|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Responding to Crises: A Test of the Situational Crisis Communication Theory by Courtney Wright A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts School of Mass Communications College of Arts and Sciences University of South Florida Major Professor: Kelly Werder, Ph.D. Kenneth Killebrew, Ph.D. Randy Miller, Ph.D. Date of Approval: December 5, 2008 Keywords: attributi on theory, corporate apologia crisis communication management, image repair discourse, neoinstitutionalism theory Copyright 2009, Courtney Wright
i Table of Contents List of Tables iii List of Figures iv Abstract v Chapter One: Introduction 1 Background 2 Purpose 2 Crisis Defined 2 Chapter Two: Literature Review 5 Crisis Communication and Crisis Management 5 Crisis Communication Management in the Beginning 6 Johnson & Johnson Tylenol Case 7 Exxon Valdez Oil Spill Case 8 Situational Crisis Communication Theory (SCCT) 10 Historical Development of SCCT 12 Benoits Image Repair Discourse 12 Image Repair Theory and Crisis Management 17 Hearits Apologia 19 SCCTs Current Articulation 21 Attribution Theory 21 Neoinstitutionalism Theory 24 The Crisis Situation 25 Crisis Response Strategies 27 Modeling the Process 29 Matching Crisis Situations to Crisis Response Strategies and Limitations 32 Terminological Changes 33 Research-Driven Changes 34 Testing the SCCT 34 Purpose and Hypotheses 37 Chapter Three: Methodology 39 Research Participants 39 Design and Materials 39 Instrumentation 40 Organizational Reputation 42 Crisis Responsibility 42 Potential Supportive Behavior 42
ii Procedures 43 Chapter Four: Results 44 Study of Participants 44 Measures of Variables of Interest 45 Test of Hypotheses 48 Exploration of Indivi dual Item Differences 49 Chapter Five: Discussion 57 Future Research 60 Conclusion 61 References 62
iii List of Tables Table 1 Sex and academic year of participants 45 Table 2 Cronbachs alpha for multiple-item indexes 46 Table 3 Organizational Reputation descriptive 46 Table 4 Crisis Respon sibility descriptive 46 Table 5 Organizationa l Control descriptive 47 Table 6 Potential Supportiv e Behavior descriptive 47 Table 7 Overall Effectiveness descriptive 47 Table 8 Mean and standard deviation for composite measures 48 Table 9 Individual item m ean and standard deviation 49 Table 10 Independent Sample t -tests 53
iv List of Figures Figure 1 Crisis types by level of crisis responsibility 26 Figure 2 Crisis response strategies by postures 28 Figure 3 Situational Crisis Communication Theory model 30 Figure 4 List of crisis response recommendations 31
v Responding to Crises: A Test of the Situational Crisis Communication Theory Courtney Wright ABSTRACT Crisis management includes efforts desi gned to prevent and to detect potential crises, and to learn from crisis experiences. The SCCT posits that ce rtain crisis responses (matched) produce better outcomes for organizations than others (unmatched), depending on the situation. In addition, the results from this study attempt to support the situational crisis communication theory in aiding crisis managers in protecting their organizations against crises.
1 Chapter One Introduction Crisis communication management is a public relations function that is increasingly important to the achievement of organizational goals. Media coverage of organizations facing crisis situations is abunda nt. During the next fi ve years, 83 percent of companies will face a crisis that will negatively impact the profitability of the company by 20 to 30 percent, according to new research by Oxford-Metrica, an independent adviser on risk, value, reputat ion and governance (Aon, 2006). For instance, the 1989 Exxon Valdez crisis cost the company billions. It is estimated that some Exxon station sales decreased by 30 percent. Th e company saw a drop of 43 percent from its 1988 profits as a result of the oil spill. In addition, th e net income per share also decreased from $1.06 to $.37 (Small, 1991). An organization may encounter a variety of crisis situations. According to Pearson and Clair (1998), a pl ethora of potential crises exist, including corporate misdeeds, product tampering, and environmental and natural disasters, to name a few. The Institute for Crisis Management stat es that crisis-prone industries include medical/surgical manufacturers, software ma nufacturers, pharmaceutical manufacturers, telecommunication companies, computer ma nufacturers, commercial banks, solid waste disposal companies, security and commodity brokers, life insurance companies, and the airline industry. Therefore, e ffective crisis management is important to a broad range of organizations.
2 Background A review of literature indicat es that minimal attention ha s been devoted to crisis management theory. According to Coombs (2008), theory development in crisis communication is behind general public re lations theory development because this specialized area of inquiry is still in its theoreti cal infancy, despite its importance in practice. This is supported by Seegar, Sell now, and Ulmer (1998), who found a lack of theory-based approaches in reviews of crisis communication literature. Crisis communication has created a la rge body of practical resear ch, but scant theory has emerged (Coombs, 2007). Purpose The purpose of this study is to further cu rrent theory-driven research in public relations by examining the per ception of the effect crisis management strategies have on organizational reputation, crisis responsibil ity, and potential suppor tive behavior of an organization. Specifically, this study seeks to test the situational crisis communication theory to determine its ability to predict effective crisis response strategies. Crisis Defined To better understand the current state of crisis management research, it is necessary to review the development of this area of public relations scholarship. To date, scholars have developed commonalities regarding cr isis situations that are used to define the concept. For example, Fearn-Banks (1996) defines a crisis as a major occurrence with a potentially negative out come affecting an organizatio n, company, or industry, as well as its publics, products, services, or good name (p. 1). Sociologist R.L. Hamblin (1958) argues a crisis is an urgent s ituation in which all group members face a
3 common threat (p. 322). According to Pa uchant and Mitroff (1992), a crisis is a disruption that physically affects a system as a whole a nd threatens its basic assumptions, its subjective sense of self, its existential core (p. 12). Fink (1986) claims a crisis is any event that may escalate in intensity; fall under close me dia and government scrutiny, interfere with normal business operations, and may affect the image and bottom line of a company. In 1993, Barton stated that a crisis [was] a major, unpredic table event that has potentially negative results. The event a nd its aftermath may significantly damage an organization and its employees, products, se rvices, financial cond ition, and reputation (p. 2). Lerbinger (1997) defines a crisis as an event that br ings, or has the potential for bringing, an organization into di srepute and imperils its futu re profitability, growth, and possibly its very survival (p. 4). Finally, Pe arson and Clair (1998) vi ew a crisis as a low-probability, high-impact even t that threatens the viabilit y of the organization and is characterized by ambiguity of cause, effect, and means of resolution, as well as by a belief that decisions must be made swiftly (p. 60). Even though these definitions are slightly different, a closer examination reveals numerous similarities. According to King (2002), a crisis has three primary characteristics. First, a crisis is an unpla nned event that has the potential to dismantle the internal and external st ructure of an organization. Second, a crisis can occur at any time. Finally, a crisis has the potential to affect th e legitimacy of an organization. The media can influence public perception in regards to issues involving cause, blame, response, resolution, and consequences. When an organi zation is presented in a negative light, its legitimacy may be threatened (King, 2002; Ray, 1999).
4 Once an organization is presented in a negative light, the probability of the organization surviving the crisis is dr amatically reduced. Fearn-Banks (1996) emphasizes, in her seminal work on crisis co mmunication, the need for organizations to develop and implement effective crisis management and communication plans. This study attempts to contribute to the current theory-driven research in crisis communication by examining crisis response strategies from a communication-centered perspective. Specifically, this study s eeks to further understanding of crisis communication by examining the effect of a crisis when a matched or unmatched response strategy is used by an organization faci ng a crisis due to an accident. The crisis type as well as the strategy used in this st udy is derived from Coom bs situational crisis communication theory. This theory and its origins will be introduced and discussed in the following literature review.
5 Chapter Two Literature Review What is to be communicated and by whom within an organization are important factors in crisis communication. Crisis communication is the communication between the organization and its pub lics prior to, during, and af ter the negative occurrence (Fearn-Banks, 1996, p. 2; King, 2002). This definition of crisis communication will be the referenced definition th roughout this study. However, in the next paragraph, several versions of the definition will be provided to show different perspectives. During the communication phase, the organiza tion must appear to be in control to members external to the [organization] (Heath, 1994; King, 2002). Such behavior will direct stakeholders physical and psychological responses, as well as impressions a bout the organization (King, 2002; Ray, 1999). To do this, crisis managers may use an array of response strategies, which will be disc ussed later in the chapter. Crisis Communication and Crisis Management Crisis management differs from crisis co mmunication in the fact that it represents a systematic attempt by organizational memb ers with external stakeholders to avert crises or to effectively manage those that do occur (Pearson & Clair, 199 8). In this case, the organization and members of the crisis ma nagement team attempt to remove some of the risk and uncertainty that would not allow the organization to be in control of its own destiny (Fearn-Banks, 1996; King, 2002). In ad dition, the crisis management team must decide what issues must be addressed with in a crisis plan. Constructing a crisis
6 management plan may be difficult due to the si tuational characteristics of a crisis and its many changing variables. Effective crisis communication management is difficult to achieve since each crisis is different from the next. What an organization says and doe s once a crisis begins (the crisis response) can have a significant effect on the succ ess of the crisis management effort (Benoit, 1997; Coombs, 1999). Res earchers have just begun to explore the dynamics of the crisis response process (B enoit, 1995, 1997; Coombs & Holladay, 1996). The next section explores crisis comm unication management from seminal work when few crisis responses were documented and made available to crisis managers to the development and testing of crisis response st rategies and their eff ectiveness in achieving organizational goals. Crisis Communication Management in the Beginning Crisis communication management, in pub lic relations research, has lagged in research and theory development (Coomb s, 2006). Although, according to Coombs (2007), crisis communication management is considered a subset of public relations, tactics of public relations are used to disseminate information to the appropriate stakeholders and publics during a crisis thus, qualifying cr isis communication management as a form of public relations. Up until the incidents of the Johnson & Johnson and the Exxon cases, pre-crisis management had been deemed symbolic because, for too many years, crisis communication research was simp ly practitioner truisms and ta les from the field: What I did during our crisis (Coomb s, 2006). However, these two crises provided benchmarks in crisis communication management research on what to do and what not to do during a
7 crisis and how to prevent it from happen ing again. Due to their influence on the development of crisis comm unication theory, these two seminal cases are briefly reviewed in the following s ection to clearly portray what happens when a crisis is handled properly and when a crisis is handled poorly. Johnson & Johnson Tylenol Case. In 1982, some Tylenol capsules were found laced with cyanide in the Chicago area. This product tampering resulted in seven deaths. To this day, the identity of the person or persons who commi tted the Tylenol product tampering is still unknown. A ll supplies of the product in st ores nationwide were pulled off the shelves by Johnson & Johnson, at a cost of $50 million. After due time and investigation, the product was reissued in tamper-resistant containers, and a sealed package of capsules was offered free to consumers who had discarded the suspect supplies in their possession (Center & Jackson, 2003). During the aftershock of the crisis, Tyle nol miraculously recovered the market share it held prior to the crisis. Despite init ial losses, the company was able to regain its credibility and reestablish pub lic trust. As a result, Johnson & Johnson set the benchmark for successful crisis management. The questi on that needs to be answered now is how was Johnson & Johnson able to gain the market and its image and reputation back after a tumultuous crisis such as the cyanide-laced capsules? The phrase, no good deed goes unpunished , holds true to Johnson & Johnson becau se the standards that the company holds itself to were the pillar s of its success in handling the crisis. The following factors, according to Center and Jackson (2003), ar e the main reasons Johnson & Johnson came out on top: 1. The company benefited from a long histor y of success and service in a field of beneficial and worthwhile healthcare products.
