<%BANNER%>

Exploring a Multi-Stage Model of Crisis Management: Utilities, Hurricanes, and Contingency


PAGE 1

EXPLORING A MULTI-STAGE MODEL OF CRISIS MANAGEMENT: UTILITIES, HURRICANES, AND CONTINGENCY By BRIAN BOUDREAUX A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Brian Boudreaux

PAGE 3

iii ACKNOWLEDGMENTS I would like to thank my chairman, Dr. Michael Mitrook, for his guidance throughout this project. His insights, support, and candor were fundamental to the completion of this study. Additionally, my committee members, Dr. Peg Hall and Dr. Jennifer Robinson, provided a refreshing view of this piece that will not soon be forgotten. Lastly, I would lik e to thank my family for th eir continued support of my academic endeavors. This would not have been possible without their never-ending support.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF FIGURES...........................................................................................................vi ABSTRACT......................................................................................................................v ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 REVIEW OF LITERATURE.......................................................................................3 Reality of a Crisis.........................................................................................................3 Issues Management.......................................................................................................4 Models of Crisis Management......................................................................................7 The Contingency Theory of Accommodation............................................................14 Criticisms of Contingency Theory......................................................................16 More on Advocacy..............................................................................................17 More on Accommodation....................................................................................18 Contingency Variables........................................................................................19 Predisposing variables..................................................................................19 Situational variables.....................................................................................22 3 METHODOLOGY.....................................................................................................26 Review of Research Questions...................................................................................26 Data Collection...........................................................................................................26 Sample........................................................................................................................2 8 Data Analysis..............................................................................................................29 4 RESULTS AND DISCUSSION.................................................................................30 Phase 1........................................................................................................................ 30 Phase 1.1..............................................................................................................31 Phase 1.2..............................................................................................................34 Phase 1 Conclusion..............................................................................................35 Phase 2........................................................................................................................ 36 Phase 2.1..............................................................................................................37

PAGE 5

v Phase 2.2..............................................................................................................38 Phase 2 Conclusion..............................................................................................40 Phase 3........................................................................................................................ 41 Phase 3.1..............................................................................................................41 Phase 3.2..............................................................................................................42 Phase 3 Conclusions............................................................................................43 5 LIMITATIONS AND FUTURE STUDY..................................................................44 6 SUMMARY AND CONCLUSIONS.........................................................................46 APPENDIX A E-MAIL RECRUITMENT LETTER.........................................................................53 B BIOGRAPHICAL INFORMATION..........................................................................54 C PRIMARY QUESTION FORMAT...........................................................................57 D FOLLOW-UP QU ESTION FORMAT.......................................................................58 E SAMPLE REMINDER E-MAIL................................................................................59 F PHASE 1.1 QUESTION AND RESPONSES............................................................60 G PHASE 1.2 QUESTION AND RESPONSES............................................................63 H PHASE 2.1 QUESTION AND RESPONSES............................................................65 I PHASE 2.2 QUESTION AND RESPONSES............................................................67 J PHASE 3.1 QUESTION AND RESPONSES............................................................69 K PHASE 3.2 QUESTION AND RESPONSES............................................................71 LIST OF REFERENCES...................................................................................................72 BIOGRAPHICAL SKETCH.............................................................................................75

PAGE 6

vi LIST OF FIGURES Figure page 1 Crisis Management: St rategic Considerations...........................................................8 2 The Crisis Life Cycle.................................................................................................9 3 Development of Issues with or Without Management Intervention..........................9 4 Advocacy-Accomodation Continuum......................................................................14 5 Contingency Variables in Crisis Management.........................................................51

PAGE 7

vii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Mast er of Arts in Mass Communication EXPLORING A MULTI-STAGE MODEL OF CRISIS MANAGEMENT: UTILITIES, HURRICANES, AND CONTINGENCY By Brian Boudreaux May 2005 Chair: Michael Mitrook Major Department: Journalism and Communications The purpose of this study was to expl ore the three-stage model of crisis management and determine what influence the contingency theory of accommodation’s highly-supported-predisposing and situational va riables have on crisis management. This qualitative study examines crisis ma nagement practices of municipal utility companies in the state of Florida. Particip ants discussed their organization’s crisis management practices in conjunction with the very active 2004 hurricane season. A modified Delphi study was used to collect data from senior-level public relations and corporate communications professionals from a cross-section of st ate municipal utility companies. Primary and secondary questions referencing each stag e of the three-stage model of crisis management were generated to identify the variables that most affected organizational crisis management. The findings expose a relationship between the highly-supported-predisposing and situational variables of the contingency theory and the three-stage model of crisis

PAGE 8

viii management. This study serves as a springboard for future quantitative research that may be able to generate a stat istical correlation between the contingency theory’s highlysupported variables and the three-st age model of crisis management.

PAGE 9

1 CHAPTER 1 INTRODUCTION Managing crises is a vital aspect of public relations practice as all organizations experience them. As one pr actitioner explains, “no one is immune. Crises are nondiscriminatory. They don’t care who gets in th e way. It’s not a matter of ‘if,’ in most cases, but simply ‘when.’” (Wilson, 2004, paragr aph 2). Likewise, L. Grunig, J. Grunig and Dozier (2002) asserted, “Reg ardless of the model of public relations practiced or the expertise of the communicator, crises inevit ably befall organizations” (p. 473). The belief that a crisis situation will not strike an organization must change; the mentality must shift from if, to when. Indeed, crises are a reality that companies, government agencies and nonprofit organizations alike must recognize. The incidence of crises is not a new phenomenon, but increasingly catastrophic results have va lidated their relevance in public relations. The ubiquitousness of crisis situations necessitates the creation of a widely accepted model of a crisis management that provides organizations the framework to assist in the survival of a crisis event. Though there are models that explore crisis management, these models have yet to include a concrete theoreti cal underpinning. This qualitative study attempts to begin grounding th e three-stage model of crisis management in the contingency theory of accommodation. The researcher believes that the contingency theory’s highly-supportedpredisposing and situational variables, identifi ed by Cancel et al. (1999), will emerge as influential to crisis management. The resear cher proposes that the predisposing variables

PAGE 10

2 will appear in the participant responses during precrisis stage while the situational variables will emerge in the crisis stage of the three-stage model of crisis management. This study will test this hypothesis while also exploring the role of the contingency theory variables in crisis management.

PAGE 11

3 CHAPTER 2 REVIEW OF LITERATURE Reality of a Crisis Although all types of organizations are vulne rable to a crisis, certain industries are inherently more prone to a cr isis event. L. Grunig et al (2002) quoted a practitioner who refers to the insurance industry as a “cri sis by definition” (p. 473). Among the many reasons cited by scholars and pr actitioners to prepare for a crisis is one advanced by Howard and Mathews (2000). Preparation is needed, they said, “. .because you cannot control the elements, human nature or the outside world” (p. 217). This statement outlines the difficulty of crisis managemen t; though practitioners and organizations can prepare for numerous scenarios, there ar e many more they have no control over. Before further discussion, the definition of a crisis must be established. Scholars have defined the term crisis in a number of ways. As de fined by Sapriel (2003), a crisis is “an event, revelation, allegation or set of circumstances which threatens the integrity, reputation, or survival of an individual or or ganisation” (p. 348). Shrivastava, Mitroff, Miller, and Miglani (1988) presented an in-depth de finition of a crisis as An organizationally-based disaster, which causes exte nsive damage and social disruption, involves multiple stakeholders, and unfolds through complex technological, organizational, a nd social processes. (p. 285) Much of the literature discusses crises at the organizational level. In doing so, the definition of a crisis has focused on only a segment of crises. Pauchant and Mitroff (1992) developed a definition of a crisis that can apply to more than just organizationally based events. The authors discuss a crisis as “a disruption that physically affects a

PAGE 12

4 system as a whole and threatens its basic as sumptions, its subjective sense of self, its existential core” (p.15). This definition encompasses non-organization crises, such as natural disasters, that have an effect, not only on individual organizations, but rather a community system as a whole. Other scholars take a more basic approach to defining a crisis. Coombs (1999a) described a crisis as something that embarra sses or challenges an or ganization’s character and that demands an explanat ion. Gonzlez-Herrero and Prat t (1996) defined a crisis as something that is simply “unwelcome and s udden” (p. 82). Lastly, Coombs and Holladay (2001) defined a crisis as “one event or inte raction within a larger relationship between an organization and its stakeholders. .[that can] damage or be a threat to a quality relationship” (p.324). Though the depths of de finitions vary, they each employ the same principle: A crisis is an event that can drastically affect the ability of an organization to sustain itself. This will serve as the defin ition of a crisis for discussion throughout this study. A danger that organizations sometimes face is the issue of a perceived crisis. Organizations may leap into crisis manageme nt protocols for mere transgressions. For example, the resignation of an organization’s Ch ief Financial Officer al one is not a crisis and does not warrant the use of crisis protocols. If the resi gnation of a CF O is the result of financial malfeasance, though, the organi zation should address the situation as a possible crisis. Organizations ne ed to be able to differentia te between perceived and real crises. Issues Management Gonzlez-Herrero and Pratt (1996) posit that a crisis is sudden. Although the onset of a crisis can be sudden, there are situations when or ganizations are aware of and

PAGE 13

5 prepared for such an event. Though many fact ors influence the outcome of a crisis, Smits and Ezzat (2003) posit that preparation is one of – if not the – most important steps in the prevention of a crisis, “Effective crisis ma nagement depends upon planning and people” (p. 2). Penrose (2000) stated, “Researchers te nd to agree that organi zations that practice proactive crisis management will lessen the damage of a crisis” (p. 155). Indeed, Gonzlez-Herrero and Pratt (1996) also explai ned that crises have early signals and, “…sensing potential problems is the first st ep toward avoiding or resolving them or minimizing their impact” (p. 82). Issues management helps an organization be come aware of possible crises and plan accordingly. One facet of this awareness involves determining those publics which could lead or contribute to a cris is. As J. Grunig and Repper (1992) explained, “Members of active publics, affect organizations more th an passive ones because they engage in individual behaviors to do so mething about the consequenc e of organizational actions” (p. 137). If an organizati on can identify these groups it can attempt to dispel or counteract damaging behaviors. Issues management helps remove the unexp ected and sudden factors of the crisis; the situation can be defused before it occurs Moreover, information gathered will not only allow practitioners to pr epare, it will allow them th e opportunity to realize and understand what preparations they cannot make “Nothing prepares you for change better than the awareness of what you can d o, and cannot do, about it” (Goodman, 2001, p. 117). The occurrence of some events is unc ontrollable, thus, pract itioners should shift their efforts away from preventing such events and towards weathering them. For instance, a state is aware of a hurricane days before it makes landfall. Though issues

PAGE 14

6 management allows the opportunity to prepar e for the possible cris is, it cannot neutralize the storm. Instead, issues management allows practitioners to prepare for the event as well as the resolution after the crisis occurs. Howard and Mathews (2000) describe issu es management as a practice that good managers have done for years; that it is a necessary part of effective planning. GonzlezHerrero and Pratt (1996) discuss issues manage ment as a tool to identify and anticipate potential issues before they are a threat. Thus, issues managers should be forwardthinking. The longer an organization is awar e of the possible issue, the better it can prepare. Indeed, Ewing (1987) posits that i ssues managers should look 12 to 36 months ahead. Issues management is often thought of as a proactive practice used to avoid a negative situation, thoug h it can also identify positive opportunities. Practicing issues management improves an organization’s aw areness of its community and ways to positively involve itself. For example, an organization can improve its standing in a community by being aware of and supporting an upcoming philanthropic event. Issues management can also present organizations wi th an opportunity to showcase its positive practices. Unfortunately, preparation and issues mana gement is sometimes not sufficient to avoid a crisis. Numerous inci dents illustrate that although organizations were prepared for an event, the crisis was still damaging. Ex amples of such scenario s can be seen in all professions: corporate, liti gation, and nonprofit. Prepara tion, though valuable, is not a guarantee of success. The difficulties associ ated with a given crisis will impact the effectiveness of the preparation. Burnett (1998) presents some difficulties crisis

PAGE 15

7 managers may face, they include too little da ta, too much data, and little planning. These variables illustrate that simply having info rmation and early preparation may not be sufficient deterrents. Models of Crisis Management “The disparate volumes of crisis management information can be overwhelming” (Coombs, 1999b, 9). Certainly there are many f actors a crisis management professional must account for in order to be successful. The organization of such information into a universally accepted model has presented pract itioners and scholars with a challenge. Burnett (1998) identifies both tasks and factors that comp romise the ability of an organization to practice crisis management (Figur e 1). First, the author cited four factors that inhibit crisis management: time pressure control issues, threat level concerns, and response option constraints. Burnett claims these factors, found on the outer-ring of the model, disrupt an organization’s ability to focus on and strategically manage a crisis situation. According to this model, only when these four factors have been addressed can the strategic management of the situation begin. Burnett (1998) divides the model’s six st ep inner-circle into three categories: identification, confrontation, and reconfiguration. The identification step is composed of goal formation and environmental analysis – th e preparation for the cr isis. Confrontation encompasses strategy formulation and st rategy evaluation – the point when an organization is involved in the crisis. Lastly, reconf iguration includes strategy implementation and strategic control – how the or ganization adapts to crisis intervention. The author posits that during a crisis the difficulty of performing well in each category increases. As illustrated by the model, employing the tasks that comprise the inner-circle provides the organization an opportunity to control and manage a crisis situation.

PAGE 16

8 Figure 1: Crisis Management: Strategic C onsiderations; Adopted from Burnett (1998) Some models make the comparison of the crisis to a lifecycle (Fink, 1986; Gonzlez-Herrero & Pratt, 1996). Inherent to this analogy is that the crisis has both a birth and a death. GonzlezHerrero and Pratt (1996) di scuss how crises follow a sequential path through four pha ses: birth, growth, maturity and decline (Figure 2). Although this is an elementary model of a cris is, it is sufficient. It divides a crisis into identifiable stages, it illustrates how a crisis changes over time, and that the cycle does not end, rather that its effects linger be yond the decline and death of the crisis. This basic model presents a simplistic, yet effective illustration of the crisis lifecycle. Gonzlez-Herrero and Pratt (1996) expanded th is model to illustra te the effect of issues management in a crisis situation. By practicing issues mana gement before crisis birth, the authors believe organizations can sh ift the outcome of the crisis (Figure 3). Previously, the crisis would ha ve reached maturity, to eventually decline into the postcrisis phase. In this adaptation of the model, issues management is shown to be effective as the planning stage results in the prevention of a crisis.

PAGE 17

9 Figure 2: The Crisis Life Cycle; Adopte d from Gonzlez-Herre ro and Pratt (1996) Figure 3: Development of Issues with or Without Management Intervention; Adopted from Gonzlez-Herrer o and Pratt (1996) Coombs (1999b) states that th e three most influential stag ed approaches to crisis management are Fink’s (1986) four-stage mode l of a crisis lifecycle, Mitroff’s (1994) five-stage model, and the basic three-stage model. The three-stage model is unique in that no single scholar is attributed with its creation. “The three-stage model is not associ ated with any particul ar theorists, but it appears to have emerged from several research efforts as a general analytical framework”

PAGE 18

10 (Seeger, Sellnow, and Ulmer, 2003, 97). Coom bs (1999b) describes the three stages of the model – precrisis, crisis, and postcrisis – as macrostages that can be applied to many models of crisis management. The precrisis stage includes all aspects of crisis prevention – issues management, planning, and other proactive steps. The crisis stage refers to the steps taken to cope with and respond to the crisis event – crisis recognition, information distribution, message development, reputation management, and e volving developments. The postcrisis stage begins when the crisis is resolved – ensu ring the crisis over, a ssuring publics of the security of the organization, and learning from the event. Coombs (1999b) and Seeger et al. (2003) contend that the three-stage model provides a framework for the incorporation of various sub-stages which change based on a multitude of variables. The type and impact of the crisis, media coverage of the event, and the size and culture of the affected orga nization can all be in fluential factors. Coombs explains that both Fink (1986) and Mitroff’s (1994) models fit into the general parameters of the three-stage model. Fink’s (1986) four-stage model examines a crisis as an extended event with sufficient warning signs that precede the event. Fink’s four stages are: the prodromal stage, the acute stage, the chronic stage, and the resolution stage. In the prodromal stage, the role of a crisis management professiona l is not reactive, but instead a proactive approach. In this stage, crisis managers at tempt to identify an impending crisis. This information can be found in various places, such as internal and external audits, government legislation, and industry publicati ons. Actions taken during the prodromal

PAGE 19

11 stage can easily be placed into the precrisis st age of the three-stage model as they address an organization’s crisis prevention. Fink (1986) argues that the actua l crisis event begins with a trigger, during what he refers to as the acute stage. This stage is characterized by the crisis event and resulting damage. The severity of the crisis and da mage are influenced by the success of the prodromal stage. Successful proactive identifi cation of a crisis can reduce the impact of the crisis in the acute stage. Failed recognition in the prodr omal stage creates a reactive situation instead of a proactive intervention. The third stage of Fink’s (1986) model is the chronic stage. This stage refers to the lasting effects of the crisis. Although individual crises may occur quickly, the lasting effects of the incident can extend the lifecycle of the crisis. Additionally, this stage may include a barrage of questions a bout the crisis which will keep the event visible to various publics. For example, an individual event, such as a natural disaster, may occur quickly, but the fallout of the incident make take week s or months to repair. Individuals and organization in Florida are still recove ring from the 2004 hurricane season. More recently, entire countries are still dealing with the massive damage and loss of life caused by a tsunami that struck areas of South east Asia December 26, 2004. Coombs (1999b) states that the acute and chronic stages act as sub-stages of the crisis stage of the threestage model. These stages include the app earance of a crisis and the steps taken to resolve the crisis event, char acteristics which are found in th e crisis stage of the threestage model.

PAGE 20

12 The final stage in Fink’s (1986) model is th e resolution stage. This stage identifies a clear end to the crisis. Alt hough organizations view this as the goal, it is not one to be rushed to. An organization’ s premature conclusion that the chronic stage has ended can leave them vulnerable to the resurgence of the crisis. Due diligence in the previous stages of the model must be practiced to in sure such a regression does not occur. This last stage of Fink’s model pa rallels the postcrisis stage of the three-stage model as it ensures the crisis has ended and distributes that message. The resolution stage does negl ect one portion of the thre e-stage model. Fink does not discuss crisis management as a cyclical pro cess. This is seen as one of the oversights of the model, that what was learned from a pr evious crisis is not expressly discussed in planning for future crises. Mitroff (1994) developed a model that divi des crisis management into five stages: signal detection, probing and prevention, damage containment, recovery, and learning. The segmentation of the crisis parallels Fink’s (1986) discussi on of the crisis lifecycle as well as the three-stage model. The first two stages – signal detection and probing and prevention – encompass the proactive steps an organization can take before a crisis event. Signal detection identifies the signs of possible crises w ithin an organization. Signal detection is much like Fink’s (1986) prodromal stage. Probing and prevention though is not addressed in Fink’s model. This stage features members of an organization seeking known crises and determining ways to prevent them. While Fink implies that crises can be prevented, Mitroff’s (1994) model actively tries to prev ent crises (Coombs, 1999b). As with Fink’s

PAGE 21

13 prodromal stage, signal detection a nd probing and prevention exemplify the characteristics of the precrisis stage. The last three stages of Mitroff’s (1994) model – damage containment, recovery, and learning – feature slight variations from Fink’s (1986) acute, chronic, and resolution stages. Like Fink, Mitroff’s stages discuss the trigger and containment of the crisis event, the arduous task of returning to the pre-crisis norm, and the reso lution of the crisis event. Damage containment, like Fink’s chronic stag e, focuses on the steps taken following the crisis event. A relationshi p can be made between damage containment and the crisis stage of the three-stage model as they both invol ve actions taken in response to the event. The differences between Fink and Mitroff’ s models are found in the recovery and learning stages. First, in the recovery stage Mitroff ( 1994) emphasizes the facilitation of the organizational recovery whereas in the chronic stage, Fink (1986) states that organizations recover at varying rates. Mitroff emphasizes oppor tunities to empower crisis managers in a particul ar crisis event while Fink fo cuses only on the timeframe of the recovery. As with damage containm ent, a relationship can be made between Mitroff’s recovery stage and the crisis stage of the three-stage model. As in the crisis stage, during the recovery st age the organization works toward the eventual end of the crisis. The second difference is that Mitroff’s (1994) model is cyclical. The learning stage allows an organization to incorporate what it has learned from the crisis into its organizational philosophy. Fink’s model simply states that resolution occurs when the crisis is no longer a concern w ith no mention of future applic ations. Mitroff’s discussion

PAGE 22

14 of crisis management as a cyclical process is significant. The learning stage is essential to the three-stage model. Like Mitroff’s five-stage model, the three-stage model is cyclical and recognizes the importance of appl ying what an organization learns during a crisis to future crisis events. The three-stage model and the five-stage model both acknowledge that a failure to l earn from a crisis can leave an organization susceptible to the crisis again. The three-stage model of crisis management is the most widely accepted model, and thus, will serve as the backdrop for this study. Whereas a more specific model may be relevant in follow-up studies this initial study to explore crisis management practices and identify the role of the contingency theory variables in crisis management will benefit from the use of th is widely accepted model. The Contingency Theory of Accommodation The contingency theory of accommodation in public relations is based on the premise that there are a number of variables that could affect the steps a public relations practitioner may take when managing a partic ular situation. Cancel, Cameron, Sallot, and Mitrook (1997) identified 86 such variables that have the potenti al to influence the decision making process, thus moving the st ance of an organizati on along a continuum ranging from pure advocacy to pure accommodation (see Figure 4). Pure | | Pure Advocacy Accommodation Figure 4: Advocacy-Accomodation Continuum; Adopted from Cancel et al. (1997) The advocacy portion of the continuum refers to those organizations or groups who strongly represent one side of an issue. Cancel et al. ( 1997) made the comparison to an

PAGE 23

15 attorney’s representation of a c lient. The inference is that an organization that aligns with the advocacy end of the continuum will main tain its stance with minimal wavering. An activist group is one such organization that could be found at the pure advocacy end of the continuum. Activist groups are characterized by st rong beliefs in a particular issue; they use a variety of methods to convi nce others that their beliefs are correct. Activists are also unlikely to waver unless th e status of their cause is improved. L. Grunig et al. (2002) defined an activist group as, A group of two or more individuals who or ganize in order to influence another public or publics through action that may include education, compromise, persuasion, pressure tactic s, or force. (p. 446) The accommodation half of the continuum is quite different (Cancel et al., 1997). It illustrates an organization’s ability to work with publics in an attempt to build trust and understanding. Depending on the role of the public relati ons practitioner, accommodation may prove difficult. In most organizations the dominant coalition determines the stance of the entire organizati on. If the organization is structured so that public relations practitioners do not have access to the dominan t coalition, practitioners may be unable to convince the organization to change its stance for a single public. L. Grunig et al. (2002) explained that public re lations departments must have access to the dominant coalition in order to contribut e to strategic management and planning. The choice to accommodate a public may involve a small change in policy or, in some cases, a change in organizational char acter that may be necessary in a given situation. Moreover, the resultant relations hip from accommodation could prove valuable in the event of a future dispute or crisis. It is at this meeting point on the continuum that effective communication occurs.

