<%BANNER%>

Public Sector Crisis Communication in Korea

Permanent Link: http://ufdc.ufl.edu/UFE0024664/00001

Material Information

Title: Public Sector Crisis Communication in Korea 2003-2008
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Lee, Jinsuk
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PUBLIC SECTOR CRISIS COMMUNICATION IN KOREA: 2003-2008 Jinsuk Lee 352-392-0466 Mass Communication Youjin Choi Master of Arts in Mass Communication August 2009 This study seeks to understand public-sector crisis and crisis communication strategies using Coombs?s crisis typology and crisis communication strategies as a theoretical framework. The study investigated 619 crisis cases in the Korean government during the Moo-Hyun Roh administration, covering February 2003 through February 2008. The study found that the Korean government employed different crisis communication strategies based on a cluster of crisis types (e.g., victim cluster, accidental cluster, preventable cluster). The research confirmed that the Korean government did not frequently employ bolstering strategies and, when they were used, they were used to supplement other strategies. Whereas the Korean government used more rebuilding strategies in crises that had a history, it used the denial strategy most frequently in nonrecurring crises. The study determined that different levels of an administrative system relate to the crisis communication strategy choices made by the Korean government.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jinsuk Lee.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2009.
Local: Adviser: Choi, Youjin.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024664:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024664/00001

Material Information

Title: Public Sector Crisis Communication in Korea 2003-2008
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Lee, Jinsuk
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PUBLIC SECTOR CRISIS COMMUNICATION IN KOREA: 2003-2008 Jinsuk Lee 352-392-0466 Mass Communication Youjin Choi Master of Arts in Mass Communication August 2009 This study seeks to understand public-sector crisis and crisis communication strategies using Coombs?s crisis typology and crisis communication strategies as a theoretical framework. The study investigated 619 crisis cases in the Korean government during the Moo-Hyun Roh administration, covering February 2003 through February 2008. The study found that the Korean government employed different crisis communication strategies based on a cluster of crisis types (e.g., victim cluster, accidental cluster, preventable cluster). The research confirmed that the Korean government did not frequently employ bolstering strategies and, when they were used, they were used to supplement other strategies. Whereas the Korean government used more rebuilding strategies in crises that had a history, it used the denial strategy most frequently in nonrecurring crises. The study determined that different levels of an administrative system relate to the crisis communication strategy choices made by the Korean government.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jinsuk Lee.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2009.
Local: Adviser: Choi, Youjin.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024664:00001


This item has the following downloads:


Full Text

PAGE 1

PUBLIC SECTOR CRISIS COMM UNICATION IN KOREA: 2003-2008 By JINSUK LEE 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 2009 1

PAGE 2

2009 by Jinsuk Lee 2

PAGE 3

This thesis is dedicated to my family 3

PAGE 4

ACKNOWLEDGMENTS I would like first to thank my advisor, Dr. Youjin Choi. She has advised me on not only my academic life, but also my life as a gradua te student and my life in the US. She always welcomes me with a shiny smile and never hesitates to help in any way she can. Whenever I was troubled with my thesis, she guided me clearly and encouraged me to proceed. My committee member, Dr. Kiousis, helped me greatly. His class on persuasion in my first semester aroused my interest in public rela tions. He has encouraged me to think differently and apply that new thinking as I wrote my th esis. He always offered proper and extensive analysis on my thesis and helped me give my thesis a unique and profound quality. I would like to thank my other committee me mber, Dr. Mitrook as well. He taught me how to manage the methodological problem in social science, so I could eas ily apply it to this thesis. He helped me develop a broader perspect ive. His warmhearted advice consistently was most encouraging, as I develope d my thesis step by step. I give a special thanks of course to my family. I offer special thanks to my father who is in Heaven. I believe that he guided my study. He w ould be most pleased with this work. I also thank my mother who prayed for me every day. Sh e is the one who I can rely on at any time. I wish her good health always. Thanks also go to my two sisters and tw o brothers and their families. I miss them very much and look forward to seeing them soon when I return to Korea. Especially, I appreciate my niece, Jiyoun Hong. We always encourage each other to study hard and succeed. My cousin and her familyEunjoo Kim, Jungsoo Kim, Charlie Kim, Shanie Kim who live in Gainesville, gave me much help. Without their wise counsel, I would not have adapted to US life. They helped in Gainesville and often invited me for meals and to thei r events. Thus, I rarely felt homesick. I also wish to thank my college friend, Sookhee Shin, who lives in Gainesville. 4

PAGE 5

What a coincidence it was that we met here. She offers me advice on life in the US and encouraged me to study at a later age. She has been a delight for my life here. A special thanks goes to Moonhee Cho. She is the one who made my graduate study possible. When I suffered from a lack of English ability, when I could not adapt to the US school system, she helped me. Without her kind assistan ce, I would not have completed my graduate work. I thank Sooyeon Kim, Jooyun Hwang, Jiyoung Cha, Yeonsoo Kim, Minji Kim, Hyunji Lim, Hyunmin Lee, Heejung Kim, Jiyoung Kim, Jungmin Lee, Bumsub Gabriel Jin, and Maria de Moya for sharing their deep knowledge and friendship. My classmates and college mates, Jiyoung Bang, Daewook Kim, Sanghoon Lee, Jae jin Lee and her husband Yongchul Kim, Gyongsub Lee, Jinha Song, Jain Rajul, Eunhwa Jung, Eun Ko, Jinhong Ha. I owe very special gratitude to the Korean Government who gave me the opportunity to study abroad. This opportunity broadened my view of the world and my field and deepened my thinking. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................8ASTRACT ....................................................................................................................... ................9CHAPTER 1 INTRODUCTION ................................................................................................................ ..112 LITERATURE REVIEW .......................................................................................................15Crisis ........................................................................................................................ ...............15Crisis Typology ......................................................................................................................18Crisis Management and Crisis Communication .....................................................................21Crisis Communication Strategy ..............................................................................................23Crisis History ..........................................................................................................................27Crisis Communication in the Public Sector ............................................................................28Different Levels of Ad ministrative System ............................................................................30Research Hypotheses and Questions ......................................................................................333 METHODS ..................................................................................................................... ........36Data Collection .......................................................................................................................36Measurement ................................................................................................................... ........37Coding and Inter-coder Reliability .........................................................................................38Data Analysis ..........................................................................................................................394 RESULTS ..................................................................................................................... ..........40Research Question 1 ...............................................................................................................40Research Question 2 ...............................................................................................................41Hypothesis 1 .................................................................................................................. .........43Research Question 3 and Research Question 4 ......................................................................43Hypothesis 2 .................................................................................................................. .........44Hypothesis 3 .................................................................................................................. .........455 DISCUSSION .................................................................................................................. .......51Implications .................................................................................................................. ..........56Limitations and Future Research ............................................................................................57 6

PAGE 7

APPENDIX A CODING SHEET FO R CONTENT ANALYSIS .................................................................60B THE CODE BOOK FOR CONTENT ANAYSIS .................................................................62LIST OF REFERENCES ...............................................................................................................66BIOGRAPHICAL SKETCH .........................................................................................................72 7

PAGE 8

LIST OF TABLES Table page 1-1 Coombs crisis typology and cr isis communicatio n strategies ..........................................144-1 List of cross tabulation of victim cl uster crisis and communication strategies .................474-2 List of chi-square test of accidental cluster crisis and commu nication strategies .............474-3 List of cross tabulation of preventable cluster and communication strategies ..................484-4 List of cross tabulation of preventable cluster and communication strategies ..................484-5 List of cross tabulation of local admi nistrative system crisis and communication strategies .................................................................................................................... ........494-6 List of cross tabulation of nationa l administrative system crisis and ................................494-7 List of cross tabulation of transnational administrative system crisis and .........................50 8

PAGE 9

Abstract Of Thesis Presented 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 PUBLIC SECTOR CRISIS COMM UNICATION IN KOREA: 2003-2008 By Jinsuk Lee August 2009 Chair: Youjin Choi Major: Mass Communication This study seeks to understand public-sector cr isis and crisis communication strategies using Coombss crisis typology a nd crisis communication strategies as a theoretical framework. The study investigated 619 crisis cases in the Korean govern ment during the Moo-Hyun Roh administration, covering February 2003 through Febr uary 2008. It identified frequently occurring crisis types and crisis response st rategies used in the Korean pub lic sector and, specifically, the relationship between crisis typologies and crisis communication st rategies. A secondary focus of this study was testing whether the crisis history of organizations was related to their choices of crisis communication strategies. The study found that the Korean government most often confronted organizational misdeeds, malevolence, and rumors. The Kor ean government employed different crisis communication strategies based on a cluster of crisis types. The government used the attack-theaccuser strategy most often in the victim cluster, the justification stra tegy most often in the accidental cluster, and the apology strategy most of ten in the preventable cluster. The research confirmed that the Korean government did not fr equently employ bolstering strategies and, when they were used, they were used to supplement other strategies. Whereas the Korean government used more rebuilding strategies in crises that had a history, it used th e denial strategy most 9

PAGE 10

10 frequently in nonrecurring crises. The study dete rmined that the different levels of an administrative system relate to the crisis co mmunication strategy choices made by the Korean government. Compensation was used most frequently when crises occurred within the control of local administrative systems. Th e denial strategy was used on national-level crises while the attacking the accuser strategy was used on international-level cr ises. The study also found that some categories of Coombs crisis types and cris is communication strategies should be modified, and suggested citizen or other actors faults as a new category for crisis types and corrective as a new category for crisis communication strategies.

