findings in studies conducted on cross national conflict shifts an d indicate d that different news sources may favor different voices. Unlike previous studies, statistical testing deemed this differences to be non significant as they were slight differences. The non significant results of testing could be attributed to th e smaller sample size used for analysis and the similarities in media coverage between the two news sources. Regarding article focus, there were differences in the article foci used by the news sources. The most prominent article focus used by both news so urces was the financial focus. The main differences in focus use centered on the foci of British Petroleum accusation and consumer impact. The Associated Press focused more on the article foci of British Petroleum and consumer impact than Reuters. Once aga in, this difference could be attributed to the proximity of the crisis to the host country. Overall, consumers in the United States felt a greater impact than those in the United Kingdom as the crisis affected a multitude of industries in the host country. In regards to the corporate crisis response, crisis response strategies were analyzed in the media coverage as well as in the news releases disseminated by British Petroleum. There were slight differences in the coverage of crisis response strategies in t he media between the two news sources. The Associated Press and Reuters focused
54 the media coverage on the enhancing and compensation strategies used by British Petroleum. Reuters covered the transferring, attack the accuser, denial, and ingratiation strate gies more heavily than the Associated Press. In stories published by the Associated Press, the excusing, justification, scapegoating, and victimage strategies were covered more heavily as compared to the stories published by Reuters. The corporate response through news releases disseminated by the corporation was analyzed for crisis response strategies. British Petroleum was consistent with crisis response messaging utilizing most frequently the enhancing and compensation strategies in their response. This finding was congruent with the recommendations made by Coombs (2007), suggesting that compensation strategy along with rebuilding s trategies be used for preventable crises. In this case, the most frequently used strategy was enhancing which Kim & Liu (in p ress) identified as an additional strategy to the rebuilding category of response. The ingratiation and transferring strategies f ollowed in frequency of usage. There were similarities and differences between the strategies disseminated by the organization and those covered by the media. The media coverage published by both news sources was congruent with the corporate response in covering the enhancing and compensati on strategies most frequently. British Petroleum did not use the attack the accuser strategy and minimally used the denial strategy in its corporate response. However, the media covered the use of these strategies by the corporation. British Petroleum used the ingratiation strategy more frequently than it was covered in the media. These findings indicate d that the
55 media may cover the corporate response in a crisis differently than intended by the organization thus, posing a threat to the reputation of the organization. Implications for Public Relations Practice This study provided reinforcement fo r public relations practitioners of transnational corporations to develop communication strategies and messages that cater to different cultures and countries. Additionally, a common communication objective between countries should be followed. The finding s of this study illustrated the differences in media coverage that can arise between two countries when covering a cross national conflict shift including: frequency of articles, article length, ar ticle focus, and sources used. Primary and secondary source s used in the media coverage differed between countries, indicating that media have different preferences when it comes to source use in coverage. In this particular case, both newswire agencies used the same primary sources but the secondary sources diffe red in variety and frequency between the agencies. Public relations practitioners must be aware of this potential difference in coverage and identify the key publics to the organization. The focus of media coverage is another consideration to be made in th e practice of international public relations as it may differ across borders. In this study, findings showed that the host country focused more on potentially negative foci related to the organization. This was attributed to the proximity of the crisis to the host country. Public relations practitioners need to consider the potential difference in article focus at is may influence the reputation of the organization and increase the need for follow up strategies to mitigate potential reputational damage.