8 2. Johnson & Johnson took pride in its public visibility and its reputation for integrity. 3. The company benefited by having had a strong founder who believed that the corporation should be socially responsible with responsibilitie s to society that went far beyond the usual sales and profit motives (Baker, 1993). Johnson & Johnson, basically, set high standards for itself to set a distinguishing tradition that shall be continued as long as the company is in business. 4. There was a credo, a For this we stand on paper, which succeeding generations of executives have built and interpreted in terms of changing times an challenges. The credo was brought out during the crisis for the world to see. 5. In its relations with employees, neighbors, investors, customers, and government agencies, there was a candor consistent w ith competitive and financial security. Company spokespeople incl uding the CEO showed l eadership and authority. 6. There was a recognition of the public intere st and its legitimate representation by news media. Information, whether good or bad, was forth coming as rapidly as it developed. 7. The corporate public re lations function was part of the management, participating in the decision process and in the implementation when communication was involved. 8. There were mechanisms for feedback from constituent publics, and a high value was placed on public input (Center & Jackson, 2003, p. 187). Johnson & Johnsons success in crisis manageme nt stands in sharp contrast to the example set by the Exxon Corporation. Exxon Valdez Oil Spill Case. It was supposed to be a rout ine trip from the Prince William Sound to Long Beach, California for the Exxon Valdez a 987-foot oil tanker. The tanker, commanded by Captain Joseph H azelwood, was longer th an three football fields, [and] loaded to the top with enough oil to fill the Rose Bowl almost halfway to the top (Turning Point, 1994). Unfortunately, the Exxon Valdez never made its des tination to Long Beach because on March 24, 1989, the oil tanker ran ag round the Bligh Reef in Prince William Sound, Alaska. The Exxon Valdez spilled 240,000 barrels (10 m illion gallons) of oil into the water. The oil slick was 10 feet wide, four miles long, and eventually covered 1,300
9 miles of shoreline. An estimated 2 million animals died as a result, including 1 million migratory fowl, a third of the sea otter popul ation, and numerous seals and sea lions. This estimate does not include the number of clams and fish on whom Alaskan fishermen depend for their livelihood (Fearn-Banks, 2007). However, the real crisis mishap occurred when Exxon refused to take responsibility for the crisis. Exxon blamed anyone and/or an ything from the captain of the Exxon Valdez to the out-of-date radar of the U.S. Coast Guard. In the wake of the crisis, Exxon blamed the Alyeska Pipeline Service Company, U.S. Coast Guard, the Alaskan government, weather, and the Depart ment of Environmental Conservation. The giant conglomerate also went as far as to blame the captain, Joseph Hazelwood, and the third mate, Gregory Cousins, because repor ts came out that Hazelwood was intoxicated during the incident and handed over the responsibility of the ship to an inexperienced Cousins (Fearn-Banks, 2007). The then-CEO of Exxon, Lawrence G. Rawl, was one of the last people to communicate about the crisis when he should have been the first. Unlike Tylenol, Exxon did not have a strategic chain of command to ha ndle the crisis. In the wake of the spill, rather than Rawl becoming the spokespe rson like a CEO should, Frank Iarossi, thenpresident of Exxon shipping, was the main repr esentative at the Valdez site (FearnBanks, 2007). There was no presence of remors e from the top executives. It took Rawl 10 days to formally issue a response to the public, which came in the form of an open letter. During the initial response to clean-up th e spill, Exxon barely participated in any efforts in containing and cleaning the crisis. Atlantic Richfield (ARCO), British
10 Petroleum (BP), and five other companies init ially participated in the response. Exxon soon after took over the clean-up efforts from the other companies (Fearn-Banks, 2007). Information about the crisis was diss eminated poorly. Iarossi did not keep Exxons publics informed and when he did make a statement, the information turned out to be erroneous the ma jority of the time. It is important to understand why the cases of Tylenol and Exxon Valdez are important to crisis communication manage ment. During a time of scant crisis communication study, these cases showed how a crisis can be handled in a successfully orchestrated manner, with positive organizati onal outcomes, or how a poor inappropriate organizational response can re sult in significant ne gative outcomes for an organization. With the occurrences of the Tylenol and Exxon Valdez cases, crisis communication management spawned into a more strategic st udy of how to handle crises to achieve more positive organizational outcomes. Through the years, crisis communication has started a slow transition from What I did during our crisis to a more theoretical approach to handling a crisis. Out of theory came the situational crisis communication theory that posits a crisis response taxonomy. Situational Crisis Communication Theory (SCCT) Crisis management includes efforts desi gned to prevent and to detect potential crises, and to learn from crisis experiences (Caponigro, 2000; Cohn, 2000; Coombs, 1999b; Mitroff, 2001). Moreover, crisis management has emphasized post-crisis communication and the use of crisis response st rategies what organizational leaders say and do after a crisis hits (Coombs, 2007). Integrated with Benoits (1995) image restoration and Hearits (1994) corporate apologia the situational crisis communications
11 theory (SCCT) model attempts to map out pos t-crisis communication. More specifically, the model illustrates how crisis response stra tegies can be used to protect reputational assets after the presentation of instructing information, wh ich is the firs t communication priority in a crisis (Sturges, 1994). According to Bergman (1994), Coombs (1998), and Sturges (1994), the SCCT attempts to explai n the type of information that instructs stakeholders what to do to protect themselv es from the crisis, the basics of what happened, and what the organization is doing to fix the situation and to prevent a recurrence of the problem. For any crisis, compassion should be the primary response strategy that organizations contemplate when the source of the crisis is uncertain. Compassion is suggested as the answer to any crisis where fault is unknown because some crises produce victims and victims place unique demands on crisis managers (Ogrizek & Guiller, 1999). Expressing compassi on involves acknowledging and expressing sympathy for victims without accepting responsibili ty or stating remorse in order to avoid litigation issues. Legal expert, J. Cohe n (1999), supports the recommendation that compassion be used when fault is unclear. Most accidents and product recalls (either human or technical error) have unclear fault at the onset of the cris is and may take weeks or months to clarify (Ray, 1999). According to Coombs (2007), the SCCT posits that the crisis situation determines which crisis response strategies will be most effective in protecting the organizations reputation. Reputational assets are important to an organizat ion and are threatened by a crisis. It follows that crisis managers s hould try to maximize the reputational protection afforded by using appropriate crisis response strategies.
12 Historical Development of SCCT The SCCT has been the focus of numer ous studies over the past (Coombs, 2007). To fully understand the development of th e SCCT, it is important to understand the theoretical perspectives of Benoits Image Repair Discourse and Hearits apologia which have informed the growth of the SCCT. Benoits Image Repair Discourse. When people, groups, and organizations are accused of objectionable behavior, reputat ions can be damaged (Benoit & Brinson, 1999). An organizations image, or reputation, is a valuable asset that represents trust, loyalty, and responsibility granted to an organization. In other words, when an organization fails to live up to its promises and expectations, negative consequences can occur for consumers, ranging from inconvenien ce to death. Whenever an organization is faced with a crisis that tarnishes its image, a domino effect of events starts to occur. When one crisis happens in an organizati on, customers immediately begin to become skeptical of how they use or buy that organi zations products or se rvices. After that initial shock of crisis, a domino effect, of sort s, starts to occur. An [organizations] image could influence how closely the governme nt regulates its actio ns. Reputation can influence the price of a companys stock. It can even influence how other companies deal with it (e.g., terms offered for loans or credit; how long a supp lier is willing to wait for payment or how much of a discount will be offered on a purchase) (Benoit & Pang, 2007). Thus, whether an organization be la rge or small, it is important for it to continuously maintain and portray a positive image. Crises and threats that have the potentia l to disrupt or tarnish an organizations image are constantly threatening organizations today. Some organizations are may face
13 several crises in a day, but they are never he ard of because of their low threat level. However, it is the major crises that make it pa st the stakeholders and into the media that many organizations dread each day. That is why it is so true when Pinsdorf (1987) acknowledged that public relations crises are no longer a matter of if, but when; no longer the exception, but the expected even the inevitable (p. 37). This proactive stance has led many companies to take extensiv e measures of crisis prevention and image preservation. This section explicates the theory of im age repair discourse, which provides the foundation for the situation crisis communicatio n theorys (SCCT) responses to organizational crisis. However, we w ill begin with developing and understanding messages used to respond to corporate image crises. To completely understand the theory, it is important to define image Image is nothing but a symbolic thought that is create d in ones mind about a representation of a certain feeling or object. However, in th e realm of corporate image, a corporations image is a subjective impression of that busin ess held by other people. A corporations (or persons or organizations) image, or reput ation, is subjective because it arises from the information held by people about that corporation. Our perceptions of an organization (or person or gr oup) are formed from the words and deeds of that organization and from what others say and do about that organization. So, in a much abridged version, an image is a subjective im pression of an organization formed through ones experience with that organization and interprete d are based on other past experiences (Benoit & Pang, 2007, pp. 244-245). Benoit and Pang (2007) also implied that an image is a subjective impression that will vary from one person to another. It is unlikely that two people will have identical
14 experiences. Furthermore, even if they did have the exact same experiences with a given organization, their other unique experien ces are likely to dominate their own interpretations of the information they shar e about the organization. Other people will often have similar impressions of an organi zation, but it is unlikely that any two people will have precisely the same impression of an organization where some people, of course, will have widely disparate impressions of a firm. So, more than likely, different people can be expected to have different images of a given corporation (Moffitt, 1994). The image restoration literature is he avy on description and retrospective sensemaking through case studies, whereas it is shor t on predictive value and causal inferences (Coombs & Schmidt, 2000). Scientific evid ence demands the process of comparison. The knowledge gained from a one-shot case study is generally illusory (Campbell & Stanley, 1963). Image restoration analysts tend to generate a list of image restoration strategies employed in the case and then speculate on how these strategies promoted success or failure of an image restoration effort (Coo mbs & Schmidt, 2000). According to Coombs and Schmidt (2000), the typical image restoratio n analysis provides limited insight into how publics reacted to the strategies and th e actual effect of the strategies on the organizational image. These limits preclude (a) developing precise additions to the body of knowledge and (b) allowing crisis managers to draw from the full benefits that image restoration theory has to offer crisis communication. In 2000, Coombs and Schmidt, made it aware that; to understand how publics react to different image restoration strategies in different types of cris es if crisis managers are to learn when certain res ponse strategies should be used or avoided, we must become
15 more prescriptive so that crisis managers ha ve clearer guidelines fo r selecting their image restoration strategies. In other words, for cr isis managers to help organizations protect their image during a crisis, it is urgent for an organization to present a clear picture of the organizations current image to its stake holders for crisis managers to provide organizations with concise information on how to continuously and actively preserve the organizations image. To do this, Coombs and Schmidt (2000) suggested that there are at least two options for developing more exact prescriptive knowledge. The first option is to execute a series of similar case studies. Using a series of similar case studies would allow the researcher to find patterns that would indicate the effect of specific strategies in a particular type of crisis. For example, a number of collapsed mines crises and the image restoration strategies that were used c ould be examined. If a specific strategy is associated with successful crisis management, then it is safe to assume that that strategy should be used in other collapsed mines cr ises. We can be more confident in the implications of a case study if a number of case studies ar e conducted and a pattern of similar results emerge (Coombs & Schmidt, 2000). The second option is to empirically examin e the effect of various strategies that were employed in a crisis case. Coombs and Schmidt (2000) suggested testable research that could come out of the descriptive list of strategies identified in a case study to determine their true effects on publics. Res pondents can read or view videotapes of the image response strategies and complete survey s designed to assess their reactions to the strategies. The claims made about the effects of strategies are tested to see if they hold true. The researcher contro ls the image response strategi es by exposing respondents to different strategies and then measuring thei r reactions for comparisons. The empirical
16 tests are another way of moving from specu lation to knowledge (Coombs & Schmidt, 2000). Analysis of image repair discourse sugge sts that there are many points of views in understanding and influencing images. For th e most part, our image of a company is based on what we have seen, heard, and read about that [organization] (Benoit & Pang, 2007). The way a stakeholder perceives a pa rticular organization depends on how that organization communicates to the stakeholder, which, in turns produces an image or impression of that organization. Other point s of views that strongly influence how we look at an organizations image are through the words and actions of other people. This means an image can be influenced (and th reatened or damaged) by the accusations, complaints, and behavior of others (Beno it & Pang, 2007). And when an organizations image is susceptible to being damaged, it is critical for that organization to take the necessary precautions to preven t a potential image distortion or take immediate action in repairing image. According to Benoit and Pang (2007), obser ving that an image is an impression also means that an image may be at odds with reality. This can be harmful to an organization because even though a particular si tuation may not seem like a threat to an organization, an organization that was or is currently perceived in a negative light may experience a biased perception. Reality and image, combined, can cause major problems for an organization because perception can be more important than reality and this is where image repair discourse meets communi cation. In a 2007 article, Benoit and Pang stated that [a] company inappropriately accused of wrongdoing mu st use communication to correct this mistaken image. They added that sometimes people see what they want to