PAGE 11

CHAPTER 1 INTRODUCTION Crises are prevalent in modern society. Ever y day people witness diverse crisis events; some are large, and some are small. There are many different kinds of cr ises in daily lives, including natural disasters, economic recession, soaring oil prices, corporation bailouts, and bribery and unethical behavior. Actually, it seems that crises are a more frequent phenomenon in modern society than ever, and seem to cause more and more devastating effects (Lerbinger, 1997). In this regard, crises ar e no longer an aberrant rare, random, or peripheral feature of todays society. They are built into the very fa bric and fiber of modern societies (Mitroff & Anagnos, 2001, p.4). Furthermore, it seems like that the nature of crises is becoming more and more complex. Sometimes crises are interwoven and connect with each other. Boin and Lagadec (2000) explain that these trends originated with globalization, increased mass communication (interwiredness), social fragmentation and the hotly disputed dissipat ion of state authority (p.185). Globalization and increased mass communication allows stakehol ders to obtain information about crises of other regions and nations. Thus crises easily diffuse through national borders. Whatever the cause of a crisis, one can negatively affect an organization if the crisis is not managed properly. The outcomes of crises can be financial, physical, and even emotional (Coombs, 2007b; Pearson & Mitroff, 1993). Crises can decrease an organi zations reputation and legitimacy (Coombs, 1998) and affect stock ma rket prices (Kim, Cha, & Kim, 2008), provoke lawsuits, and causes distrust of organizations. As a consequence of crises, some corporations perish in the market, while other corporations receive a chance to strengthen their market position after successfully ma naging the crisis. 11

PAGE 12

In the public sector, crises are also prev alent in various forms: natural disasters, epidemics, economic fluctuations international disput es, terrorism, labor disputes, and major accidents. Every local, regional, and national aut hority must deal with these kinds of crises. However, the outcomes of public sector crises are greater than those of private sector crises because public sector crises can affect not only public organizations, but also the fundamental social and political system of that societ y (t Hart, Rosenthal, & Kouzmin, 1993). In the government setting, stakeholders do not view crises simply as crisis events that stand alone. The stakeholders connect a crisis with the larger and more profound overall picture. They see crises as a failure of their government system or a failure of leader ship. Kouzmin, Alexander, Jarman, and Alan (1989) emphasize that crises in the public sector raise questions about government ineffectiveness. Lee (20 08) states that stakeholders s ee such a crisis as the product of the government and its leadership rather than in relation to a spec ific episode (pp. 1-2). Today, crises are inevitable, but they are mana geable. Crisis management is an effort to reestablish the image and legitimacy of an organization (Lee, 2004), and many scholars, including Coombs, Bradford, Garnett, Rosentha l and Kouzmin, from diff erent disciplines in politics, public administration, communication, a nd public relations, have developed crisis theories to help crisis managers manage cr ises. Communication and public relations research especially focus on crisis communication among diverse crisis management areas (Seeger, Sellnow, & Ulmer, 2001; Marra, 1998). Coombs (1998, 2006, 2007) proposes that crisis communication strategies be used for different types of crisis situations. He classifies ten crisis types according to the level of responsibility attributed to organizations. He defines ten crisis-communication strategies on a continuum of defensive and accommodative responses based on the level of responsibility that 12

PAGE 13

organizations show. His crisis typology and cris is communication strategi es have been tested with actual crisis cases (Englehardt, Sallo t, & Springston, 2004; Vlad et al., 2006). Research has been conducting on finding the most effective crisis communication response for various private sector cases. Even though the public sector suffers from frequent crises, research on public sector crises is rare. A few scholars, such Lee and Liu, have researched government crisis communication. Lee exam ined the 2003 SARS and the Hong Kong governments crisis management. He (2008) f ound that the Government Information Service (GIS), which is in charge of public relations wa s apart from the SARS crisis scene, and no single designated spokesperson existed to respond to in quiries about SARS. Liu (2007) also studied President Bushs speeches for Katrina and found that Bush use six main themes in his speech: compassionate president, major devastation, op timism, commander-in-chief, locals-know-best, and religious invocation (p. 42). Most of the studies, including the above, use case studies with a focus on how one organization responds to one crisis case when studying crisis management (Coombs & Holladay, 2008). However, it is also useful to investigat e how an organization responds to various crises over a certain period of time. Because such research can help organizations identify frequent crisis types, effective crisis re sponse strategies, and factors which affect the selection of crisis response strategies. This current study focused on Korean public sect or crises using Coombss crisis typology and crisis communication strategies (2007) as a theoretical framework for investigating public sector crises (See Table 1-1). Th e study investigated th e crises in the Korean Government from February 2003 to February 2008 during the MooHyun Rohs administration. The aim of this study was to identify frequent crisis types and crisis response strategies in the Korean public 13

PAGE 14

14 sector and especially the rela tionship between crisis typol ogies and crisis communication strategies. In particular, the study examined whether the Korean Government had used different crisis communication strategies for crisis situat ions with different le vels of organizational responsibility. Another focus of th is study was to test whether the crisis history of organizations was related to the crisis communication strate gy choices of individual organizations. The study investigated whether different levels of an administrative system relate to the crisis communication strategy choices of the Korean Government during the period under study. Table 1-1. Coombs crisis typology a nd crisis communica tion strategies Crisis Types Crisis Co mmunication Strategies Victim Cluster Natural Disasters De nial Posture Attacking the accuser Workplace Violence Denial strategy Rumors Scapegoating Malevolence Diminishment Posture Excusing Accidental Cluster Challenges Technical-error accidents Justification Technical-error product harm Rebuilding Posture Compensation Preventable Cluster Human-error accidents Apology Human-error product harm accidents Bolstering Posture Reminding Organizational misdeeds Ingratiation Victimage

PAGE 15

CHAPTER 2 LITERATURE REVIEW Crisis The term crisis is used in day-to-day conve rsation and in news media reporting. However, a consistent definition of crisis is lacking. Scholars who study cr isis present vari ous definitions for the term. Seeger et al. (2001) define cris is as the perception of an unpr edictable event that threatens important expectancies of stakeholders and can seriously impact an organizations performance and generate negative outcomes (pp. 2-3). They (1998) see crisis as a specific, unexpected, and nonroutine event or series of even ts that has a negative impact on an organization (p.233). This view is similar to Coombss (2007b) definition of crisis as a sudden and unexpected event that threatens to disrupt an orga nizations operation and poses bot h a financial and reputational threat (p.164). Pearson and Clai r (1998) also suggest that an organizational crisis is a lowprobability, high-impact event that threatens the vi ability of the organization and is characterized by ambiguity of cause, effect, and means of resolution, as well as by a belief that decisions must be made swiftly (p. 60). Thus, crisis can be de fines as an unexpected event that has serious impact on the organization. Several characteristics of crisis can be inferred from the definitions of crisis as: suddenness, uncertainty, time compression, and severe threat (Lerbinger, 1997; Rosenthal, t Hart & Charles, 1989). First, a crisis is an event having low probability and unexpectedness (Crandall & Mensah, 2008). A cris is is not a daily event that an organization expects to encounter. Crisis managers in organizations do not deal with crises every day. Crises also occur suddenly even though there is a sign or antecedent event that precedes a crisis (Lerbinger, 1997). According to Fishman (1999), modern societys very technological innovation increases the 15