56 In this particular case, the response disseminated directly from the corporation was consistent. The media covered the dominant strategies used by the organization but also covered strategies that were used minimally or not used at all by the organization. Th ese findings indicate d that the media may not necessarily cover the response disseminated directly from the corporation as it was intended. Public relations practitioners should be aware of this issue as the media coverage has the potential to misrepresent Theoretical Implications The case of the Deepwater Horizon oil spill is an example of a cross national conflict shift and this study provided further support for the cross national co nflict shifting theory. This case was justified as a cross national conflict shift through the application of the theoretical propositions to the case and the subsequent case study that described the crisis within the context of the theory. This was the fi rst cross national conflict shifting study to focus on the Deepwater Horizon oil spill and to analyze the combination of news sources from the United States and the United Kingdom. The findings of this study were congruent to several of the previous cross national conflict shifting studies. The case study and statistical findings illustrated all of the intricacies that are included in a cross national conflict shift including the media coverage across borders, the corporate response and the publics involved This study depicted the connection between these three aspects of a cross national conflict shift and contributed to gl obal public relations research. This study also provided theoretical implications for the growing body of knowledge in crisis response. The findings supported the situational crisis communication theory and the additional response strategies that have been suggested
57 to this theory. These findings indicated that research on crisis response strategies outside of the traditional theory may b e useful as it provides new insight into the situational crisis communication theory. Limitations and Further Research The limitations of this study mainly pertain to the timeframe of the crisis. The Deepwater Horizon oil spill is an ongoing crisis and wil l last well beyond the timeframe selected for study. This timeframe of study was an appropriate selection but caused the long term depth of the crisis to not be included. The use of a quantitative content analysis methodology also prese nted limitations to the study. The disadvantage of conducting this kind of study quantitatively is that the results may lack the depth that it would otherwis e have in a qualitative study. The majority of previous cross national conflict shifting studies have utilized quantita tive content analysis method s. It is suggested based on this limitation that qualitative studies on the cross national conflict shifting theory be conducted to c ontribute to theory building. Based on the timeframe li mitation described follow up studies on the media coverage and corporate response of this particular case would be useful to capture possible changes over time. Additional news source types including newspapers and television may provide a more detailed understanding of the media coverage of th is case. The selection of additional corporate response resources beyond the news releases may also provide a wider scope to the study of crisis response. In regards to the use of additional news source types and corporate response resource s it is importan t to consider emergent channels of communi cation including social media. This channel of communication is providing a new dynamic approach for
58 organizations publics to communicate regarding an issue and allows the opportunity f or the corporation to respon d. Social media has the capabilities to create a conversation between an organization and its publics without the limitations of time or geographic borders. For that reason, social media should be a consideration to be used in future cross national conflic t shifting studies.
59 APPENDIX A ARTICLE CODING SHEET Variable Name Variable Label Value Label ID Identification Number 0=122 DATE Story Date MM/DD/YYYY LENGTH Word Count 0000 9999 ORIGIN Country of Origin of News Source 1=USA 2=UK NEWSSOURCE News S ource Type 1=AP 2=Reuters FOCUS Story Focus 1=Financial 2=Legal 3=BP Accusation 4=Recovery 5=Consumer Impact 6=Government TNCEXHL TNC executive mentioned in headline 0=NO 1=YES TNCHL TNC mentioned in headline 0=NO 1=YES USGOVHL United States Government body mentioned in headline 0=NO 1=YES UKGOVHL United Kingdom Government body mentioned in headline 0=NO 1=YES USGOVOHL United States Government Official mentioned in headline 0=NO 1=YES UKGOVOHL United Kingdom Government Official mentioned in headline 0=NO 1=YES NGOHL NGO mentioned in headline 0=NO 1=YES INDMEMHL Industry member mentioned in headline 0=NO 1=YES CITIZENHL U.S. citizen mentioned in headline 0=NO 1=YES INVESTORHL Investor mentioned in HL 0=NO 1=YES EXPERTHL Expert mentioned in HL 0=N O 1=YES TNCEXLP TNC executive mentioned in lead paragraph 0=NO 1=YES