17 see and fail to see what they do not want to see. The truth can repair an image, but if the relevant audience refuses to accept the trut h, reality cannot help the unfairly damaged image. So image and image repair both arise from reality but must be shaped through communication. Furthermore, reality clearly influences images, but rarely do people have a complete knowledge of the facts, and wh at they do know is filtered or interpreted by their personal attitudes and expe riences (Benoit & Pang, 2007). Image repair discourse does not pertain solely to organizations, but discourse may also be implemented with individuals as we ll. Even though organizations have better resources than individuals to repair their im age, the basic options are the same for both individual and corporate image repair effort s (Benoit & Pang, 2007). To really protect a brand, a corporate official, at times, must act as an individual to reac h out to stakeholders rather than speak for the organization all of the time. Thus, writing on both corporate and individual image repair can be useful. Image Repair Theory and Crisis Management. Benoit (2007) and his colleagues have developed the theory of image repair discourse and applied it in a variety of contexts: corporate (Benoit 1995a, 1 995b; 1998; Benoit & Brinson, 1999; Benoit & Czerwinski, 1997; Benoit & Hirson, 2001; Be noit & Pang, 2007; Blaney, Benoit, & Brazeal, 2002; Brinson & Benoit, 1996); politic al (Benoit 1995a, 1999; Benoit, Gullifor & Panici, 1991; Benoit & McHale, 1999; Benoit & Nill, 1998a; Benoit & Pang, 2007; Benoit & Wells, 1998; Kennedy & Benoit, 1997; Len-Rios & Benoit, 2004); international (Benoit & Brinson, 1999; Benoit & Pang, 2007; Drumheller & Benoit, 2004; Zhang & Benoit, 2004; Zhang & Benoit, in press); a nd other contexts (Benoit, 1997a; Benoit & Anderson, 1996; Benoit & Hanczor, 1994; Be noit & Nill, 1998b; Benoit & Pang, 2007;
18 Blaney & Benoit, 1997-2001). The next section explains the relationship of image repair discourse and crisis communication. As stated by Benoit and Pang (2007), imag es [can] be threatened when another person obtains information that creates an unfavorable impression about another person or organization and when images are damage d undeservedly. An organization runs a great risk of its image becoming tarnished wh en an organization is falsely accused of a crisis in a malicious manner, whether it be in disregard of the truth or by mistake. Threats to image that are not based in reality can be just as dama ging as threats arising from the accuseds harmful actions (B enoit & Pang, 2007). And this is when communication is important to crisis management and image repair discourse. It is with communication that we are successful in repairing false accusations. Another philosophy that should be followe d when dealing with image repair theory is that image, and threat to image, arise from the perceptions of the audience (Benoit & Pang, 2007). This philosophy allows that one person can have a totally different perception of a situati on than that of the person stan ding next to them. That is why it is crucial to take different perspect ives into account when dealing with image restoration. Benoit and Pang ( 2007) were lead to another im portant observation: It is vital to realize that businesses frequently must deal with several audiences. Identification of the key audience or audiences is importa nt because different audiences often have diverse interests, concerns, and goals (Benoit & Pang, 2007). When a crisis communicator has preserved an organizations image amongst their diverse stakeholders, Hearits apologia should be incorporated with the im age repair discourse to provide the
19 proper message that should be disseminated amongst the stakeholders to preserve the established positive image of the organization. Hearits Apologia. It is important to understand th at when apologia is used by an organization, it does not necessarily mean th at an organization is apologizing for its actions in a crisis by accepting full responsib ility. The organizati ons effort may deny, explain, or apologize for the action thr ough communication discourse (Fearn-Banks, 2007). Corporate apologia has been deployed fo r decades by organizations when dealing with organizational crises. Th is next section will define apologia as well as explicate its conceptual fundamentals. The first theoretical framework for apol ogia was introduced in an essay by Ware and Linkugel (1973). During their research a nd studies of apologia, Ware and Linkugel (1973), observed four factors and four postu res used in apologetic speaking. These factors and postures, which will be discussed next, were discovered in social scientific research on the resolution of belief d ilemmas, mastered by Abelson (1959. The first identified factor, denial, oc curs when one denies alleged facts, sentiments, objects, or relationships (Ware & Linkugel, 1973, p. 275). The second factor is bolstering, which is the obverse of denial because denial involves negation and bolstering involves identificati on. Bolstering refers to a ny rhetorical strategy which reinforces the existence of a fact, sentimen t, object, or relationship (p. 277). The speaker attempts to identify [themselves] with something viewed favorably by the audience (p. 277). Denial and bolstering ar e known as reformative strategies because they do not attempt to change the audience s meaning or affect for whatever is in
20 question (pp. 275-276); they simply revise or amend the cognitions of the audience (p. 276). The third factor is differentiation, which serves the purpose of separating some fact, sentiment, object, or relationship fr om some larger context within which the audience presently views that attribute (W are & Linkugel, p. 278). The fourth factor, transcendence, is the obverse of different iation because whereas differentiation moves toward the less abstract, transcendence moves toward the more abstract. Transcendence cognitively joins some fact, sentiment, objec t or relationship with some larger context within which the audience does not presen tly view that attribute (p. 280). Differentiation and transcendence are transforma tive strategies that involve a change in meaning. When the factors for apologia were id entified and described, Ware and Linkugel (1973) identified the four postures that a pologetic speakers can refer to: absolution, vindication, explanation, and justification. Each of th ese postures involves the combination of one transformative factor with one reformative factor. The first posture is absolution, which results from the union of the differentiation and denial factors and it is one in which the speaker seeks acquittal (p. 282). Vindication, the second posture, involves denial and transcendence and aims not only at the preservation of the accuseds reputation, but also at the rec ognition of his grater worth as a human being relative to the worth of his accusers (p. 283). The third posture is explanation, which combines bolstering and differentiation. I n the explanative addr ess, the speaker assumes that if the audience understands his motives, actions, belief s, or whatever, they will be unable to condemn him (p. 283). The fourth posture is justification, which o ccurs when bolstering
21 and transcendence are joined. Justification asks not only for understanding, but also for approval (p. 283). Although the exigencies for apologia are et hical in nature, Hoff (1980) notes that management should ask if the corporation did so mething wrong. Is there a need to justify or defend an action? SCCTs Current Articulation The situational crisis communication th eory was developed after J.A. Benson (1998) challenged the crisis communication fiel d to support his theory that there are a set number of crisis types and cris is response strategies. The S CCT matches each crisis to its most appropriate response stra tegy. Coombs (2007) stated that meaningful matching is possible only if there is some corresponde nce/link between crisis types and crisis response strategies. To create this link, SCCT drew from attribution theory and neoinstitutional th eory, which are reviewed below. Attribution Theory. This theory is a useful fram ework for explaining the relation between a situation and the selection of communication strategies (Coombs & Holladay, 1996). Research demonstrates that people sear ch for causes of events in a variety of domains (Weiner, Perry, & Magnusson, 1988) McAuley, Duncan and Russell (1992) identified four causal dimensions people might use when making attribution: stability, external control, personal control, and locus. Stability assesses if the events cause happens frequently (stable) or infrequently (unstable). External control indicates whether or not the events caus e is controllable. Personal control assesses whether or not the events cause is controllable by the actor. Locus reflects if the events cause is something
22 about the actor or something about the s ituation (McAuley et al., 1992; Russell, 1982; Wilson, 1993). Research indicates a substantial over lap between personal control and locus (Coombs & Holladay, 1996). Wilson (1993) sugg ests that the two cau sal dimensions be taken as one dimension that reflects intenti onality of an act. Wh en both are high in an actor, perceptions of intentional actions ar e created, while unintentional actions are created when both are low in an actor. Alt hough measures have been developed for four dimensions (McAuley et al., 1992), functiona lly there are three causal dimensions: stability, external control, and locu s/personal control (locus, for short). The judgments people make about these th ree causal dimensions influence their feelings and behaviors toward the actor (Weiner, 1985; Weiner et al., 1988; Wilson et al., 1993). Peoples attributions to an event can be changed in two ways depending on explanations given by the actors. First, th e messages can shape how people perceive the three attribution dimensions. Second, the messa ges can affect the feelings created by the attributions (Weiner et al., 1988). People make attributions about an organization for a crisis when they determine the cause of the crisis. Greater attributions of responsibility lead to stronger feelings of anger and a more negative view of an actors image (Weiner, Amirhan, Folkes, & Verette, 1987). The three causal dimensions of attr ibution should aff ect evaluations of organizational responsibility for a crisis in predictable ways (Coombs & Holladay, 1996). According to Coombs and Holladay (1996), organi zational crisis responsibility should be perceived as strongest if the cau se is stable (i.e., the organizat ion has a history of crisis),
23 external control (controlled by others outside the organization) is low, and the locus is strongly internal (unint entionality is high). Coombs and Holladay (1996) also stipulate that when a crisis event is repeated (stable), publics are more likely to attribute re sponsibility to the organization. Moreover, attributions of low external control indicate that the crisis was not under the control of groups outside of the organization; thus, the cr isis should not be attr ibuted to external agents. Attributions that entail a strong in ternal locus/intentionality suggest that the organization could have done something to prevent the crisis (Coombs & Holladay, 1996). This type of attributi on suggests that the organizatio n knew what to do to prevent the crisis and any steps needed to contain it. Coombs and Holladay (1996) stated that organizational cris is responsibility should be weakest when attributions suggest th e cause is unstable (i .e., the crisis is an exception in the organizations performance hi story), with strong external control and weak internal locus (low intentionality). Attributions, as found by Coombs and Holladay (1996), reflecting strong external control and low intentional ity (weak internal locus) suggest that factors outside th e organization and its control ar e responsible for the crisis event. An unstable crisis creates weak at tributions of organi zation responsibility (Coombs & Holladay, 1996), such as when circumstances are beyond the organizations control and the crisis cannot be prevented. One objective of crisis management is to prevent or lessen reputational damage to an organization (Barton, 1993; Pearson & Mitroff, 1993; Stur ges, 1994). According to Coombs and Holladay (1996), attr ibutions of organizational crisis responsibility should precipitate reputational damage If communication can alter publics causal attributions
24 or affect feelings generated by these attributions, crisis respon se strategies can be used to reduce reputational damage. Neoinstitutionalism Theory. An organization is granted le gitimacy if stakeholders believe an organization is good and/or has a right to c ontinue operations (Allen & Cailouet, 1994; Bedeian, 1989). Legitimacy is critical to the successf ul operation of an organization (DiMaggio & Powell, 1991) because it is a conformation of social rules and expectations established by stakeholders. On the contrary, when crises occur, stakeholders question the organi zations ability to conform to the stakeholders rules and expectations. In turn, this questions an or ganizations legitimacy. An organization will use communication strategically as a response to legitimacy threats because corporate discourse does shape how stakeholders view an organization (Alle n & Caillouet, 1994; Marcus & Goodman, 1991). The crisis response strategy can be used to a) show the challenge is invalid or b) attempt to get stakeholders to judge the cris is more mildly and ev aluate the organization more positively (Allen & Caillouet, 1994). From the neoinstitutional perspective, organizations should favor the use of crisis resp onse strategies that reflect efforts to reestablish legitimacy (Coombs & Holladay, 1996). Neoinstitutional research consistently indicates corporate actors use mechanisms and procedures to convey conformity with their institutional environment to enhance legitimacy and survival chances (Allen & Caillouet, 1994, p. 48). Coombs and Holladay (1996) argue that or ganizations must shift the focus from the violation of social norms (the crisis) to efforts designed to repair the violation
25 because, if a crisis cannot be shown to be i nvalid, crisis managers should use strategies that show how the organization has returned to the norms held by its stakeholders. Attribution and neoinstitutional theories provide the framework for the SCCT, which is organized into three parts: the crisis situation, crisis res ponse strategies, and the matching recommendations (Coombs, 2007). The Crisis Situation A crisis situation will generate particular attributions of crisis responsibilities or the degree to which the organization is percei ved to be responsible for the crisis event (Coombs, 2007). According to Coombs (2007), the level of crisis responsibility is a primary indicator of how much a threat the crisis is to th e organizations reputation and what crisis response strategies are necessary to address that threat. The crisis situation basically involves crisis types and threat intens ifiers. Threat intensifiers will be discussed and described shortly, however, crisis types simply provide categories for crises. Each crisis type is different in that it situati onally provokes different stakeholder views and evaluation of responsibility. Each crisis t ype will generate a specific level of crisis responsibility. The SCCT is built on a taxonom y of thirteen crisis types, which have been divided into three clusters (Coombs, 2007) Each of the crisis types in a cluster shares a similar level of crisis responsibil ity with the others (Coombs & Holladay, 2002).