PAGE 16

unpredictability of modern society, particularly, because of high-risk a nd complexity of that technology (Perrow, 1984). Nuclear power plants or DNA facilities that employ high-risk and complex technologies need complicated safety devices. However, when communications and interactions among related employees regarding such safety devices fa il, crises can happen. Second, when a crisis occurs, the organiza tion falls into an un certain situation. A crisis is literally an unstable time or stat e of affairs (Fink, 1986, p. 15) or a state of uncertainty (Ho & Hallhan, 2004, p. 367). Rosenthal, Boin, and Comfort (2001) argue that the uncertainty of a crisis may pertain to the specif ic nature of the threat to peoples initial and emergent responses, to the dynamics of the situati on and to the future consequences of the crisis (p. 7). Decision makers and public relations practitioners find it difficult to recognize accurate causes and effects of a crisis, and information about the crisis environment, such as stakeholders perceptions and news media atti tudes during a crisis. Continually changing environments that include new technology, government regulations, co mpetition, consumer attitudes, and social movements make it difficult for organizations to recognize and collect a ccurate information. Third, organizations have only a limited period of time to respond to a crisis (Barton 1993; Pearson & Clair, 1998; Crandall & Mensah, 2008; Rosenthal et al., 1989). Crises demand urgent decision-making and immediate response by the organizations (Rosenthal et al., 1989, p. 14). During a crisis, some decisions regard ing life and death of stakeholders and the organizations must be made within a split moment or short period of time, even hours or minutes (Rosenthal et al., 2001). This urgency occurs not onl y in reacting to an actu al crisis, but also in explaining the crisis situation to stakeholders, such as media, stakeholders, and stockholders (Fishman, 1999). Suddenness and uncertainty hinde rs the decision making process that already 16

PAGE 17

suffers from urgency. Lack of quick and proper decision making can lead organizations toward an even more aggravating situation (Lerbinger, 1997). Fourth, crises threaten critical values of organizations (Ros enthal et al., 1989). Incidents and crises can be differentiated by this criterion. Crises can be differentiated from incidents in terms of the magnitude and severity of their impact on organizational resources and reputation. For example, incidents do not impact more th an a local operation and do not interrupt normal organizational operations (Coombs, 2007). A crisis, which has four characteristics of suddenness, uncertainty, time compression, and severe threat, produces negative consequences on an organizati on. Lerbinger (1997) notes that crisis brings, or has the poten tial for bringing, an or ganization into disrepute and imperils its future profitability, growth, and, pos sibly, its very survival (p. 4). Among negative outcomes, financial damage is the most obvious one (Lerbinger, 1997). Reputation is also in danger during a crisis (Coombs, 2007b; Coombs & Holladay, 2005). According to Tucker and Melewar (2005), reputati on is the perception of an organization based on its stakeholders interpretation of that organizations past, present and future activities and the way in which these are communicated (p. 379). To the organization, re putation is a critical factor that determines competitiveness and most important assets (Stubbs, 2002). However, reputation, normally seen as an invisible factor, al so has a perishable aspect, and crisis is often the event that endangers a peri shable reputation. Crises can ru in the relationship between an organization and its stakeholde rs (Heath & Millar, 2004; Coombs & Holladay, 2001). A crisis can damage organizations relationship with employees, family members, stockholders, suppliers, community members, and customers (Coombs, 2007b). 17

PAGE 18

In conclusion, crisis can be defined as thos e unpredictable and sudden events that place an organization into uncertain and time compressing situations, resulting in serious and negative impact to that organization. Crisis Typology Crises do appear in diverse shapes and sizes and it is impossible to make response plans for every crisis case. In both crisis management and crisis communication, diagnosis of the type of crisis is important. When one can classify crises into several categories, crisis managers or crisis communication pract itioners can apply the best applic able response. Gundel (2005) points out that the first step in managing a crisis is th e classification of the crisis. As a matter of fact, crisis typology is related to the eventual interpretation of the crisis (Coombs, 2007). Typology of crisis not only affects the interp retation of a crisis, but also the adoption of precise crisis communication strate gies. That is, the criteria used to classify crisis types can affect how to view the crisis types and how to respond to them. Coombs and Holladay (1996) found that one constant argument fr om crisis communication research is that different situations influence the choice of co mmunication strategies. Mitroff and Anagnos (2001) divi de crises into seven types: economic, e.g., labor strikes, labor shortage, and fluctuation of stock pric e; informational, e.g., loss of confidential information, and false information; physical, e.g., loss of key equipment and plants; human resource, e.g., loss of CEO, and workplace violence; reputational crisis, e.g., rumors and slander; psychopathic act, e.g., product tampering and kidna pping; and natural disa ster, e.g., earthquake and hurricanes. This study uses Coombss crisis typology which uses responsibility of the organization for classification. Coombs (1998) declares that crisis responsibility is a n atural link between a crisis situation and crisis comm unication strategy (p. 180). Since cr isis responsibility means the 18

PAGE 19

degree to which stakeholders blame the organi zation for the crisis event (Coombs, 1998, p. 180), an organizations crisis communication st rategy should be chosen based the degree of responsibility accepted for the crisis. Responsibility is closely related to the control of events: attr ibution of events to internal or external forces. Weiner (1995) states that control means responsibility. According to McAuley, Duncan, and Russel (1992), the term c ontrol should be divided into two meanings: (a) controllable or uncontrollable by the person and (b) c ontrollable or uncontrollable by other people. They label the former as personal control and the latter as external control (p. 567). Coombs (1998) incorporates McAuley et al.s th eory into the responsib ility of organization for the crisis. He states that stronger perceptions of external control s hould lessen crisis responsibility and image damage because the organization could do little or nothing to prevent the crisis. Stronger perceptions of personal control/locus of causal ity should increase cris is responsibility and image damage because the organization could have acted to prevent the crisis (p. 182). Benoit (1997) also pays attention to the responsibility issue in a crisis, saying that the reason that stakeholders attack an organization under cris is is related to two components. First the accused is held responsible for an action, and second, that act is co nsidered offensive (p. 178). Mitroff and Anagnos (2001) also emphasized the importance of assuming responsibility in a crisis. Coombs (2007) clusters crises into three groups, based on the levels of an organizations crisis responsibility: the victim cluster (natur al disasters, rumors, workplace violence, and malevolence), the accidental cris is cluster (challenges, technical -error accidents, and technicalerror product harm), and the preventable crisis cluster (human-error accidents, human-error product harm, and organizational misdeeds). 19

PAGE 20

In the victim cluster, stakeholders view the or ganization as the victim of the crisis because external forces caused the crisis, and it was beyond the organizations managerial control (Coombs, Hazleton, Holladay, & Chandler, 1995). Therefore, the crisis produces very little attribution of crisis responsibi lity (Coombs, 2007, p 141). Natural disasters are crises resulting from acts of God, for example, earthquakes, hurricanes, tornados, floods, mud slides, and blizzards (Seeger et al., 2001; Coombs, 2007). Usually, preparing for and managing natural disasters are the responsibility of governments (Seeger et al., 2001). Workplace violence is violence committed by an employee or a former employee of an organization to other employees, for example, shooting coworkers (Coombs & Holladay, 2002; Coombs, 2007). Rumors are false information inte ntionally distributed ab out an organization and its products to damage the organization (Coom bs, 2007). Malevolence is an opponent or a third partys attack on the organization, such as tampering, kidnapping, terrorism, or computer hacking (Coombs, 2007, p. 65). In the accidental crisis clus ter, stakeholders perceive the organization as having only limited control of a crisis. Ther efore, it produces a l ow attribution of cr isis responsibility (Coombs, 2007, p 141). Challenges are caused by conf rontations with discontented stakeholders who claim that the organization operate inappr opriately. An example is the picketing by the American Family Associations (AFA) to Waldenbook because of pornography peddling (Coombs & Holladay, 2002; Coombs, 2007). Techni cal-error accidents ar e industrial accidents caused by the failure of technology or equipmen t utilized or supplied by the organization (Coombs, 2007, p. 65). Power plant blackouts caused by worn-out components is an example. Technical-error product harm is when the technology utilized or supplied by the organization 20