60 TNCLP TNC mentioned in lead paragraph 0=NO 1=YES USGOVLP United States Government body mentioned in lead paragraph 0=NO 1=YES UKGOVLP United Kingdom Government body mentioned in lead paragraph 0=NO 1=YES USGOVOLP United States Government Official mentioned in lead paragraph 0=NO 1=YES UKGOVOLP United Kingdom Government Official mentioned in lead paragraph 0=NO 1=YES NGOLP NGO mentioned in lead paragraph 0=NO 1=YES INDMEMLP Industry member mentioned in lead paragraph 0=NO 1=YES CITIZENLP U.S. citizen mention in lead paragraph 0=NO 1=YES INVESTORLP Investor mentioned in lead paragraph 0=NO 1=YES EXPERTLP Expert mentioned in lead paragraph 0=NO 1=YES ATTACK Organization confronts t he person/group that claims crisis exist 0=NO 1=YES DENIAL Organization denies crisis exists 0=NO 1=YES SCAPEGOATING Organization evades responsibility by placing crisis blame on another party 0=NO 1=YES EXCUSING Diminish perceived responsibility attri bution of crisis 0=NO 1=YES JUSTIFICATION Minimize perceived damage of crisis 0=NO 1=YES COMPENSATION Organization offers compensation to victims 0=NO 1=YES APOLOGY Organization claims 0=NO
61 responsibility for crisis and asks for forgiveness 1=YES REMIN DING Organization reminds publics of past good deeds 0=NO 1=YES ENHANCING Organization discusses current good deeds 0=NO 1=YES INGRATIATION Organization praises the involved publics in crisis 0=NO 1=YES VICTIMAGE Organization plays role of victim in cr isis 0=NO 1=YES TRANSFERRING Organization uses third party support in hopes of transferring third party credibility on to them 0=NO 1=YES
62 APPENDIX B NEWS RELEASE CODING SHEET Variable Name Variable Label Value Label DATE Story Date MM/DD/YYYY ATTAC K Organization confronts the person/group that claims crisis exist 0=NO 1=YES DENIAL Organization denies crisis exists 0=NO 1=YES SCAPEGOATING Organization evades responsibility by placing crisis blame on another party 0=NO 1=YES EXCUSING Diminish perc eived responsibility attribution of crisis 0=NO 1=YES JUSTIFICATION Minimize perceived damage of crisis 0=NO 1=YES COMPENSATION Organization offers compensation to victims 0=NO 1=YES APOLOGY Organization claims responsibility for crisis and asks for for giveness 0=NO 1=YES REMINDING Organization reminds publics of past good deeds 0=NO 1=YES ENHANCING Organization discusses current good deeds 0=NO 1=YES INGRATIATION Organization praises the involved publics in crisis 0=NO 1=YES VICTIMAGE Organization plays role of victim in crisis 0=NO 1=YES TRANSFERRING Organization uses third party support in hopes of transferring third party credibility on to them 0=NO 1=YES
63 LIST OF REFERENCES Allen, M.W. & Calliouet, R. H. (1994). Legitimation endeavors: Impre ssion management strategies used by an organization in a crisis. Communication Monograph, 61 (1), 44 62. Assinder, N. (2010 June 11). Britain to Obama: Stop Bullying Us Over BP. TIME Retrieved August 10, 2011, from http://www.time.com/world/article/0,8599,1996056,00.html Baltimore, C. (2010, June 8). Update 3 U.S. experts confirm oil from BP well. Reuters. Retrieved August 9, 2011, from http://www.reuters.com/article/2010/06/08/oil spill noaa idUSN0825637820100608 Benoit, W. L. (1997). Image repair discourse and crisis communication Public Relations Review, 23 (2), 177 186. Benoit, W. L., & B Insensitive royal or compassionate queen? Public Relations Review, 25 (2), 145 146. Hurricane Katrina. Public R elations Review, 35 40 46. BP oil spill timeline. (2010, July 22). The Guardian Retrieved February 18, 2011, from http://www.guardian.co.uk/environme nt/2010/jun/29/bp oil spill timeline deepwater horizon British Petroleum. (2010 April 21). BP confirms that Transocean Ltd. issued the following statement today [Press Release]. Retrieved August 9, 2011, from http://www.bp.com/genericarticle.do?categoryId=2012968&contentId=7061443 British Petroleum. (2010 August 23). BP reports nearly $400 million in claim payments as program transitions to GCCF [Press Release]. Retrie ved August 9, 2011 from http://www.bp.com/genericarticle.do?categoryId=2012968&contentId=7064597 Coombs, W. T. (1995). Choosing the right words: the development of gu idelines for the response strategies. Management Communication Quarterly, 8 (4), 447 476. Coombs, W. T. (2007). Ongoing crisis communication: Planning, managing, and responding Thousand Oaks, CA: Sage Publications, Inc Coombs, W. T. (2007). Protecting organization reputations during a crisis: The development and application of situational crisis communication theory. Corporate Reputation Review, 10 163 176.