PAGE 24

16 Criticisms of Contingency Theory J. Grunig and Hunt (1984) identified the four models of public relations: press agentry, public information model, two-way asymmetrical, and two-way symmetrical. These were the initial models developed for th e field of public relations. As a result, the models were the subject of criticism from public relations scholars. The contingency theory is one of many theori es that expand these models. Of the four models, the two-way sy mmetrical model of communication has received the greatest a ttention. Indeed, J. Grunig and L. Grunig (1992) stated that this was both a descriptive and normative model. They believe that two-way communication between an organization and its publics epitomi zes the ideal practice of public relations. J. Grunig (2001) stated that a dialogue must be initiated in order to determine the stance of the individual public, “…neither side can really know the morality or reasonableness of the other side ’s interests without talking with its representatives” (p. 15). The author further posits that acco mmodative techniques should only be used during this conversation. Only upon a failed a ttempt to accommodate does Grunig explain, “…the symmetrical approach suggests that advocacy of [one side’s] interests or withdrawal from the dialogue is ethically reasonable” (p. 16). Two-way symmetrical comm unication, though, may not always be feasible as Cancel et al. (1999) stated in paraphrasing Cancel et al. ( 1997) “a monolithic approval of two-way symmetrical public relations is not sustainable given numerous publics—nor likely to lead to effectiv e public relations” (p. 178). Thomas and Kilmann (1974) discussed th e dimension of negotiation within the Dual Concern Model as a “party’s desire for own concern” and “party ’s desire to satisfy other’s concern.” The balance of the Dual Concern Model is what Thomas and Kilmann

PAGE 25

17 referred to as the compromise stage. In th is stage, the organization’s concerns – the points they advocate – influence negotiation, and ultimately the level of accommodation. Moreover, it is incorrect to assume that an organization can completely abandon their stance when attempting to accommodate a public, especially as they remain accountable to other internal a nd external publics. More on Advocacy The description of an activist group by L. Grunig et al. (2002) supports the very notion of the contingency theory of accomm odation. The authors established that an activist group can consist of tw o or more individuals. This factor will greatly influence how much a company accommodates the particular situation. Consider the request of a small, unknow n activist group to meet with a large company to discuss possible changes in th e company’s environmental policy. Company managers agreeing to meet with the group is unlikely, based solely on the activist group’s lack of stature. If managers of a large co mpany were to meet with every activist group, there would be little time to run the company. They would need an overwhelming reason to hold the meeting. Secondly, if in fact the two individu als were granted the meeting, their lack of prominence as an activist group would minimize the reverence they garner. Though the organization has met with them, the individuals would need to be extremely convincing to receive si gnificant accommodation. The reason is simple. In order to accomm odate requests, the company may need to make concessions that could be detrimental; profit margins could drop, jobs could be lost, or business relationships coul d dissolve. A company has to weigh this effect on other stakeholders and whether it will risk these nega tives to fulfill the need of a small public.

PAGE 26

18 Likewise, failure to accommodate could pr ove detrimental. The backlash from maintaining an organizational pos ition could be hurtful. If the concerns of a group are reasonable and they are ignored, th e repercussions could be damaging. If it was determined that accommodating the activist group proved too costly, the organization would remain at the advocacy end of the continuum, though the position could shift. If before the meeting the or ganization was positioned at “pure advocacy,” it may shift slightly toward the accommodation end. This could be the result of agreeing to future meetings or other small concessions. If no such concessions were made, the organizational position woul d remain at pure advocacy. More on Accommodation The activist example serves to also expl ore the concept of ac commodation. In the previous scenario the company was unlikel y to practice accommodation with the twoperson activist group based on the cost-benefit of doing so. Conversely, when the activist is a large and very reputable group, a company will be more likely to consider accommodation. For instance, as stated on its Web site, th e Sierra Club is America's oldest, largest and most influential grassroots environmen tal organization. The organization boasts 700,000 members. By contrasting the Sierra Cl ub with the two-person activist group that wants to protect the environment, it is clea r that the response would be very different. If a company were approached to discuss an environmental policy, the clout of the Sierra Club would increase the likelihood of the meeting. The company would have to weigh the same factors it did for the two-pe rson activist example, though in this case the public is much different. The influence of an organization such as the Sierra Club is large. The company must be aware that the numerous members could have a sufficient

PAGE 27

19 impact on its product or service. Demonstrat ions, boycotts, and media coverage could all affect the success of the compa ny. The severity of these even ts could prove very costly. In the previous example, the organizati on was unlikely to accommodate due to the cost-benefit ratio. In this case though, th e cost of not accommodating could be much larger. The company would weigh these new variables to make a decision to accommodate some of the Sierra Club’s requests. In doing so, the company reduces the risk of retribution, improves its relationship with the Sierra Club, and aligns itself with an organization with a positive reputation. Contingency Variables The contingency theory of accommodati on identified 86 variables that could influence an organization’s level of accommodation. Cancel, Mitrook, and Cameron (1999) tested the theory and variables to dete rmine those that were most relevant. In a qualitative study, researchers in terviewed 18 public relations practitioners to determine “whether the contingency theory makes se nse to them” (Cancel et al., 1999, p. 172). Interview participants worked for nationwid e corporations and each had between 20 and 40 years of experience. The study’s findings divided the variables into two major groups, predisposing and situational variables. Each group consisted of variables th at were found to be “highly supported” by the practitioners (Cancel et al., 1999). Examination of these variables will provide knowledge pertinent to ground crisis management in the contingency theory. Predisposing variables Cancel et al. (1999) define predisposing variables as “those variables which have their greatest influence on an organiza tion by helping to shape the organization’s predisposition towards relations with external publics” (p. 177). Th ese variables are part

PAGE 28

20 of the foundation of a company. They are st eadfast characteristics that define what the corporation stands for; they gi ve the company a personality. In their study, Cancel et al. (1999) found that, “predisposing variables in fluence location along th e continuum before the corporation enters into a particular s ituation involving an external public” (p. 190). From the interviews, Cancel et al. (1999) identified five highly-supportedpredisposing variables: (a) corporation busin ess exposure, (b) public relations access to dominant coalition, (c) dominant coaliti on’s decision power and enlightenment, (d) corporate size, and (e) indivi dual characteristics of involved persons. These variables will be examined further in this study. Though future research exploring contingency theory may examine all of the predisposing variables, this study will focus on only those variables identified by Cancel et al. (1999) as highly-supported. By examining these variables the researcher attempts to support previous findings. The goal is a structured analysis of crisis management utilizing the framework of the contingency theory. Corporation business exposure was cited as influential to relations with external publics in 13 of the 18 interviews (Cancel et al., 1999). Resear chers discovered that participants whose company produces consumer or necessary goods needed to be very responsive to their publics. Moreover, comp anies that produce consumer products were also found to have greater exposure to public s than companies in the commercial or industrial markets. Respondent s believe that as a result of the higher exposure, these companies’ expectation to accomm odate to publics was higher. Public relations access to dominant coalition “enthusiastically” received support in all interviews as influential of how organi zations deal with their publics (Cancel et al., 1999). Respondents explained that the dominant coalition had the final say in what was

PAGE 29

21 allocated to publics. This supports a common theme explained by Grunig and Grunig (1992) and a belief of practitioners and scholar s alike: public relations must have a place in the dominant coalition. Participants explai ned that they were used as an advisor to management about opportunities to accommodate publics, but felt that in order to be successful they required access. L. Grunig et al. (2002) also suppor ts the findings of Cancel et al. (1999) that the structure of the public relations department influences access to the dominant coalition. The authors pos it that when public relations is autonomous from other departments, such as marke ting, their access to the dominant coalition increases allowing for a greater exam ination of public relations issues. Dominant coalition’s decisi on power and enlightenment takes the variable of access to the dominant coalition one step further. This vari able was cited most often by interview subjects as influential (Cancel et al 1999). Subjects agr eed that the dominant coalition must understand the importance and value of public relations to the organization. Many practitioners explained th at there were times they had to educate management about their role and value. Corporate size was cited by many subjects as causi ng high visibility (Cancel et al, 1999). As found in the variable of business exposure, high visibil ity can lead to the expectation that a company will be accomm odative. Also, respondents explained they felt the size of their companies made them targets of activist groups. As discussed earlier, special intere st and activist groups can influen ce the accommodative practices of a company. Subjects also stated that a la rge corporation could have so many diverse publics that they cannot respond to all of them.

PAGE 30

22 Individual characteristic s of involved persons was mentioned by 15 of the 18 respondents as influential to relations with exte rnal publics (Cancel et al., 1999). The individuals referred to were members of the dominant coalition, not public relations managers. This variable also illustrates how important it is for public relations to be included in the dominant coalition. Speci fic characteristics mentioned were openmindedness, the ability to screen out pers onal bias, and past tr aining and education. Situational variables Situational variables, as defined by Cancel et al. (1999), “are th e specific and often changing dynamics at work during particular situations involving an organization and an external public” (p. 177). These variables present a great challe nge to crisis managers as they appear only as the crisis presents itsel f. As the authors e xplained, these variables impact an organization’s position “as the situation plays out” (p. 177). Thus, these variables appear during or immediately follo wing a crisis and exert their influence on the predisposing variables. The situational variables can largely impact the position of a company along the conti ngency continuum. Whereas the company’s predisposing variables may place it on the advocacy end of the continuum, a powerful situational variable could shif t that position. For instance, an automobile company may claim its new sport utility vehi cle is safe. If an independe nt safety test shows the SUV has the propensity to flip when drivers are fo rced to swerve at high speeds, the threat to the company could result in a shift from advocacy of their safety stance to accommodation of new safety recommendations. Cancel et al. (1999) identif ied five situational variab les as receiving high support from respondents: (a) urgency of situation, (b) characteristics of exte rnal public’s claims

PAGE 31

23 or requests, (c) characteristic s of external public, (d) potential or obvious threats, and (e) potential cost or benefit for a corpor ation from choosing various stances. Urgency of situation was cited in all interviews and strongly associated with influencing the accommodation of an exte rnal public (Cancel et al, 1999). Many respondents stated that this variable must be establishe d soon after determining the specific details of an incident. The degree of urgency in a given situation is based on a number of factors; internal, external, and pr edisposing factors can influence the level of urgency. These variable factor s support the notion that not all situations are similar and practitioners must be prepared to react differently in each situation. Characteristics of external public’s claims or requests were among the most frequently mentioned variables (Cancel et al, 1999). Participants stated that the perception of the external publics’ claim or request influence how they proceeded. Truth or falsity, as well as the existence of re sponsibility for the negative claim held an important role in the decision to accommodate. In addition, the reasonability of a claim also influenced accommodation. If a claim or request is reasonable, corporations were found to be more accommodative. Characteristics of external public refers to the corporation’s interaction with and perceptions of an external public. The characteristic most frequently noted in interviews was how well the public could aff ect the company’s ability to do business and make money, positively or negatively (Cancel et al., 1999). Respondents also cited the public’s size, organization, and ab ility to garner media covera ge as characteristics they weigh. Indeed, the capabilities of extern al publics can influence the decision to accommodate an external public.

PAGE 32

24 Potential or obvious threats were cited in all interviews as a factor the practitioners faced at one point (Cancel et al, 1999). A larger threat was associated with a quicker response and higher likelihood to accommodate. The threat level can be associated to a number of fact ors including the power of the external public. A powerful public presents a greater threat to a company as it has greater resour ces and capabilities at its disposal. Possible media coverage was the most frequently cited threat, though the threat of economic loss was underlying to each threat. One practit ioner explained that impacting a company’s ability to do business and make a profit is a quintessential piece of a threat. Potential cost or benefit is fundamental in business. In regard to the contingency theory, cost-benefit is how a company’s posi tion along the continuum affects its costs and benefits. For instance, if accommodating an ex ternal public will be a costly pursuit, the company must determine if the benefits will counteract some of the expense. Benefits of accommodation include improved public perception, alignment with a positive organization, and the development of an adva ntageous relationship. Not surprisingly, the subjects explained their company strives for a situation with the lowest cost and highest benefit (Cancel et al., 1999). The researcher believes that the high ly-supported-predisposing and situational variables will likely have a role in crisis management. In an attempt to illustrate the relationship between these variables and the th ree-stage model of cris is management, four research questions have been developed to guide the study: RQ1: How will the predisposing variables emerge from the three-stage model of crisis management?

PAGE 33

25 RQ2: How will the situational variables emerge from the three-stage model of crisis management? RQ3: At what point in the model will the predisposing variables emerge from the three-stage model of crisis management? RQ4: At what point in the model will th e situational variables emerge from the three-stage model of crisis management?

PAGE 34

26 CHAPTER 3 METHODOLOGY Review of Research Questions Four research questions were developed to test the influence of contingency variables on a crisis situation. The resear ch questions will provide a groundwork that demonstrates the use of contingency theory in crisis management practice. The study first examined whether the contingency theory variables identified by Cancel et al. (1999) as highly-supported emerged during crisis ma nagement. The study then attempted to determine at what point those vari ables affected crisis management. Questions one and two examine the role of the highly-supported-predisposing and situational variables in a crisis. Questions three and four study at what point the highlysupported-predisposing and situational va riables influence crisis management. RQ1: Will the predisposing variables emerge from the three-stage model of crisis management? RQ2: Will the situational variables emerge from the three-stage model of crisis management? RQ3: At what point in the model might th e predisposing variables emerge from the three-stage model of crisis management? RQ4: At what point in the model might th e situational variables emerge from the three-stage model of crisis management? Data Collection Data collection for this study used a modifi ed Delphi technique. This qualitative interactive research method allowed the research er to gather a great deal of information from individuals with excep tional knowledge about a s ubject (Clare, 1994). The

PAGE 35

27 technique allowed the researcher to prompt pa rticipants with a question. Upon receipt of all answers, the res earcher posed a supporting question based on the initial findings for the panel to respond. The process served as a way to structure group communication (Turoff & Hiltz, 1996). Beyond the receipt of indi vidual feedback, the Delphi technique allowed respondents to react to their colleagues while guaranteeing anonymity. Responses were not immediately provided to all of the particip ants. During analysis of the responses to determine a follow-up question, the researcher removed any information that may have identified the respondent in the answer. The Delphi system allowed for geographica lly distributed individuals to interact while discussing their crisis management expe riences. Moreover, this technique helped remove negative dynamics that may inhibi t communication in a group setting, such as hostility toward other participants and do mination of the discus sion by a participant (Clare, 1994; Wakefield, 2000). A limitation of the Delphi technique is that participants needed to be reminded and motivated to provide feedback. Whereas in inte rviews the researcher is present to ensure respondent interaction, the De lphi method only allowed in teraction along mediated channels. In an attempt to maintain particip ation, the resear cher sent periodic reminder emails and made telephone calls to participants who did not respond to the question by the requested deadline. The questions examined the role that pred isposing and situational variables have in the crisis management process and attempted to determine other variables that influenced the respondents’ crisis response. Respondent s were asked to discuss those variables

PAGE 36

28 which most influenced their crisis manageme nt response. The researcher prompted the respondents with questions that addressed the predisposing an d situational va riables that the participants did not discuss. Variab les mentioned by the subjects that were not highly-supported by Cancel et al (1999) were also documented. Sample The study used a purposive sample of communication experts from municipal utility companies in the state of Florida. The municipal utilities companies recruited for this study were electric companies that are community-owned and locally managed. The effects of the very active, 2004 hurricane seas on on the state of Flor ida supports the belief that these individuals have e xperience in natural disaster, crisis situations and related communication functions. Candidates for participation in this st udy included a cross-section of municipal utility companies throughout the state of Florida. The study ai med to target local level, senior members of the Florida municipal u tility community. Contact information for senior members at the various organizations was provided by Florida Municipal Energy Association (FMEA) Executiv e Director, Barry Moline. Following a phone call placed to Mr. Molin e discussing the study, the researcher was provided with six senior individuals at m unicipal utilities across the state of Florida that were impacted by the 2004 hurricane season. The six organizational contact s were sent an initial recruitment e-mail in early January, 2005 (APPENDIX A). A followup telephone call was made to those candidates who did not respond within one wee k. Initially, five of the six individuals agreed to participate in the study – the sixth candidate explained his organization was still too busy with recovery efforts from the hurricane season to participate in the study.

PAGE 37

29 Data Analysis The findings were analyzed to determine the role of predispos ing and situational variables in crisis management. Throughout th e research process data was analyzed in order to generate follow-up questions for partic ipants. The analysis followed the constant comparative method. The individual responses were compared to reveal the common themes presented by participan ts. Information that did no t support the cate gorization of the highly-supported-predis posing and situational variables was documented.

PAGE 38

30 CHAPTER 4 RESULTS AND DISCUSSION Data collection occurred over a six-w eek period. From the original five participants, three remained at the end of the study. Two of th e three participants responded to all phases of the study. The th ird individual was unable to participate in phase 1.1 but responded to each subsequent phase. Of the two individuals who withdrew from participati on, both expressed a lack of time to dedicate to the study while in the firs t phase. One particip ant withdrew before responding to the first primary question while the other withdrew after responding to the first primary question. In each phase of the study, the researcher received responses from three participants. Participants were info rmed to use the 2004 hurricane season or a particular storm from that hurricane season that affected their organi zation as the basis of their response. The background of the participants wa s quite different. Two of the three participants are employed by small city govern ments while the third worked for a utility company servicing a large city. The indivi dual who withdrew after answering the first primary question worked for a utility company in a small city. Responses from each participant are reflected in the analysis of each phase of the study. Phase 1 Phase 1 of the study was developed to e xpose a possible relationship between the highly-supported-predisposing variables id entified by Cancel et al. (1999) and the precrisis stage of the thr ee-stage crisis management model. The highly-supported-

PAGE 39

31 predisposing variables are co rporate business exposure, public relations access to the dominant coalition, the dominant coalit ion’s decision power and enlightenment, corporate size, and the indivi dual characteristics of involved persons. Phase 1 provided data that allows the researcher to answer th e first and third resear ch questions of this study: RQ1: How will the predisposing variables emerge from the three-stage model of crisis management? RQ3: At what point in the model will the predisposing variables emerge from the three-stage model of crisis management? The data suggest that the predisposing vari ables do influence crisis management. Each respondent addressed the role of the vari ables, to some degree, in their organization. Discussion centered on how the predisposi ng variables influenced the preparatory actions of participant organizations and th e established role of participants in organizational decision-making. Moreover, th ere was a relationship between participant responses regarding the predisposing variables and the precrisis stag e of the three-stage model. The precrisis stage addresses the proa ctive steps an organization takes to avoid a crisis. Likewise, in phase 1 the respondents discussed the existen ce of the predisposing variables prior to the crisis event. Phase 1.1 The first question was meant to determine what precrisis infrastructure existed at each respondent’s organization. The phase 1.1 question read: “What preparatory steps has your organization taken in an attempt to avoid a crisis event?” The question was purposefully vague as to not influence the responses of participants, but rather to gauge the relevance of varying facets of their crisis response. The purpose was for those items they viewed as most important to surface in the

PAGE 40

32 responses. Participants seeking further clarif ication were provided an addendum to phase 1.1 asking the respondents to discuss the charac teristics of their organization that shape its crisis response. Phase 1.1 responses dealt with the proac tive steps taken for a crisis event. Preparation of various crisis plans and manuals had been completed by all respondents. Participants 1 and 3 discusse d the impact of the Septembe r 11, 2001 attacks on their crisis preparation. Each participant stated that si nce the attacks there has been a new emphasis on the security of their facil ities. Participants 1 and 2 di scussed the use of government sanctioned training or meetings to support cr isis preparation. A ll three participants discussed the importance of public safety. The highly-supported-predisposing variables played only a minor role in phase 1.1. However, one participant did address two of them in their res ponse. Participant 3 discussed the variables corporate size and access to the dominant coalition. In this industry corporate size is determined by the number of clients se rviced – a greater demand for power necessitates a larger organiza tion. This participant stated that only after power outages dropped below a “reas onable number” did twice daily briefings cease. The reasonable number of outages wa s 40,000. This figure is a reflection of the company’s size as it can service up to this number of outages without the implementation of increased protocols. This supports the findings of Cancel et al (1999) that the size of an organization influences its ability to acco mmodate its publics. This larg e utility company was able to accommodate a large number of outages before enacting new protocols. A smaller organization would not be able to handle su ch a large number of out ages with the same

PAGE 41

33 ease. Whereas the large utility may not deviate from normal protocol, the smaller organization may have to alter its communi cation response and community information efforts. Participant 2 is a member of a seven person team that oversees power distribution activities (Appendix G). For a wo rkforce of this size, implementation of special operations would have likely occurr ed well before 40,000 outages. Therefore, there is support that corporate size influences th e precrisis stage of the three-stage model. Additionally, access to the dominant coaliti on was addressed briefly in phase 1.1. Participant 3 explained that following the massive outages, twice daily briefings were held to address the situation. Many depa rtments, including communications, were represented in the meetings [The meetings] involved the CEO, the Electric, Water and Sewer Operations Department, Internal and External Comm unications, Public Outreach, Facilities Maintenance, Client Relationships, Securi ty, Finance, Risk Management, Safety, Engineering, Technology Services, and Governmental Relations. All vice presidents on the executive management team were at the meetings, along with many of the field operations supervisors to ensure direct information flow from the field. (Appendix F) Public relations practitioners in this organization – internal and external communication, public outreach, client relati onships, and government relations – all had access to the dominant coalition. Though it is unclear whether this access is permitted when the organization is not facing a crisis the inclusion of p ublic relations in discussions about crises iden tifies the active role the department has within the organization. The amount of access public rela tions has to the dominant coalition in this organization supports the study of Cancel et al. (1999) as well as the variable’s influence on the crisis stage of the threestage model. The access of pub lic relations and the size of the organization are discussed furt her in the analysis of phase 1.2.

PAGE 42

34 Phase 1.2 Phase 1.2 delved into the part icipant’s role in the orga nization. The researcher wanted to learn how their ideas were inco rporated into the or ganization’s decisionmaking process. Additionally, the question at tempted to determine the other sources of information the dominant coalition recognize d. The phase 1.2 question read: “How has your input been incorporated into your organization’s current crisis management protocols? Who else’s input, internal or external, was taken into account?” This question attempted to elicit responses addressing multiple variables identified in the contingency theory. As mentioned, organizational size was found to be an influential variable in phase 1.2. Unlike the phase 1.1 response, partic ipant 2 is from “a very small organization” (Appendix G). The participant explained one objective the organization recognized was the need to educate the public about crisis situations. The small size of the organization necessitated support from local agencies – police, fire, and emergency management – to successfully educate the public. Without ex ternal support the smaller organization may not have been able to addre ss the needs of their public. As in phase 1.1, the predisposing variable of corporate size identified by Cancel et al. (1999) is supported. The organization’s small size influenced its ab ility to provide publics with important information in the precrisis stage. Moreover, the size of the organization al so influenced access to the dominant coalition. Participant 2 explai ned that the department that handled power-related issues consisted of seven individuals, “1 superint endent, 3 linemen, 2 apprentices, plus me” (Appendix G). As a result of the small depart ment, input from all i ndividuals – including

PAGE 43

35 the respondent who frequently serves as ci ty spokesperson for public services – was included in the decision-making process. Lastly, this organization also supported th e notion that characte ristics of involved individuals influence the pr ecrisis stage. Respondent 2 explained that although everyone’s suggestions were encouraged, the mo re experienced individuals were able to provide greater insight. It is likely that an individual in a small department with more experience will influence the choices made in crisis preparation. This experience relates to discussion in Cancel et al. (1999) of past tr aining or education as a characteristic frequently mentioned by respondents, as infl uential to the organi zation’s accommodation of publics. Phase 1 Conclusion Phase 1 identified support for the predispos ing variables of co rporate size, public relations access to th e dominant coalition, and indivi dual characteristics of involved persons. There was only marginal support fo r the variables of dominant coalition enlightenment and corporate business exposure. In phases 1.1 and 1.2 respondents addre ssed the importance of public safety, government training courses, and resource av ailability as well as highly-supportedpredisposing variables cited by Cancel et al. ( 1999). In order for th ese opportunities to be available for public relations pr actitioners, it is likely that both funding and support of the dominant coalition were achieved. This likelihood provides marginal support for dominant coalition enlightenment of the role of public relations. Additionally, corporate business exposure was not directly mentioned as influential to cris is preparation, though public access to the organization and visibili ty in the media were addressed once the crisis occurred and will be discussed in the analysis of phase 2.