PAGE 21

fails and results in a defect of a potentially harmful product (Coombs, 2007, p. 65). Television recalls due to a defective connector is an example (Coombs & Holladay, 2002). In the preventable crisis cluster, stakeholders view organizations as intentionally violating laws and regulations (Coombs & Holladay, 2002) and, therefore, producing strong attributions of crisis responsibil ity (Coombs, 2007, p. 141). Human-erro r accidents are caused by human errors, such as failed judgment, negligence, blunders, or sabotag e (p. 8). Research has revealed that a human error is inevitable and occurs in every conceivable settin g (Rosenthal et al., 2001, p. 8). Workers accidentally eradicating a line supplying phosphoric acid to a container holding sodium nitrate are an example of human error (Coombs & Holladay, 2002). Human-error product harm is a product defect or harm to consumers caused by human error (Coombs, 2007). An example is the recent E. coli tainted hambur ger recall, caused by workers placing contaminated beef into the grinding process by mistake (Coombs & Holladay, 2002). Organizational misdeeds occur when mana gement takes actions it knows may place stakeholders at risk or knowingly violates the law (Coombs, 2007, p. 65) like when top managers are accused of inappropriate sexual harassment of female employees. (Coombs & Holladay, 2002). Crisis Management and Crisis Communication Crisis is unexpected and unpredictable, but th at does not mean that it cannot be managed. Crisis must be managed and contained (K ersten, 2005, p. 245). More and more organizations set up a crisis management plan in order to reduce damages caused by the current crisis and avoid crisis recurrence. According to Pearson and Clair (1998), crisis management is an organized effort by an organization with its stakeholders to avoid crises or to control crises effectively. Recent crisis management research shows that crisis mana gement includes such diverse areas as the 21

PAGE 22

development of crises over stages, cris is planning, crisis decision-making, and communication strategies availabl e to organizations in the afte rmath of a crisis (Ulmer, 2001, pp. 591-592). When a crisis is controlled and managed by an organization, it is no longer an abnormal situation; it becomes a normal process in organizational operati on (Kersten, 2005). Wellmanaged crises can produce financial and repu tational benefits to an organization, and an organization can have the opportu nity to grow though crisis. Then, what is successful crisis management ? Pearson and Clair (1998) stress that the effectiveness of crisis management occurs wh en potential crises ar e avoided or when key stakeholders appreciate that the positive outcomes from managing a crisis in the shortand longterms outweigh any negative outcomes. For example, Johnson & Johnson suffered from a Tylenol incident in 1982, which caused seven deaths by cyanide poisoning. Yet the company fully recovered its previous market share in just one year (Banks, 2005). The Washington Post (October 11, 1982) praised the company and how Johnson & Johnson has effectively demonstrated how a major business ought to ha ndle a disaster. (p.1). Successful crisis management not only recovers physical and fi nancial damages, but also restores the organizations reputation and legitimacy. Crisis communication can play a critical role in successful crisis management. As seen in the Exxon Valdez oil spill, organizations that have poor and inadequa te communication with their stakeholders have crises situations deteriorated. Thr ough crisis communication, however, organizations can diminish the uncertainty of a cr isis to the public and recover control of the crisis situation. 22

PAGE 23

Crisis communication has four functions duri ng a crisis: environmental scanning, crisis response, crisis resolution, and organizational learning (Seeger et al., 2003, p. 66). Environmental scanning is where organizations monitor and maintain relationships with external environment. Crisis response is planning a communication and action plan for crisis. Crisis resolution is restructuri ng relationships with stakeholders after the crisis, and it includes offering apologies and plausible accounts, etc. O rganizational learning is the acquisition of new knowledge and skills that emerge from the crisis. Seeger et al.s four crisis communication func tions imply that crisis communication plays a critical role in every stage of a crisis before, during, and after the crisis stage. In every stage, crisis communication make an organizations more alert to environmental changes, facilitates the maintenance of a positive relationship with stakeholders, such as employees, customers, suppliers, and shareholders, and responds appropriately to th e evolution of the crisis. Crisis Communication Strategy Crisis communication strategies are an essent ial part of effective crisis management. Simply delivering an apology in every crisis event is not recomm ended (Coombs, 1999). Apology sometimes becomes the cause of lawsu its from stakeholders and sometimes sounds clichd to stakeholders, sometimes hinderi ng the organization from winning back public confidence. An inappropriate comm unication strategy can even evoke severe public anger. If an organization delivers a justifica tion strategy despite stakeholders anticipation of an apology, the justification strategy will provoke distrust and even rage from stakeholders. Therefore, a manageable crisis can become a fatal event that th reatens the actual survival of the organization. When a crisis occurs, an organization should ta ke the right action to defend and recover its image. While restoring and defending its image is the most important goal of an organization confronted with crisis (Beno it, 1995; Fishman, 1999), selection of proper crisis communication 23

PAGE 24

strategies is the key to restor ing the organizations damaged image and reputation (Benoit, 1997; Coombs 1998; Seeger et al., 2001) Benoit (1997) offers comprehens ive image restoration crisis strategies, presenting five categories and ten substrategies: denial (simple denial, and shift the blame), evasion of responsibility (provocati on, defeasibility, accident, and good intentions), reducing offensiveness of the event (bolsteri ng, minimization, differentiation, transcendence, attack accuser, and compensation), correc tive action and finally, mortification. Denial and evasion of responsibility mean th at an organization atte mpts to reject or reduce their accused responsibility of crisis (Benoit, p. 179). Whereas denial is denying any responsibility for the crisis evasion of responsibility is trying to diminish the responsibility for the crisis. Reducing offensiveness of the event and corrective action refer to an organizations attempt to decrease the degree of perceived offense among accusers (Benoit, p. 179). While the reducing offensiveness of the even t strategy emphasizes that the crisis is not serious, the corrective action strategy proposes a plan to solve the problem. The last category of mortification is an attemp t to repair the organizations im age through a plea for forgiveness. This study utilizes Coombss crisis comm unication strategies. In line with Benoits strategy, Coombs (2007) categorized ten types of crisis communicat ion strategies and clustered them into four groups: denial posture (attack ing the accuser, denial, and scapegoating), diminishment posture (excusing and justifi cation). rebuilding posture (compensation and apology), and bolstering posture (remi nding, ingratiation. and victimage). A denial posture eliminates any association between the organization and the crisis. (pp. 139-141). With a strategy of attacking the accuse r, an organization confronts the person or organizations arguing for the existence of a crisis The organization can threaten the accuser to use force, such as a lawsuit. With the denial strategy, the organization denies the existence of 24

PAGE 25

the crisis. The organization states that there is no crisis and sometimes discusses the rationale why the crisis does not exist. With scapegoating, the organization tries to shift the blame from the organization to other person or or ganizations (Coombs 2007; Benoit, 1997). The diminishment posture tries to lessen the attribution of organizational control over the crisis (Coombs, 2007, p. 139). Excusing is wh ere the organization tries to diminish its responsibility for the cris is. This strategy denies that the organization had any intention to do harm or has not the power to control the crisis. Excusing is related to Benoits (1997) concept of defeasibility where organizations claim that they are short of in formation about the situation or do not have control over the crisis Justification is similar to Benoits minimization strategy as well (Seeger et al., 2001). Justif ication is where the crisis manager tries to minimize the perceived damage associated with the crisis (C oombs, 2007, p. 140). The response can include stating that there was no severe damage or injuries or stating that th e people affected by the crisis deserved what they received. However, jus tification can provoke th e perception that the organization is only trivializing victim concerns (Coombs, 2007, p. 141). The rebuilding posture is where organizations attempt to improve their reputations. In this posture, organizations offer words and actions for the benefit of stakeholders and compensate for the negative outcomes from crisis. The comp ensation strategy means that the accused organization provides various benefits, such as mone y, gifts, or services, to reduce or assuage the negative emotions of the victims toward the organization. With the apology strategy, the organization admits full responsibility for th e crisis and requests forgiveness publicly. The bolstering posture complements the other th ree postures. The bolstering posture has an organization-centered characterist ic which does not consider th e interest of stakeholders, particularly when this strategy is used alone (Coombs, 2007). Among the bolstering cluster, the 25