64 Cutler, D. (2010, June 3). Timeline Gulf of Mexico oil spill Reuters. Retrieved February 18, 2010, from http://www.reuters.com/article/2010/06/03/oil spill events idUSN0322326220100603 Deepwater Horizon Incident Joint Information Center. (2010). Consolidated fish and wildlife collection report. Retrieved August, 9 2011, from http://www.restorethegu lf.gov/sites/default/files/documents/pdf/Consolidated%20W ildlife%20Table%20110210.pdf Deepwater Horizon Incident Joint Information Center. (2011, April 19). Last fisheries re opening today [Press Release]. Retrieved August 9, 2011, from http://www.restorethegulf.gov/release/2011/04/19/last fisheries re opening today Efstathiou, J., & Plungis, J. (2010, June 7). Hayward slammed by lawmakers for stonewalling on s pill (Update 1). Bloomberg Businessweek Retrieved February 19, 2011, from http://www.bloomberg.com/apps/news?pid=2065100&sid=ato9epKGrQ14 Experts double estimate BP oil spill size. (2010, June 11). BBC. Retrieved August 10, 2011, from http://www.bbc.co.uk/news/10290238 water separation machines help with clean up. T he Guardian. Retrieved August 9, 2011, from http://www.guardian.co.uk/environment/2010/jun/16/kevin costner oil spill machines Grant, S. (2010, June 7). Can the BP brand survive Tony Hayward? PRSAY. Retrieved September 7, 2010 from, http://prsay.prsa.org/index.php/2010/06/07/can the bp brand survive tony hawyard/ Harlow, W. F., Brantley, B. C., & Harlow, R. M. (2011). BP initial image repair strategies after the Deepwater Horizon spill. Public Relations Review, 37 80 83. Hearit, K. M. (1996). The use of counter attack in apologetic public relations crises: The c ase of General Motos vs. Dateline NBC. Public Relations Review, 25 (3), 291 308. Hearit, K. M., & Brown, J. (2004). Merril Lynch: Corporate apologia and business fraud. Public Relations Review, 30 459 466. Hinckley, D. (2010, June 2). James Cameron lends h and to help solve Gulf oil spill at meeting with scientist and engineers. NY Daily News Retrieved August 9, 2011, from http://articles.nydail ynews.com/2010 06 02/news/27066022_1_oil spill scientists and engineers gulf Hofstede, G. (2001). Thousand Oaks, CA: Sage Publications, Inc.