26 Crisis types by level of crisis responsibility Attributions of crisis responsibility, high: Preventable cluster Human breakdown accidents: Human error causes an industrial accident. Human breakdown recalls: Human error causes a product to be recalled. Organizational misdeed with no injuries: Stakeholders are deceived without injury. Organizational misdeed management misconduct: Laws are regulations are violated by management. Organizational misdeed with injuries: Stakeholders are placed at risk by management and injuries occur. Attributions of crisis responsibilities, moderate: Accidental cluster Challenges: Stakeholders claim an organization is operating in an inappropriate manner. Megadamage: A technical accident where the focus is on the environmental damage from the accident. Technical breakdown accidents: A technology or equipment failure causes an industrial accident. Technical breakdown recalls: A technology or equipment failure causes a product to be recalled. Attributions of crisis responsibilities, low: Victim cluster Natural disaster: Acts of nature that damage an organization, such as an earthquake. Rumors: False and damaging information about an organization is being circulated. Workplace violence: Current or former employee attacks current employees onsite. Product tampering/malevolence: External agent causes damage to an organization. FIGURE 1 From The Development of the Situational Crisis Communication Theory, by T. Coombs, 2007. In T. L. Hansen-Horn and B. D. Neff (Ed.), Public Relations: From Theory to Practice (pp. 262-277). Upper Saddle River, NJ: Pearson. Threat intensifiers strengthen the reput ation damage a crisis can cause on an organization and include crisis history, relati onship history, and severi ty (Coombs, 2007). Coombs (2007) defines the threat intensifiers as follows: Crisis history includes similar crises an organization has had in the past. News stories often include reports if an organization has experienced similar crises. Relationship history indicates if the organization has had a record of good works or bad behavior. Rela tionship history is concerned with how the organization has tr eated its stakeholders in the past. Organizational behavior is a key factor in determining reputati ons. Stakeholders feel it is important for an organizations words and d eeds to match (Herbig, Milewicz, & Golden,
27 1994). Together, crisis and re lationship history are known as performance history because they are indicators of how th e organization has acted in the past. Severity is the amount of damage inflicted by th e crisis, including injuries, loss of lives, financial loss, and environmental destruction. If an organization has a negative cris is history, negative relationship history, and/or the crisis damage is se vere, the reputational damage of the crisis type is intensified (Coombs, 2007). As seen in Table 1, the crisis types are divided into three clusters based on similar, initial attributions of crisis responsibility: preven table, accidental, and victim. It is suggested, that if a crisis intensifies, it should move a crisis type to the next level. According to Coombs (2007), the threat inte nsifiers indicate what crisis response strategy is appropriate. According to Blazer and Sulsky (1992), a favorable relationship history should produce a halo effect, acting as a shield that protects the organization from the reputational damage of a crisis. So, a favorable pre-cris is relationship with stakeholders should benefit an organizati on (Birch, 1994; Fearn-Banks, 1 996; Siomkos & Shrivastava, 1993). When a positive relationship is establis hed with stakeholders, a crisis that sheds negative light on an organization or when th e cause of a crisis is still unknown, the attributions of the crisis re sponsibility and the organization s reputation will be affected for a more positive outcome for the organization. Crisis Response Strategies In its current form, the SCCT includes 10 crisis response strategies, grouped into three postures. A posture represents a set of strategies that shar e similar communicative goals and vary in terms of their focus on prot ecting the crisis victims (victim orientation)
28 and taking responsibility for the crisis (Coombs 2007). It is importa nt to understand that the key word that links crisis type s and crisis response strategies is responsibility The attribution and neoinstitutional theoretical concepts are reflected in the three postures that represent the three basic communicative options. Crisis response strategies by postures Deny posture (low concern for vi ctim and responsibility acceptance) Attack the accuser: Crisis manager confronts the pers on or group claiming something is wrong with the organization. Denial: Crisis manger claims that there is no crisis. Scapegoat: Crisis manger blames some person or group outside the organization for the crisis. Diminish posture Excuse: Crisis manger minimizes organization responsibility by denying intent to do harm and/or claiming inability to contro l the events that triggered the crisis. Justification: Crisis manager minimizes the percei ved damage caused by the crisis. Deal posture (high concern for victim and responsibility acceptance) Ingratiation: Crisis manager praises stakeholders and/or reminds them of past good works by the organization. Concern: Crisis manger expresses concern for the victims. Compensation: Crisis manager offers money or other gifts to victims. Regret: Crisis manager indicates the orga nization feels bad about the crisis. Apology: Crisis manager indicates the organization takes full responsibility for the crisis and asks stakeholders for forgiveness. FIGURE 2 From The Development of the Situational Crisis Communication Theory, by T. Coombs, 2007. In T. L. Hansen-Horn and B. D. Neff (Ed.), Public Relations: From Theory to Practice (pp. 262-277). Upper Saddle River, NJ: Pearson. According to Coombs (2007), the deny posture represents a se t of strategies that claim no crisis occurred or that the accuse d organization has no re sponsibility for the crisis. If there is no crisis, there can be no organizational responsibility for a crisis (attribution theory) and no violation of legitimacy (neoinstitutional theory). The diminish posture reflects a set of strategies that attempt to alter stakeholder attributions by
29 reframing how stakeholders should interpret the crisis (attributi on theory). Crisis managers might try to place distance between the organization and the crisis, thereby seeking to reduce responsibi lity for the crisis. The deal posture represents a set of strategies that seek to impr ove the organizations reputation in some way. By protecting victims and accepting responsibility, crisis mana gers encourage stakeholders to judge the organization more positively or less negatively. An organization in an intentional crisis is expected to address victim con cerns, so the crisis response strategy must demonstrate the organization is meeting expectations/adhering to social norms, which is a tenant of neoinstitutiona l theory. The deal posture includes the concer n strategy that is an expression of compassion. Although automatically used when th ere are victims, the concern strategy is optional when no one seems to be harmed. Moreover, the grouping shows that an expression of concern is viewed very simila rly to apology and regret, which are the two crisis response strategies that can open an organization to legal liability. Modeling the Process To obtain a better understanding of the rela tionship between vari ables, the SCCT is shown in model form.
30 Situational Crisis Communication Theory model FIGURE 3 From The Development of the Situational Crisis Communication Theory, by T. Coombs, 2007. In T. L. Hansen-Horn and B. D. Neff (Ed.), Public Relations: From Theory to Practice (pp. 262-277). Upper Saddle River, NJ: Pearson. The relationships are based on the propositions below (Coombs, 2007, pp. 268-269). 1. Organizational reputation proposition: There is a strong, negative correlation between organizational reputation and crisis responsibility. Attributions of crisis responsibility have a strong effect on pe rceptions of organizational reputation. The stronger the attribution of crisis responsibility, the more the crisis can damage the organizational reputation and, in turn, affect future interactions with the organization (potential supportive behavior). 2. Potential supportive behavior proposition: A strong, positive co rrelation exists between organizational reputation and poten tial supportive behavior, intentions to engage in acts that woul d help an organization. A negative reputation should result in less supportive behavior from stakeholders, while a positive reputation should engender more. 3. Severity proposition: Severity has a significant in tensifying effect on crisis responsibility and damage to organization reputation. As the crisis increases in severity (inflicts greater damage), attr ibutions of crisis responsibility will intensify. Severity of an incident tends to increase pe rceptions of responsibility among individuals. The same dynamic is be lieved to hold true for organizations in crisis. Severity is also an indica tion of deviation from the norm. Greater severity suggests a greater violation of the expected norms and could result in direct damage to the or ganizations reputation. 4. Crisis history proposition: An unfavorable crisis history has a significant intensifying effect on crisis responsibil ity and damage to organizational reputation. Organizations that have experienced similar cr ises in the past will be attributed greater crisis re sponsibility and suffer more direct reputational damage
31 than an organization with no history of cris es. The history of crises indicates that the crisis is part of a pa ttern of behavior by the orga nization, another negative act by the organization and not an anomaly. 5. Relationship history proposition A: An unfavorable relationship history has a significant intensifyi ng effect on crisis respons ibility and damage to the organizational reputation. An organization that has treated stakeholders badly in the past will be attributed greater cris is responsibility an d suffer more direct reputational damage than an organization with a neutral or positive relationship history. 6. Relationship history proposition B: a favorable relationship history has a significant reducing effect on crisis responsibility and damage to the organizational reputation. Organizations that have maintained favorable relationships with stakeholders will see w eak attributions of crisis responsibility and less reputational damage for a crisis than those with neutral or unfavorable ones. 7. Crisis response strategy selection proposition: Organizations will suffer less reputational damage from a crisis a nd experience greater potential supportive behavior if they match the crisis response strategy to the reputa tional threat of the crisis. See Figure 4 for a list of gene ral recommendations. These propositions emulate how SCCT forms a relationship be tween crisis situations and crisis response strategies. List of crisis response recommendations Rumor: Use any of the denial strategies. Natural disaster: Use instructing information. Workplace violence: Use instructing information. Product tampering: Use instructing information. Product recall, technical error, megadamage; and Accidents, technical error: Use excuse and/or justification. History, relationship history, and/or severe damage: Use any of the deal strategies. Product recall, human error and accidents, human error: Use any of the deal strategies. Organizational misdeeds: Use any of the deal strategies. When victims occur: Use the concern crisis response strategy in combination with other recommended strategy (ies). FIGURE 4 From The Development of the Situational Crisis Communication Theory, by T. Coombs, 2007. In T. L. Hansen-Horn and B. D. Neff (Ed.), Public Relations: From Theory to Practice (pp. 262-277). Upper Saddle River, NJ: Pearson.