PAGE 26

reminding strategy lets a crisis manager re mind stakeholders about the organizations good deeds in the past, while the ingratiation strategy praises the stakeholders as the most important asset to the organization. With the victimage st rategy, the organization insi sts that they are also a victim of the crisis. Scholars (Coombs, 1999; Marcus & Good man, 1991; Siomkos & Shrivastava, 1993) emphasize the utility of using the continuum con cept in organizational cr isis response strategy. Considering that there are so many crisis commun ication strategies, introd ucing the concept of a continuum to crisis management strategy typology is useful. It allows scholars to organize extensively used crisis communi cation strategies and categorize those strategies (Coombs, 1999). Coombs (1998) suggests that the concept of conti nuum is a useful connection to use in order to match a crisis situation and cr isis communicatio n strategies. Marcus and Goodman (1991) use a defe nsive-accommodative continuum when investigating shareholder responses to crises. They found that there is a si gnificant difference in the responses of shareholders to accommodative and defensive po licies from an organization in crisis: while shareholders show a positive response to an acco mmodative strategy in the case of scandal, they show a positive response to use of a defensive strategy in the case of an accident. In cases of product safety and h ealth incidents shareholders show no significant differences between accommodative and defensive policies. Siomkos and Shrivast ava (1993) propose a defensive-accommodative continuum in a corporati ons response strategy when confronted with a product liability crisis. They s uggest four levels of organiza tional response continuum, ranging from low to high levels of response, namely, d enial, involuntary recall, voluntary recall, and super effort (p. 74). 26

PAGE 27

Coombs adopts the continuum concept in cr isis communication stra tegy. He proposes that crisis communication strategies be arranged along a defensive-accommodative continuum. Defensive strategies put an or ganizations interests first and accommodative strategies puts the victims interest first (Coombs & Holladay, 2002) Accommodative strategies imply that an organization admits that the strong perceptions of stakeholders are an organizations responsibility in a crisis s ituation (Coobms, 1998). Susskind a nd Field (1996) recommend that the responses of avoiding organi zational responsibility, i.e., defe nsive strategies, have often led to lessened confidence in public organizations. Crisis History Evaluation of past crises is a considerably important factor in the interpretation of recent crises (Coombs, 2004). The news of oil spills or lead contamination of toys by certain companies instantly reminds the public of past oil spills or toy contamination cases. McAuley, Duncan, and Russell (1992)s stabil ity concept explains why the history of events so often reminds people of past events. The term stability refers to whether the perceived causality of a crisis changes or does no t change over time (McAuley et al., 1992). If the same mistake happens frequently, it can be ca lled stable; but if the mistake is unique, it can be called unstable (Coombs, 2004). Coombs (2007) suggests that crisis type and cr isis history can be applied in a two-step evaluation of reputational threats: 1) decide the crisis type with the viewpoint of crisis responsibility, and 2) adjust th e initial evaluation of the crisis according to the performance history of the organization crisis history is the one of the performance history. Coombs (1998, 2004) found that the history of crisis affects th e publics perception regarding the responsibility of an organization and its negativ e impact on image, and lowers the 27

PAGE 28

organizations reputation because a history of cris is tells the public that the organization has had and will continue to have similar and ongoing crises. Prior crisis history can worsen public evalua tion of the crisis type Coombs and Holladay (1996, 2001, 2002, 2004) argue that if an organization ha s a prior crisis hist ory, stakeholders will see a victim crisis as an accidental crisis, a nd an accidental crisis will be treated as an preventable or intentional crisis because people would put more or ganizational responsibility for repeated crises. They recommend that public relations practitioners should choose an accommodative strategy with repeated crises and assume strong responsibility even if the organization has just weak re sponsibility for those crises. Crisis Communication in the Public Sector Most crisis types are not different between th e private sector and the public sector. Crises, such as terrorism, management-staff conflict, and communication breakdown, can happen in either sector. Rosenthal and Kouzmin (1993) insist that crises in the private sector occur in the public sector as well, and the sepa ration of private and public sector crises is not absolute. Still, some crises are more apt to occur in the sphere of the public sector, such as environmental crises, epidemics, nuclear crises, et hnic tensions, population and migr ation flows, riots and sociopolitical turmoil, and dramatic institutional change and conflict (R osenthal & Kouzmin, 1993, pp. 1-2). According to Liu and Horsley (2007), previous public relations literat ure did not separate public sector relations from privat e sector relations. Ma ny public relations theories excellence theory, relationship theory, and cr isis communication theory are ap plicable to both private and public sectors. As such, crisis communication for the private sector can be recommended for the public sector. 28

PAGE 29

Horsley and Barker (2002) argue that thei r crisis model of co mmunication derived from corporate crisis communication is potentially applicable to the public sector (p. 416). They created five interconnected stages in their model. The first stag e is the ongoing public relations effort, which means most successful communicati on plans should begin before a crisis actually occurs. The second stage is iden tification of and preparation fo r potential crisis, in which efficient public relations practiti oners detect potential problems a nd prepare for crisis. The next stage is internal training and rehearsal, where the crisis te am prepares for crisis through training and practicing the skills necessary for good crisis management. The fourth stage is the crisis event. In this stage, speedy communicatio n response is used to stop spreading rumors. Finally, the last stage is evalu ation and revision of public relati ons efforts which means that after the crisis passes, public relations practiti oners should evaluate the crisis plan and provide feedback to improve it (p. 416). Baker (1997) advises public rela tions practitioners in the public sector regarding what to do or what not to do in crisis. The do list is 1. Do protect the legal investigative process. 2. Do protect the privacy rights of individuals and their families. 3. Do have a crisis response and coordina tion team. This team must contain all the important players, such as lawyers, legisla tive affairs, public affa irs, and investigative and technical experts. 4. Do have a media training team ready to prepare leaders for each major media event with a full dress rehearsal before the crisis. A tip here is to use news desk officers to play the role of re porters: since they handle hot i nquires, and know the issues and questions to ask. 5. Do stay cool under fire from the media and critics. 6. Do show respect for people who become critics. 7. Do be prepared for the unexpected (p. 469). The dont list is 1. Dont be a party to sp eculation in the media. 2. Dont deal with hypothetical questions. 3. Dont get emotionally invol ved in a story (p. 469). 29

PAGE 30

Research (Horsely & Baker, 2002) shows that the public sector is not well prepared for crises, considering the actual frequent occurrences of crises in that sect or. According to Horsley and Bakers survey (2002) of senior-level public officials in charge of communication activities of one mid-Atlantic state, half of respondents experien ced a crisis w ithin the past two years. However, only 28% of government agencies hire full-time public relations practitioners who can produce crisis communication plans. These agencies have no budget to employ public relation consultants for crisis communica tion. Additionally, 37% of the ag encies never rehearse their crisis responses. Their major type s of crisis were disasters, such as bomb threats, weather, fire, health, and safety. These findings indicate that the public sector needs to pay more attention to both crisis management a nd crisis communication. Different Levels of Ad ministrative System Even though private sector cris is communication models or strategies can apply to public sector crises indiscriminately, this research considers the unique environment of the public sector. Rosenthal and Kouzmin (1997) argue that different levels of an administrative system should be examined to understand the dynamics of governmental crisis responses. An administrative system is the government system at local, regional, and national levels and even at the transnational administrative level. In a democratic society, governmental power is divided into central and noncentral governmental elements (Lijphart, 1999, p. 185). The criteria dividing governments into central and noncentral governments are territory (Wo llmann, 2003). The central government governs the entire territory of a nation, and noncentral governments, such as regional or local governments, govern limited parts of a nation, usually under the supervision of the central government In a crisis, the power and activ ities are concentrated on the ce ntral government rather than other territorial administrative entities (t Hart, Rosenthal & Kouzmin, 1993). A central 30