PAGE 44

36 Phase 1 also included some support for a contingency variables not highlysupported in Cancel et al. (1999) The role of legal counsel in crisis preparation was addressed by one respondent. In phase 1.1 the respondent explai ned that full-time general counsel was hired in 2002 to aid in cr isis preparedness. Th is individual’s role was to assist in the proactive identification of organizational vulnerab ilities. This was the only mention of legal couns el throughout the study. Phase 2 The second phase of the study transitioned to examine the relationship between the highly-supported-situational va riables identified by Cancel et al. (1999) and the crisis stage of the three-stage model. The highly-supported-situa tional variables are: urgency of the situation, characteristics of external public’s claims or requests, characteristics of external publics, pote ntial or obvious threats, and the potential cost or benefit of a particular action. The data collected in phase 2 provided the researcher with information to answer the second and fourth re search questions of this study: RQ2: How will the situational variables emerge from the three-stage model of crisis management? RQ4: At what point in the model will th e situational variables emerge from the three-stage model of crisis management? In this phase, the data illustrated the role of the situational variables in crisis management. Participant responses focused on the existence of the situational variables relative to the crisis event – the discussion addressed the organizat ional response to the crisis as well as the factors that influenced the response. In ma ny cases, the situational variables provided insight as to why the organi zations responded in a particular manner. Participants often addressed their organization’s response to the crisis when commenting on the role of the situational variables. The discussion of the variables’ influence

PAGE 45

37 concentrated on the crisis stage of three-stage model. As a result, the researcher can infer that the situational variables exert influe nce in the crisis st age of the model. Phase 2.1 The question in phase 2.1 identified the variables that influenced organizational response to the 2004 hurricane season – referred to as the crisis event. The phase 2.1 question read: “What characteristics of the crisis event influen ce your organization’s crisis response?” The characteristics of exte rnal publics’ claims and requests was an influential variable in phase 2.1. Respondents 2 and 3 expl ained that their organizations received a lot of pressure from publics to restore power Respondent 3 explained that with more than half of their customer s without power, the company re ceived as many calls in one day as they typically do in a month. Considering the number of phone calls the organization had to modify how they comm unicated with publics and the media. The organization’s “blast fax” to the media a nd members of the city council helped the organization distribute information to the public (Appendix H). Thus, the number of requests for service influenced this organization’s response in the crisis stage of the threestage model. The variable of corporate size was revisi ted in conjunction with the claims and requests of publics. Participant 3 explai ned that once outages were reduced to a manageable amount the use of blast faxes ceas ed and the organization returned to its normal communicative practices. As addresse d in phase 1, the number of manageable outages was 40,000. A smaller organization ma y not consider that number of outages low enough to resume normal protocols. Respondent 4 works for a much smaller organization and explained that due to its size, the organization was forced to work

PAGE 46

38 closely with the county emerge ncy management team to rest ore power. The crisis stage of the three-stage model concludes when the crisis ends. Though a reduction to 40,000 outages did not end the crisis, an organizati on’s size and ability to handle a large volume of outages will expedite the crisis stage. Public outcry supports the existence of anot her situational variab le’s role in the crisis stage of the three-stage model – potenti al cost or benefit. Again, respondents 2 and 3 explained that as a result of the massive pow er outages their organization was forced to determine the best way to restore power. E ach respondent explained that they organized repairs from the power source and major pow er lines to the nei ghborhoods in order to most efficiently restore power to the most people. This sequence did not follow standard procedure – in fact it was the exact opposite in one case. The organization determined the cost of following standard protocol wa s too high – it would have taken too long to restore power – and that a different approach produced a greater bene fit in the situation. Additionally, respondent 2 briefly discusse d that there was great importance placed on restoration of power to critic al facilities. The need to re store power to these facilities quickly relates with the vari ables of urgency of the situ ation and potential and obvious threats. Medical facilities need power to provide treatment for patients. Without power, patients may not be able to receive care – in some cases urgent, lifesaving care. Moreover, it is likely that these facilities received an increase in patients as a result of the storm, thus power restoration was essential. These variables were further addressed by respondents in phase 2.2. Phase 2.2 Phase 2.2 further examined a point raised in phase 2.1. As a result of the power outages many publics contacted or attempted to contact their utility pr ovider. This phase

PAGE 47

39 asked if any particular public contacted the organization more than others. The phase 2.2 question read, “Were certain publics more vocal regarding their need for power restoration? Did their requests in fluence your organi zation’s response?” This question directly addre ssed the situational variable of characteristic of an external public. The responses to this question varied, as each respondent had a different experience with the distinguishin g characteristics. Participant 2 explained the most vocal group in the community belonged to a more affluent, gated community who felt they were not receiving the service they deserve d. The respondent explained that only rarely did any calls influence power restoration pro cedure. Instead, the organization prioritized restoration from important community serv ices down to individual members of the community. There were a few instances wher e external publics affected restoration. These situations were limited to unknown crit ical needs facilities and “quick easy fixes that would restore power to many customers” (Appendix I). Thus, certain characteristics of external publics and the cost-benefit of “ quick fixes” influenced this organization’s response in the crisis stage. Respondent 3 stated that parents became most vocal in the days following a power outage. Parents with jobs had to stay home to watch their children or were forced to pay for daycare due to school closures. Additi onally, some parents c ontacted city council members and the mayor in an attempt to expe dite school power restoration. The pressure from this public placed more focus on school restoration. Support is again found for the Cancel et al. (1999) variable, characteristics of external publ ics. Outcry for a focus of restoration efforts at school proved succe ssful in shifting focus to that task.

PAGE 48

40 Participant 4 explained that calls requesting power came from all members of the community. The first groups to have power re stored were the police, fire, and emergency services followed by downtown businesses and medical facilities. Again, characteristics of the publics influence the cr isis response. Those servi ces and facilities that were deemed most important to receive power restorat ion were addressed first. In reference to the general public, the respondent claimed th at no group influenced their response. Instead, power restoration crews were dispatched in order of request. Phase 2 Conclusion Phase 2 identified much support for the Cancel et al. (1999) highly-supportedsituational variables – each va riable was addressed by respon dents. The variable most supported was characteristics of external publics. Each respondent e xplained that certain publics – medical facilities, schools, or emer gency services – received a greater amount of organizational resources. The variable which received the least support was potential or obvious threats. The only rela tionship with this variable can be with the threat of not immediately restoring power to medical facilities or emergency services. Additionally, the applications of lessons learned during this crisis were briefly discussed in phase 2. Partic ipant 2 explained that his or ganization “insert[s] lessons learned as needed to our preparations” (A ppendix H). Participant 3 discussed how his organization “develop[ed] a contact/location list for all schools we serve” (Appendix I). These two tasks support the premise of the thre e-stage model that crisis management is cyclical. Lessons learned during one crisis should be included in future crisis management planning and preparation so as to av oid or reduce the risk of another crisis. The application of lessons learned during a cris is will be discussed further in the analysis of phase 3.

PAGE 49

41 Phase 3 The third phase of this study did not serve to relate the continge ncy theory with the three-stage model. Instead, this phase ex amined whether the participants’ crisis management practices are cycli cal. The three-stage model is a cyclical model of crisis management – it explains that what an or ganization learns duri ng a crisis should be applied to future crisis events. The re searcher wanted to probe whether or not participants practiced cyclical crisis ma nagement and how it affected their attitudes regarding possible future crises Unexpectedly, the data presented a reemergence of the predisposing variables. As a result, phase th ree of the study also, li ke phase one, served to answer the first and third research questions. Phase 3.1 The phase 3.1 question examined what in fluence the 2004 hurricane season had on the future of the organization. The ques tion read: “How did th e 2004 hurricane season influence your organization’s pr eparation for future crises?” Respondents discussed a variety of to pics. Respondent 2 emphasized the importance of the experience his organization ga ined. The participant expressed that this experience would reduce the need for future “reinvention” of techniques and policies (Appendix J). Moreover, the individual e xplained the role inte raction with other organizations played in thei r learning. This point wa s echoed by participant 4 who claimed their organization “had a new understand ing” of what they must prepare for in the future (Appendix J). Respondent 3 listed th e major changes this crisis prompted in the organization. The list included changes in protocol, communica tion techniques, staff assignments, and the receipt of community feedback.

PAGE 50

42 The experience gained by the respondents fr om the crisis event can be valuable in future crises. As a result, the highly-suppor ted-predisposing variable characteristics of involved persons, reemerged. The first-hand experience familiarized them with crisis management practices. The knowledge gained fr om this crisis even t has helped prepare these individuals for future crises. Whereas before they may have never dealt with a crisis, the 2004 hurricane seas on likely made them more aw are of such a situation and provided them with experience to react to future events. Phase 3.2 Phase 3.2 asked if the crisis and what the organization learned from the crisis has prepared them for similar events in the future as well as what concerns still exist. The phase 3.2 questions read: “C onsidering what your organi zation learned from the 2004 hurricane season, do you feel prepared for th e 2005 hurricane season? What concerns does your organization have?” The responses to this question were mi xed. Participant 3, from a large utility company, explained that the e xperience has prepared the orga nization for future events. The respondent explained that the first-hand experience was much more valuable than any drill or exercise in teach ing the organization how to cope with such situations. In contrast, the participant did express concern if the commun ity and organization were to take a direct hit from a storm. The participant stated that the additional damage and outages caused by a direct hit are of great concern to the company and the community. Respondent 2 had quite the opposite re sponse regarding his organization’s preparation. This individual explained that his organization does not feel completely prepared for a hurricane season. The participan t cited the impact of the cost of the 2004 season on the small utility as too great to overcome in one year. As a result, this

PAGE 51

43 organization will triage their needs as the 2005 hurricane season begins. This response marks the reemergence of the highly-supportedpredisposing variable corporate size and illustrates how the variable will impact this organization’s ability to manage a future crisis event. Participant 4 explained that although an organization can prepare, the 2004 hurricane season was something for which you c ould not prepare. This individual stated that, “you may think you are prepared, but when you have three successive hurricanes wreak havoc on your community in six weeks, you can’t get that prepared” (Appendix K). As a result of the 2004 hurricanes, this participant feels that hurricane season will always concern her community. Phase 3 Conclusions From the phase three responses, the resear cher can conclude that each respondent has learned from the experience of the 2004 hurricane season and will apply those lessons to future crisis management practices. The respondents, therefore, did use a cyclical model of crisis management – a vital aspect of the three-stage model. The knowledge gained and resulting changes from the 2004 hurricane season will only increase the ability of these organizations to prepare for cr isis events as well as reduce the impact of the crises. Additionally, predispos ing variables emerged in pha se three responses. These variables signal the creation of the new precrisis stage that occurs when new information, gained during a crisis, is incorporated into existing precrisis protocol. The outcome of this synthesis is a new precrisis norm, exhibiting highly-s upported-predisposing variables.

PAGE 52

44 CHAPTER 5 LIMITATIONS AND FUTURE STUDY This qualitative study does not allow the researcher to make generalizations into the relationship between the contingenc y theory of accommodation and crisis management practices. Moreover, the small num ber of participants may slightly detract from the relevance of the study. This study provides a springboard for further study. Future studies that attempt to relate the contingency theory and crisis management practices should soon transition from a qualitative approach to a quantitative methodology. As a follow-up to this study scholars should us e depth interviews to collect data in an attempt to relate the highl y-supported variable s identified by Cancel et al. (1999) to crisis management. The Delphi approach us ed in this study proved problematic at times in recruitment, sustained participation, a nd depth of information. Though this study did provide purposeful data, an interview format would allow the rese archer the opportunity to probe each respondent further and receiv e greater clarificati on on the role of the highly-supported variables. As a result of the Delphi methodology’s limited opportunity for follow-up questions, there was likely valu able information that the respondents did not convey. The depth intervie w will provide the wealth of knowledge that can precede a quantitative study. Upon analysis of detailed qualitative data that supports a rela tionship between the contingency theory and crisis management, a quantitative study may generate statistically

PAGE 53

45 relevant data. From this data, contin gency theory scholars can provide crisis management practitioners with a theo retical grounding of their practice.

PAGE 54

46 CHAPTER 6 SUMMARY AND CONCLUSIONS Crisis management literature is rich with models that various organizations can follow to help minimize the impact of a crisis Each model has its individual merits, but the three-stage model of crisis management is the most widely accepted model (Coombs, 1999b; Seeger et al., 2003). Th e acceptance of the model is the result of its overarching applicability to a variety of organizational crisis situat ions. The three-stage model divides a crisis into macros tages – precrisis, crisis, and postcrisis – which allows for analysis of the event in a logical flow. Data collection in this study was modeled after the three-stag e model of crisis management. Participants responded to a seri es of questions which that followed the lifecycle of the crisis. The respondents, who were public relations and corporate communications practitioners at municipal utility companies in Florida, discussed their organization’s crisis management practices during the 2004 hurricane season. Participant responses were analyzed with those variable s that influenced the crisis management process categorized according to the thre e-stage model of crisis management. The first phase of the study addressed the pr ecrisis stage of the three-stage model. This stage is characterized by the proactive steps taken as crisis preparation of an organization – the proactive steps (Coombs, 1999b). Respondents addressed a number of variables that influenced thei r crisis preparation, including fa cility security standards, protocol changes resulting from the 9/11 attacks, and information gained from

PAGE 55

47 organizational crisis management audits. Of th e variables influencing this stage, the size of the organization revealed itself to be th e most apparent influence cited by respondents. The size of the participants’ organization emerged as highly influential to their crisis preparation. The smaller organizations did not have the benef it of resources that available to larger utilities had. Moreover, the protocols that sma ller organizations did have in place were much less effective as a resu lt of resulting from a smaller staff, a less an inadequately equipped response network, a nd a much smaller operating budget. It was found that although a large organization can sust ain a large number of outages (40,000), a smaller organization does not have the resources that allow it to effectively address such an outage. The variable of corporate size was addr essed by Cancel et al. (1997) in early contingency theory work. The researchers iden tified this variable as influential in the accommodation of publics. Moreover, corporate size was identified by Cancel et al. (1999) as a highly-supported-pr edisposing variable in furthe r development and testing of the contingency theory. The discussion of this variable was the first in a series of highlysupported contingency theory variables th at emerged during data collection. The emergence of this variable This served as the first indicator of a potential relationship between contingency theory va riables and the three-stage mo del of crisis management that appeared in the data. The contingency theory variables start to emerge emerged as a way to understand the three three-stage model of crisis management. This proposition is supported by the presence of additional highly-supported contingency variables in participant responses regarding th eir crisis management practices.

PAGE 56

48 Additional highly-supported-predisposi ng variables that emerged during the precrisis portion of data co llection include public relations’ access to the dominant coalition and the characteris tics of involved persons. Participants explained that members of the public relations function of the organization were in place prior to the crisis event. Also, the predisposing va riable of employment tenure within the organization emerged characteristics of i nvolved persons emerged as participants explained that the input of the more tenur ed members of the organization was very valuable as a result of their greater insight into the crisis. The appearance of these variables in the da ta further supports the notion that there is some a relationship between the hi ghly-supported-predisposing variables of contingency theory and the prec risis stage of the three-stage model of crisis management. Moreover, the predisposing variables of domi nant coalition power and enlightenment and organizational business exposure were also addressed by participants. Although these variables emerged to a lesser extent, their ex istence in the data provides supplementary support to a relationship between crisis management and the variables identified in the contingency theory. The second phase of the study shifted questi on the focus to the crisis stage of the three-stage model of crisis management. The crisis phase is characterized by the crisis event and its lasting effects, and continue s until the crisis has ended. Participants discussed a number of variables that infl uenced how their organization responded to a crisis. As was found in the precrisis que stion responses, highly -supported-predisposing variables from the contingency theory also em erged from in responses to crisis crisisphase questioning.

PAGE 57

49 In the crisis phase responses, highly-s upported-situational vari ables identified by Cancel et al. (1999) emerged. The variables – characteristic s of external publics, the characteristics of external publ ics’ requests, the cost-benefit of particular actions, the urgency of the situation, and potential or obvious threats – were all addressed by respondents. Of these variable s, the characteristics of ex ternal publics appeared most often. Each participant provided an example of how a particular public influenced their organization’s crisis response. The appearance of highly-supported-situa tional variables in the crisis phase responses, in addition to the emergence of the highly-supported-pred isposing variables in the precrisis responses, strengthens the likelihood of a relati onship between the contingency theory variables and the thre e-stage model of crisis management. Additionally, the appearance of the predispos ing variables followed by the situational variables mimics the classic progression of the crisis. The last phase of questions addressed the pos tcrisis stage of the three-stage model. The postcrisis stage begins once the crisis a nd its effects have has ended and posits that organizations learn from the crisis in orde r to apply that knowledge to future crisis preparation. A fluid transition from the postcrisi s stage to the precrisis stage occurs in the three-stage model of crisis management – the new postcrisis knowledge is synthesized with existing knowledge to become the new precrisis norm. Th e responses to the postcrisis phase questions examined whether th e organizations practiced cyclical crisis management, if that is, whether the synthesi s of new information into existing protocols occurred.

PAGE 58

50 Each participant explained that the crisis event allowed their organization to learn and prepare for future crises. Respondents explained that the firs t-hand knowledge they received was valuable and that the knowledge they gained al lowed them to update crisis protocols. Additionally, highl y-supported-predisposing variab les from the precrisis stage of the model reemerged in question responses to the postcrisis stag e. The predisposing variables of corporate size, characteristics of involved persons, and organizational business exposure during a crisis each appear ed were present in pa rticipant responses. The emergence of the predisposing variables in the postcrisis stage parallels the accepted principle that effective crisis ma nagement is a cyclical process. The study identified a potent ial relationship between crisis management and the highly-supported contingency th eory variables identified by Ca ncel et al. (1999). Figure 5 illustrates the contingency theory variab les identified by respondents overlaid on the three-stage model of crisis management. A lthough this study cannot generate a statistical correlation between the three-st age model of crisis management and the highly-supported contingency variables identified by Cancel et al. ( 1999), it does identify that a potential relationship between crisis management and th ese variables exists. Figure 5 illustrates the emergence of contingency theory variables from the three-stage model of crisis management.

PAGE 59

51 Figure 5: Contingency Variab les in Crisis Management Although this study cannot generate a statis tical correlation betw een the three-stage model of crisis management and the highlysupported contingency theory variables, future studies that examine the three-stage model of crisis management may verify the principles that emerged in this study’s fi ndings. Additional qual itative research can provide both scholars and pract itioners with a greater under standing of the relationship between crisis management and the conti ngency theory’s highly-supported-predisposing and situational variables can be achieved thr ough additional qualitative research. Future quantitative studies may be able to generate a statistically relevant correlation between model and theory, thereby illustrating how the three-stage model of crisis management uses the contingency theory variables to sh ape crisis preparation, response, and learning, may emerge from future qualitative studies of crisis management and contingency theory variables. Precrisis Stage (Emergence of Predisposing Variables) Postcrisis Stage (Re-Emergence of Predisposing Variables) Crisis Stage (Emergence of Situational Variables) Supported: Corporate Size Public Relations’ Access to Dominant Coalition Individual Characteristics of Involved Persons Some Support: Business Exposure Dominant Coalition’s Decision Power and Enlightenment Supported: Characteristics of External Public Urgency of Situation Characteristics of External Public’s Claim or Request Potential Cost or Benefit Potential or Obvious Threats Synthesis of Knowledge Gained from Crisis into Precrisis Norm Supported: Corporate Size Individual Characteristics of Involved Persons Business Exposure

PAGE 60

52 This researcher feels that there is a conc rete link between the contingency theory of accommodation and crisis management. The highly-supported variables identified by Cancel et al. (1999) act only as a starting point for future research. This study has generated support for this beli ef but it has only begun to uncover the relationship between crisis management and the contingency theory variables. The opportunity to identify and define this relationship between theory and pr actice serves as an avenue for future public relations scholarship.

PAGE 61

53 APPENDIX A E-MAIL RECRUITMENT LETTER Dear (Participant Name): I am a master’s student in pub lic relations at the University of Florida and am conducting my thesis on the crisis management proce ss in Florida municipal utility companies. The 2004 hurricane season was a blow to our st ate that I would like to study and learn how members of your industry r eacted. As a dedicated memb er of this industry that serves all Floridians I was hoping you would be interested in participating in my study. I have discussed my study with FMEA Executiv e Director Barry Moline to help ensure meaningful findings. The study is based on the Delphi technique. Pa rticipants will be sent questions regarding their crisis management response. Prior to th e study, participants will be asked a series of questions to generate a prof ile of each respondent. Upon co mpletion of the profiles, the study will then begin its three phases, each phase consisting of two questions. In each phase an initial question will be sent to pa rticipants followed by a second question based on the collective response of the participants. The duration of the study will not exceed six weeks and will not require more than three hours of participation over that time. The identity of participants will remain confidential th roughout the process. If upon completion of the study you would like a c opy of the findings I would be happy to provide them to you. Recruitment of participants will be comple ted no later than January 10, 2005. Following the recruitment process you w ill receive the respondent profile questions and the first phase of the study. From the receipt of each phase of ques tions you will have one week to respond. If you would like to participate in this study please sign the attached informed consent form and fax it to my research supervisor, Dr. Michael A. Mitrook, at (352) 392-3952. If you cannot open the attachment or have any questions please contact me at bboudreaux@jou.ufl.edu or (352) 262-1086. Thank you for your consideration. Sincerely, Brian Boudreaux

PAGE 62

54 APPENDIX B BIOGRAPHICAL INFORMATION Participant #1 1. How many years have you been involved in corporate communication? In what capacities have you done corporate communication? I have 20 years of corporate communicat ions experience, including corporate, government and nonprofit. In addition, I wo rked as a television new producer at Orlando’s CBS affiliate. 2. In your career have you ever been faced w ith a crisis situation? If so, what was your role in the crisis response? Yes, I have served as primary spoke sman through numerous utility crises, including two Amtrak train derailments, wild fires, terrorist thr eats and a customer service robbery. 3. What is your current title? Manager of Corporate Communications. Participant #2 I am an engineer. I am involved in cor porate communications only as a necessary function of my job. I am not formally trai ned in communications. My previous employer (GRU) offered communications tr aining classes which I attended. My current employer [City Name] has not co rporate communications personnel. As the Director of Public Services, I am often requi red to be the City's spokesperson on Public Services related issues. I have been involved in numer ous crisis situations including last year's hurricanes which affected our area. My current position is Director of Public Services for [City Name].