32 Matching Crisis Situations to Crisis Response Strategies and Limitations The SCCT maintains that as attributions of crisis responsibilitie s and/or the threat of reputational damage increases, crisis managers must use crisis response strategies that reflect a greater concern for vi ctims and take more responsibility for the crisis (Coombs, 2007). Figure 4 provides a list of crisis response recommenda tions to serve as guidelines provided by the SCCT. For example, a low orga nizational crisis such as a technical-error would require nothing more than an excuse to justify the error. However, if the same error keeps occurring, crisis ma nagers would have to implement strategies from the deal posture because stakeholders may look at the crisis as something that is preventable rather than an accidental technical error. Coombs (2007) argues that the concern crisis response strategy should be applied to any crisis with victims to express compassion. Like any other developing theory, the S CCT has its limitations. It posits that by matching the proper crisis type to the pr oper crisis response strategy, an optimum solution will be achieved in resolving a crisis However, Coombs (2007) states that legal and/or financial liabilities can restrict what an organization can and cannot say and sometimes organizations may become inoperable if such heavy liabiliti es are taken by an organization. Apologies, and in some states e xpression of regret, will result in significant legal liabilities and financial costs (France, 2002). In order to le ssen an organizations liability for a crisis, many cris is managers use a less effective diminish strategy. SCCT allows crisis manage rs to understand the effect of choosing a nonmatching strategy by indicating why the effectiveness of th e response is reduced (Coombs, 2007).
33 Terminological Changes. Since its appearance on a flowchart in 1995, SCCT has gone through many changes, terminologica lly and technically. The historical development can be divided into terminologi cal changes, research-driven, and theory testing (Coombs, 2007). The S in the acronym SCCT, stands for situational However, when the theory was first introduced, the S stood for symbolic The term symbolic was chosen because the crisis response strategies were viewed as symbolic resources that could be employed during a crisis because words are symbols, hence, crisis response strategies were symbolic resources (Coombs, 2007). La ter, many researchers thought the term symbolic to be problematic because the word symbolic represents a type of action and nothing substantive. Eventually, symbolic was repla ced with situational because the theory is premised on the crisis situation (Coombs, 2007). The threat intensifiers began as modifi ers. However, research showed that negative performance histories drove the e ffect of the modifiers, termed the Velcro effect (Coombs & Holladay, 2001). Coombs (2007) found that negative performance history served to intensify the attributions of cr isis responsibility and the damage to the organizational reputation, so the name was changed. Another notable change was that the names technical error and human error of the crisis type category evolved from technical breakdown and human breakdown. It was observed that breakdown was too cumbersome and error captured the essence of the crisis types more parsimoniously (Coombs, 2007).
34 Research-Driven Changes. Perhaps the most substantive growth of the SCCT was the change in representation of crisis ty pes on a continuum rather than a grid. The information entered in the grid was easily entered. However, when all information was in its proper place, the grid of information did very little to explain the effects of the crisis types. Originally, crisis t ypes were viewed as a 2-by-2 grid representing personal control (whether the organization coul d control the source of the crisis) and external control (whether an external agent was in control of the source of the crisis) (Coombs, 1995; Coombs, Hazleton, Holladay, & Chandler, 1995) However, research has found that external control contributes little to the explanation of variance for SCCT (Coombs, 2007). In the early development of SCCT, it wa s thought that crisis responsibility was predicted by personal control. However, later research show ed that personal control and crisis responsibility were essentially the same variable (Coombs & Holladay, 2001, 2002). Personal control and crisis responsibilit y were later combined together to form one category. Testing the SCCT To test SCCT, measures were develope d to test the central concepts of organizational reputation, crisis responsibi lity, and potential s upportive behavior. The organizational reputation concept was deri ved from McCroskeys (1966) measure of character (trust and past and current conceptualizations of reputation). His original model had 10-items, but was later scaled down to five items and still held its reliability range of .80 to .92.
35 Crisis responsibility was measured by using two types of scales: McAuley, Duncan, and Russells (1992) Causal Dimensi on Scale II (CDSII) and Griffin, Babin, and Dardens (1992) blame scale. The Causal Dimension Scale II, or CDSII, assesses attributions of controllability of an ev en while the blame scale assesses who is responsible for the event. While the s cales went through wording and evaluative modifications to fit the mold of the orga nization (Coombs & Holladay, 2002), the final seven-item scale demonstrated reliabilities sim ilar to that of the organizational reputation measure with a range of .80 to .91. The potential supportive be havior measure is the only scale that was developed from scratch by Coombs. The idea of this measure is to find out how a stakeholder might act toward an organization after a crisis and if people intend to behave in ways that are favorable to the organization after the cr isis (Coombs, 2007). The potential supportive behavior scale demonstrated reliabilities between .81 to .87. Out of all of the scales, organizational reputation has been the most extensively tested. According to Coombs (2007), five students have found a significant negative correlation between crisis responsibility and organization reputation. The average correlation across these studies r = -.415. The correlations were found across the entire range of crisis types, including organizati on misdeeds, human-error accidents, technical errors accidents, technicalerror recalls, workplace viol ence, and product tampering (Coombs, 1998, 1999a; Coombs & Holladay, 2001, 2002; Coombs & Schmidt, 2000). Three studies used organizational misdeed s, human-error crisis, technical-error crisis, workplace violence, product tampering, and technical-error recall to test the crisis history proposition. Crisis history was found to have a significant effect on organization
36 reputation for all but the technical-error recall and a significant effect on crisis responsibility for all but product tamperi ng and technical-error recall (Coombs, 1998, 2002b; Coombs & Holladay, 2001). This finding was termed the Velcro effect, as the unfavorable condition attracted and snagged additional repu tational damage (Coombs & Holladay, 2002). A human-error accident crisis was us ed to test the relationship history propositions of the theory. The results stipul ate that the unfavorable relationships history was found to have a significant effect on organizational reputation and crisis responsibility which showed its support for A. On the contrary, however, B was not supported because a favorable relationships history was no different than having no relationship history (Coombs, 2007). As with a crisis history, a Velcro effect was observed; only the negative condition ha d an effect (Coombs & Holladay, 2001). The severity proposition was tested in one study using tec hnical-error accident and organizational misdeed crisis types. The results indicated that severity of the crisis damage did not affect either organizationa l reputation or crisis responsibility as anticipated (Coombs, 1998). The potential supportive behavior propos ition has been examined in two studies using organizational misdeed and human-error accident crisis types. Organizational reputation, r =.37, and potential supportiv e behavior correlated, r =.48. The crisis response strate gy selection proposition was te sted in two studies using organizational misdeed and techni cal-error accident cris is types. According to the studies conducted, the matched strategies (those r ecommended by SCCT) performed better than the mismatched strategies. The SCCT matc hes each crisis to its most appropriate
37 response strategy. Coombs (2007) stated that meaningful matching is possible only if there is some correspondence/ link between crisis types and crisis response strategies `while the mismatched conditions include d using responses that accepted greater responsibility than recommended by SCCT, to prevent finding an effect by simply using lower, less-effective crisis response stra tegies (Coombs & Holladay, 1996; Coombs & Schmidt, 2000). Purpose and Hypotheses This study attempts to contribute to cu rrent theory-driven research in crisis communication by examining crisis response strategies from a communication-centered perspective. Specifically, this study seeks to further understa nding of situational crisis communication theory by examining the effect of crisis response strategies on public perceptions of an organization in an accide nt crisis situation. To accomplish this objective, three hypotheses were developed. H1: In an accident crisis situati on, a matched organizational response message will produce a more positiv e perception of organizational reputation than an unmatched or ganizational response strategy. H2: In an accident crisis situati on, a matched organizational response message will produce a more posit ive perceptions of crisis responsibility for an organi zation than an unmatched organizational response strategy. H3: In an accident crisis situati on, a matched organizational response message will produce a more positive potential supportive behavior than an unmatched or ganizational response strategy. It is a primary goal of this study to learn more about th e effectiveness of crisis response strategies in producing the desired ou tcome for organizations in crisis. This study seeks to test the propositions of the s ituational crisis communication theory and identify the message strategies most effec tive in producing positive public perceptions of
38 organizational reputation, crisis responsibi lity, and potential s upportive behavior of publics toward an organization in crisis, specifically an accide nt. To the test the proposed hypotheses, the methodology section will provid e, in detail, how each hypotheses will be tested.
39 Chapter Three Methodology Research Participants The participants for this study were 90 undergraduate students enrolled in an introductory mass communication class at a large southeastern university. Crises produce unique forms of publics for organizations, vi ctims, and nonvictims. Victims are those who are directly affected in some way (e.g., evacuated from an area, injured physically, or lost property). The focus of this pres ent study is on nonvictims. Nonvictims are not affected by the crisis but follow the crisis in the news media because they are part of the organization where the possible crisis could occu r. The students fit th e parameters of the nonvictim population because they share the characteristics of being directly associated with an organization where a possible crisis could occur and they hold strong perceptions of the organization prior to a possible crisis. Of the participants, 63.3% were female ( n =57) and 36.7% were male ( n =33). The respondents ranged in age from 18 to 40 years old (M=19.93, SD=2.96). Design and Materials The Campbell & Stanley one-shot case st udy (intervention and post-test) required the development of one scenario. A limitation of this design can occur when treatments are not randomized, however, treatments were randomized in this study to address this limitation. One response to one particular crisis was developed for the study. The matched response had the organization taking full responsibility of the crisis and
40 conveying an apologetic response to its stakeholders. An accident was selected for the crisis type for two reasons. First, accident s are a common type of cr isis (Irvine & Miller, 1996). The results of applied research are more beneficial when they address issues that have a direct impact on someone and/or some thing. The lessons are more valuable when they can be applied more frequently. S econd, accidents provide a greater variance in perceptions (Coombs & Hollada y, 2001). Accidents vary in terms of the perceptions formed about the crisis and the organization in crisis. For instance, technical breakdowns and workplace violence produce minimal percepti ons of crisis respons ibilities, whereas human error produces fairly stronger percep tions of crisis responsibilities (Coombs, 1999a). Even though there are ways to help in the prevention of accidents, stakeholders can perceive accidents very differently. That is why it is important to utilize a crisis type that demonstrates variance on many of the perceptual variables used in a study. Instrumentation For this study, a hypothetical crisis wa s developed based on a possible crisis situation that could have potenti al serious repercussions for stak eholders if ever this crisis occurred. In this case, the budge t cuts that are affecting all pu blic universitie s in Florida, especially the University of South Florida and the USF Sc hool of Mass Communications, is expected to reduce many of its communicatio ns class offerings. As a result, students would be required to partic ipate in random drawings fo r the classes they need. Regardless of tenure, all students are ente red into the drawing equally. Depending on how some students names are randomly sel ected, this process has the potential to dramatically delay students course work which means more money and more time spent at the university because ther e is not enough faculty to teach all of the classes needed to
41 accommodate all of the students. The fo llowing message was used as the matched response: For the first time in 35 years, Florida is experiencing a severe economic downturn, resulting in a drop in tax collection and a decrease of the amount of state money availabl e to State University System Institutions, i.e., USF. The budge t cut strips the University of approximately $51 million or 15% of our overall budget. Because of these large budget cuts, the unive rsity has to reduce many of its current course offerings because the university is not able to employ enough faculty to accommodate current students. Unfortunately, many courses at USF will be part of a new course enrollment process. This new course enrollment proce ss enters students names into a lottery for needed courses. Name s will be selected randomly until each course is full. Due to the overwhe lming number of students and limited course availability, all students ar e entered into the drawing equally, regardless of academic level or area of study. It is important to understand that USFs overall concern is to our students. The school has always made it a priority to accommodate its students and your overall collegiate success. We know how this budget cut is going to dramatically dist urb a large number of your graduation plans and USF is fully committed to doing whatever is necessary so that this will not happen. USF may have to limit the number of new students and/or eliminate some programs, but your graduation and academic plans will not be affected by this economic crisis. If you have any questions or concerns, please do not hesitate to contact myself or my staff. The next message shows the unmatched response that was used in the instrumentation: For the first time in 35 years, Florid a is experiencing a severe economic downturn, resulting in a drop in tax collection and a decrease of the amount of state money available to State University System Institutions, i.e., USF. The budget cut strips th e University of approximately $51 million or 15% of our overall budget. Because of these large budget cuts, the university has to reduce many of its current course offerings because the university is not able to employ enough faculty to accommodate current students. Unfortunately, many courses at USF will be part of a new course enrollment process. This new course enrollment proce ss enters students names into a lottery for needed courses. Names will be selected randomly until each course is full. Due to the overwhe lming number of students and limited course availability, all students ar e entered into the drawing equally, regardless of academic level or area of study. More information about this course enrollment system will be announced in late November.