PAGE 31

government has the power to mandate evacuation or implement crisis-rela ted law and authorize specific overriding activities, such as helping refuges or checking facilities. For example, the United States has a formal system of central government that inte rvenes in a loca l crisis area. Once the local area is declared a disaster area, the Federal Emergency Management Agency (FEMA) actively coordinates rescue and relief activities and crisis communication (Petak, 1985; t Hart et al, 1993). Friesema et al. (1979) explain why a central government should be involved in the case of natural disasters. The financial cost of natura l disasters is too large to be handled by local governments alone; most of the costs are then burdened to the larger entity, the central government. In some countries, the need fo r concerted and hierarchically coordinated administrative action forms an integral part of the official legal-admini strative definition of disaster (t Hart et al., 1993, p. 16). The level of intervention can be different in different countries. According to Rosenthal and Kouzmin (1997), certain factors affect the in tervention of central gove rnment: discrepancies between central and local governments in terms of capability and resources; structure and culture of a nation that stresses centrali zation or decentralization; politic al risk appreciation by central and local governments; prior experience. The centralization of crisis responses result s in both positive and negative outcomes. Centralization of crisis can mobilize appropriate resources and responses to a crisis and produce positive and effective results. However, sometimes those interventions provoke tensions between central and local governments (H orsley & Barker, 2002). For loca l governments, intervention of a central government implies an encroachme nt on their power and supervision by central government officers. 31

PAGE 32

The administrative system of the Korean government has three layers: 1) a central government; 2) higher-level local governments, such as a Metropolitan City, a Province (Do) followed in the hierarchy by a Small City (Si), County (Gun), and District (Gu), which serve as mediators between the central and lower-level local government s; and 3) lower-level local governments, such as eup, myeon, and dong. The Ko rean government operates a Central Safety Management Committee, which supervises overall disaster and safety policies and promotes coordination among relevant departments. The Mini ster of Culture, Sports and Tourism, who is in charge of domestic and international govern ment public relations is a member of this committee providing public relations advice. Unde r the Central Safety Management Committee, the Central Disaster Safety M easures Headquarters is devoted to ensuring both quick and efficient reaction to large-scale disasters (National Emerge ncy Management Agency, 2008). These days, more and more crises tend to tr anscend the boundaries of one nation. Crises that occur in one nation can be another nation s crisis at any time. The global society is witnessing an economic recession that started in the U.S. in 2008 and has spread worldwide, and a war in the Arab region can a ffect the rest of th e world. Rosenthal and Kouzmin (1993) argue that environmental and technological crises can ea sily spill over on to the entire global society. They found that large-scale crises, such as the Sahei draughts and feminine in Somalia; Bhopal; Chernobyl; Sandoz and Lockerbie ar e likely to become transnatio nal (p. 2). Reflecting on these trends, Rosenthal et al. (2001) argue that transnationalization of crises is irreversible. Transnational crises require tr ansnational cooperation and he lp. Especially, environmental problems such as global warming, protection of rain forest, and ozone layer depletion demand international cooperation. The tsunami in Sout heast Asia in 2005 is a typical example of 32

PAGE 33

international cooperation to cope with large-scale crises. As cris es become more transnational, the problem solving of cr ises should become mo re transnational too. Research Hypotheses and Questions In crisis communication theory, crisis t ypology has great valu e (Gundel, 2005). Gundel (2005) claims that classifying cris es is the first step to keep them under control since they can be named and analyzed (p. 106). This research examined what types of crises frequently occurred in Korea during the period under study ba sed on Coombss three clusters and the ten types of crisis. RQ1: What types of crisis did the Korean government face most during this period? Coombs (1998, 2006, 2007) suggests that different crisis communicati on strategies should be used on crisis typology. If a crisis is attributed to an or ganization, the organization should choose an accommodating strategy. Coombs and Holladay (1996) found that matching crisis communication strategies to a crisis type based on organizational responsibility brings about a more positive image of an organization th an no response at all or mismatching the communication strategies with the crisis type. For example, if a crisis is perceived as preventable, the organization should use rebuilding strategies, such as apology and compensation, for more effective crisis res ponse and outcomes. Accommodative strategies suggest that an organization admit its crisis responsibility and then take action to repair crisis damage. This study identified what crisis co mmunication strategies the Korean government frequently used and whether the Korean governme nt responded to crises based on the level of governmental responsibility. RQ2: What types of crisis communication strategi es were used in the victim cluster, accidental crisis cluster, and preventable crisis cluster respectively by the Korean government? 33

PAGE 34

H1: The Korean government was expected to us e rebuilding strategy for crises in the preventable crisis cluster. Coombs (2007) suggests that bolstering strategi es should not be used alone and be used with other crisis communication strategies. This study will identify whether the Korean government has frequently used bolstering strategies the kinds of bolstering strategies that were used most often, and how they were used as a supplement to the other th ree types of postures. Thus the following research question can be asked: RQ3: Did the Korean government use bolstering strategies and what kinds of bolstering strategies were used most often? RQ4: Did the Korean government use bolstering strategies as a suppl ement to the other three postures? Coombs (2008) suggests that crisis managers utilize a two-step decision-making process when selecting crisis communica tion strategies. The first step is deciding the proper crisis communication strategy in conformity with the cris is typology. The second step is modifying the initial assessment of the appropria te crisis communication strategy, considering the crisis history and the prior reputation of the organization. If the organization has a history of crisis, stakeholders will put more responsibility on th e organization. Therefore, organizations should use more accommodative strategi es for repeated crises. H2: The Korean government was expected to use more accommodative strategies for recurring crises compared to non-recurring crises. Government crises research should consider di fferent levels of an administrative system. Whereas the central government of Korea is in ch arge of policy for foreign relations, defense, and the overall economic and education policy, Ko rean local governments are in charge of 34

PAGE 35

citizen service, administration, and industrial development promotion, etc., within certain territory of administration (Ministry of Government Affairs and Home Affairs 2008a; Ministry of Government Affairs and Home Affairs 2008b). The concentration of po in a central government in times of crisis makes that central government lead a crisis response, regardless of the geographic origin of the cr isis. This study identif ied whether a central government uses different crisis communication strategies in cris es which were originated at different levels of the ad wer ministrative system. H3: The Korean government was expected to us e different communication strategies in crises that originated within the control of local, national, or transnational administrative systems. 35

PAGE 36

CHAPTER 3 METHODS Content analysis was employed in this study to address the research questions. According to Krippendorff (2004), content analysis is not only a resear ch method that takes meaning seriously, but also a method that is both powerful and unobtrusive (p. xiii). He says that it makes sense of what is mediated between people-textual matter, symbols, messages, information, mass-media content, and technol ogy-supported social interactionswithout perturbing or affecting those who handl e that textual matter (p. xiii). Data Collection The study analyzed various crisis press rel eases of the Korean government. The press release is raw material that c ontains information about how the government interprets and reacts to a crisis. To determine clearly the overall features of government crisis response, this study collected data from one administration, that of Moo-Hyun Roh. The research period ran from February 25, 2003, to February 24, 2008. Rohs admi nistration was selected as the subject for this research because it was the most recent ad ministration to have completed a full term in office. This study examined press releases issued by agencies that represent the Korean central government. The data were collected from Korean policy information news release archives on the official Korean government website (http://www.Korea.kr), which stores press releases dating from 2003. The Web site is operated by the Ministry of Culture, Sports, and Tourism, which manages domestic and overseas public relations for the Korean government. There were 41,233 press releases from Febr uary 25, 2003, to February 24, 2008, available on this Web site. A preliminary study was done to determine whether a press release is a crisisannouncing press release. The total data wa s consisted of a total of 1,375 Web pages 36

PAGE 37

containing 41,233 press releases, an d 1 Web page contains 30 pre ss releases. Systematic random sampling was used for selecting the research pop ulation. From every 10th page, 20 press releases from the top of the list were selected; theref ore, the total of 27,480 data elements became the population. Ten percent of that population or 2,740 press releases were analyzed during the preliminary test to determine whether a pre ss release was a crisis-a nnouncing press release. Based on Coombs crisis typology, Rosenthal and Kouzmins crisis types (which are more prevalent in the public sector), and others crisis typology, press releases were designated as crisis releases if they contained any of the following words: cris is, disaster, epidemic, turmoil, crime, tension, conflict, violence, rumors, ki dnapping, terrorism, hacking, strike, policy failure, technical failure, product failure, violation of law, special favor, financial damage, death toll, victim, and public distrust. The author and another coder re ad the title and first paragrap h of the press releases to identify crisis releases out of the 2,740 press release sample, and Holstis reliability reached 0.943. Measurement The unit of analysis for this study is a press release; each press release was examined to determine administrative system, crisis typology, crisis communication st rategies, and crisis history. Out of 27,480 data 20 press releas es per Web page multiplied by 1,374 were classified as crisis, and this research analy zed the identified 619 crisis press releases. The administrative system was analyzed to determine whether the crisis originally occurred within (1) the control of a metropolitan city, province ( do), city (si), county (gun), or district (gu) or of a eup, mye on, or dong; (2) the control of the central government; or (3) the control or territory of other nations. The study did not separate local and regional government levels because Korea has a more centralized government system than Western societies. If the 37