PAGE 63

55 Participant #3 1. How many years have you been involved in corporate communication? In what capacities have you done corporate communication? 18 years Media Relations Representati on, Media Relations Manager, Public Relations Manager, and as a privat e consultant for seven years. 2. In your career have you ever been faced w ith a crisis situation? If so, what was your role in the crisis response? Yes. Typcially handlin g media relations and as an agency/organization's spokesperson with the media. 3. What is your current title? Media Relations Coordinator Participant #4 1. How many years have you been involved in corporate communication? In what capacities have you done corporate communication? I have been employed for 23 years with th e City of [City Name]. I started as Secretary to the City Manager in 1981 and became Deputy City Clerk in 1994. Our City is unique to just a dozen or so cities in Florida where the City Manager is also the City Clerk. Each City Manager has been different regarding how much communication comes from the City Clerk. 2. In your career have you ever been faced w ith a crisis situation? If so, what was your role in the crisis response? When Hurricane Charley blew through our town on Friday, August 13th, myself and two other city hall employees manne d the darkened, power-less city hall on the following Monday, handling questions from citizens, calls from out of towners who wanted to assist in any wa y possible, vendors who wanted to come to our aid, emergency workers who wanted to know where to go and taking utility payments from our customers. Only the phones were worked inside city hall that day. .power was off for days around our ci ty. Decisions had to be made on how we could help. 3. What is your current title? Deputy City Clerk

PAGE 64

56 Participant #5 1. How many years have you been involved in corporate communication? In what capacities have you done corporate communication? 19 years all with munici pal utilities: Memphis Li ght, Gas and Water Division, Ft. Pierce Utilities Authority and [C urrent Employer]. I have been a communications specialist, senior comm unications specialist, communications coordinator, manager and now, director wh ich is part of our senior management team. 2. In your career have you ever b een faced with a crisis situ ation? If so, what was your role in the crisis response? Yes, I oversee the communications effort fo r these. Hurricane Jeanne and Frances are the largest. I also participated in e fforts during ice storms, gas explosions and electrocution. 3. What is your current title? Marketing and Commun ications Director

PAGE 65

57 APPENDIX C PRIMARY QUESTION FORMAT Dear (Participant Name): Thank you for your response to th e participant profile e-mail. This e-mail will begin the first phase of the study. The primary question is as follows: What preparatory steps has your organizatio n taken in an attempt to avoid a crisis event? Please be as specific as po ssible when describing the safeguards your organization employs. Feel free to draw on experiences in your current position as well as identify incidents from throughout your career. There is no minimum or maximum length requirement. Please respond no later than Tuesday, January 18th. Once all responses have been received a compilation will be se nt to you for further comment. Sincerely, Brian Boudreaux

PAGE 66

58 APPENDIX D FOLLOW-UP QUESTION FORMAT Dear (Participant Name): Thank you for your response to phase 1.1 of th is study. Attached to this e-mail is a compilation of responses from all current part icipants. Please review the responses of your colleagues before answering: How has your input been incorporated into your organization’s current crisis management protocols? Who else’s input, internal or external, was taken into account? Please be as specific as possible in your answer. There is no minimum or maximum length requirement to your answer. Additi onally, feel free to comment on any points raised by your colleagues. Please respond with your answer no later than Wednesday, January 26th. Sincerely, Brian Boudreaux

PAGE 67

59 APPENDIX E SAMPLE REMINDER E-MAIL Dear (PARTICIPANT NAME): This e-mail is to remind you that your respons e to phase 2.1 is requested by Wednesday, February 2nd. Once I have received answers from all me mbers of the study I will distribute the responses for further comment. As always, no names of individuals or companies will appear in the responses to maintain anonymity. If you need another copy of the question or ha ve any other questions please contact via email ( bboudreaux@jou.ufl.edu ) or call me at (352) 262-1086. Sincerely, Brian Boudreaux

PAGE 68

60 APPENDIX F PHASE 1.1 QUESTION AND RESPONSES Phase 1.1: What preparatory steps has your organizatio n taken in an attempt to avoid a crisis event? Addendum What characteristics of your organizat ion influence your crisis management practice? Participant #1 [Company Name] continually looks for vulnera bilities within its organization… whether they involve infrastructure, employees and/ or risk management issues. [Company Name] hired a full-time general counsel in 2002 to assist in this effort. From a planning perspective, [Company Name ] conducts an annual disaster drill that involves all employees of the utility. The pur pose of the drill is to test emergency equipment, measure response time, and evalua te communication within the utility. Past drills have included a wide ra nge of activities, including a simulation of cyber terrorism, a hazardous chemical spill, power line and substa tion damage, repair of utility vehicles in the field, power outages, producti on of sandbags at both power plants, and the testing of all emergency generators. During the drill, an evaluation team also throws out unexpected scenarios to test the read iness of utility employees. [Company Name] has sent employees to se veral FEMA-sponsored training courses on disasters. Since September 11, [Company Name] has upgrade d perimeter security at its substations and power plants as well as building securi ty at its service center and administrative offices. I.D. badges are now worn daily by all employees. [Company Name] has also purchased and installed surveillance cameras, electronic perimeter fence sensors, and proximity card readers on all exterior doors. [Company Name] produces an annual safety cal endar with safety tips to educate children on the dangers of coming into contact with elec trical equipment. This is to prevent injury or death caused by electrocution. In addition, si gnage has been installed at all substations providing notice of high voltage.

PAGE 69

61 In 2003, [Company Name] installed portable defi brillators for public use at all of its facilities in [local] County. The defibrilla tor is a lifesaving device for individuals suffering from cardiac arrest or other medical emergencies. In 2003, [Company Name] Board of Director s approved the purchase of terrorism insurance to protect the u tility's $360 million infrastructure from a potential attack. Participant #2 Develop a written Vulnerability Assessment Plan. Develop a written Emergency Response Plan. Meet periodically with staff with the objective of identifying potential areas susceptible to impact. We then attempt to eliminate or minimize the exposed area. If that is impractical, then we develop a pl an of action to addre ss restoration in the event impact occurs. Meet periodically with a statewide associa tion of other electric utilities to do the same. We also discuss mutual aid (ass isting other utilitie s that experience impacts). Review crisis events experienced by other utilities to learn what works and what does not. This also helps you consider al l the possible events that occur. Meet periodically with lo cal government emergency response team to discuss emergency response issues. Maintain an emergency preparedness manual with emergency contact numbers and procedures. The numbers are of other u tilities (mutual aid), material suppliers, employee rosters, fuel and food suppliers communication equipment (radios, cell phones suppliers), government agencies, medical treatment facilities, key customers (water treatment facilities, wa stewater treatment facilities, hospitals, food suppliers, etc.), cust omers with special medical needs that depend upon electric service, contractors that may offer assistance, food suppliers, emergency shelters, etc. Prior to an anticipated storm event, meet with all employees to develop plan of action (work responsibilities, when to report, where to report, when to safe up facilities, etc). We make post-storm wo rk assignments (material procurement, damage assessment, telephone switchboard, m eal preparation, etc.) We make sure all critical facilities have auxiliary power and all res ponse vehicles and fuel are topped off. We contact all numbers on the c ontacts list to ensure the numbers are accurate and the contacts are aware of our status. We secure buildings and facilities (board them up). We notify the pub lic of our status an d preparations and request that they prepare for the storm. We advertise the telephone number for our emergency response attendant. We ensure our emergency supply cabinets are fully stocked. We ensure all radios ar e accounted for, have adequate spare batteries, and are fully charged. Participant #3

PAGE 70

62 In 2001, [Company Name] went through a reorga nization and moved to a process-based structure. Therefore, during the recent hurricane events, many departments had to coordinate to ensure all goals and objectives were met in restoring power as quickly as possible. We conducted two meetings daily, one in th e morning and one in the afternoon, until the outage figures dropped to a reasonable number (b elow 40,000). At that point, we went to one meeting per day, and these meetings we re conducted via confer ence calls over the weekend. These involved the CEO, the Electric, Water and Sewer Operations Department, Internal and External Communications, Public Outr each, Facilities Maintenance, Client Relationships, Security, Finance, Risk Management, Safety, Engineering, Technology Services, and Governmental Relations All vice presidents on the executive management team were at the meetings, along with many of the field operations supervisors to ensure direct information flow from the field. Since a major disruption of electric service can jeopardize public safety and public health (affecting police, fire and hospital operations ), restoring power to those operations takes precedence in a major outage. With [Company Name ], another priority is either restoring power via electric wire or gene rators to ensure sewer pumping stations can continue to pump to avoid SSOs, Sanitary Sewer Overflows. For many years, and with a new emphasis after the September 11t h attacks, [Company Name] developed a substantial Emergency Pl an Book that sets forth all operating procedures to be implemented should a disaster natural or otherwise, occur in the service area.

PAGE 71

63 APPENDIX G PHASE 1.2 QUESTION AND RESPONSES Phase 1.2 How has your input been incorporated into your organization’s current crisis management protocols? Who else’s input, internal or external, was taken into account? Participant #2 We are a very small organization 1 superi ntendent, 3 linemen, 2 apprentices, plus me. We all discuss emergency preparation forma lly in meetings called for the purpose or informally everyday. Our more experienced workers have greater insights, but everyone is free to make suggestions. I am primarily responsible for ensuring that we are adequately prepared. It is my job to make sure we think about it, learn as much as we can from other organizations, and meet to discuss our response. We work with other agencies in our community local police, fire rescue, and emergency management. We also attempt to educate our public. Any input received from those organizations is incorporated to the extent it makes sense for us to do so. We rely heavily on the information we pick up from the Florida Municipal Electric Association (FMEA) and the Florida Munici pal Power Agency (FMPA) regarding storm preparation and emergency response. Participant #3 When the hurricanes began making landfall, we were handling outage information by request. We realized that, while we wanted to remain accessible to the media, this was becoming extremely difficult. Therefore, once the major impact of the storm had passed, we implemented a twice daily communicati on to the media and other stakeholders (the Emergency Operations Center, City Council Members, Mayor, neighboring county and city representatives, and to all internal contacts, etc.) on updated outage information and where our crews were working. We e-mailed and faxed to all the groups mentioned at 5 a.m. and at 4 p.m. daily to en sure morning and afternoon television broadcasts, drive time radio and newspapers would have access to current information

PAGE 72

64 when they went on the air. This also freed up our operations staff to stay focused on the work of restoring power rather than res ponded to requests for information throughout the day. Also, we would communicate rumor-cont rol information, based on questions we were receiving from the media and the public. Most of our external stakeholders were pleased with the move, especially city counc il members and city hall, where constituents were calling when they couldn't reach [Company Name]. One serious complaint was that customers could not get through to [Company Name] on the phone. We were receiving more phone calls in one day than we usually received in a month, so there was no way we were capable to respond to everyone. Therefore, based on input from the communications and custom er outreach personnel, we set up teams of three-to-five employees to go out to the most affected areas and talk to customers one-onone. This was a highly effective move in ra ising our customer satisfaction levels and the general public relations boost we received from the effort. Participant #4 Our representative meets with the local gove rnment emergency response team to discuss issues and make plans.

PAGE 73

65 APPENDIX H PHASE 2.1 QUESTION AND RESPONSES Phase 2.1 What characteristics of the crisis even t influenced your organization’s crisis response? Participant #2 The most significant storm characteristic that impacted my area was simply the amount of damage incurred. We watch for storms and prepare for impact pr etty much the same way. We insert lessons learned as needed to our preparation. Damage assessment and restoration of servi ce were the big tasks we faced post-storm. There was a lot of pressure to restore power ASAP. We prioritized ou r major circuits and critical facilities. Th e effort expended by everyone w ithin our organization to recover from the storm's impacts is most memorable. Participant #3 Unlike other smaller storms and events, the fi rst major hurricane (Frances) forced us to modify our normal procedures for communi cating with customers and the media. Our typical procedure is for customers to contact [Company Name] by phone in the case of an outage. With more than half of our customers out of power on Labor Day morning, our phone lines received as many calls in one day as [Company Name] typically receives in a month. Therefore, many customers could not contact us. (At th is point, their calls were not necessary, since so many circuits were out and we already knew where they were). Under normal circumstances, we also ask customers to report downed power lines to use through the customer care cen ter (our main switchboard number). Therefore, through the media we communicated to customers not to contact [Company Name] with outage information. ..to instea d wait for us to communicate through the media when that customer communication woul d be necessary. We also asked customers to contact 911 to report downed power lines. Th is allowed fire offici als to tape off areas where lines were down to allow our linemen to work on the main circuits that would bring more customers back on line quicker.

PAGE 74

66 We resumed normal customer communication pr ocedures once the number of outages got down to a minimal amount (about 40,000 without power). To assist in communicating with customers who could not reach us and might not have access to a radio, we assembled three-person te ams to visit the most affected areas to hang door-hangers regarding the restoration pr ocess and when crews would be in their neighborhoods. With our two-time-per-day blast fax to the media, which we also shared with area cities/council members so they could assi st in communicating up-to-date and accurate information, we also added a rumor-control narr ative. Based on feedback received firsthand and through our customer care consultant s, we would address rumors (rich people are getting power back on first, the utility is supplying free gene rators to wealthy customers, etc.) and quell some of the misi nformation that would spring up in different areas of town. In terms of the restoration pr ocess itself, our normal procedur e is to work on distribution lines (lines within the nei ghborhoods) first, then move out toward the substations. However, with so many people without power that was no an effective way to go. Therefore, we worked the opposite direction -from transmission lines to substations, substations to distributions, and into the neighborhoods last. This allowed [Company Name] to get the major circuits, thus the la rgest number of customers, back on quicker, then move into the smaller pockets (nei ghborhoods) and then ha ndling the individual outage problems last. While there were many minor changes we ma de in our operations and communications procedures, these were the major changes we administered within two days of Hurricane Frances. .and employed throughout the subs equent strike of Hurricane Jeanne. Participant #4 We are a small city and work closely w ith the county emergency management team.

PAGE 75

67 APPENDIX I PHASE 2.2 QUESTION AND RESPONSES Phase 2.2 Were certain publics more vocal regarding their need for power restoration? Did their requests influence your organization’s response? Participant #2 The most vocal group in our co mmunity was a more affluent gated community that felt they were not receiving adequate attention. Our priorities for re storation were major feeders, critical f acilities, major commercial, everyone else. We were contacted by people for every part of the community wanting, and in same cases demanding, their power restored. Rarely did it affect our plans for restorat ion the exceptions we re previously unknown critical or medical needs and quick easy fixes that would restore power to many customers. Participant #3 While there is no precise data on this, my pe rception would be that parents a few days into the outages began to express concer n over getting the schools opened as soon as possible. Pressure came from the mayor's office, vis-a-vis the city council members offices through their cons tiuents, that [Company Name] needed to make school restoration a priority. Obviously, reasons for this were many -ki ds home from school over extended days begin the affect the ability of parents to get into their work routine (if their employers had power), costing them additional money for da y care or requiring them to stay home. Also, having no power at home would just irr itate an already stressf ul situation at home. This pressure did influence our organization' s response somewhat, but did was difficult to focus on a school's power supply when the en tire circuit serving the school and the adjacent neighborhoods were out of power. We did identify that school administrators did not have a good handle on how many sc hools did not have power, and [Company Name] did develop a contact/location list fo r all schools we serve -which we did not compile prior to the hurricanes.

PAGE 76

68 Participant #4 Our electric department got the police, fire and emergency services running, then tackled the downtown business and medical clinic. Re sidents called in from all over town with requests for power. Three of us in city ha ll worked with no power and flashlights when upstairs to fill out sheets of re quests for service and faxed them to the electric department (the cell phones the electric crews had were not working). Our city had electric crews from four other cities assisting us with rebuilding the lines. Our electric crews were not in fluenced by who called, but rath er in order of the request.

PAGE 77

69 APPENDIX J PHASE 3.1 QUESTION AND RESPONSES Phase 3.1 How did the 2004 hurricane season influe nce your organization’s preparation for future crises? Participant #2 It provided us with valuable experience. There is no substi tute for experiencing events. It provided us with a compilation of inform ation related to emerge ncy response that we will incorporate into future emergency re sponse manuals and minimize the amount of "reinventing" that is required. It generated much, much disc ussion and dialogue related to emergency response. I / we participated in as much of this dialogue as possible to learn from the experiences of others. Participant #3 Brian, I've touched on some of these previous ly, but in a nutshell, here are the major changes: Strengthening our electrical mutual aid agreements Initiating a more formal water and sewer mutual aid program Upgrading some of our coastal-reliabili ty-standards equipment near the beach Upgrading insulators, cut-outs, arrestors, etc in areas that experienced problems Documenting recovery expenditures in a “FEMA-friendly” format Assigning office/engineering st aff (to handle customer inte ractions and non-repair type work) to ride with repair crews Including door-to-door employee-custom er communications with TV, radio, newspaper, internet communications plan As much as practicable, beginning detail ed, systematic engineering assessments prior to restoration work Changing initial recovery response to con centrate on main lines, substations, etc o i.e., start with transmission and substations…then feeders…then laterals…then services Changing our customer information procedures o E.g., customer will not be asked to report outages during initial heavy recovery work

PAGE 78

70 Participant #4 After dealing with three hurricanes in one season, each city department had a new understanding of what they would need to do to be prepared for the 2005 season.

PAGE 79

71 APPENDIX K PHASE 3.2 QUESTION AND RESPONSES Phase 3.2 Considering what your organization learned from the 2004 hurricane season, do you feel prepared for the 2005 hu rricane season? What concerns does your organization have? Respondent #2 No, we do not feel fully prepared for a hu rricane season. The cost involved with fully preparing to respond to hurricanes like those th at hit central Florida are too great for a small utility to overcome in a single year. We will list our needs, prioritize them, and address them when possible. Our great concerns are with inventory of material items, communication equipment, and obtaining assistance through mutual aid. Respondent # 3 Yes. Since all of us went through the actual event, as opposed to doi ng a drill or tabletop exercise, we all have first-hand experience on dealing with the major impact of hurricane winds and the aftermath. We now know what procedures need to be modified, compared to normal operating procedures that we have for dealing with typical spring/summer storms, and how all agency personnel can be used more effectiv ely to work with customers and communicate with them and the media. [City Served] actually did not take a direct hit, but was affected by the strong winds on the northeast side of both hurricanes. Base d on what we saw last summer, there is a concern on the additional amount of damage that would be caused to [Company Name] and the city, and the time it would take to rest ore the area to normal, if we were ever to receive a direct hit from the eye of the storm. Respondent # 4 Our organization learned that you may think you are prepared, but when you have three successive hurricanes wreck havoc on your community in six weeks, you can't get that prepared. We will always have concerns when Hurricane Season approaches. We know how hard we worked to tackle whatever was thrown our way in 2004 putting our city back together.

PAGE 80

72 LIST OF REFERENCES Burnett, J.J. (1998). “A strategi c approach to managing crises.” Public Relations Review, 24 (4), 475-488. Cancel, A.E., Cameron, G.T., Sallot, L.M ., and Mitrook, M.A. (1997). “It depends: a contingency theory of accommodation in public relations.” Journal of Public Relations Research, 9 (1), 31-63. Cancel, A.E., Mitrook, M.A., and Cameron, G.T. (1999). “Testing the contingency theory of accommodation in public relations.” Public Relations Review, 25 (2), 171-197. Clare, M. (1994). “Delphi Technique.” Cornell University. Accessed December 9, 2004. http://www.cce.cornell.edu/admi n/program/documents/delphi.htm Coombs, W.T. (1999a). “Information and compa ssion in crisis responses: a test of their effects.” Journal of Public Re lations Research, 11 (2), 125-142. Coombs, W.T. (1999b). Ongoing crisis communication: Planning, managing, and responding. Thousand Oaks, CA: Sage Publications, Inc. Coombs, W.T. and Holladay, S.J. (2001). “An extended examination of crisis situations: A fusion of the relational manageme nt and symbolic approaches.” Journal of Public Relations Research, 13 (4), 321-340. Ewing, R.P. (1987). Managing the new bottom line: Issues management for senior executives. Homewood, IL: Dow Jones-Irwin. Fink, S. (1986). Crisis management: Planning for the inevitable. New York: AMACOM. Gonzlez-Herrero, A. and Pratt, C.B. (1996) “An integrated sy mmetrical model for crisis-communication management.” Journal of Public Relations Research, 8(2). pp. 79-105. Goodman, M.B. (2001). “Current tre nds in corporate communication.” Corporate Communications, 6 (3), 117-123. Grunig, J.E. (2001). “Two-way symmetrical public relations: Past, pres ent, and future.” Handbook of public relations. R. Heath and G. Vasquez. Thousand Oaks, Sage Publications, Inc.: 11-30.

PAGE 81

73 Grunig, J.E. and Grunig, L.A. (1992). “Models of public relations and communication.” In J.E. Grunig (Ed.), Excellence in public relations and communication management. pp. 285-326. Hillsdale, NJ: Lawrence Erlbaum Associates. Grunig, J.E. and Hunt, T. (1984). Managing public relations. New York: Holt, Rinehart and Winston. Grunig, J.E. and Repper, F.C. (1992). “Strateg ic management, publics, and issues.” In J.E. Grunig (Ed.), Excellence in public relations and communication management. pp. 31-64. Hillsdale, NJ: Lawrence Erlbaum Associates. Grunig, L.A., Grunig, J.E. and Dozier, D.M. (2002). Excellent public relations and effective organizations: A study of comm unication management in three countries. Mahwah, NJ: Lawrence Erlbaum Associates. Howard, C.M. and Mathews, W.K. (2000). “Chapter 10 – Crisis planning: how to anticipate and manage em ergency situations.” On deadline: Managing media relations. Third edition. pp. 215-269. Long Grove, IL:Waveland Press, Inc. Mitroff, I.I. (1994). Crisis management and environmentalism: A natural fit. California Management Review, 36 (2), 101-113. Pauchant, T.C. and Mitroff, I.I. (1992). Transforming the crisis prone organization. San Francisco, CA: Jossey-Bass Publishers. Penrose, J.M. (2000). “The role of perception in cr isis planning.” Public Relations Review 26 (2), 155-171. Sapriel, C. (2003). “Effective crisis manage ment: tools and best practice for the new millennium.” Journal of Communi cation Management 7 (4), 348-355. Seeger, M.W., Sellnow, T.L., and Ulmer, R.R. (2003). Communication and organizational crisis. Westport, CT: Praeger Publishers. Shrivastava, P., Mitroff, I. I., Miller, D. and Miglani, A. (1988). “Understanding industrial crises.” Journal of Management Studies 25 (4), 285-303. Smits, S.J. and Ezzat, N. (2003). “‘Thinking the unthinkable’ – leadership’s role in creating behavioral readiness for crisis management.” Competitiveness Review, 13 (1), 1-23. Thomas, K.W. and Kilmann, R.H. (1974). Thomas-Kilmann conflict MODE instrument. Tuxedo, NY: Xicom. Turoff, M. and Hiltz, S.R. (1996). “Computer based Delphi process.” New Jersey Institute of Technology. Accessed December 9, 2004. http://eies.njit.edu/~turoff/Papers/delphi3.html

PAGE 82

74 Wakefield, R.I. (2000). “Prelimi nary Delphi research on inte rnational pub lic relations programming.” In Moss, D., Verc ic, D., and Warnaby, G. (Eds.), Perspectives of public relations research pp. 179-208. New York: Routledge. Wilson, S. (2004). “The myths and reality of crisis communication.” Accessed October 15, 2004. http://www.wilson-group.com/articles/myths.shtml

PAGE 83

75 BIOGRAPHICAL SKETCH Brian Boudreaux completed this study as the capstone of his candidacy for a Master of Arts in Mass Comm unication, specializing in public relations at the University of Florida. In 2003, Boudreaux received hi s Bachelor of Arts degree in speech communication from the University of Illinoi s. Boudreaux’s research interests include crisis management and strategic planning. Upon graduation, the author will begin his professional career in the cor porate or government sector.