42 This experimental design involved the manipulation of crisis history for the scenario. Three scales were employed to te st the dependent variables in the study: (a) organizational reputation, (b) crisis responsib ility, (c), potential supportive behavior. Organizational Reputation. Organizational reputation was measured using five items from Coombs and Holladays (1996) 10-item Organizationa l Reputation Scale, which is an adaptation of McCroskeys (1966) scale for measuring ethos. Specifically, an adaptation of the Character subscale of McCroskeys Ethos Scale was used to assess organization reputation (Coombs & Holladay, 2002). In this in stance, charac ter is used as a trustworthiness and good will of the source, that is, an assessment of the degree to which USF is concerned about its students. McCroskeys items were modified by simply replacing the term speaker with USF The five items used in the present study were: (a) USF is concerned with the well-being of its publics, (b) USF is honest and open to its publics, (c) I trust USF to be honest and open about the situation, (d) Under most circumstances, I would be likely to believe what USF says, (e) USF is not concerned with the well-being of its students. Crisis Responsibility. Crisis responsibility was m easured using Griffin, Babin, and Dardens (1992) three-item scale for blame. In this measure, organization was replaced with the term USF The three items were: (a) Circumstances, not USF, are responsible for the crisis, (b) The blame fo r the crisis lies with USF, (c) The blame for the crisis lies in the circumstances, not USF. Potential Supportive Behavior. This final measurement was adapted from a Coombs (1999a) 8-item scale comprised of a list of actions an or ganization might ask
43 stakeholders to perform. However, the scale was modified into a 5-item version that used two items from the original scale where the term organization was replaced with USF The items are: (a) I intend to say nice th ings about USF to other people and (b) I intend to call or email my state government o fficials to voice my c oncern on this crisis that is affecting USF. Th e last three items were specifically developed for this experiment. The items are: (c) I plan to continue my college education at USF, (d) intend to support USFs administ rative decision, and (e) I pl an to transfer to another academic institution to complete my degree. Procedures Each respondent received a packet containing a cover pa ge with instructions, one stimulus crisis case, and a copy of the survey instrument. The orde r of the materials in the packet was 1) a cover page, 2) a stimulus and 3) a copy of the survey instrument. Each respondent read the s timulus and completed the surv ey instrument accordingly. Respondents were verbally instructed to read carefully each case and then respond to the questions following the case. The admini stration required about 10 to 15 minutes.
44 Chapter Four Results This study attempts to contribute to cu rrent theory-driven research in crisis communication by examining crisis response strategies from a communication-centered perspective. Specifically, this study seeks to further understa nding of situational crisis communication theory by examining the effect of crisis response strategies on public perceptions of an organization in an accide nt crisis situation. To accomplish this objective, three hypotheses were developed. H1: In an accident crisis situati on, a matched organizational response message will produce a more positiv e perception of organizational reputation than an unmatched or ganizational response strategy. H2: In an accident crisis situation, a matched organizational response message will produce a more posit ive perceptions of crisis responsibility for an organi zation than an unmatched organizational response strategy. H3: In an accident crisis situati on, a matched organizational response message will produce more positive potential supportive behavior than an unmatched organi zational response strategy. Study of Participants To test the hypotheses posited by this study, a 1x2 factoral experiment was conducted. The experiment utilized a balanc ed design that include d 90 college students enrolled at a large southeastern universit y. The participants included 33 males and 57 males. Participants ranged in age from 18 to 40, with an average age of 19.93 years. Table 1 presents the frequencies and percen tages of the sex and academic year of the participants.
45 Table 1. Sex and academic year of participants Variable Frequency Percentage Sex male female 33 57 36.7 63.3 Year freshman sophomore junior senior 37 18 32 3 41.1 20.0 35.6 3.3 Measures of Variables of Interest Prior to the hypotheses testing, the intern al consistency of the multi-item scales used to test the variables of interest wa s assessed using Cronbachs alpha to find the reliability coefficients. The five items incl uded to test organizational reputation produced a reliability coefficient of .79. The three it ems included to test crisis responsibility produced a reliability coefficient of .86. The four items included to test organizational control produced a re liability coefficient of .78. Five items were included to test supportive behavior, and the al pha indicated scale reliabil ity by dropping the item I intend to call or email my state government o fficials to voice my c oncern on this crisis that is affecting USF. Af ter this item was omitted, the f our remaining items produced a reliability coefficient of .87. Overall, the Cronbachs alpha scores, show n in Table 2, indicate reliability for the multi-item scales used to measure the variables of interest. According to Berman (2002), alpha values between .80 and 1.00 indicate high reliability. It is also agreed that the lower limit of .70 is stil l a useful measure of constr ucts (Broom & Dozier, 1990; Stacks, 2002).
46 Table 2. Cronbachs alpha for multiple-item indexes Variable Number of items Organizational Reputation .79 5 Crisis Responsibility .86 3 Organizational Control .78 4 Supportive Behavior .87 4 Next, an evaluation of means for each vari able of interest wa s performed prior to creating composite measures fo r hypothesis testing. An evaluation of the items used to measure organizational reputation produced me an scores ranging from 3.34 to 3.80. The mean and standard deviation for each item us ed to measure organizational reputation is shown in Table 3. Table 3. Organizational Reputation descriptives Variable N Mean Std. Deviation 1orgrep1 90 3.31 .830 2orgrep2 90 3.80 .914 3orgrep3 90 3.34 1.007 4orgrep4 90 3.51 .890 5orgrep5 90 3.46 .950 An evaluation of the items used to meas ure crisis responsibility produced mean scores ranging from 2.30 to 2.54. The mean and standard deviation for each item used to measure crisis responsibility is shown in Table 4. Table 4. Crisis Responsibility descriptive Variable N Mean Std. Deviation 6cr1 90 2.54 .996 7cr2 90 2.30 .854 8cr3 90 2.30 .893 An evaluation of the items used to m easure organizational control produced mean scores ranging from 2.49 to 2.72. The mean and standard deviation for each item used to measure organizational cont rol is shown in Table 5.
47 Table 5. Organizational Control descriptive Variable N Mean Std. Deviation 9oc1 90 2.49 1.063 10oc2 90 2.50 1.008 11oc3 90 2.72 1.017 12oc4 90 2.53 .877 An evaluation of the items used to measure potential supportive behavior produced mean scores ranging from 2.61 to 3.80. The mean and standard deviation for each item used to measure potential s upportive behavior is shown in Table 6. Table 6. Potential Suppor tive Behavior descriptive Variable N Mean Std. Deviation 13sb1 90 3.80 1.210 14sb2 90 3.67 1.122 15sb3 90 3.06 1.193 16sb4 90 3.39 1.371 18sb5 90 2.61 .991 An evaluation of the items used to m easure overall effectiveness produced mean scores ranging from 2.46 to 2.50. The mean and standard deviation for each item used to measure overall effectiveness is shown in Table 7. Table 7. Overall Effectiveness descriptive Variable N Mean Std. Deviation 19eff1 90 2.48 .951 20eff2 90 2.46 1.007 21eff3 90 2.50 1.030 Finally, items used to measure each vari able of interest were collapsed into composite measures for each variable. The mean scores for the composite measures ranged from 2.38 to 3.48. The mean and standa rd deviation for the composite measures are shown in Table 8.
48 Table 8. Mean and standard deviation for composite measures Composite Measure N Mean Std. Deviation Organizational Reputation 90 3.48 .678 Crisis Responsibility 90 2.38 .808 Organizational Control 90 2.56 .767 Potential Supportive Behavior 90 3.48 1.041 Overall Effectiveness 90 2.51 .812 Test of Hypotheses Hypothesis 1 posited that in an accident crisis situation, a matched organizational response message will produce a more positiv e perception of organizational reputation that an unmatched organizational response strategy. To test this hypothesis, an independent-samples t test was conducted. The results indicated that an unmatched response (N=45, M=3.57, SD=.692) produced a higher mean score than the predicted matched response (N=45, M=3.40, SD=.65). Howe ver, the difference in means was not significant, t (88) = -.506, p = .793. These results indicate the matched response does not yield a more positive perception of organizational reputation; therefore H1 is not supported. Hypothesis 2 posited that in an accident crisis situation, a matched organizational response will produce more positive percep tions of crisis re sponsibility for an organization than an unmatched organizational response strategy. To test this hypothesis, an independent-samples t -test was conducted as well. The results indicated that a matched response (N=45, M=2.45, SD=.832) produced a higher mean score than an unmatched response (N=45, M=2.31, SD=.786) as initially predicted. However, the difference in means was not significant, t (88) = -.527, p = .691. These results indicate that the matched response yielded a more positive perception of crisis responsibility; therefore H2 is not supported.