PAGE 38

crisis originally occurred in an area under the control of the central governme nt, that crisis was coded as having a presence in the national administration. If the cris is originally occurred within the territory or contro l of another nations government, it wa s coded as having a presence in the transnational administration. Crisis typology and crisis communication stra tegies followed Coombss definition, as explained in the literature revi ew, and a code book provided detailed guidelines for each item. Crisis typology and crisis communi cation strategies were marked as mutually exclusive and exhaustive. Crisis typology and cr isis communication strategies ar e coded as yes/no questions. To determine the crisis communication strategies employed, the code r checked all strategies that applied to each press release becau se an organization can use differe nt strategies in a single press release or in several consecutive press releases. Crisis history was determined by two coders through discussion of whether the government experi enced similar crises dur ing the five years of the Roh administration. Coding and Inter-coder Reliability Two coders who can read Korean, including the researcher, coded the data according to the code book (Appendix B). To increase inter-coder reliability, these coders received thorough training on how to use both the code sheet and the code book. A pre-test evaluated the coding scheme for relia bility in both the coding book and the code sheet. Seventy five crises were pre-tested. Scotts pi reliability method was used to compute inter-coder reliability. Although Holstis reliability formula is easy to apply (Wimmer & Dominick, 2006), it has a deficiency in that it doesnt consider the agreement between coders that can occur by chance. Scotts pi reliability method can correct this deficiency. Two coders worked independently, and Scotts pi reliability was in an accept able range of .70 to 1.00, except for two variables. Technical product harm accident and victimage reached 1.00. Two 38

PAGE 39

variables showed reliability scores lower than .70. The other category of crisis types had reliability of .00. However, only one case out of th e seventy five pretested releases was marked in this category. The other ca tegory of crisis communication st rategies had reliability of .51. After coding disagreements were resolved betw een two coders, inter-co der reliability in both categories reached more than 0.78. Data Analysis The Statistical Package for the Social Sc iences (SPSS) 13.0 for Windows was used for statistical computing in the study. The following statistica l analyses were app lied to the research questions. To test RQ1, RQ3, and RQ4, frequency statistics were used because the research questions primarily address the presence or abse nce of items. To test RQ2, H1, and H3, the chisquare test was employed. Chi-squa re analysis can test whether one factor is significantly more important than others. To verify H2, a t -test was conducted because the main focus of H2 is to compare two classificationsrecurring and nonrecu rring crises. RQ2, H1, H2, and H3 used only 559 press releases to exclude overlapping crisis communication strategies; several crisis communication strategies can apply to one crisis. 39

PAGE 40

CHAPER 4 RESULTS The study proposed four research questions and three hypotheses. Cont ent analysis of 619 crises cases was conducted to examine the rese arch questions and hypotheses. This chapter summarizes the results of the content analysis. Research Question 1 Research question 1 asked what types of cris is the Korean government faced most during the researched peri od. Challenges (26.5%, n = 164)) were the most frequent type of crisis, followed by organizational misdeeds (23.4%, n = 145), malevolence (12.1%, n = 75), rumors (11.0%, n = 68), human-error accident (10.7%, n = 66), natural disasters (8.7%, n =54), humanerror product harm (1.5%, n = 9), technical-erro r product harm (1.3%, n = 8), technical accident (1.1%, n = 7), workplace violence (1.0%, n = 6), and others (2.7%, n = 17). An example of crisis types from the data included: challenge (a general strike by the national labor union protesting Koreas involvemen t in the USA FTA agreement), organizational misdeed (one ministry funding its employees trip abroad with public money when the trip was not wholly for business purposes), malevolence (w hen Koreans were kidnapped in other nations and when Korean army camps in another nation were attacked by terrorists), rumor (suspicion that the government was eavesdropping on peopl es cell phone calls), human-error accident (when the government neglected the appeal for help by a North Korean who had escaped from North Korea), natural disaster (typhoon, bird flu, red tide, etc.), human-error product harm (error in public scholastic aptitude exam s that resulted in fury among testers), technical-error product harm (the collapse of a bridge as the result of faulty design or construction), technical accident (the shutdown of a nuclear power plant caused by the malfunction of a cooling system), and workplace violence (an incidence of one soldier s hooting or beating other soldiers). The others 40

PAGE 41

category (2.7%) encompasses events whose cause cannot be attributed to the government, such as arson committed by an individual citizen. If the results of arson affected the nation, the government would become involved in solving the crisis. When crises were further categorized into victim cluster, acci dental cluster, and preventable cluster, most of the crises occurred in the prev entable cluster (35.5%, n = 220), followed by the victim cluster (32.8%, n = 203), accidental cluster (28.9%, n = 179), and others (2.7%, n = 17). Malevolence (12.1%) was the most fre quent cause of crises occurring in the victim cluster, the frequency of challenges (26.5 %) were high in the accidental cluster, and organizational misdeeds (23.4%) were hi gh in the preventable cluster. Research Question 2 Research question 2 examined the relati onship between crisis type and crisis communication strategies: What type s of crisis communica tion strategies were used in the victim cluster, accidental crisis cluster, and preven table crisis cluster, re spectively, by the Korean government? Before analyzing research question 2, overall make-up of crisis communication strategies are as follows: the Korean governme nt used denial most often (23.9%, n = 166), followed by attacking the accuser (19.6%, n = 136), justification (17.1%, n = 119), apology (12.0%, n = 83), compensation (9.1%, n = 63), excusing (6.6%, n = 46), others (5.8%, n = 40), ingratiation (2.9%, n = 20), reminding (1.7%, n = 12), victimage (1.2%, n = 8), and scapegoating (0.1%, n = 1). The government used denial stra tegies in a case of secret negotiations between Korea and the USA about the cow meats imports, and it deni ed the existence of th e negotiations. Attacking the accuser was used against various protes ts blaming the government for wrongdoing. The government usually warned protesters not to bre ach the law during the protest, and it accused the protesters if they violated th e law. Justification was used wh en the government attempted to 41

PAGE 42

minimize the compensation and claim that there were no damages to civilian despite public request to compensate damages done on civilians For example, when Korea was handed over a U.S. army base in Korea, the U.S. didnt cure environmental damage, and therefore it will cost several billions to cure and will have very negati ve effects on the health of citizens. Government used an apology for government officials bribery, violence in jail, and an accident in a power plant, and compensation was given in case of su ch disasters. The Others category included a case in which the government took correctiv e action but did not provide an apology. Scapegoating was used when a government-financed, small-sized banking in stitution for mutual savings and loans had huge damage from wrongful investment; the government said it was not the governments fault but the inevitable re sult of a problematic financial market. The usage of crisis communication strategies was different depending on whether the crisis was a victim cluster crisis or not ( (10) = 130.002, p < .001). While attacking the accuser, compensation, and ingratiation were significantly more likely to be used in victim cluster compared to other types of cris is clusters, excusing, justification, and apology were significantly less likely to be used in victim cluster. Attacking the accuser (34.6%, n = 65) was the most frequently used strategy when crises occu rred in victim clusters (see Table 4-1). The accidental cluster showed different usage of crisis communication strategies from other types of crisis clusters ( (10) = 66.179, p < .001). While justification was more likely to be used in accidental cluster, compensation and apology were less likely to be used in accidental cluster. Justification (33.3%, n = 52) was the most frequently used strategy when crises occurred in accidental clusters, according to cross tabulations (see Table 4-2). In the preventable cluster, th e usage of crisis communication strategies was different from other types of crisis clusters ( (10) = 128.402, p < .001). Whereas apology was significantly 42