Permanent Link: http://ufdc.ufl.edu/UFE0010486/00001

Material Information

Title: Exploring a Multi-Stage Model of Crisis Management: Utilities, Hurricanes, and Contingency
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010486:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010486/00001

Material Information

Title: Exploring a Multi-Stage Model of Crisis Management: Utilities, Hurricanes, and Contingency
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010486:00001


This item has the following downloads:


Full Text












EXPLORING A MULTI-STAGE MODEL OF CRISIS MANAGEMENT:
UTILITIES, HURRICANES, AND CONTINGENCY













By

BRIAN BOUDREAUX


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN MASS COMMUNICATION

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Brian Boudreaux















ACKNOWLEDGMENTS

I would like to thank my chairman, Dr. Michael Mitrook, for his guidance

throughout this project. His insights, support, and candor were fundamental to the

completion of this study. Additionally, my committee members, Dr. Peg Hall and Dr.

Jennifer Robinson, provided a refreshing view of this piece that will not soon be

forgotten. Lastly, I would like to thank my family for their continued support of my

academic endeavors. This would not have been possible without their never-ending

support.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF FIGURES ....................................... .. ......... ............................ vi

ABSTRACT .................................................. .................... vii

CHAPTER

1 IN T R O D U C T IO N ................................................. .............................................. .

2 R E V IEW O F L ITER A TU R E ........................................... ....................... ............... 3

R reality o f a C crisis ................................................................................................. 3
Issues Management......... ........... ............................. ...............4
M odels of C risis M anagem ent................................................................. ...............7...
The Contingency Theory of Accommodation...................................................... 14
Criticisms of Contingency Theory ................................................................. 16
M ore on A advocacy .............. .................. ............................................. 17
More on Accommodation......................................................... ............... 18
Contingency Variables ............................................................................ 19
P redisposing variables............................................................. ............... 19
Situ ation al v ariab les ....................................... ...................... ................ 22

3 M E T H O D O L O G Y ................................................... ............................................ 26

R eview of R research Q questions ...................................... ..................... ................ 26
D ata C o llectio n ........................................................................................................... 2 6
S am p le ........................................................................................................ ....... .. 2 8
D ata A n aly sis .............................................................................................................. 2 9

4 RESULTS AND DISCUSSION .............................................................................. 30

P h a se 1 ....................................................................................................... ......... 3 0
P h a se 1 .1 .......................................................................................................... . 3 1
P h a se 1 .2 .......................................................................................................... . 3 4
Phase 1 Conclusion................................................... ............. ..35
P h a se 2 ....................................................................................................... ........ .. 3 6
P h a se 2 .1 .......................................................................................................... . 3 7









P h a se 2 .2 ..............................................................................................................3 8
Phase 2 Conclusion...................................................... ............ 40
P h a s e 3 ........................................................................................................................4 1
P h a s e 3 .1 ..............................................................................................................4 1
P h a se 3 .2 ..............................................................................................................4 2
Phase 3 C conclusions .........................................................................................43

5 LIMITATIONS AND FUTURE STUDY ................................................ ...............44

6 SUMM ARY AND CONCLUSIONS......................................................................46

APPENDIX

A E-M AIL RECRUITM ENT LETTER .........................................................................53

B BIOGRAPHICAL INFORM ATION.......................................................................54

C PRIM ARY QUESTION FORM AT ........................................................................57

D FOLLOW-UP QUESTION FORMAT ..................................................... ...............58

E SAM PLE REM INDER E-M AIL.............................................................................59

F PHASE 1.1 QUESTION AND RESPONSES.......................................... ...............60

G PHASE 1.2 QUESTION AND RESPONSES.......................................... ...............63

H PHASE 2.1 QUESTION AND RESPONSES.......................................... ...............65

I PHASE 2.2 QUESTION AND RESPONSES.......................................... ...............67

J PHASE 3.1 QUESTION AND RESPONSES.......................................... ...............69

K PHASE 3.2 QUESTION AND RESPONSES.......................................... ...............71

L IST O F R E F E R E N C E S ................................................................................................... 72

BIO GRAPH ICAL SK ETCH ..........................................................................................75















v















LIST OF FIGURES

Figure page

1 Crisis M anagem ent: Strategic Considerations ...................................... ...............8...

2 T he C risis L ife C ycle ............................... .............................................9....

3 Development of Issues with or Without Management Intervention .......................9...

4 Advocacy-A ccom odation Continuum ................................................. ............... 14

5 Contingency Variables in Crisis M anagement.................................... ................ 51















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts in Mass Communication

EXPLORING A MULTI-STAGE MODEL OF CRISIS MANAGEMENT:
UTILITIES, HURRICANES, AND CONTINGENCY

By

Brian Boudreaux

May 2005

Chair: Michael Mitrook
Major Department: Journalism and Communications

The purpose of this study was to explore the three-stage model of crisis

management and determine what influence the contingency theory of accommodation's

highly-supported-predisposing and situational variables have on crisis management.

This qualitative study examines crisis management practices of municipal utility

companies in the state of Florida. Participants discussed their organization's crisis

management practices in conjunction with the very active 2004 hurricane season. A

modified Delphi study was used to collect data from senior-level public relations and

corporate communications professionals from a cross-section of state municipal utility

companies. Primary and secondary questions referencing each stage of the three-stage

model of crisis management were generated to identify the variables that most affected

organizational crisis management.

The findings expose a relationship between the highly-supported-predisposing and

situational variables of the contingency theory and the three-stage model of crisis









management. This study serves as a springboard for future quantitative research that may

be able to generate a statistical correlation between the contingency theory's highly-

supported variables and the three-stage model of crisis management.














CHAPTER 1
INTRODUCTION

Managing crises is a vital aspect of public relations practice as all organizations

experience them. As one practitioner explains, "no one is immune. Crises are non-

discriminatory. They don't care who gets in the way. It's not a matter of 'if,' in most

cases, but simply 'when."' (Wilson, 2004, paragraph 2). Likewise, L. Grunig, J. Grunig

and Dozier (2002) asserted, "Regardless of the model of public relations practiced or the

expertise of the communicator, crises inevitably befall organizations" (p. 473). The

belief that a crisis situation will not strike an organization must change; the mentality

must shift from if, to when. Indeed, crises are a reality that companies, government

agencies and nonprofit organizations alike must recognize. The incidence of crises is not

a new phenomenon, but increasingly catastrophic results have validated their relevance in

public relations.

The ubiquitousness of crisis situations necessitates the creation of a widely

accepted model of a crisis management that provides organizations the framework to

assist in the survival of a crisis event. Though there are models that explore crisis

management, these models have yet to include a concrete theoretical underpinning. This

qualitative study attempts to begin grounding the three-stage model of crisis management

in the contingency theory of accommodation.

The researcher believes that the contingency theory's highly-supported-

predisposing and situational variables, identified by Cancel et al. (1999), will emerge as

influential to crisis management. The researcher proposes that the predisposing variables






2


will appear in the participant responses during precrisis stage while the situational

variables will emerge in the crisis stage of the three-stage model of crisis management.

This study will test this hypothesis while also exploring the role of the contingency

theory variables in crisis management.














CHAPTER 2
REVIEW OF LITERATURE

Reality of a Crisis

Although all types of organizations are vulnerable to a crisis, certain industries are

inherently more prone to a crisis event. L. Grunig et al. (2002) quoted a practitioner who

refers to the insurance industry as a "crisis by definition" (p. 473). Among the many

reasons cited by scholars and practitioners to prepare for a crisis is one advanced by

Howard and Mathews (2000). Preparation is needed, they said, ". .because you cannot

control the elements, human nature or the outside world" (p. 217). This statement

outlines the difficulty of crisis management; though practitioners and organizations can

prepare for numerous scenarios, there are many more they have no control over.

Before further discussion, the definition of a crisis must be established. Scholars

have defined the term crisis in a number of ways. As defined by Sapriel (2003), a crisis

is "an event, revelation, allegation or set of circumstances which threatens the integrity,

reputation, or survival of an individual or organisation" (p. 348). Shrivastava, Mitroff,

Miller, and Miglani (1988) presented an in-depth definition of a crisis as

An organizationally-based disaster, which causes extensive damage and social
disruption, involves multiple stakeholders, and unfolds through complex
technological, organizational, and social processes. (p. 285)

Much of the literature discusses crises at the organizational level. In doing so, the

definition of a crisis has focused on only a segment of crises. Pauchant and Mitroff

(1992) developed a definition of a crisis that can apply to more than just organizationally

based events. The authors discuss a crisis as "a disruption that physically affects a









system as a whole and threatens its basic assumptions, its subjective sense of self, its

existential core" (p. 15). This definition encompasses non-organization crises, such as

natural disasters, that have an effect, not only on individual organizations, but rather a

community system as a whole.

Other scholars take a more basic approach to defining a crisis. Coombs (1999a)

described a crisis as something that embarrasses or challenges an organization's character

and that demands an explanation. Gonzalez-Herrero and Pratt (1996) defined a crisis as

something that is simply "unwelcome and sudden" (p. 82). Lastly, Coombs and Holladay

(2001) defined a crisis as "one event or interaction within a larger relationship between

an organization and its stakeholders. .[that can] damage or be a threat to a quality

relationship" (p.324). Though the depths of definitions vary, they each employ the same

principle: A crisis is an event that can drastically affect the ability of an organization to

sustain itself. This will serve as the definition of a crisis for discussion throughout this

study.

A danger that organizations sometimes face is the issue of a perceived crisis.

Organizations may leap into crisis management protocols for mere transgressions. For

example, the resignation of an organization's Chief Financial Officer alone is not a crisis

and does not warrant the use of crisis protocols. If the resignation of a CFO is the result

of financial malfeasance, though, the organization should address the situation as a

possible crisis. Organizations need to be able to differentiate between perceived and real

crises.

Issues Management

Gonzalez-Herrero and Pratt (1996) posit that a crisis is sudden. Although the onset

of a crisis can be sudden, there are situations when organizations are aware of and









prepared for such an event. Though many factors influence the outcome of a crisis, Smits

and Ezzat (2003) posit that preparation is one of- if not the most important steps in the

prevention of a crisis, "Effective crisis management depends upon planning and people"

(p. 2). Penrose (2000) stated, "Researchers tend to agree that organizations that practice

proactive crisis management will lessen the damage of a crisis" (p. 155). Indeed,

Gonzalez-Herrero and Pratt (1996) also explained that crises have early signals and,

"... sensing potential problems is the first step toward avoiding or resolving them or

minimizing their impact" (p. 82).

Issues management helps an organization become aware of possible crises and plan

accordingly. One facet of this awareness involves determining those publics which could

lead or contribute to a crisis. As J. Grunig and Repper (1992) explained, "Members of

active publics, affect organizations more than passive ones because they engage in

individual behaviors to do something about the consequence of organizational actions"

(p. 137). If an organization can identify these groups it can attempt to dispel or

counteract damaging behaviors.

Issues management helps remove the unexpected and sudden factors of the crisis;

the situation can be defused before it occurs. Moreover, information gathered will not

only allow practitioners to prepare, it will allow them the opportunity to realize and

understand what preparations they cannot make. "Nothing prepares you for change better

than the awareness of what you can do, and cannot do, about it" (Goodman, 2001, p.

117). The occurrence of some events is uncontrollable, thus, practitioners should shift

their efforts away from preventing such events and towards weathering them. For

instance, a state is aware of a hurricane days before it makes landfall. Though issues









management allows the opportunity to prepare for the possible crisis, it cannot neutralize

the storm. Instead, issues management allows practitioners to prepare for the event as

well as the resolution after the crisis occurs.

Howard and Mathews (2000) describe issues management as a practice that good

managers have done for years; that it is a necessary part of effective planning. Gonzalez-

Herrero and Pratt (1996) discuss issues management as a tool to identify and anticipate

potential issues before they are a threat. Thus, issues managers should be forward-

thinking. The longer an organization is aware of the possible issue, the better it can

prepare. Indeed, Ewing (1987) posits that issues managers should look 12 to 36 months

ahead.

Issues management is often thought of as a proactive practice used to avoid a

negative situation, though it can also identify positive opportunities. Practicing issues

management improves an organization's awareness of its community and ways to

positively involve itself. For example, an organization can improve its standing in a

community by being aware of and supporting an upcoming philanthropic event. Issues

management can also present organizations with an opportunity to showcase its positive

practices.

Unfortunately, preparation and issues management is sometimes not sufficient to

avoid a crisis. Numerous incidents illustrate that although organizations were prepared

for an event, the crisis was still damaging. Examples of such scenarios can be seen in all

professions: corporate, litigation, and nonprofit. Preparation, though valuable, is not a

guarantee of success. The difficulties associated with a given crisis will impact the

effectiveness of the preparation. Burnett (1998) presents some difficulties crisis









managers may face, they include too little data, too much data, and little planning. These

variables illustrate that simply having information and early preparation may not be

sufficient deterrents.

Models of Crisis Management

"The disparate volumes of crisis management information can be overwhelming"

(Coombs, 1999b, 9). Certainly there are many factors a crisis management professional

must account for in order to be successful. The organization of such information into a

universally accepted model has presented practitioners and scholars with a challenge.

Burnett (1998) identifies both tasks and factors that compromise the ability of an

organization to practice crisis management (Figure 1). First, the author cited four factors

that inhibit crisis management: time pressure, control issues, threat level concerns, and

response option constraints. Burnett claims these factors, found on the outer-ring of the

model, disrupt an organization's ability to focus on and strategically manage a crisis

situation. According to this model, only when these four factors have been addressed can

the strategic management of the situation begin.

Burnett (1998) divides the model's six step inner-circle into three categories:

identification, confrontation, and reconfiguration. The identification step is composed of

goal formation and environmental analysis the preparation for the crisis. Confrontation

encompasses strategy formulation and strategy evaluation the point when an

organization is involved in the crisis. Lastly, reconfiguration includes strategy

implementation and strategic control how the organization adapts to crisis intervention.

The author posits that during a crisis the difficulty of performing well in each category

increases. As illustrated by the model, employing the tasks that comprise the inner-circle

provides the organization an opportunity to control and manage a crisis situation.



























Figure 1: Crisis Management: Strategic Considerations; Adopted from Burnett (1998)

Some models make the comparison of the crisis to a lifecycle (Fink, 1986;

Gonzalez-Herrero & Pratt, 1996). Inherent to this analogy is that the crisis has both a

birth and a death. Gonzalez-Herrero and Pratt (1996) discuss how crises follow a

sequential path through four phases: birth, growth, maturity and decline (Figure 2).

Although this is an elementary model of a crisis, it is sufficient. It divides a crisis

into identifiable stages, it illustrates how a crisis changes over time, and that the cycle

does not end, rather that its effects linger beyond the decline and death of the crisis. This

basic model presents a simplistic, yet effective illustration of the crisis lifecycle.

Gonzalez-Herrero and Pratt (1996) expanded this model to illustrate the effect of

issues management in a crisis situation. By practicing issues management before crisis

birth, the authors believe organizations can shift the outcome of the crisis (Figure 3).

Previously, the crisis would have reached maturity, to eventually decline into the post-

crisis phase. In this adaptation of the model, issues management is shown to be effective

as the planning stage results in the prevention of a crisis.


















.ATURrY


GROWTH


DECLINE


BIRTH


Figure 2: The Crisis Life Cycle; Adopted from Gonzalez-Herrero and Pratt (1996)


CRISIS









S" I ...ANN.... .. ,,






Figure 3: Development of Issues with or Without Management Intervention; Adopted
from Gonzalez-Herrero and Pratt (1996)

Coombs (1999b) states that the three most influential staged approaches to crisis

management are Fink's (1986) four-stage model of a crisis lifecycle, Mitroffs (1994)

five-stage model, and the basic three-stage model.

The three-stage model is unique in that no single scholar is attributed with its

creation. "The three-stage model is not associated with any particular theorists, but it

appears to have emerged from several research efforts as a general analytical framework"









(Seeger, Sellnow, and Ulmer, 2003, 97). Coombs (1999b) describes the three stages of

the model precrisis, crisis, and postcrisis as macrostages that can be applied to many

models of crisis management.

The precrisis stage includes all aspects of crisis prevention issues management,

planning, and other proactive steps. The crisis stage refers to the steps taken to cope with

and respond to the crisis event crisis recognition, information distribution, message

development, reputation management, and evolving developments. The postcrisis stage

begins when the crisis is resolved ensuring the crisis over, assuring publics of the

security of the organization, and learning from the event.

Coombs (1999b) and Seeger et al. (2003) contend that the three-stage model

provides a framework for the incorporation of various sub-stages which change based on

a multitude of variables. The type and impact of the crisis, media coverage of the event,

and the size and culture of the affected organization can all be influential factors.

Coombs explains that both Fink (1986) and Mitroff's (1994) models fit into the general

parameters of the three-stage model.

Fink's (1986) four-stage model examines a crisis as an extended event with

sufficient warning signs that precede the event. Fink's four stages are: the prodromal

stage, the acute stage, the chronic stage, and the resolution stage. In the prodromal stage,

the role of a crisis management professional is not reactive, but instead a proactive

approach. In this stage, crisis managers attempt to identify an impending crisis. This

information can be found in various places, such as internal and external audits,

government legislation, and industry publications. Actions taken during the prodromal









stage can easily be placed into the precrisis stage of the three-stage model as they address

an organization's crisis prevention.

Fink (1986) argues that the actual crisis event begins with a trigger, during what he

refers to as the acute stage. This stage is characterized by the crisis event and resulting

damage. The severity of the crisis and damage are influenced by the success of the

prodromal stage. Successful proactive identification of a crisis can reduce the impact of

the crisis in the acute stage. Failed recognition in the prodromal stage creates a reactive

situation instead of a proactive intervention.

The third stage of Fink's (1986) model is the chronic stage. This stage refers to the

lasting effects of the crisis. Although individual crises may occur quickly, the lasting

effects of the incident can extend the lifecycle of the crisis. Additionally, this stage may

include a barrage of questions about the crisis which will keep the event visible to various

publics.

For example, an individual event, such as a natural disaster, may occur quickly, but

the fallout of the incident make take weeks or months to repair. Individuals and

organization in Florida are still recovering from the 2004 hurricane season. More

recently, entire countries are still dealing with the massive damage and loss of life caused

by a tsunami that struck areas of Southeast Asia December 26, 2004. Coombs (1999b)

states that the acute and chronic stages act as sub-stages of the crisis stage of the three-

stage model. These stages include the appearance of a crisis and the steps taken to

resolve the crisis event, characteristics which are found in the crisis stage of the three-

stage model.









The final stage in Fink's (1986) model is the resolution stage. This stage identifies

a clear end to the crisis. Although organizations view this as the goal, it is not one to be

rushed to. An organization's premature conclusion that the chronic stage has ended can

leave them vulnerable to the resurgence of the crisis. Due diligence in the previous

stages of the model must be practiced to insure such a regression does not occur. This

last stage of Fink's model parallels the postcrisis stage of the three-stage model as it

ensures the crisis has ended and distributes that message.

The resolution stage does neglect one portion of the three-stage model. Fink does

not discuss crisis management as a cyclical process. This is seen as one of the oversights

of the model, that what was learned from a previous crisis is not expressly discussed in

planning for future crises.

Mitroff (1994) developed a model that divides crisis management into five stages:

signal detection, probing and prevention, damage containment, recovery, and learning.

The segmentation of the crisis parallels Fink's (1986) discussion of the crisis lifecycle as

well as the three-stage model.

The first two stages signal detection and probing and prevention encompass the

proactive steps an organization can take before a crisis event. Signal detection identifies

the signs of possible crises within an organization. Signal detection is much like Fink's

(1986) prodromal stage. Probing and prevention though is not addressed in Fink's

model. This stage features members of an organization seeking known crises and

determining ways to prevent them. While Fink implies that crises can be prevented,

Mitroff's (1994) model actively tries to prevent crises (Coombs, 1999b). As with Fink's









prodromal stage, signal detection and probing and prevention exemplify the

characteristics of the precrisis stage.

The last three stages of Mitroff's (1994) model damage containment, recovery,

and learning feature slight variations from Fink's (1986) acute, chronic, and resolution

stages. Like Fink, Mitroff's stages discuss the trigger and containment of the crisis event,

the arduous task of returning to the pre-crisis norm, and the resolution of the crisis event.

Damage containment, like Fink's chronic stage, focuses on the steps taken following the

crisis event. A relationship can be made between damage containment and the crisis

stage of the three-stage model as they both involve actions taken in response to the event.

The differences between Fink and Mitroff s models are found in the recovery and

learning stages.

First, in the recovery stage Mitroff (1994) emphasizes the facilitation of the

organizational recovery whereas in the chronic stage, Fink (1986) states that

organizations recover at varying rates. Mitroff emphasizes opportunities to empower

crisis managers in a particular crisis event while Fink focuses only on the timeframe of

the recovery. As with damage containment, a relationship can be made between

Mitroff's recovery stage and the crisis stage of the three-stage model. As in the crisis

stage, during the recovery stage the organization works toward the eventual end of the

crisis.

The second difference is that Mitroff's (1994) model is cyclical. The learning stage

allows an organization to incorporate what it has learned from the crisis into its

organizational philosophy. Fink's model simply states that resolution occurs when the

crisis is no longer a concern with no mention of future applications. Mitroff s discussion









of crisis management as a cyclical process is significant. The learning stage is essential

to the three-stage model. Like Mitroff's five-stage model, the three-stage model is

cyclical and recognizes the importance of applying what an organization learns during a

crisis to future crisis events. The three-stage model and the five-stage model both

acknowledge that a failure to learn from a crisis can leave an organization susceptible to

the crisis again.

The three-stage model of crisis management is the most widely accepted model,

and thus, will serve as the backdrop for this study. Whereas a more specific model may

be relevant in follow-up studies, this initial study to explore crisis management practices

and identify the role of the contingency theory variables in crisis management will

benefit from the use of this widely accepted model.

The Contingency Theory of Accommodation

The contingency theory of accommodation in public relations is based on the

premise that there are a number of variables that could affect the steps a public relations

practitioner may take when managing a particular situation. Cancel, Cameron, Sallot,

and Mitrook (1997) identified 86 such variables that have the potential to influence the

decision making process, thus moving the stance of an organization along a continuum

ranging from pure advocacy to pure accommodation (see Figure 4).

Pure -------------------- Pure
Advocacy Accommodation

Figure 4: Advocacy-Accomodation Continuum; Adopted from Cancel et al. (1997)


The advocacy portion of the continuum refers to those organizations or groups who

strongly represent one side of an issue. Cancel et al. (1997) made the comparison to an









attorney's representation of a client. The inference is that an organization that aligns with

the advocacy end of the continuum will maintain its stance with minimal wavering.

An activist group is one such organization that could be found at the pure advocacy

end of the continuum. Activist groups are characterized by strong beliefs in a particular

issue; they use a variety of methods to convince others that their beliefs are correct.

Activists are also unlikely to waver unless the status of their cause is improved. L.