49 Hypothesis 3 posited that a matched or ganizational response message will produce more positive potentia l supportive behavior than an unmatched organization response strategy. An independent-samples t -test was also used to test this hypothesis. The results stipulated that an unmatch ed response (N=45, M=3.55, SD=.997) produced a higher mean score than the predicted matched response (N=45, M=3.41, SD=1.09). However, the difference in means wa s not significant, t (88) = -1.129, p = .417. These results indicate that the matched response yielded a more positive perception of crisis responsibility; therefor e H3 in not supported. Exploration of Individual Item Differences Following the hypothesis testing, explorati on analysis was conducted to determine if significant differences existed between the matched and unmatched crisis responses for each individual item. Table 9 shows the mean and standard deviation for each item used to measure the variables of interest in this study. Table 9. Individual item mean and standard deviation Crisis Type N Mean Std. Deviation 1 Org ReP: USF is concerned with the wellbeing of its publics matched 45 3.27 .837 unmatched 45 3.36 .830 2 Org Rep: USF is basically dishonest matched 45 3.71 .920 unmatched 45 3.89 .910 3 Org Rep: I do NOT trust USF to tell the truth about the incident matched 45 3.20 1.014 unmatched 45 3.49 .991
50 4 Org Rep: Under most circumstances, I would be likely to believe what USF says matched 45 3.40 .863 unmatched 45 3.62 .912 5 Org Rep: USF is not concerned with the well-being of its students matched 45 3.40 .889 unmatched 45 3.51 1.014 6 CR: Under most circumstances, I would be likely to believe what USF says matched 45 2.49 1.014 unmatched 45 2.60 .986 7 CR: The blame for the crisis lies with USF matched 45 2.47 .919 unmatched 45 2.13 .757 8 CR: The blame for the crisis lies in the circumstances, not USF matched 45 2.40 .915 unmatched 45 2.20 .869 9 OC: The cause of the crisis was something that USF could control matched 45 2.31 .973 unmatched 45 2.67 1.128
51 10 OC: The cause of the crisis is something over which USF had no power matched unmatched 45 45 2.42 2.58 1.011 1.011 11 OC: The cause of the crisis is something that was manageable by USF matched 45 2.67 1.066 unmatched 45 2.78 .974 12 OC: The cause of the crisis is something over which USF had power matched 45 2.60 .939 unmatched 45 2.47 .815 13 SB: I plan to continue my college education at USF matched 45 3.80 1.254 unmatched 45 3.80 1.179 14 SB: I intend to say nice things about USF to other people matched 45 3.53 1.160 unmatched 45 3.80 1.079 15 SB: I intend to support USFs administrative decision matched 45 2.93 1.195 unmatched 45 3.18 1.193 16 SB: I plan to transfer to another academic institution to complete my degree matched 45 3.36 1.464 unmatched 45 3.42 1.288
52 17 SB: I intend to call or email my state government officials to voice my concern on this crisis that is affecting USF matched 45 3.31 1.427 unmatched 45 2.56 1.324 18 EFF: I am satisfied by the way USF has managed this crisis matched unmatched 45 45 2.69 2.53 .973 1.014 19 EFF: I believe that USFs response to this crisis is effective matched 45 2.44 .893 unmatched 45 2.51 1.014 20 EFF: I do not believe USF has handled this crisis as well as it could have matched 45 2.49 1.079 unmatched 45 2.42 .941 21 EFF: I am comfortable that USF responded to the crisis in a manner that accommodated to the students best interest. matched 45 2.38 1.072 unmatched 45 2.62 .984 Separate independent samples t -tests were conducted on ea ch item to determine if any of the mean scores for the matched and unm atched crisis responses were significantly different. Only the item, I intend to call or email my state government officials to voice
53 my concern on this crisis that is affecting USF, indicated a significant different between matched (N=45, M=3.31, SD=1.43) and unmatched (N=45, M=2.56, SD=1.32) responses. The Results are shown in Table 10. Table 10. Independent Sample t -tests Levene's Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) 1 Org ReP*: USF is concerned with the well-being of its publics Equal variances assumed .069 .793 -.506 88 .614 Equal variances not assumed -.506 87.994 .614 2 Org Rep*: USF is basically dishonest Equal variances assumed .119 .731 -.922 88 .359 Equal variances not assumed -.922 87.990 .359 3 Org Rep*: I do NOT trust USF to tell the truth about the incident Equal variances assumed .029 .866 -1.367 88 .175 Equal variances not assumed -1.367 87.957 .175 4 Org Rep*: Under most circumstances, I would be likely to believe what USF says Equal variances assumed .131 .719 -1.187 88 .238 Equal variances not assumed -1.187 87.740 .238 5 Org Rep*: USF is not concerned with the well-being of its students Equal variances assumed 1.834 .179 -.553 88 .582 Equal variances not assumed -.553 86.527 .582
54 6 CR*: Under most circumstances, I would be likely to believe what USF says Equal variances assumed .159 .691 -.527 88 .600 Equal variances not assumed -.527 87.932 .600 7 CR*: The blame for the crisis lies with USF Equal variances assumed 4.989 .028 1.878 88 .064 Equal variances not assumed 1.878 84.862 .064 8 CR*: The blame for the crisis lies in the circumstances, not USF Equal variances assumed 1.213 .274 1.064 88 .290 Equal variances not assumed 1.064 87.768 .290 9 OC*: The cause of the crisis was something that USF could control Equal variances assumed 1.803 .183 -1.601 88 .113 Equal variances not assumed -1.601 86.138 .113 10 OC*: The cause of the crisis is something over which USF had no power Equal variances assumed .018 .893 -.730 88 .467 Equal variances not assumed -.730 88.000 .467 11 OC*: The cause of the crisis is something that was manageable by USF Equal variances assumed 1.517 .221 -.516 88 .607 Equal variances not assumed -.516 87.299 .607 12 OC*: The cause of the crisis is something over which USF had power Equal variances assumed .738 .393 .719 88 .474 Equal variances not assumed .719 86.280 .474 13 SB*: I plan to continue my college education at USF Equal variances assumed .066 .798 .000 88 1.000 Equal variances not assumed .000 87.670 1.000
55 14 SB*: I intend to say nice things about USF to other people Equal variances assumed .665 .417 -1.129 88 .262 Equal variances not assumed -1.129 87.540 .262 15 SB*: I intend to support USFs administrative decision Equal variances assumed .180 .672 -.971 88 .334 Equal variances not assumed -.971 88.000 .334 16 SB*: I plan to transfer to another academic institution to complete my degree Equal variances assumed 1.627 .205 -.229 88 .819 Equal variances not assumed -.229 86.592 .819 17 SB*: I intend to call or email my state government officials to voice my concern on this crisis that is affecting USF Equal variances assumed .642 .425 2.604 88 .011 Equal variances not assumed 2.604 87.506 .011 18 EFF*: I am satisfied by the way USF has managed this crisis Equal variances assumed .029 .866 .743 88 .460 Equal variances not assumed .743 87.853 .460 19 EFF*: I believe that USFs response to this crisis is effective Equal variances assumed 1.143 .288 -.331 88 .741 Equal variances not assumed -.331 86.622 .741 20 EFF*: I do not believe USF has handled this crisis as well as it could have Equal variances assumed .448 .505 .312 88 .756 Equal variances not assumed .312 86.403 .756
56 21 EFF*: I am comfortable that USF responded to the crisis in a manner that accommodated to the students best interest. Equal variances assumed .250 .618 -1.127 88 .263 Equal variances not assumed -1.127 87.356 .263 Org Rep = Organizational Reputation; CR = Crisis Res ponsibility; OC = Organizational Control; SB = Potential Supportive Behavior; EFF = Overall Effectiveness Each of the hypotheses and the meaning of the corresponding results will be discussed in the following chapter. From the examination, conclusions are formed and recommendations are offered for future research.
57 Chapter Five Discussion This study examined the situational crisis communication theory to test if the theory would preserve an organizations repu tation in the wake of a crisis. The SCCT offers a set of principles that guide the selection of the cris is response strategies in order to maximize reputational protection (Coombs & Holladay, 2002). SCCT provides crisis managers with guidelines for understandi ng which response strategies are most appropriate for a given crisis type (Coombs, 1995). In an attempt to contribute to the development and refinement of the SCCT, th ree hypotheses were tested in this study. H1, which states that in an accident cr isis situation, a matched organizational response message will produce a more positiv e perception of organizational reputation than an unmatched organizational response st rategy, was not supported by the results of this study. This finding does not confirm the premise of the situational crisis communications theory and, therefore, does not add to the validity of the theory. H2, which states that in an accident cr isis situation, a matched organizational response message will produce a more positive pe rception of crisis responsibility for an organization than an unmatched organiza tional response strate gy, was not supported by the results of this study. Although not s upported, this was the only hypothesis that suggested slight support for the matched re sponse. It was not a significant amount; however, there was opportunity that, with mo re research, it significant results could possibly emerge. Per the results, this fi nding does not confirm the premise of the
58 situational crisis communications theory and, therefore does not add to the validity of the theory. H3, which states that in an accident cr isis situation, a matched organizational response message will produce a more positiv e potential supportive behavior than an unmatched organizational response strategy. This finding does not confirm the premise of the situational crisis communications theory and, therefore, does not add to the validity of the theory. It was hypothesized that in an accident cr isis situation, a matched organizational response message will produce a more positiv e perception of organizational reputation, crisis responsibility, and pot ential supportive behavior than an unmatched organizational response strategy. All hypotheses were not supported by the resu lts of the study. It should also be discussed, the item th at the alpha indica ted that should be dropped from organizational reputation to give a better scale reliability. By dropping the item I intend to call or email my state gove rnment officials to voice my concern on this crisis that is affecting USF, a better coe fficient was achieved. W ith all other responses suggesting low involvement from the particip ants, what provoked the students to want to go above and beyond to contact the government? When all previous studies on the SCCT i ndicate that a particular result should occur, but the results come back inconclusive it must be taken into consideration that there were unfactored variab les that were not considered while conducting the study. The first variable that should be discussed is the lack of apathy from the participants. The study did not take the participants presum ptive opinion about USF into consideration which could have made the study clearer to why the participants answered
59 the questions in the manner that they did. Knowing the participants presumptive opinion about USF is a key factor in this study because each stakeholders view on an organization varies and some of the participants in the study could have had a predisposed opinion that did not favor USF which may have caused heavily opinionated responses that ultimately prove d every hypothesis inconclusive. The next variable up for discussion is lo w involvement from participants. This should not be ruled out for the cause of insignifi cant data because it could simply be that the participants did not care to be highly involved in the study because, even though the crisis was purely hypothetical, it did not affect their academ ic plans and if the study had nothing to do with them, then it is very clea r to see why the particip ants would not be as involved in something that did not have a direct affect on their academic careers. Besides the prediction of low involvement from the participants a valid limitation on the results that is beyond th e researchers control could st em from the fact that the participants were part of a large class that is constantly saturated with surveys and experiments from graduate students because the class produces a large number for a diverse outcome of results. Another factor that was discovered while analyzing the results was that too much information was disseminated before the actual experiment was administered. The participants of the study were alerted that the crisis in the experiment was purely hypothetical even before the pa rticipants received the survey to complete. With this major factor, it should be made aware that th e results of this study should be assessed skeptically and that this coul d be one of the main reasons why the results came out the way they did.
60 Although the results from the study di d not produce significance that was supposed to further validate the SCCT, a more significant discovery was found that is more important than if this study actually fu rthered SCCTs validation. While trying to group a hypothetical crisis with USF in the cen ter, it was very difficult to categorize the crisis type because the crisis types list did not accommodate the type of crisis that may occur at an educational institution or organizatio n of the sort. This is crucial to the SCCT because if it were an actual implemented th eory, it would be very difficult for a crisis manager to craft an effective response for a cr isis type that does not have a category. As described on page 25, it would appear th at a hypothetical crisis of this caliber would fit in organizational misdeed with no injuries, organizational misdeed management misconduct, or challenges. However, if ex amined closer, in a crisis like this; stakeholders are never deceived, laws or re gulations are not viol ated by management (administration), and the educational instit ution is not operating in an inappropriate manner no matter the responsibility acceptance level. If this crisis di d not fit any of the predisposed crisis types, then it must be conc luded that another crisis type and/or a more detailed description must be added to accommodate this type of crisis. Future Research Research should begin to assess how cr isis types are perceived. Through the scant research on the SCCT are researchers paying enough attention to make sure that there is a category for every type of possible crisis? Because my study was a hypothetical crisis that could severely affect the educational institutions stakeholders and it did not fit any of the 13 different types of crises. If there is not room for a crisis type of this caliber, then one must be conjured because all assumptions in crisis management
61 must be accurate if these assumptions are being used to recomme nd plans of action for crisis managers. Conclusion Despite limitations, the present study offers some insight on the effect of a crisis on perceptions organizations involved in crisis. It is important that the SCCT be tested to build reliable social science a nd to contribute to the consta nt changing and developing of crisis communication management. Results fr om this study do not support the SCCT in validating that a matched response is more effective than an unmatched response. However, it should be taken into serious account that if this study we re to be duplicated and all limitations were covered and/or co rrected, the results would support the SCCT which will provide more information to crisis managers in crafting more strategic messages that will more effectively protect an organizations reputation.
62 References AON. (2006). AON. Retr ieved Oct. 7, 2007, from  ( http://www.aon.com/risk_management/cri sis_management/crisis_mgmt_plannin g.jsp ) Abelson, R. P. (1959). Modes of resolution of belief dilemmas. Journal of Conflict Resolution 3, 343-352. Allen, M. W., & Caillouet, R. H. ( 1994). Legitimation endeavors: Impression management strategies used by an organization in crisis. Communication Monographs 61, 44-62. Balzer, W. K., & Sulsky, L. M. (1992). Ha lo and performance a ppraisal research: A critical examination. Journal of Applied Psychology. 77, pp. 975-985. Barton, L. (1993). Crisis in organizations: Managing and communicating in the heat of chaos Cincinnati, OH: College Divisi ons South-Western Publishing Bedeian, A. G. (1989). Management (2nd ed.). Chicago: Dryden. Benoit, W. L. (1995a). Accounts, excuses, apologies: A theory of image restoration discourse. Albany: State University of New York Press. Benoit, W. L. (1995b). Sears repair of its auto service image: Image restoration discourse in the corporate sector. Communication Studies, 46 pp. 89-105. Benoit, W. L. (1997a). Hugh Grants image restoration discourse: An actor apologizes. Communication Quarterly, 45, pp. 251-267. Benoit, W. L. (1998). Merchants of death: Persuasive defeses by the tobacco industry. In J. F. Klumpp (Ed.), Argument in a time of change : Definition, frameworks, and critiques (pp. 220-225). Annandale, VA: NCA. Benoit, W. L. (2007). President Bush's major post-Katrina speeches: Enhancing image repair discourse theory app lied to the public sector. Public Relations Review 33, pp 40-48. Benoit, W. L., & Anderson, K. K. (1996). Blending politics and entertainment: Dan Quayle versus Murphy Brown. Southern Communication Journal, 62, pp. 73-85.