65 Holsti, O. (1969). Content analysis for the social sciences and humanities. Reading, MA: Addison Wesley. BBC News. Retrieved February 18, 2011, from http://www.bbc.co.uk/news/10282448 Jarvis, A. A. (2010, September 14). BP Oil Spill: Disaster by Numbers. The Independent Retrieved August 10, 2011, from http://www.independent.co.uk/environment/bp oil spill disaster b y numbers 2078396.html : A test of the attribution theo ry and situational crisis communication theory. Public Relations Review, 35 (3), 307 309. Kim, J. R., & Molleda, J. C. (2005, March). Cros s national conflict shifting: an analysis of Paper presented at the 8 th International Public Relations Research Conference in Miami, FL. Kim, S., & Liu, B. F. (in p ress). Are all crises opportunities? A comparis on of how corporate and government organizations responded to the 2009 flu pandemic. Manuscript accepted to the Journal of Public Relations Research. Lim, H., & Molleda, J. C. (2009, May). The influence of a cross national conflict shift on a transnationa Paper presented at the International Communication Association 59 th annual conference, Division of Public Relations, Chicago, IL. The Guardian. Retrieved February 18, 2011 from http://www.guardian.co.uk/environment/2010/nov/11/tony hayward bp oil spill Molleda, J. C. (2010). Cross national confl ict shifting: a transnational crisis perspective in global public relations. In R. L. Heath (Ed.), The SAGE Handbook of Public Relations (pp.679 690). Thousand Oaks, CA: Sage Publications, Inc. Molleda, J. C. (2011). Advancing the theory of cross national conflict shifting: a case coverage. International Journal of Strategic Communication, 5 (1), 49 70. Molleda, J. C., & Connolly Ahern, C. (2002, August). Cross national conflict shifting: Conceptualization and expansion in a n international public relations context. Paper presented to the convention of the Association for Education in Journalism and Communication, Miami, FL. Molleda, J. C., Connolly Ahern, C., & Quinn, C. (2005). Cross national conflict shifting: Expanding a theory of global public relations management through quantitative content analysis. Journalism Studies, 6 (1), 87 102.
66 Molleda, J.C., & Laskin, A. (2010). Coordination and control of global public relations to m anage cross national conflict shifts: A multidisciplinary perspective for research in practice. In G. J. Golan, T. J. Johnson, & W. Wanta (Eds.), International media communication in a global age (pp. 319 344). New York, NY: Routledge. Molleda, J. C., & Qu inn, C. (2004). Cross national conflict shifting: A global public relations dynamic. Public Relations Review, 30 1 9. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling (2011, January 11 ). Deep Water: The Gulf Oil Disaster and th e Future of Offshore Drilling. Retrieved August 9, 2011, from http://www.oilspillcommission.gov/sites/default/files/documents/DEEPWAT ER_Re porttothePresident_FINAL.pdf Reuben, A. (2010, June 10). Why is BP important to UK economy ? BBC News. Retrieved October 15, 2011, from http://www.bbc.co.uk/news/10282777 Riffe, D., Lacy, S., &Fico, F G. (2005). Analyzing media messages: Using quantitative content analysis for research. Mahwah, NJ: Lawrence Erlbaum Associates. Taylor, M. (2000). Cultural variance as a challenge to global public relations: A case study of the Coca Cola scare in Europe. Public Relations Review, 26 277 293. Transocean. (2010 April 20). Transocean Ltd. reports fire on semisubmersible drilling rig Deepwater Horizon [Press Release]. Retrieved August 9, 2011, from http://www.deepwater.com/fw/main/News 748.html?c=113031&p=irol news&nyo=0 Wang, Y. (2005). Cross national conflict shifting: a case study of the DuPont Teflon crisis. Unpublished masters thesis, University of Florida, Gainesvil le, FL. situation worse. The Guardian. Retrieved February 18, 2011 from http: //www.guardian.co.uk/business/2010/jun/01/bp response oil spill tony hayward Wimmer, R. D., & Dominick, J. R. (2006). Mass Media Research. Boston, MA: Wad s worth.
67 BIOGRAPHICAL SKETCH Sarah Gravina enrolled in the Burnett Honors College at the Universit y of Central Florida and received her B.A. in Advertising/Public Relations in 2009. Her interest in international public relations was developed during study abroad programs in Spain and Ireland. In the f all of 2009, she was admitted to the graduate progra m in the College of Journalism and Mass Communication at the University of Florida. In the summer of 2011, she completed an internship in the public affairs department at Mayo Clinic in Jacksonville, Florida and had the opportunity to apply her academic ex perience with her participati on in a variety of internal external and crisis communication activities In December 2011, she graduated with her Master of Arts in Mass Communication with a specialization in public relations. Her research interests include international public relations, crisis communications, and health/science communications. After graduation, she plans to pursue a career in the health communications field.