Grunig et al. (2002) defined an activist group as,

A group of two or more individuals who organize in order to influence another
public or publics through action that may include education, compromise,
persuasion, pressure tactics, or force. (p. 446)

The accommodation half of the continuum is quite different (Cancel et al., 1997).

It illustrates an organization's ability to work with publics in an attempt to build trust and

understanding. Depending on the role of the public relations practitioner,

accommodation may prove difficult. In most organizations the dominant coalition

determines the stance of the entire organization. If the organization is structured so that

public relations practitioners do not have access to the dominant coalition, practitioners

may be unable to convince the organization to change its stance for a single public. L.

Grunig et al. (2002) explained that public relations departments must have access to the

dominant coalition in order to contribute to strategic management and planning.

The choice to accommodate a public may involve a small change in policy or, in

some cases, a change in organizational character that may be necessary in a given

situation. Moreover, the resultant relationship from accommodation could prove valuable

in the event of a future dispute or crisis. It is at this meeting point on the continuum that

effective communication occurs.









Criticisms of Contingency Theory

J. Grunig and Hunt (1984) identified the four models of public relations: press

agentry, public information model, two-way asymmetrical, and two-way symmetrical.

These were the initial models developed for the field of public relations. As a result, the

models were the subject of criticism from public relations scholars. The contingency

theory is one of many theories that expand these models.

Of the four models, the two-way symmetrical model of communication has

received the greatest attention. Indeed, J. Grunig and L. Grunig (1992) stated that this

was both a descriptive and normative model. They believe that two-way communication

between an organization and its publics epitomizes the ideal practice of public relations.

J. Grunig (2001) stated that a dialogue must be initiated in order to determine the

stance of the individual public, "...neither side can really know the morality or

reasonableness of the other side's interests without talking with its representatives" (p.

15). The author further posits that accommodative techniques should only be used during

this conversation. Only upon a failed attempt to accommodate does Grunig explain,

"...the symmetrical approach suggests that advocacy of [one side's] interests or

withdrawal from the dialogue is ethically reasonable" (p. 16).

Two-way symmetrical communication, though, may not always be feasible as

Cancel et al. (1999) stated in paraphrasing Cancel et al. (1997) "a monolithic approval of

two-way symmetrical public relations is not sustainable given numerous publics-nor

likely to lead to effective public relations" (p. 178).

Thomas and Kilmann (1974) discussed the dimension of negotiation within the

Dual Concern Model as a "party's desire for own concern" and "party's desire to satisfy

other's concern." The balance of the Dual Concern Model is what Thomas and Kilmann









referred to as the compromise stage. In this stage, the organization's concerns the

points they advocate influence negotiation, and ultimately the level of accommodation.

Moreover, it is incorrect to assume that an organization can completely abandon their

stance when attempting to accommodate a public, especially as they remain accountable

to other internal and external publics.

More on Advocacy

The description of an activist group by L. Grunig et al. (2002) supports the very

notion of the contingency theory of accommodation. The authors established that an

activist group can consist of two or more individuals. This factor will greatly influence

how much a company accommodates the particular situation.

Consider the request of a small, unknown activist group to meet with a large

company to discuss possible changes in the company's environmental policy. Company

managers agreeing to meet with the group is unlikely, based solely on the activist group's

lack of stature. If managers of a large company were to meet with every activist group,

there would be little time to run the company. They would need an overwhelming reason

to hold the meeting. Secondly, if in fact the two individuals were granted the meeting,

their lack of prominence as an activist group would minimize the reverence they garner.

Though the organization has met with them, the individuals would need to be extremely

convincing to receive significant accommodation.

The reason is simple. In order to accommodate requests, the company may need to

make concessions that could be detrimental; profit margins could drop, jobs could be lost,

or business relationships could dissolve. A company has to weigh this effect on other

stakeholders and whether it will risk these negatives to fulfill the need of a small public.









Likewise, failure to accommodate could prove detrimental. The backlash from

maintaining an organizational position could be hurtful. If the concerns of a group are

reasonable and they are ignored, the repercussions could be damaging.

If it was determined that accommodating the activist group proved too costly, the

organization would remain at the advocacy end of the continuum, though the position

could shift. If before the meeting the organization was positioned at "pure advocacy," it

may shift slightly toward the accommodation end. This could be the result of agreeing to

future meetings or other small concessions. If no such concessions were made, the

organizational position would remain at pure advocacy.

More on Accommodation

The activist example serves to also explore the concept of accommodation. In the

previous scenario the company was unlikely to practice accommodation with the two-

person activist group based on the cost-benefit of doing so. Conversely, when the activist

is a large and very reputable group, a company will be more likely to consider

accommodation.

For instance, as stated on its Web site, the Sierra Club is America's oldest, largest

and most influential grassroots environmental organization. The organization boasts

700,000 members. By contrasting the Sierra Club with the two-person activist group that

wants to protect the environment, it is clear that the response would be very different.

If a company were approached to discuss an environmental policy, the clout of the

Sierra Club would increase the likelihood of the meeting. The company would have to

weigh the same factors it did for the two-person activist example, though in this case the

public is much different. The influence of an organization such as the Sierra Club is

large. The company must be aware that the numerous members could have a sufficient









impact on its product or service. Demonstrations, boycotts, and media coverage could all

affect the success of the company. The severity of these events could prove very costly.

In the previous example, the organization was unlikely to accommodate due to the

cost-benefit ratio. In this case though, the cost of not accommodating could be much

larger. The company would weigh these new variables to make a decision to

accommodate some of the Sierra Club's requests. In doing so, the company reduces the

risk of retribution, improves its relationship with the Sierra Club, and aligns itself with an

organization with a positive reputation.

Contingency Variables

The contingency theory of accommodation identified 86 variables that could

influence an organization's level of accommodation. Cancel, Mitrook, and Cameron

(1999) tested the theory and variables to determine those that were most relevant. In a

qualitative study, researchers interviewed 18 public relations practitioners to determine

"whether the contingency theory makes sense to them" (Cancel et al., 1999, p. 172).

Interview participants worked for nationwide corporations and each had between 20 and

40 years of experience.

The study's findings divided the variables into two major groups, predisposing and

situational variables. Each group consisted of variables that were found to be "highly

supported" by the practitioners (Cancel et al., 1999). Examination of these variables will

provide knowledge pertinent to ground crisis management in the contingency theory.

Predisposing variables

Cancel et al. (1999) define predisposing variables as "those variables which have

their greatest influence on an organization by helping to shape the organization's

predisposition towards relations with external publics" (p. 177). These variables are part









of the foundation of a company. They are steadfast characteristics that define what the

corporation stands for; they give the company a personality. In their study, Cancel et al.

(1999) found that, "predisposing variables influence location along the continuum before

the corporation enters into a particular situation involving an external public" (p. 190).

From the interviews, Cancel et al. (1999) identified five highly-supported-

predisposing variables: (a) corporation business exposure, (b) public relations access to

dominant coalition, (c) dominant coalition's decision power and enlightenment, (d)

corporate size, and (e) individual characteristics of involved persons. These variables

will be examined further in this study. Though future research exploring contingency

theory may examine all of the predisposing variables, this study will focus on only those

variables identified by Cancel et al. (1999) as highly-supported. By examining these

variables the researcher attempts to support previous findings. The goal is a structured

analysis of crisis management utilizing the framework of the contingency theory.

Corporation business exposure was cited as influential to relations with external

publics in 13 of the 18 interviews (Cancel et al., 1999). Researchers discovered that

participants whose company produces consumer or necessary goods needed to be very

responsive to their publics. Moreover, companies that produce consumer products were

also found to have greater exposure to publics than companies in the commercial or

industrial markets. Respondents believe that as a result of the higher exposure, these

companies' expectation to accommodate to publics was higher.

Public relations access to dominant coalition "enthusiastically" received support

in all interviews as influential of how organizations deal with their publics (Cancel et al.,

1999). Respondents explained that the dominant coalition had the final say in what was









allocated to publics. This supports a common theme explained by Grunig and Grunig

(1992) and a belief of practitioners and scholars alike: public relations must have a place

in the dominant coalition. Participants explained that they were used as an advisor to

management about opportunities to accommodate publics, but felt that in order to be

successful they required access. L. Grunig et al. (2002) also supports the findings of

Cancel et al. (1999) that the structure of the public relations department influences access

to the dominant coalition. The authors posit that when public relations is autonomous

from other departments, such as marketing, their access to the dominant coalition

increases allowing for a greater examination of public relations issues.

Dominant coalition's decision power and enlightenment takes the variable of

access to the dominant coalition one step further. This variable was cited most often by

interview subjects as influential (Cancel et al, 1999). Subjects agreed that the dominant

coalition must understand the importance and value of public relations to the

organization. Many practitioners explained that there were times they had to educate

management about their role and value.

Corporate size was cited by many subjects as causing high visibility (Cancel et al,

1999). As found in the variable of business exposure, high visibility can lead to the

expectation that a company will be accommodative. Also, respondents explained they

felt the size of their companies made them targets of activist groups. As discussed

earlier, special interest and activist groups can influence the accommodative practices of

a company. Subjects also stated that a large corporation could have so many diverse

publics that they cannot respond to all of them.









Individual characteristics of involved persons was mentioned by 15 of the 18

respondents as influential to relations with external publics (Cancel et al., 1999). The

individuals referred to were members of the dominant coalition, not public relations

managers. This variable also illustrates how important it is for public relations to be

included in the dominant coalition. Specific characteristics mentioned were open-

mindedness, the ability to screen out personal bias, and past training and education.

Situational variables

Situational variables, as defined by Cancel et al. (1999), "are the specific and often

changing dynamics at work during particular situations involving an organization and an

external public" (p. 177). These variables present a great challenge to crisis managers as

they appear only as the crisis presents itself. As the authors explained, these variables

impact an organization's position "as the situation plays out" (p. 177).

Thus, these variables appear during or immediately following a crisis and exert

their influence on the predisposing variables. The situational variables can largely impact

the position of a company along the contingency continuum. Whereas the company's

predisposing variables may place it on the advocacy end of the continuum, a powerful

situational variable could shift that position. For instance, an automobile company may

claim its new sport utility vehicle is safe. If an independent safety test shows the SUV

has the propensity to flip when drivers are forced to swerve at high speeds, the threat to

the company could result in a shift from advocacy of their safety stance to

accommodation of new safety recommendations.

Cancel et al. (1999) identified five situational variables as receiving high support

from respondents: (a) urgency of situation, (b) characteristics of external public's claims









or requests, (c) characteristics of external public, (d) potential or obvious threats, and (e)

potential cost or benefit for a corporation from choosing various stances.

Urgency of situation was cited in all interviews and strongly associated with

influencing the accommodation of an external public (Cancel et al, 1999). Many

respondents stated that this variable must be established soon after determining the

specific details of an incident. The degree of urgency in a given situation is based on a

number of factors; internal, external, and predisposing factors can influence the level of

urgency. These variable factors support the notion that not all situations are similar and

practitioners must be prepared to react differently in each situation.

Characteristics of external public's claims or requests were among the most

frequently mentioned variables (Cancel et al, 1999). Participants stated that the

perception of the external publics' claim or request influence how they proceeded. Truth

or falsity, as well as the existence of responsibility for the negative claim held an

important role in the decision to accommodate. In addition, the reasonability of a claim

also influenced accommodation. If a claim or request is reasonable, corporations were

found to be more accommodative.

Characteristics of external public refers to the corporation's interaction with

and perceptions of an external public. The characteristic most frequently noted in

interviews was how well the public could affect the company's ability to do business and

make money, positively or negatively (Cancel et al., 1999). Respondents also cited the

public's size, organization, and ability to garner media coverage as characteristics they

weigh. Indeed, the capabilities of external publics can influence the decision to

accommodate an external public.









Potential or obvious threats were cited in all interviews as a factor the

practitioners faced at one point (Cancel et al, 1999). A larger threat was associated with a

quicker response and higher likelihood to accommodate. The threat level can be

associated to a number of factors including the power of the external public. A powerful

public presents a greater threat to a company as it has greater resources and capabilities at

its disposal. Possible media coverage was the most frequently cited threat, though the

threat of economic loss was underlying to each threat. One practitioner explained that

impacting a company's ability to do business and make a profit is a quintessential piece

of a threat.

Potential cost or benefit is fundamental in business. In regard to the contingency

theory, cost-benefit is how a company's position along the continuum affects its costs and

benefits. For instance, if accommodating an external public will be a costly pursuit, the

company must determine if the benefits will counteract some of the expense. Benefits of

accommodation include improved public perception, alignment with a positive

organization, and the development of an advantageous relationship. Not surprisingly, the

subjects explained their company strives for a situation with the lowest cost and highest

benefit (Cancel et al., 1999).

The researcher believes that the highly-supported-predisposing and situational

variables will likely have a role in crisis management. In an attempt to illustrate the

relationship between these variables and the three-stage model of crisis management, four

research questions have been developed to guide the study:

* RQ 1: How will the predisposing variables emerge from the three-stage model of
crisis management?






25


* RQ2: How will the situational variables emerge from the three-stage model of
crisis management?

* RQ3: At what point in the model will the predisposing variables emerge from the
three-stage model of crisis management?

* RQ4: At what point in the model will the situational variables emerge from the
three-stage model of crisis management?














CHAPTER 3
METHODOLOGY

Review of Research Questions

Four research questions were developed to test the influence of contingency

variables on a crisis situation. The research questions will provide a groundwork that

demonstrates the use of contingency theory in crisis management practice. The study

first examined whether the contingency theory variables identified by Cancel et al. (1999)

as highly-supported emerged during crisis management. The study then attempted to

determine at what point those variables affected crisis management.

Questions one and two examine the role of the highly-supported-predisposing and

situational variables in a crisis. Questions three and four study at what point the highly-

supported-predisposing and situational variables influence crisis management.

* RQ 1: Will the predisposing variables emerge from the three-stage model of crisis
management?

* RQ2: Will the situational variables emerge from the three-stage model of crisis
management?

* RQ3: At what point in the model might the predisposing variables emerge from the
three-stage model of crisis management?

* RQ4: At what point in the model might the situational variables emerge from the
three-stage model of crisis management?

Data Collection

Data collection for this study used a modified Delphi technique. This qualitative

interactive research method allowed the researcher to gather a great deal of information

from individuals with exceptional knowledge about a subject (Clare, 1994). The









technique allowed the researcher to prompt participants with a question. Upon receipt of

all answers, the researcher posed a supporting question based on the initial findings for

the panel to respond. The process served as a way to structure group communication

(Turoff & Hiltz, 1996).

Beyond the receipt of individual feedback, the Delphi technique allowed

respondents to react to their colleagues while guaranteeing anonymity. Responses were

not immediately provided to all of the participants. During analysis of the responses to

determine a follow-up question, the researcher removed any information that may have

identified the respondent in the answer.

The Delphi system allowed for geographically distributed individuals to interact

while discussing their crisis management experiences. Moreover, this technique helped

remove negative dynamics that may inhibit communication in a group setting, such as

hostility toward other participants and domination of the discussion by a participant

(Clare, 1994; Wakefield, 2000).

A limitation of the Delphi technique is that participants needed to be reminded and

motivated to provide feedback. Whereas in interviews the researcher is present to ensure

respondent interaction, the Delphi method only allowed interaction along mediated

channels. In an attempt to maintain participation, the researcher sent periodic reminder e-

mails and made telephone calls to participants who did not respond to the question by the

requested deadline.

The questions examined the role that predisposing and situational variables have in

the crisis management process and attempted to determine other variables that influenced

the respondents' crisis response. Respondents were asked to discuss those variables









which most influenced their crisis management response. The researcher prompted the

respondents with questions that addressed the predisposing and situational variables that

the participants did not discuss. Variables mentioned by the subjects that were not

highly-supported by Cancel et al. (1999) were also documented.

Sample

The study used a purposive sample of communication experts from municipal

utility companies in the state of Florida. The municipal utilities companies recruited for

this study were electric companies that are community-owned and locally managed. The

effects of the very active, 2004 hurricane season on the state of Florida supports the belief

that these individuals have experience in natural disaster, crisis situations and related

communication functions.

Candidates for participation in this study included a cross-section of municipal

utility companies throughout the state of Florida. The study aimed to target local level,

senior members of the Florida municipal utility community. Contact information for

senior members at the various organizations was provided by Florida Municipal Energy

Association (FMEA) Executive Director, Barry Moline.

Following a phone call placed to Mr. Moline discussing the study, the researcher

was provided with six senior individuals at municipal utilities across the state of Florida

that were impacted by the 2004 hurricane season.

The six organizational contacts were sent an initial recruitment e-mail in early

January, 2005 (APPENDIX A). A follow-up telephone call was made to those

candidates who did not respond within one week. Initially, five of the six individuals

agreed to participate in the study the sixth candidate explained his organization was still

too busy with recovery efforts from the hurricane season to participate in the study.






29


Data Analysis

The findings were analyzed to determine the role of predisposing and situational

variables in crisis management. Throughout the research process data was analyzed in

order to generate follow-up questions for participants. The analysis followed the constant

comparative method. The individual responses were compared to reveal the common

themes presented by participants. Information that did not support the categorization of

the highly-supported-predisposing and situational variables was documented.














CHAPTER 4
RESULTS AND DISCUSSION

Data collection occurred over a six-week period. From the original five

participants, three remained at the end of the study. Two of the three participants

responded to all phases of the study. The third individual was unable to participate in

phase 1.1 but responded to each subsequent phase.

Of the two individuals who withdrew from participation, both expressed a lack of

time to dedicate to the study while in the first phase. One participant withdrew before

responding to the first primary question while the other withdrew after responding to the

first primary question. In each phase of the study, the researcher received responses from

three participants. Participants were informed to use the 2004 hurricane season or a

particular storm from that hurricane season that affected their organization as the basis of

their response.

The background of the participants was quite different. Two of the three

participants are employed by small city governments while the third worked for a utility

company servicing a large city. The individual who withdrew after answering the first

primary question worked for a utility company in a small city. Responses from each

participant are reflected in the analysis of each phase of the study.

Phase 1

Phase 1 of the study was developed to expose a possible relationship between the

highly-supported-predisposing variables identified by Cancel et al. (1999) and the

precrisis stage of the three-stage crisis management model. The highly-supported-









predisposing variables are corporate business exposure, public relations access to the

dominant coalition, the dominant coalition's decision power and enlightenment,

corporate size, and the individual characteristics of involved persons. Phase 1 provided

data that allows the researcher to answer the first and third research questions of this

study:

* RQ 1: How will the predisposing variables emerge from the three-stage model of
crisis management?

* RQ3: At what point in the model will the predisposing variables emerge from the
three-stage model of crisis management?

The data suggest that the predisposing variables do influence crisis management.

Each respondent addressed the role of the variables, to some degree, in their organization.

Discussion centered on how the predisposing variables influenced the preparatory

actions of participant organizations and the established role of participants in

organizational decision-making. Moreover, there was a relationship between participant

responses regarding the predisposing variables and the precrisis stage of the three-stage

model. The precrisis stage addresses the proactive steps an organization takes to avoid a

crisis. Likewise, in phase 1 the respondents discussed the existence of the predisposing

variables prior to the crisis event.

Phase 1.1

The first question was meant to determine what precrisis infrastructure existed at

each respondent's organization. The phase 1.1 question read: "What preparatory steps

has your organization taken in an attempt to avoid a crisis event?"

The question was purposefully vague as to not influence the responses of

participants, but rather to gauge the relevance of varying facets of their crisis response.

The purpose was for those items they viewed as most important to surface in the









responses. Participants seeking further clarification were provided an addendum to phase

1.1 asking the respondents to discuss the characteristics of their organization that shape

its crisis response.

Phase 1.1 responses dealt with the proactive steps taken for a crisis event.

Preparation of various crisis plans and manuals had been completed by all respondents.

Participants 1 and 3 discussed the impact of the September 11, 2001 attacks on their crisis

preparation. Each participant stated that since the attacks there has been a new emphasis

on the security of their facilities. Participants 1 and 2 discussed the use of government

sanctioned training or meetings to support crisis preparation. All three participants

discussed the importance of public safety.

The highly-supported-predisposing variables played only a minor role in phase 1.1.

However, one participant did address two of them in their response. Participant 3

discussed the variables corporate size and access to the dominant coalition. In this

industry corporate size is determined by the number of clients serviced a greater

demand for power necessitates a larger organization. This participant stated that only

after power outages dropped below a "reasonable number" did twice daily briefings

cease. The reasonable number of outages was 40,000. This figure is a reflection of the

company's size as it can service up to this number of outages without the implementation

of increased protocols.

This supports the findings of Cancel et al. (1999) that the size of an organization

influences its ability to accommodate its publics. This large utility company was able to

accommodate a large number of outages before enacting new protocols. A smaller

organization would not be able to handle such a large number of outages with the same









ease. Whereas the large utility may not deviate from normal protocol, the smaller

organization may have to alter its communication response and community information

efforts. Participant 2 is a member of a seven person team that oversees power

distribution activities (Appendix G). For a workforce of this size, implementation of

special operations would have likely occurred well before 40,000 outages. Therefore,

there is support that corporate size influences the precrisis stage of the three-stage model.

Additionally, access to the dominant coalition was addressed briefly in phase 1.1.

Participant 3 explained that following the massive outages, twice daily briefings were

held to address the situation. Many departments, including communications, were

represented in the meetings

[The meetings] involved the CEO, the Electric, Water and Sewer Operations
Department, Internal and External Communications, Public Outreach, Facilities
Maintenance, Client Relationships, Security, Finance, Risk Management, Safety,
Engineering, Technology Services, and Governmental Relations. All vice
presidents on the executive management team were at the meetings, along with
many of the field operations supervisors to ensure direct information flow from the
field. (Appendix F)

Public relations practitioners in this organization internal and external

communication, public outreach, client relationships, and government relations all had

access to the dominant coalition. Though it is unclear whether this access is permitted

when the organization is not facing a crisis, the inclusion of public relations in

discussions about crises identifies the active role the department has within the

organization. The amount of access public relations has to the dominant coalition in this

organization supports the study of Cancel et al. (1999) as well as the variable's influence

on the crisis stage of the three-stage model. The access of public relations and the size of

the organization are discussed further in the analysis of phase 1.2.









Phase 1.2

Phase 1.2 delved into the participant's role in the organization. The researcher

wanted to learn how their ideas were incorporated into the organization's decision-

making process. Additionally, the question attempted to determine the other sources of

information the dominant coalition recognized. The phase 1.2 question read: "How has

your input been incorporated into your organization's current crisis management

protocols? Who else's input, internal or external, was taken into account?"

This question attempted to elicit responses addressing multiple variables identified

in the contingency theory. As mentioned, organizational size was found to be an

influential variable in phase 1.2. Unlike the phase 1.1 response, participant 2 is from "a

very small organization" (Appendix G).

The participant explained one objective the organization recognized was the need

to educate the public about crisis situations. The small size of the organization

necessitated support from local agencies police, fire, and emergency management to

successfully educate the public. Without external support the smaller organization may

not have been able to address the needs of their public. As in phase 1.1, the predisposing

variable of corporate size identified by Cancel et al. (1999) is supported. The

organization's small size influenced its ability to provide publics with important

information in the precrisis stage.

Moreover, the size of the organization also influenced access to the dominant

coalition. Participant 2 explained that the department that handled power-related issues

consisted of seven individuals, "1 superintendent, 3 linemen, 2 apprentices, plus me"

(Appendix G). As a result of the small department, input from all individuals including









the respondent who frequently serves as city spokesperson for public services was

included in the decision-making process.