63 Benoit, W. L., & Brinson, S. L. (1999). Queen Elizabeths image repair discourse: Insensitive royal or compassionate queen? Public Relations Review, 25, pp. 145156. Benoit, W. L., & Czerwinski, A. (1997). A critical analysis of USAirs image repair discourse. Business Communication Quarterly, 60, pp. 38-57. Benoit, W. L., & Pang, A. (2007). Crisis communication and image repair discourse. Public relations: from theo ry to practice, pp. 244-259. Benoit, W. L., Gullifor, P., & Panici, D. (1991). President Reagans defensive discourse on the Iran-contra affair. Communication Studies, 42, pp. 272-294. Benoit, W. L., & Hanczor, R. (1994). The Tonya Harding controvers y: An analysis of image repair strategies. (Communication Quarterly, 42, pp. 416-433. Benoit, W. L., & Hirson, D. (2001). Doonesbury versus the Tobacco Institute: The Smoke Starters Coupon. Communication Quarterly, 49, pp. 279-294. Benoit, W. L., & McHale, J. P. (1999). Kenne th Starrs image repair discourse viewed in 20/20 Communication Quarterly, 47, pp. 265-280. Benoit, W. L., & Nill, D. M. (1998a). A cr itical analysis of Judge Clarence Thomass statement before the Senate Judiciary Committee. Communication Studies, 49, pp. 179-195. Benoit, W. L., & Wells, W. T. (1998). An an alysis of three image restoration discourses on Whitewater. Journal of Public Advocacy, 3, pp. 21-37. Benson, J. A. (1988). Crisis revisited: An analysis of the strategies used by Tylenol in the second tampering episode. Central States Speech Journal, 39, pp. 49-66. Bergman, E. (1994). Crisis? What crisis? Communication World, 11, pp. 9-13. Birch, J. (1994). New factors in crisis planning and response. Public Relations Quarterly, 39(1), pp. 31-34. Blaney, J. R., Benoit, W. L. (1997). The persuasive defense of Jesus in the Gospel according to John. Journal of Communication and Religion, 20, pp. 25-30. Blaney, J. R., Benoit, W. L., & Brazeal, L. M. (2002) Blowout! Firestones image restoration campaign. Public Relations Research, 28, pp. 379-392. Brinson, S., & Benoit, W. L. (1996). Dow Cornings image repair strategies in the breast implant crisis. Communication Quarterly, 44, pp. 29-41.
64 Campbell, D. T., & Stanley, J. C. (1963). Experiment and quasi-experiment designs for research. Chicago: Rand McNally College Publishing Company. Caponigro, J. R. (2000). The crisis counselor: A step-by-step guide to managing a business crisis. Chicago: Contemporary Books. Center, A. H., & Fellow, P. J. (2003). Public relations practices: Managerial case studies and problems Prentice Hall. Cohen, J. R. (1999). Advising clients to apologize. S. California Law Review, 72, pp. 1009-131. Cohen, R. (2000). The PR crisis bible. New York: St. Martins. Coombs, W. T. (1995). Choosing the right words: The development of guidelines for the selection of the appropriate crisis response strategies. Management Communication Quarterly, 8, pp. 447-476. Coombs, W. T. (1998). An an alytic framework for crisis situations: Bett er responses from a better understanding of the situation. Journal of Public Relations Research, 10, pp. 177-191. Coombs, W. T. (1999a). Information and comp assion in crisis responses: A test of their effects. Journal of Public Relations Research, 11, pp. 125-142. Coombs, W. T. (1999b). Ongoing crisis communication Planning, managing, and responding. Thousand Oaks, CA: Sage. Coombs, W. T. (2002b, November). Further testing of the situational crisis communication theory: An extended examinatio n of crisis history as a modifier. Paper presented at the annual mee ting of the National Communication Association, New Orleans, LA. Coombs, W. T. (2006). Attr ibution Theory as a guide for post-crisis communication research. Public Relations Review 33, pp. 135-139 Coombs, W. T. (2007). The development of the situation crisis communication theory. Pearson. Coombs, W. T., Hazleton, V., holladay, S. J., & Chandler, R. C. (1995). The crisis grid: Theory and application in crisis management. In L. Barton (Ed.), New avenues in risk and crisis management (Vol. 4, pp. 30-39). La s Vegas, NV: UNLV Small Business Development Center.
65 Coombs, W. T., & Holladay, S. J. (1996). Communication and attrib utions in a crisis: An experimental study of crisis communication. Journal of Public Relations Research, 8, pp. 279-295. Coombs, W. T., & Holladay, S. J. (2001). An extended examination of the crisis situation: A fusion of the relational management and symbolic approaches. Journal of Public Relations Research, 13, pp. 321-340. Coombs, W. T., & Holladay, S. J. (2002). He lping crisis managers protect reputational assets: Initial tests of the situational crisis communication theory. Management Communication Quarterly, 16, pp. 165-186. Coombs, W. T., & Schmidt, L. (2000). An empirical analysis of image restoration: Texacos racism crisis. Journal of Public Relations Research, 12(2), pp. 163178. Dimaggio, P. J., & Powell, W. W. (1991). The iron cage revisited: Institutional isomorphism and collective rationality in or ganization fields. In W. W. Powell & P J Dimaggio (Eds.), The new institutionalism in organizational analysis (pp. 63-82). Chicago: University of Chicago Press. Drumheller, K., & Benoit, W. L. (2004). USS Greenville collides with Japans Ehime Maru: Cultural issues in image repair discourse. Public Relations Review, 30, pp. 177-185. Fearn-Banks, K. (1996). Crisis communications: A casebook approach. Mahwah, NJ: Lawrence Erlbaum Associates. Fearn-Banks, K. (2007). Crisis communications: A casebook approach. Mahwah, NJ: Lawrence Erlbaum Associates. Fink, S. (1986) Crisis management Planni ng for the inevitable New York. AMACOM General Dynamics lands F-16 contract (1986, October) Dallas Morning News P A1. France, M. (2002, August 26). The mea culpa defense. Business Week, 3796 pp. 7678. Griffin, M., Babin, B. J., & Darden, W. R. (1992). Consumer assessments of responsibility for product-rela ted injuries: The impact of regulations, warnings, and promotional policies. Advances in Consumer Research, 19, pp. 870-877. Hamblin, R. L. (1958). Leadership and crises. Sociometry 21 pp. 322-335. Hearit, K. M. (1994, Summer). Apologies and public relations cr ises at Chrysler, Toshiba, and Volvo. Public Relations Review, 20, pp. 113-125.
66 Heath, R. L. (1994). Management of corporate communi cation: From interpersonal contacts to external affairs. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Herbig, P., Milewicz, J., & Golden, J. (1994). A model of re putation building and destruction. Journal of Business Research 31, pp. 23-31. Hoff, R. (1980, January). Corporate adve rtising: Too many objectives equal mush. Duns Review, 127-129. Irvine, R. B., & Millar, D. P. (1996). De bunking the stereotypes of crisis management: The nature of business crises in the 1990s. In L. Barton (Ed.), New avenues in risk and crisis management, Volume V (pp. 51-63). Las Vegas: University of Nevada, Las Vegas Small Business Development Center. Kennedy, K. A., & Benoit, W. L. (1997). The Newt Gingrich book d eal controversy: A case study in self-defense rhetoric. Southern Communication Journal, 63, pp. 197-216. Len-Rios, M., & Benoit, W. L. (2004). Gary Condits image repair strategies: Squandering a golden opportunity. Public Relations Research, 50, pp. 95-106. Lerbinger, O. (1997). The crisis manager: Facing risk and responsibility. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Marcus, A. A., & Goodman, R. S. (1991). Victims and shareholders: The dilemmas of presenting corporate policy during a crisis. Academy of Management Journal, 34, 281-305. McAuley, E., Duncan, T. E., & Russell, D. W. (1992). Measuring causal attributions: The revised causal dimension scale (CDII). Personality and Social Psychology Bulletin, 18, 566-573. McCroskey, J. C. (1966). An introduction to rhetorical communication. Englewood Cliffs, NJ: Prentice Hall. Moffitt, M. A. (1994). Collapsing and integr ating concepts of public and image into a new theory. Public Relations Research, 20, pp. 159-170. Mitroff, I. I. (2001). Managing crises before they happen. What every executive and manger needs to know about crisis management. New York: AMACOM. Ogrizek, M., & Guillery, J. M. (1999). Communicating in crisis New York: Aldine De Gruyter.
67 Pauchant, T. C., & Mitroff, I. I. (1992). Transforming the crisis-prone organization: Preventing individual, organizational, and environmental tragedies. San Francisco: Jossey-Bass. Pearson, C. M., & Clair J. A. (1998) Reframing crisis management. The Academy of Management review 23, pp. 59-76. Pearson, C. M., & Mitroff, I. I. (1993). From crisis prone to crisis prepared: A framework for crisis management. The Executive 7, pp. 48-59. Pinsdorf, M. K. (1987). Communicating when your company is under siege: Surviving public crisis. Lexington, MA: Heath. Ray, S. J. (1999). Strategic communication in crisis management. Lessons from the airline industry. Westport, CT: Quorum Books. Russell, D. (1982). The causal dimension scal e: A measure of how individuals perceive causes. Journal of Personality and Social Psychology, 42, pp. 1137-1145. Seegar, M. W., Sellnow, T. L., & Ulmer, R. R. (1998). Communi cation, organization, and crisis. In M. E. Roloff (Ed.), Communication yearbook 21 (pp. 231-276). Thousand Oaks, CA: Sage. Siomkos, G. & Shrivastava, P. (1993). Responding to product liability crises. Long Range Planning 26(5), 72-79. Sturges, D. L. (1994). Communication through crisis: A strategy for organizational survival. Management Communication Quarterly, 7, pp. 297-316. Ware, B. L., & Linkugel, W. A. (1973). Th ey spoke in defense of themselves: On the generic criticism of apologia. Quarterly Journal of Speech, 59, pp. 273-283. Weiner, B. (1985). An attributional theo ry of achievement motivation and emotion. Psychology Review, 92, pp. 548-573. Weiner, B., Amirkan, J., Folkes, V. S., & Verette J. A. (1987). An attribution analysis of excuse giving: Studies of a nave theory of emotion. Journal of Personality and Social Psychology, 53, pp. 316-324. Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attribution anal ysis of reactions to stigmas. Journal of Personality and Social Psychology, 55, pp. 738-748. Wilson, S. R. (1993). An attribution anal ysis of compliance ga ining interactions. Communication Monographs 60, pp. 352-372.
68 Zhang, J., & Benoit, W. L. (2004). Messa ge strategies of Saudi Arabias image restoration campaign after 9/11. Public Relations Review 30, pp. 161-167.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002028692
007 cr bnu|||uuuuu
008 090911s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002802
Responding to crises :
b a test of the situational crisis communication theory
h [electronic resource] /
by Courtney Wright.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 68 pages.
Thesis (M.A.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Crisis management includes efforts designed to prevent and to detect potential crises, and to learn from crisis experiences. The SCCT posits that certain crisis responses (matched) produce better outcomes for organizations than others (unmatched), depending on the situation. In addition, the results from this study attempt to support the situational crisis communication theory in aiding crisis managers in protecting their organizations against crises.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Kelly Werder, Ph.D.
Crisis communication management
Image repair discourse
x Mass Communications
t USF Electronic Theses and Dissertations.