Lastly, this organization also supported the notion that characteristics of involved

individuals influence the precrisis stage. Respondent 2 explained that although

everyone's suggestions were encouraged, the more experienced individuals were able to

provide greater insight. It is likely that an individual in a small department with more

experience will influence the choices made in crisis preparation. This experience relates

to discussion in Cancel et al. (1999) of past training or education as a characteristic

frequently mentioned by respondents, as influential to the organization's accommodation

of publics.

Phase 1 Conclusion

Phase 1 identified support for the predisposing variables of corporate size, public

relations access to the dominant coalition, and individual characteristics of involved

persons. There was only marginal support for the variables of dominant coalition

enlightenment and corporate business exposure.

In phases 1.1 and 1.2 respondents addressed the importance of public safety,

government training courses, and resource availability as well as highly-supported-

predisposing variables cited by Cancel et al. (1999). In order for these opportunities to be

available for public relations practitioners, it is likely that both funding and support of the

dominant coalition were achieved. This likelihood provides marginal support for

dominant coalition enlightenment of the role of public relations. Additionally, corporate

business exposure was not directly mentioned as influential to crisis preparation, though

public access to the organization and visibility in the media were addressed once the

crisis occurred and will be discussed in the analysis of phase 2.









Phase 1 also included some support for a contingency variables not highly-

supported in Cancel et al. (1999). The role of legal counsel in crisis preparation was

addressed by one respondent. In phase 1.1 the respondent explained that full-time

general counsel was hired in 2002 to aid in crisis preparedness. This individual's role

was to assist in the proactive identification of organizational vulnerabilities. This was the

only mention of legal counsel throughout the study.

Phase 2

The second phase of the study transitioned to examine the relationship between the

highly-supported-situational variables identified by Cancel et al. (1999) and the crisis

stage of the three-stage model. The highly-supported-situational variables are: urgency

of the situation, characteristics of external public's claims or requests, characteristics of

external publics, potential or obvious threats, and the potential cost or benefit of a

particular action. The data collected in phase 2 provided the researcher with information

to answer the second and fourth research questions of this study:

* RQ2: How will the situational variables emerge from the three-stage model of
crisis management?

* RQ4: At what point in the model will the situational variables emerge from the
three-stage model of crisis management?

In this phase, the data illustrated the role of the situational variables in crisis

management. Participant responses focused on the existence of the situational variables

relative to the crisis event the discussion addressed the organizational response to the

crisis as well as the factors that influenced the response. In many cases, the situational

variables provided insight as to why the organizations responded in a particular manner.

Participants often addressed their organization's response to the crisis when commenting

on the role of the situational variables. The discussion of the variables' influence









concentrated on the crisis stage of three-stage model. As a result, the researcher can infer

that the situational variables exert influence in the crisis stage of the model.

Phase 2.1

The question in phase 2.1 identified the variables that influenced organizational

response to the 2004 hurricane season referred to as the crisis event. The phase 2.1

question read: "What characteristics of the crisis event influence your organization's

crisis response?"

The characteristics of external publics' claims and requests was an influential

variable in phase 2.1. Respondents 2 and 3 explained that their organizations received a

lot of pressure from publics to restore power. Respondent 3 explained that with more

than half of their customers without power, the company received as many calls in one

day as they typically do in a month. Considering the number of phone calls the

organization had to modify how they communicated with publics and the media. The

organization's "blast fax" to the media and members of the city council helped the

organization distribute information to the public (Appendix H). Thus, the number of

requests for service influenced this organization's response in the crisis stage of the three-

stage model.

The variable of corporate size was revisited in conjunction with the claims and

requests of publics. Participant 3 explained that once outages were reduced to a

manageable amount the use of blast faxes ceased and the organization returned to its

normal communicative practices. As addressed in phase 1, the number of manageable

outages was 40,000. A smaller organization may not consider that number of outages

low enough to resume normal protocols. Respondent 4 works for a much smaller

organization and explained that due to its size, the organization was forced to work









closely with the county emergency management team to restore power. The crisis stage

of the three-stage model concludes when the crisis ends. Though a reduction to 40,000

outages did not end the crisis, an organization's size and ability to handle a large volume

of outages will expedite the crisis stage.

Public outcry supports the existence of another situational variable's role in the

crisis stage of the three-stage model potential cost or benefit. Again, respondents 2 and

3 explained that as a result of the massive power outages their organization was forced to

determine the best way to restore power. Each respondent explained that they organized

repairs from the power source and major power lines to the neighborhoods in order to

most efficiently restore power to the most people. This sequence did not follow standard

procedure in fact it was the exact opposite in one case. The organization determined

the cost of following standard protocol was too high it would have taken too long to

restore power and that a different approach produced a greater benefit in the situation.

Additionally, respondent 2 briefly discussed that there was great importance placed

on restoration of power to critical facilities. The need to restore power to these facilities

quickly relates with the variables of urgency of the situation and potential and obvious

threats. Medical facilities need power to provide treatment for patients. Without power,

patients may not be able to receive care in some cases urgent, lifesaving care.

Moreover, it is likely that these facilities received an increase in patients as a result of the

storm, thus power restoration was essential. These variables were further addressed by

respondents in phase 2.2.

Phase 2.2

Phase 2.2 further examined a point raised in phase 2.1. As a result of the power

outages many publics contacted or attempted to contact their utility provider. This phase









asked if any particular public contacted the organization more than others. The phase 2.2

question read, "Were certain publics more vocal regarding their need for power

restoration? Did their requests influence your organization's response?"

This question directly addressed the situational variable of characteristic of an

external public. The responses to this question varied, as each respondent had a different

experience with the distinguishing characteristics. Participant 2 explained the most vocal

group in the community belonged to a more affluent, gated community who felt they

were not receiving the service they deserved. The respondent explained that only rarely

did any calls influence power restoration procedure. Instead, the organization prioritized

restoration from important community services down to individual members of the

community. There were a few instances where external publics affected restoration.

These situations were limited to unknown critical needs facilities and "quick easy fixes

that would restore power to many customers" (Appendix I). Thus, certain characteristics

of external publics and the cost-benefit of "quick fixes" influenced this organization's

response in the crisis stage.

Respondent 3 stated that parents became most vocal in the days following a power

outage. Parents with jobs had to stay home to watch their children or were forced to pay

for daycare due to school closures. Additionally, some parents contacted city council

members and the mayor in an attempt to expedite school power restoration. The pressure

from this public placed more focus on school restoration. Support is again found for the

Cancel et al. (1999) variable, characteristics of external publics. Outcry for a focus of

restoration efforts at school proved successful in shifting focus to that task.









Participant 4 explained that calls requesting power came from all members of the

community. The first groups to have power restored were the police, fire, and emergency

services followed by downtown businesses and medical facilities. Again, characteristics

of the publics influence the crisis response. Those services and facilities that were

deemed most important to receive power restoration were addressed first. In reference to

the general public, the respondent claimed that no group influenced their response.

Instead, power restoration crews were dispatched in order of request.

Phase 2 Conclusion

Phase 2 identified much support for the Cancel et al. (1999) highly-supported-

situational variables each variable was addressed by respondents. The variable most

supported was characteristics of external publics. Each respondent explained that certain

publics medical facilities, schools, or emergency services received a greater amount

of organizational resources. The variable which received the least support was potential

or obvious threats. The only relationship with this variable can be with the threat of not

immediately restoring power to medical facilities or emergency services.

Additionally, the applications of lessons learned during this crisis were briefly

discussed in phase 2. Participant 2 explained that his organization inserts[] lessons

learned as needed to our preparations" (Appendix H). Participant 3 discussed how his

organization developede] a contact/location list for all schools we serve" (Appendix I).

These two tasks support the premise of the three-stage model that crisis management is

cyclical. Lessons learned during one crisis should be included in future crisis

management planning and preparation so as to avoid or reduce the risk of another crisis.

The application of lessons learned during a crisis will be discussed further in the analysis

of phase 3.









Phase 3

The third phase of this study did not serve to relate the contingency theory with the

three-stage model. Instead, this phase examined whether the participants' crisis

management practices are cyclical. The three-stage model is a cyclical model of crisis

management it explains that what an organization learns during a crisis should be

applied to future crisis events. The researcher wanted to probe whether or not

participants practiced cyclical crisis management and how it affected their attitudes

regarding possible future crises. Unexpectedly, the data presented a reemergence of the

predisposing variables. As a result, phase three of the study also, like phase one, served

to answer the first and third research questions.

Phase 3.1

The phase 3.1 question examined what influence the 2004 hurricane season had on

the future of the organization. The question read: "How did the 2004 hurricane season

influence your organization's preparation for future crises?"

Respondents discussed a variety of topics. Respondent 2 emphasized the

importance of the experience his organization gained. The participant expressed that this

experience would reduce the need for future "reinvention" of techniques and policies

(Appendix J). Moreover, the individual explained the role interaction with other

organizations played in their learning. This point was echoed by participant 4 who

claimed their organization "had a new understanding" of what they must prepare for in

the future (Appendix J). Respondent 3 listed the major changes this crisis prompted in

the organization. The list included changes in protocol, communication techniques, staff

assignments, and the receipt of community feedback.









The experience gained by the respondents from the crisis event can be valuable in

future crises. As a result, the highly-supported-predisposing variable, characteristics of

involved persons, reemerged. The first-hand experience familiarized them with crisis

management practices. The knowledge gained from this crisis event has helped prepare

these individuals for future crises. Whereas before they may have never dealt with a

crisis, the 2004 hurricane season likely made them more aware of such a situation and

provided them with experience to react to future events.

Phase 3.2

Phase 3.2 asked if the crisis and what the organization learned from the crisis has

prepared them for similar events in the future as well as what concerns still exist. The

phase 3.2 questions read: "Considering what your organization learned from the 2004

hurricane season, do you feel prepared for the 2005 hurricane season? What concerns

does your organization have?"

The responses to this question were mixed. Participant 3, from a large utility

company, explained that the experience has prepared the organization for future events.

The respondent explained that the first-hand experience was much more valuable than

any drill or exercise in teaching the organization how to cope with such situations. In

contrast, the participant did express concern if the community and organization were to

take a direct hit from a storm. The participant stated that the additional damage and

outages caused by a direct hit are of great concern to the company and the community.

Respondent 2 had quite the opposite response regarding his organization's

preparation. This individual explained that his organization does not feel completely

prepared for a hurricane season. The participant cited the impact of the cost of the 2004

season on the small utility as too great to overcome in one year. As a result, this









organization will triage their needs as the 2005 hurricane season begins. This response

marks the reemergence of the highly-supported-predisposing variable corporate size and

illustrates how the variable will impact this organization's ability to manage a future

crisis event.

Participant 4 explained that although an organization can prepare, the 2004

hurricane season was something for which you could not prepare. This individual stated

that, "you may think you are prepared, but when you have three successive hurricanes

wreak havoc on your community in six weeks, you can't get that prepared" (Appendix

K). As a result of the 2004 hurricanes, this participant feels that hurricane season will

always concern her community.

Phase 3 Conclusions

From the phase three responses, the researcher can conclude that each respondent

has learned from the experience of the 2004 hurricane season and will apply those lessons

to future crisis management practices. The respondents, therefore, did use a cyclical

model of crisis management a vital aspect of the three-stage model. The knowledge

gained and resulting changes from the 2004 hurricane season will only increase the

ability of these organizations to prepare for crisis events as well as reduce the impact of

the crises.

Additionally, predisposing variables emerged in phase three responses. These

variables signal the creation of the new precrisis stage that occurs when new information,

gained during a crisis, is incorporated into existing precrisis protocol. The outcome of

this synthesis is a new precrisis norm, exhibiting highly-supported-predisposing

variables.














CHAPTER 5
LIMITATIONS AND FUTURE STUDY

This qualitative study does not allow the researcher to make generalizations into

the relationship between the contingency theory of accommodation and crisis

management practices. Moreover, the small number of participants may slightly detract

from the relevance of the study. This study provides a springboard for further study.

Future studies that attempt to relate the contingency theory and crisis management

practices should soon transition from a qualitative approach to a quantitative

methodology.

As a follow-up to this study scholars should use depth interviews to collect data in

an attempt to relate the highly-supported variables identified by Cancel et al. (1999) to

crisis management. The Delphi approach used in this study proved problematic at times

in recruitment, sustained participation, and depth of information. Though this study did

provide purposeful data, an interview format would allow the researcher the opportunity

to probe each respondent further and receive greater clarification on the role of the

highly-supported variables. As a result of the Delphi methodology's limited opportunity

for follow-up questions, there was likely valuable information that the respondents did

not convey. The depth interview will provide the wealth of knowledge that can precede a

quantitative study.

Upon analysis of detailed qualitative data that supports a relationship between the

contingency theory and crisis management, a quantitative study may generate statistically






45


relevant data. From this data, contingency theory scholars can provide crisis

management practitioners with a theoretical grounding of their practice.














CHAPTER 6
SUMMARY AND CONCLUSIONS

Crisis management literature is rich with models that various organizations can

follow to help minimize the impact of a crisis. Each model has its individual merits, but

the three-stage model of crisis management is the most widely accepted model (Coombs,

1999b; Seeger et al., 2003). The acceptance of the model is the result of its overarching

applicability to a variety of organizational crisis situations. The three-stage model

divides a crisis into macrostages precrisis, crisis, and postcrisis which allows for

analysis of the event in a logical flow.

Data collection in this study was modeled after the three-stage model of crisis

management. Participants responded to a series of questions which that followed the

lifecycle of the crisis. The respondents, who were public relations and corporate

communications practitioners at municipal utility companies in Florida, discussed their

organization's crisis management practices during the 2004 hurricane season. Participant

responses were analyzed with those variables that influenced the crisis management

process categorized according to the three-stage model of crisis management.

The first phase of the study addressed the precrisis stage of the three-stage model.

This stage is characterized by the proactive steps taken as crisis preparation of an

organization the proactive steps (Coombs, 1999b). Respondents addressed a number of

variables that influenced their crisis preparation, including facility security standards,

protocol changes resulting from the 9/11 attacks, and information gained from









organizational crisis management audits. Of the variables influencing this stage, the size

of the organization revealed itself to be the most apparent influence cited by respondents.

The size of the participants' organization emerged as highly influential to their

crisis preparation. The smaller organizations did not have the benefit of resources that

available to larger utilities had. Moreover, the protocols that smaller organizations did

have in place were much less effective as a result of resulting from a smaller staff, a less

an inadequately equipped response network, and a much smaller operating budget. It was

found that although a large organization can sustain a large number of outages (40,000), a

smaller organization does not have the resources that allow it to effectively address such

an outage.

The variable of corporate size was addressed by Cancel et al. (1997) in early

contingency theory work. The researchers identified this variable as influential in the

accommodation of publics. Moreover, corporate size was identified by Cancel et al.

(1999) as a highly-supported-predisposing variable in further development and testing of

the contingency theory. The discussion of this variable was the first in a series of highly-

supported contingency theory variables that emerged during data collection. The

emergence of this variable This served as the first indicator of a potential relationship

between contingency theory variables and the three-stage model of crisis management

that appeared in the data. The contingency theory variables start to emerge emerged as a

way to understand the three three-stage model of crisis management. This proposition is

supported by the presence of additional highly-supported contingency variables in

participant responses regarding their crisis management practices.









Additional highly-supported-predisposing variables that emerged during the

precrisis portion of data collection include public relations' access to the dominant

coalition and the characteristics of involved persons. Participants explained that

members of the public relations function of the organization were in place prior to the

crisis event. Also, the predisposing variable of employment tenure within the

organization emerged characteristics of involved persons emerged as participants

explained that the input of the more tenured members of the organization was very

valuable as a result of their greater insight into the crisis.

The appearance of these variables in the data further supports the notion that there

is some a relationship between the highly-supported-predisposing variables of

contingency theory and the precrisis stage of the three-stage model of crisis management.

Moreover, the predisposing variables of dominant coalition power and enlightenment and

organizational business exposure were also addressed by participants. Although these

variables emerged to a lesser extent, their existence in the data provides supplementary

support to a relationship between crisis management and the variables identified in the

contingency theory.

The second phase of the study shifted question the focus to the crisis stage of the

three-stage model of crisis management. The crisis phase is characterized by the crisis

event and its lasting effects, and continues until the crisis has ended. Participants

discussed a number of variables that influenced how their organization responded to a

crisis. As was found in the precrisis question responses, highly-supported-predisposing

variables from the contingency theory also emerged from in responses to crisis crisis-

phase questioning.









In the crisis phase responses, highly-supported-situational variables identified by

Cancel et al. (1999) emerged. The variables characteristics of external publics, the

characteristics of external publics' requests, the cost-benefit of particular actions, the

urgency of the situation, and potential or obvious threats were all addressed by

respondents. Of these variables, the characteristics of external publics appeared most

often. Each participant provided an example of how a particular public influenced their

organization's crisis response.

The appearance of highly-supported-situational variables in the crisis phase

responses, in addition to the emergence of the highly-supported-predisposing variables in

the precrisis responses, strengthens the likelihood of a relationship between the

contingency theory variables and the three-stage model of crisis management.

Additionally, the appearance of the predisposing variables followed by the situational

variables mimics the classic progression of the crisis.

The last phase of questions addressed the postcrisis stage of the three-stage model.

The postcrisis stage begins once the crisis and its effects have has ended and posits that

organizations learn from the crisis in order to apply that knowledge to future crisis

preparation. A fluid transition from the postcrisis stage to the precrisis stage occurs in the

three-stage model of crisis management the new postcrisis knowledge is synthesized

with existing knowledge to become the new precrisis norm. The responses to the

postcrisis phase questions examined whether the organizations practiced cyclical crisis

management, if that is, whether the synthesis of new information into existing protocols

occurred.









Each participant explained that the crisis event allowed their organization to learn

and prepare for future crises. Respondents explained that the first-hand knowledge they

received was valuable and that the knowledge they gained allowed them to update crisis

protocols. Additionally, highly-supported-predisposing variables from the precrisis stage

of the model reemerged in question responses to the postcrisis stage. The predisposing

variables of corporate size, characteristics of involved persons, and organizational

business exposure during a crisis each appeared were present in participant responses.

The emergence of the predisposing variables in the postcrisis stage parallels the accepted

principle that effective crisis management is a cyclical process.

The study identified a potential relationship between crisis management and the

highly-supported contingency theory variables identified by Cancel et al. (1999). Figure

5 illustrates the contingency theory variables identified by respondents overlaid on the

three-stage model of crisis management. Although this study cannot generate a statistical

correlation between the three-stage model of crisis management and the highly-supported

contingency variables identified by Cancel et al. (1999), it does identify that a potential

relationship between crisis management and these variables exists. Figure 5 illustrates

the emergence of contingency theory variables from the three-stage model of crisis

management.












SSynthesis of Knowledge Gained from Crisis into Precrisis Norm



Precrisis Stage Crisis Stage Postcrisis Stage
(Emergence of (Emergence of (Re-Emergence of
Predisposing Variables) Situational Variables) Predisposing Variables)

Supported: Supported: Supported:
Corporate Size Characteristics of External Corporate Size
Public Relations' Access to Public Individual
Dominant Coalition Urgency of Situation Characteristics of
Individual Characteristics Characteristics of External Involved Persons
of Involved Persons Public's Claim or Request Business Exposure
Potential Cost or Benefit
Some Support: Potential or Obvious
Business Exposure Threats
Dominant Coalition's
Decision Power and
Enlightenment


Figure 5: Contingency Variables in Crisis Management

Although this study cannot generate a statistical correlation between the three-stage

model of crisis management and the highly-supported contingency theory variables,

future studies that examine the three-stage model of crisis management may verify the

principles that emerged in this study's findings. Additional qualitative research can

provide both scholars and practitioners with a greater understanding of the relationship

between crisis management and the contingency theory's highly-supported-predisposing

and situational variables can be achieved through additional qualitative research. Future

quantitative studies may be able to generate a statistically relevant correlation between

model and theory, thereby illustrating how the three-stage model of crisis management

uses the contingency theory variables to shape crisis preparation, response, and learning,

may emerge from future qualitative studies of crisis management and contingency theory

variables.






52


This researcher feels that there is a concrete link between the contingency theory of

accommodation and crisis management. The highly-supported variables identified by

Cancel et al. (1999) act only as a starting point for future research. This study has

generated support for this belief but it has only begun to uncover the relationship between

crisis management and the contingency theory variables. The opportunity to identify and

define this relationship between theory and practice serves as an avenue for future public

relations scholarship.














APPENDIX A
E-MAIL RECRUITMENT LETTER

Dear (Participant Name):

I am a master's student in public relations at the University of Florida and am conducting
my thesis on the crisis management process in Florida municipal utility companies.

The 2004 hurricane season was a blow to our state that I would like to study and learn
how members of your industry reacted. As a dedicated member of this industry that
serves all Floridians I was hoping you would be interested in participating in my study. I
have discussed my study with FMEA Executive Director Barry Moline to help ensure
meaningful findings.

The study is based on the Delphi technique. Participants will be sent questions regarding
their crisis management response. Prior to the study, participants will be asked a series of
questions to generate a profile of each respondent. Upon completion of the profiles, the
study will then begin its three phases, each phase consisting of two questions. In each
phase an initial question will be sent to participants followed by a second question based
on the collective response of the participants.

The duration of the study will not exceed six weeks and will not require more than three
hours of participation over that time.

The identity of participants will remain confidential throughout the process. If upon
completion of the study you would like a copy of the findings I would be happy to
provide them to you.

Recruitment of participants will be completed no later than January 10, 2005. Following
the recruitment process you will receive the respondent profile questions and the first
phase of the study. From the receipt of each phase of questions you will have one week
to respond.

If you would like to participate in this study please sign the attached informed consent
form and fax it to my research supervisor, Dr. Michael A. Mitrook, at (352) 392-3952. If
you cannot open the attachment or have any questions please contact me at
bboudreaux@jou.ufl.edu or (352) 262-1086. Thank you for your consideration.

Sincerely,

Brian Boudreaux














APPENDIX B
BIOGRAPHICAL INFORMATION

Participant #1

1. How many years have you been involved in corporate communication? In what
capacities have you done corporate communication?

I have 20 years of corporate communications experience, including corporate,
government and nonprofit. In addition, I worked as a television new producer at
Orlando's CBS affiliate.

2. In your career have you ever been faced with a crisis situation? If so, what was your
role in the crisis response?

Yes, I have served as primary spokesman through numerous utility crises,
including two Amtrak train derailments, wildfires, terrorist threats and a customer
service robbery.

3. What is your current title?

Manager of Corporate Communications.


Participant #2

I am an engineer. I am involved in corporate communications only as a necessary
function of my job. I am not formally trained in communications. My previous employer
(GRU) offered communications training classes which I attended.

My current employer [City Name] has not corporate communications personnel. As the
Director of Public Services, I am often required to be the City's spokesperson on Public
Services related issues.

I have been involved in numerous crisis situations including last year's hurricanes which
affected our area.

My current position is Director of Public Services for [City Name].









Participant #3

1. How many years have you been involved in corporate communication? In what
capacities have you done corporate communication?

18 years Media Relations Representation, Media Relations Manager, Public
Relations Manager, and as a private consultant for seven years.

2. In your career have you ever been faced with a crisis situation? If so, what was your
role in the crisis response?

Yes. Typcially handling media relations and as an agency/organization's
spokesperson with the media.

3. What is your current title?

Media Relations Coordinator

Participant #4

1. How many years have you been involved in corporate communication? In what
capacities have you done corporate communication?

I have been employed for 23 years with the City of [City Name]. I started as
Secretary to the City Manager in 1981 and became Deputy City Clerk in 1994.
Our City is unique to just a dozen or so cities in Florida where the City Manager
is also the City Clerk.

Each City Manager has been different regarding how much communication comes
from the City Clerk.

2. In your career have you ever been faced with a crisis situation? If so, what was your
role in the crisis response?

When Hurricane Charley blew through our town on Friday, August 13th, myself
and two other city hall employees manned the darkened, power-less city hall on
the following Monday, handling questions from citizens, calls from out of
owners who wanted to assist in any way possible, vendors who wanted to come
to our aid, emergency workers who wanted to know where to go and taking utility
payments from our customers. Only the phones were worked inside city hall that
day.. .power was off for days around our city. Decisions had to be made on how
we could help.

3. What is your current title?


* Deputy City Clerk









Participant #5

1. How many years have you been involved in corporate communication? In what
capacities have you done corporate communication?

19 years all with municipal utilities: Memphis Light, Gas and Water Division,
Ft. Pierce Utilities Authority and [Current Employer]. I have been a
communications specialist, senior communications specialist, communications
coordinator, manager and now, director which is part of our senior management
team.

2. In your career have you ever been faced with a crisis situation? If so, what was your
role in the crisis response?

Yes, I oversee the communications effort for these. Hurricane Jeanne and Frances
are the largest. I also participated in efforts during ice storms, gas explosions and
electrocution.

3. What is your current title?


* Marketing and Communications Director














APPENDIX C
PRIMARY QUESTION FORMAT

Dear (Participant Name):

Thank you for your response to the participant profile e-mail.

This e-mail will begin the first phase of the study. The primary question is as follows:

What preparatory steps has your organization taken in an attempt to avoid a crisis
event?

Please be as specific as possible when describing the safeguards your organization
employs. Feel free to draw on experiences in your current position as well as identify
incidents from throughout your career.

There is no minimum or maximum length requirement.

Please respond no later than Tuesday, January 18th. Once all responses have been
received a compilation will be sent to you for further comment.


Sincerely,

Brian Boudreaux















APPENDIX D
FOLLOW-UP QUESTION FORMAT

Dear (Participant Name):

Thank you for your response to phase 1.1 of this study. Attached to this e-mail is a
compilation of responses from all current participants. Please review the responses of
your colleagues before answering:

How has your input been incorporated into your organization's current crisis
management protocols? Who else's input, internal or external, was taken into
account?

Please be as specific as possible in your answer. There is no minimum or maximum
length requirement to your answer. Additionally, feel free to comment on any points
raised by your colleagues.

Please respond with your answer no later than Wednesday, January 26th


Sincerely,

Brian Boudreaux














APPENDIX E
SAMPLE REMINDER E-MAIL

Dear (PARTICIPANT NAME):

This e-mail is to remind you that your response to phase 2.1 is requested by Wednesday,
February 2nd.

Once I have received answers from all members of the study I will distribute the
responses for further comment. As always, no names of individuals or companies will
appear in the responses to maintain anonymity.

If you need another copy of the question or have any other questions please contact via e-
mail (bboudreaux@jou.ufl.edu) or call me at (352) 262-1086.

Sincerely,

Brian Boudreaux














APPENDIX F
PHASE 1.1 QUESTION AND RESPONSES

Phase 1.1:
What preparatory steps has your organization taken in an attempt to avoid a crisis
event?

Addendum
What characteristics of your organization influence your crisis management
practice?



Participant #1

[Company Name] continually looks for vulnerabilities within its organization... whether
they involve infrastructure, employees and/or risk management issues. [Company Name]
hired a full-time general counsel in 2002 to assist in this effort.

From a planning perspective, [Company Name] conducts an annual disaster drill that
involves all employees of the utility. The purpose of the drill is to test emergency
equipment, measure response time, and evaluate communication within the utility. Past
drills have included a wide range of activities, including a simulation of cyber terrorism,
a hazardous chemical spill, power line and substation damage, repair of utility vehicles in
the field, power outages, production of sandbags at both power plants, and the testing of
all emergency generators. During the drill, an evaluation team also throws out unexpected
scenarios to test the readiness of utility employees.

[Company Name] has sent employees to several FEMA-sponsored training courses on
disasters.

Since September 11, [Company Name] has upgraded perimeter security at its substations
and power plants as well as building security at its service center and administrative
offices. I.D. badges are now worn daily by all employees. [Company Name] has also
purchased and installed surveillance cameras, electronic perimeter fence sensors, and
proximity card readers on all exterior doors.

[Company Name] produces an annual safety calendar with safety tips to educate children
on the dangers of coming into contact with electrical equipment. This is to prevent injury
or death caused by electrocution. In addition, signage has been installed at all substations
providing notice of high voltage.









In 2003, [Company Name] installed portable defibrillators for public use at all of its
facilities in [local] County. The defibrillator is a lifesaving device for individuals
suffering from cardiac arrest or other medical emergencies.

In 2003, [Company Name] Board of Directors approved the purchase of terrorism
insurance to protect the utility's $360 million infrastructure from a potential attack.

Participant #2

Develop a written Vulnerability Assessment Plan.
Develop a written Emergency Response Plan.
Meet periodically with staff with the objective of identifying potential areas
susceptible to impact. We then attempt to eliminate or minimize the exposed area.
If that is impractical, then we develop a plan of action to address restoration in the
event impact occurs.
Meet periodically with a statewide association of other electric utilities to do the
same. We also discuss mutual aid (assisting other utilities that experience
impacts).
Review crisis events experienced by other utilities to learn what works and what
does not. This also helps you consider all the possible events that occur.
Meet periodically with local government emergency response team to discuss
emergency response issues.
Maintain an emergency preparedness manual with emergency contact numbers
and procedures. The numbers are of other utilities (mutual aid), material suppliers,
employee rosters, fuel and food suppliers, communication equipment (radios, cell
phones suppliers), government agencies, medical treatment facilities, key
customers (water treatment facilities, wastewater treatment facilities, hospitals,
food suppliers, etc.), customers with special medical needs that depend upon
electric service, contractors that may offer assistance, food suppliers, emergency
shelters, etc.
Prior to an anticipated storm event, meet with all employees to develop plan of
action (work responsibilities, when to report, where to report, when to safe up
facilities, etc). We make post-storm work assignments (material procurement,
damage assessment, telephone switchboard, meal preparation, etc.) We make sure
all critical facilities have auxiliary power and all response vehicles and fuel are
topped off. We contact all numbers on the contacts list to ensure the numbers are
accurate and the contacts are aware of our status. We secure buildings and
facilities (board them up). We notify the public of our status and preparations and
request that they prepare for the storm. We advertise the telephone number for our
emergency response attendant. We ensure our emergency supply cabinets are
fully stocked. We ensure all radios are accounted for, have adequate spare
batteries, and are fully charged.


Participant #3









In 2001, [Company Name] went through a reorganization and moved to a process-based
structure. Therefore, during the recent hurricane events, many departments had to
coordinate to ensure all goals and objectives were met in restoring power as quickly as
possible.

We conducted two meetings daily, one in the morning and one in the afternoon, until the
outage figures dropped to a reasonable number (below 40,000). At that point, we went to
one meeting per day, and these meetings were conducted via conference calls over the
weekend.

These involved the CEO, the Electric, Water and Sewer Operations Department, Internal
and External Communications, Public Outreach, Facilities Maintenance, Client
Relationships, Security, Finance, Risk Management, Safety, Engineering,
Technology Services, and Governmental Relations. All vice presidents on the executive
management team were at the meetings, along with many of the field operations
supervisors to ensure direct information flow from the field.

Since a major disruption of electric service can jeopardize public safety and public health
(affecting police, fire and hospital operations), restoring power to those operations takes
precedence in a major outage. With [Company Name], another priority is either restoring
power via electric wire or generators to ensure sewer pumping stations can continue to
pump to avoid SSOs, Sanitary Sewer Overflows.

For many years, and with a new emphasis after the September 11th attacks, [Company
Name] developed a substantial Emergency Plan Book that sets forth all operating
procedures to be implemented should a disaster, natural or otherwise, occur in the service
area.














APPENDIX G
PHASE 1.2 QUESTION AND RESPONSES

Phase 1.2
How has your input been incorporated into your organization's current crisis
management protocols? Who else's input, internal or external, was taken into
account?

Participant #2

We are a very small organization 1 superintendent, 3 linemen, 2 apprentices, plus me.

We all discuss emergency preparation formally in meetings called for the purpose or
informally everyday.

Our more experienced workers have greater insights, but everyone is free to make
suggestions.

I am primarily responsible for ensuring that we are adequately prepared. It is my job to
make sure we think about it, learn as much as we can from other organizations, and meet
to discuss our response.

We work with other agencies in our community local police, fire rescue, and emergency
management. We also attempt to educate our public. Any input received from those
organizations is incorporated to the extent it makes sense for us to do so.

We rely heavily on the information we pick up from the Florida Municipal Electric
Association (FMEA) and the Florida Municipal Power Agency (FMPA) regarding storm
preparation and emergency response.


Participant #3

When the hurricanes began making landfall, we were handling outage information by
request. We realized that, while we wanted to remain accessible to the media, this was
becoming extremely difficult. Therefore, once the major impact of the storm had passed,
we implemented a twice daily communication to the media and other stakeholders
(the Emergency Operations Center, City Council Members, Mayor, neighboring county
and city representatives, and to all internal contacts, etc.) on updated outage information
and where our crews were working. We e-mailed and faxed to all the groups
mentioned at 5 a.m. and at 4 p.m. daily to ensure morning and afternoon television
broadcasts, drive time radio and newspapers would have access to current information









when they went on the air. This also freed up our operations staff to stay focused on the
work of restoring power rather than responded to requests for information throughout the
day. Also, we would communicate rumor-control information, based on questions we
were receiving from the media and the public. Most of our external stakeholders were
pleased with the move, especially city council members and city hall, where constituents
were calling when they couldn't reach [Company Name].

One serious complaint was that customers could not get through to [Company Name] on
the phone. We were receiving more phone calls in one day than we usually received in a
month, so there was no way we were capable to respond to everyone. Therefore, based
on input from the communications and customer outreach personnel, we set up teams of
three-to-five employees to go out to the most affected areas and talk to customers one-on-
one. This was a highly effective move in raising our customer satisfaction levels and the
general public relations boost we received from the effort.


Participant #4

Our representative meets with the local government emergency response team to discuss
issues and make plans.














APPENDIX H
PHASE 2.1 QUESTION AND RESPONSES

Phase 2.1
What characteristics of the crisis event influenced your organization's crisis
response?

Participant #2

The most significant storm characteristic that impacted my area was simply the amount
of damage incurred.

We watch for storms and prepare for impact pretty much the same way. We insert lessons
learned as needed to our preparation.

Damage assessment and restoration of service were the big tasks we faced post-storm.
There was a lot of pressure to restore power ASAP. We prioritized our major circuits and
critical facilities. The effort expended by everyone within our organization to recover
from the storm's impacts is most memorable.


Participant #3

Unlike other smaller storms and events, the first major hurricane (Frances) forced us to
modify our normal procedures for communicating with customers and the media.

Our typical procedure is for customers to contact [Company Name] by phone in the case
of an outage. With more than half of our customers out of power on Labor Day morning,
our phone lines received as many calls in one day as [Company Name] typically receives
in a month. Therefore, many customers could not contact us. (At this point, their calls
were not necessary, since so many circuits were out and we already knew where they
were). Under normal circumstances, we also ask customers to report downed power lines
to use through the customer care center (our main switchboard number).

Therefore, through the media we communicated to customers not to contact [Company
Name] with outage information. ..to instead wait for us to communicate through the
media when that customer communication would be necessary. We also asked customers
to contact 911 to report downed power lines. This allowed fire officials to tape off areas
where lines were down to allow our linemen to work on the main circuits that would
bring more customers back on line quicker.









We resumed normal customer communication procedures once the number of outages got
down to a minimal amount (about 40,000 without power).

To assist in communicating with customers who could not reach us, and might not have
access to a radio, we assembled three-person teams to visit the most affected areas to
hang door-hangers regarding the restoration process and when crews would be in their
neighborhoods.

With our two-time-per-day blast fax to the media, which we also shared with area
cities/council members so they could assist in communicating up-to-date and accurate
information, we also added a rumor-control narrative. Based on feedback received first-
hand and through our customer care consultants, we would address rumors (rich people
are getting power back on first, the utility is supplying free generators to wealthy
customers, etc.) and quell some of the misinformation that would spring up in different
areas of town.

In terms of the restoration process itself, our normal procedure is to work on distribution
lines (lines within the neighborhoods) first, then move out toward the substations.

However, with so many people without power, that was no an effective way to go.
Therefore, we worked the opposite direction -- from transmission lines to substations,
substations to distributions, and into the neighborhoods last. This allowed [Company
Name] to get the major circuits, thus the largest number of customers, back on quicker,
then move into the smaller pockets (neighborhoods) and then handling the individual
outage problems last.

While there were many minor changes we made in our operations and communications
procedures, these were the major changes we administered within two days of Hurricane
Frances.. .and employed throughout the subsequent strike of Hurricane Jeanne.


Participant #4

We are a small city and work closely with the county emergency management team.














APPENDIX I
PHASE 2.2 QUESTION AND RESPONSES

Phase 2.2
Were certain publics more vocal regarding their need for power restoration? Did
their requests influence your organization's response?

Participant #2

The most vocal group in our community was a more affluent gated community that felt
they were not receiving adequate attention. Our priorities for restoration were major
feeders, critical facilities, major commercial, everyone else.

We were contacted by people for every part of the community wanting, and in same cases
demanding, their power restored.

Rarely did it affect our plans for restoration the exceptions were previously unknown
critical or medical needs and quick easy fixes that would restore power to many
customers.


Participant #3

While there is no precise data on this, my perception would be that parents a few days
into the outages began to express concern over getting the schools opened as soon as
possible. Pressure came from the mayor's office, vis-a-vis the city council members
offices through their constituents, that [Company Name] needed to make school
restoration a priority.

Obviously, reasons for this were many -- kids home from school over extended days
begin the affect the ability of parents to get into their work routine (if their employers had
power), costing them additional money for day care or requiring them to stay home.
Also, having no power at home would just irritate an already stressful situation at home.

This pressure did influence our organization's response somewhat, but did was difficult to
focus on a school's power supply when the entire circuit serving the school and the
adjacent neighborhoods were out of power. We did identify that school administrators
did not have a good handle on how many schools did not have power, and [Company
Name] did develop a contact/location list for all schools we serve -- which we did not
compile prior to the hurricanes.






68


Participant #4

Our electric department got the police, fire and emergency services running, then tackled
the downtown business and medical clinic. Residents called in from all over town with
requests for power. Three of us in city hall worked with no power and flashlights when
upstairs to fill out sheets of requests for service and faxed them to the electric department
(the cell phones the electric crews had were not working). Our city had electric crews
from four other cities assisting us with rebuilding the lines.

Our electric crews were not influenced by who called, but rather in order of the request.














APPENDIX J
PHASE 3.1 QUESTION AND RESPONSES

Phase 3.1
How did the 2004 hurricane season influence your organization's preparation for
future crises?

Participant #2

It provided us with valuable experience. There is no substitute for experiencing events.

It provided us with a compilation of information related to emergency response that we
will incorporate into future emergency response manuals and minimize the amount of
"reinventing" that is required.

It generated much, much discussion and dialogue related to emergency response. I / we
participated in as much of this dialogue as possible to learn from the experiences of
others.


Participant #3

Brian, I've touched on some of these previously, but in a nutshell, here are the major
changes:

Strengthening our electrical mutual aid agreements
Initiating a more formal water and sewer mutual aid program
Upgrading some of our coastal-reliability-standards equipment near the beach
Upgrading insulators, cut-outs, arrestors, etc in areas that experienced problems
Documenting recovery expenditures in a "FEMA-friendly" format
Assigning office/engineering staff (to handle customer interactions and non-repair
type work) to ride with repair crews
Including door-to-door employee-customer communications with TV, radio,
newspaper, internet communications plan
As much as practicable, beginning detailed, systematic engineering assessments
prior to restoration work
Changing initial recovery response to concentrate on main lines, substations, etc
o i.e., start with transmission and substations...then feeders...then
laterals.. .then services
Changing our customer information procedures
o E.g., customer will not be asked to report outages during initial heavy
recovery work






70



Participant #4

After dealing with three hurricanes in one season, each city department had a new
understanding of what they would need to do to be prepared for the 2005 season.














APPENDIX K
PHASE 3.2 QUESTION AND RESPONSES

Phase 3.2
Considering what your organization learned from the 2004 hurricane season, do you
feel prepared for the 2005 hurricane season? What concerns does your organization
have?

Respondent #2

No, we do not feel fully prepared for a hurricane season. The cost involved with fully
preparing to respond to hurricanes like those that hit central Florida are too great for a
small utility to overcome in a single year.

We will list our needs, prioritize them, and address them when possible.

Our great concerns are with inventory of material items, communication equipment, and
obtaining assistance through mutual aid.

Respondent # 3

Yes. Since all of us went through the actual event, as opposed to doing a drill or tabletop
exercise, we all have first-hand experience on dealing with the major impact of hurricane
winds and the aftermath. We now know what procedures need to be modified, compared
to normal operating procedures that we have for dealing with typical spring/summer
storms, and how all agency personnel can be used more effectively to work with
customers and communicate with them and the media.

[City Served] actually did not take a direct hit, but was affected by the strong winds
on the northeast side of both hurricanes. Based on what we saw last summer, there is a
concern on the additional amount of damage that would be caused to [Company Name]
and the city, and the time it would take to restore the area to normal, if we were ever to
receive a direct hit from the eye of the storm.

Respondent # 4

Our organization learned that you may think you are prepared, but when you have three
successive hurricanes wreck havoc on your community in six weeks, you can't get that
prepared.

We will always have concerns when Hurricane Season approaches. We know how hard
we worked to tackle whatever was thrown our way in 2004 putting our city back together.















LIST OF REFERENCES


Burnett, J.J. (1998). "A strategic approach to managing crises." Public Relations Review,
24(4), 475-488.

Cancel, A.E., Cameron, G.T., Sallot, L.M., and Mitrook, M.A. (1997). "It depends: a
contingency theory of accommodation in public relations." Journal of Public
Relations Research, 9(1), 31-63.

Cancel, A.E., Mitrook, M.A., and Cameron, G.T. (1999). "Testing the contingency theory
of accommodation in public relations." Public Relations Review, 25(2), 171-197.

Clare, M. (1994). "Delphi Technique." Cornell University. Accessed December 9, 2004.
http://www.cce.cornell.edu/admin/program/documents/delphi.htm.

Coombs, W.T. (1999a). "Information and compassion in crisis responses: a test of their
effects." Journal of Public Relations Research, 11(2), 125-142.

Coombs, W.T. (1999b). Ongoing crisis communication: Planning, muniut,'giig. and
responding. Thousand Oaks, CA: Sage Publications, Inc.

Coombs, W.T. and Holladay, S.J. (2001). "An extended examination of crisis situations:
A fusion of the relational management and symbolic approaches." Journal of
Public Relations Research, 13(4), 321-340.

Ewing, R.P. (1987). Managing the new bottom line: Issues management for senior
executives. Homewood, IL: Dow Jones-Irwin.

Fink, S. (1986). Crisis management: Planning for the inevitable. New York: AMACOM.

Gonzalez-Herrero, A. and Pratt, C.B. (1996). "An integrated symmetrical model for
crisis-communication management." Journal of Public Relations Research, 8(2).
pp. 79-105.

Goodman, M.B. (2001). "Current trends in corporate communication." Corporate
Communications, 6(3), 117-123.

Grunig, J.E. (2001). "Two-way symmetrical public relations: Past, present, and future."
Handbook ofpublic relations. R. Heath and G. Vasquez. Thousand Oaks, Sage
Publications, Inc.: 11-30.









Grunig, J.E. and Grunig, L.A. (1992). "Models of public relations and communication."
In J.E. Grunig (Ed.), Excellence in public relations and communication
management. pp. 285-326. Hillsdale, NJ: Lawrence Erlbaum Associates.

Grunig, J.E. and Hunt, T. (1984). Managing public relations. New York: Holt, Rinehart
and Winston.

Grunig, J.E. and Repper, F.C. (1992). "Strategic management, publics, and issues." In
J.E. Grunig (Ed.), Excellence in public relations and communication management.
pp. 31-64. Hillsdale, NJ: Lawrence Erlbaum Associates.

Grunig, L.A., Grunig, J.E. and Dozier, D.M. (2002). Excellent public relations and
effective organizations: A study of communication management in three countries.
Mahwah, NJ: Lawrence Erlbaum Associates.

Howard, C.M. and Mathews, W.K. (2000). "Chapter 10 Crisis planning: how to
anticipate and manage emergency situations." On deadline: Managing media
relations. Third edition. pp. 215-269. Long Grove, IL:Waveland Press, Inc.

Mitroff, I.I. (1994). Crisis management and environmentalism: A natural fit. California
Management Review, 36(2), 101-113.

Pauchant, T.C. and Mitroff, I.I. (1992). Transforming the crisis prone organization. San
Francisco, CA: Jossey-Bass Publishers.

Penrose, J.M. (2000). "The role of perception in crisis planning." Public Relations
Review 26(2), 155-171.

Sapriel, C. (2003). "Effective crisis management: tools and best practice for the new
millennium." Journal of Communication Management 7(4), 348-355.

Seeger, M.W., Sellnow, T.L., and Ulmer, R.R. (2003). Communication and
organizational crisis. Westport, CT: Praeger Publishers.

Shrivastava, P., Mitroff, I. I., Miller, D. and Miglani, A. (1988). "Understanding
industrial crises." Journal of Management Studies 25(4), 285-303.

Smits, S.J. and Ezzat, N. (2003). "'Thinking the unthinkable' leadership's role in
creating behavioral readiness for crisis management." Competitiveness Review, 13
(1), 1-23.

Thomas, K.W. and Kilmann, R.H. (1974). Thomas-Kilmann conflict MODE instrument.
Tuxedo, NY: Xicom.

Turoff, M. and Hiltz, S.R. (1996). "Computer based Delphi process." New Jersey
Institute of Technology. Accessed December 9, 2004.
http://eies.njit.edu/-turoff/Papers/delphi3.html.






74


Wakefield, R.I. (2000). "Preliminary Delphi research on international public relations
programming." In Moss, D., Vercic, D., and Warnaby, G. (Eds.), Perspectives of
public relations research. pp. 179-208. New York: Routledge.

Wilson, S. (2004). "The myths and reality of crisis communication." Accessed October
15, 2004. http://www.wilson-group.com/articles/myths.shtml.















BIOGRAPHICAL SKETCH

Brian Boudreaux completed this study as the capstone of his candidacy for a

Master of Arts in Mass Communication, specializing in public relations at the University

of Florida. In 2003, Boudreaux received his Bachelor of Arts degree in speech

communication from the University of Illinois. Boudreaux's research interests include

crisis management and strategic planning. Upon graduation, the author will begin his

professional career in the corporate or government sector.