PAGE 43

more likely used in preventable cluster, attacking the accuser and compensation were significantly less likely used in preventable cluste r. Apology (29.0%, n = 58) was the most frequently used strategy when crises occurred in preventable cluste rs, according to cross tabulations (see Table 4-3). Hypothesis 1 Hypothesis 1 addressed the Korean governments expected use of a rebuilding strategy for crises in the preventable cris is cluster. To examine the hypothesis 1, the study compared the denial posture, diminishment posture, rebuild ing posture, and bolstering posture of crisis communication strategies within the preventable crisis cluster. The usage of crisis communication postures was differe nt depending on whether the cris is was preventable cluster crisis or not ( (4) = 21.919, p < .001). The relationship between the preventable cluster crisis and rebuilding posture was significant ( (1) = 15.663, p < .001). The relationship between denial posture and the preventable crisis cluster also were significant ( (1) = 16.250, p < .001). While rebuilding strategy was significantly more likely to be used in preventable cluster, denial strategy was significantly less likely to be us ed in preventable clus ter (see Table 4-4). Research Question 3 and Research Question 4 Research question 3 asked whether the Korean government used bolst ering strategies and, if so, what kinds of bolstering strategies were used often. Among a ll crisis communication strategies, the Korean government used bolstering strategies 5.8% ( n = 40) and other strategies 94.2% ( n = 654) of the time. Among bolstering strate gies, ingratiation was used most (50.0%, n = 20), followed by reminding (30%, n = 12) and victimage (20%, n = 8). Research question 4 asks whether the Korean government used bolstering strategies as a supplement to other communication postures. According to the fr equency test, 85% (n = 34) of 43

PAGE 44

bolstering strategies were used to supplement other communication strategies, whereas 15% (n = 6) of bolstering strategi es were used alone. The government used the ingra tiation strategy mainly in times of disasters or kidnapping cases; it ingratiated itself with the people in an effort to recover from disasters and with the family of those kidnapped in other nations. Th e government used reminding strategies when it confronted a vote of confidence; the president reminded people of policies he had successfully implemented. Victimage was used in cases where governmental budget spending was delayed because of political conflict in congress and subsequent vote delay. Hypothesis 2 Hypothesis 2 assumed that the Korean government was expected to use more accommodative strategies for recurring crises th an for non-recurring crises. Independent-samples t -tests and a chi-square test were used to determine whether a difference existed between recurring and nonrecurring crises. The analysis showed significant differences between recurring and non-recurring crises (t = 4.47, p < .001). The average number of accommodative strategies used in recurring crises (M = 2.83, sd = 1.241) were higher than non-recurring crises (M = 1.85, sd = 1.082). The analysis showed significant differences between recurring and non-recurring crises in denial posture (t = -2.54, p < .05) The average number of denial strategies used in non-recurring crises (M = .50, sd = .501) were higher than for recurring crises (M = .36, sd = .484). The differences between recurring and non-recurring crises in diminishment posture were also significant (t = -4.17, p < .001). The average numbe r of diminishment strategies used in nonrecurring crises (M = .27, sd = .446) were higher than recurring crises (M = .08, sd = .279). The analysis showed significant differences between recurring and non-recurri ng crises in rebuilding 44

PAGE 45

posture (t = 6.21, p < .001). The average number of diminishment strategies used in recurring crises (M = .44, sd = .489) were higher than non-recurring crises (M = .17, sd = .376). Recurring crises included abuse of human rights in the army and jails, general strikes, typhoons, kidnappings, bribery, errors in exams, and wrongful use of budget, among others. Hypothesis 3 Hypothesis 3 examined whether the Korean government was expected to use different communication strategies in crises that originat ed within the control of local, national, or transnational administrative systems. Among 599 crises, 24 (4.3%) origina lly occurred within the control of a local administra tive system, 486 (87.5%) within the control of a national administrative system, and 46 (8.2%) within the cont rol of a transnational ad ministrative system. The usage of crisis communication strategies was different depending on whether the crisis originated from a local administrative systems or not ( (10) = 82.344, p < 0.001). Whereas compensation strategy was more likely to be us ed in a local administrative system, denial strategy was less likely to be used in a lo cal administrative system (see Table 4-5). Compensation (62.5%, n=15) was used the most fr equently when crises occurred within the control of local administrative syst ems, according to cross tabulations. In the national administrative system, the us age of crisis communication strategies was different compared to a local and tr ansnational administrative systems ( (10) = 65.399, p < .001). While denial strategy was more likely to be used in national administrative system, attacking the accuser, compensation, and ingratiati on strategy were less likely to be used in national administrative system (s ee Table 4-6). Denial (28.8%, n = 141) was used the most frequently when crises occurred within the co ntrol of the national administrative system, according to cross tabulations. 45

PAGE 46

The usage of crisis communication strategies was different depending on whether the crisis originated from transnational ad ministrative systems or not ( (10) = 46.228, p < .001). While attacking the accuser and ingratiatio n strategies were more likely to be used in a transnational administrative system, denial and apology strategies were less likely to be used in a transnational administrative system (see Table 4-7). Attacking the accuser (52.2%, n = 24) was used the most frequently when crises occurred within the co ntrol of transnational administrative systems, according to cross tabulations. Crises originating in the transnational administrative system were the kidnapping of a Korean, disput es with neighboring nations, a nd dispatching the army to Iraq, to name a few. If the crisis originated within a local admi nistrative system, compensation was used most frequently; if the crisis originated within the national administrative system, denial was used most frequently; and if the crisis originated in transnational administrative systems, attacking the accuser was the most commonly used strate gy. Thus, hypothesis 3 was confirmed: The government used different strategies according to administrative systems. 46

PAGE 47

Table 4-1. List of cross tabul ation of victim cluster crisis and communication strategies Victim Total Absence Presence Attacking the accuser*** 55 (14.8%) 65 (34.6%) 120 (21.5%) Denial 94 (25.3%) 50 (26.6%) 144 (25.8%) Scapegoating 1 (0.3%) 0 (0.0%) 1 (0.2%) Excusing 29 (7.8%) 5 (2.7%) 34 (6.1%) Justification*** 89 (24.0%) 8 (4.3%) 97 (17.4%) Compensation*** 12 (3.2%) 43 (22.9%) 55 (9.8%) Apology*** 62 (16.7%) 6 (3.2%) 68 (12.2%) Reminding 1 (.3%) 0 (.0%) 1 (.2%) Ingratiation 1 (.3%) 3 (1.6%) 4 (0.7%) Victimage 1 (.3%) 0 (.0%) 1 (.2%) Others 26 (7.0%) 8 (4.3%) 34 (6.1%) Total 371 (100.0%) 188 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 Table 4-2. List of chi-square test of accident al cluster crisis and communication strategies Accidental Total Absence Presence Attacking the accuser 82(20.3%) 38(24.4%) 120(21.5%) Denial 108(26.8%) 36 (23.1%) 144 (25.8%) Scapegoating 0 (.0%) 1 (.6%) 1 (.2%) Excusing 23 (5.7%) 11 (7.1%) 34 (6.1%) Justification *** 45 (11.2%) 52 (33.3%) 97 (17.4%) Compensation** 50 (12.4%) 5(3.2%) 55 (9.8%) Apology*** 64 (15.9%) 4 (2.6%) 68 (12.2%) Reminding 1 (.2%) 0 (.0%) 1 (.2%) Ingratiation 3 (.7%) 1 (.6%) 4 (.7%) Victimage 0 (.0%) 1 (.6%) 1 (.2%) Others 27 (6.7%) 7 (4.5%) 34 (6.1%) Total 403 (100.0%) 156 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 47

PAGE 48

Table 4-3. List of cross ta bulation of preventable cluste r and communication strategies Preventable Total Absence Presence Attacking the accuser*** 105 (29.2%) 15 (7.5%) 120 (21.5%) Denial 87 (24.2%) 57 (28.5%) 144 (25.8%) Scapegoating 1 (.3%) 0 (.0%) 1 (.2%) Excusing 18 (5.0%) 16 (8.0%) 34 (6.1%) Justification 61 (17.0%) 36 (18.0%) 97 (17.4%) Compensation*** 51 (14.2%) 4 (2.0%) 55 (9.8%) Apology*** 10 (2.8%) 58 (29.0%) 68 (12.2%) Reminding 0 (.0%) 1 (.5%) 1 (.2%) Ingratiation 4 (1.1%) 0 (.0%) 4 (.7%) Victimage 1 (.3%) 0 (.0%) 1 (.2%) Others 21 (5.8%) 13 (6.5%) 34 (6.1%) Total 359 (100.0%) 200 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 Table 4-4. List of cross ta bulation of preventable cluste r and communication strategies Preventable Total Absence Presence Denial*** 193 (53.8%) 72 (36.0%) 265 (47.4%) Diminishment 79 (22.0%) 52 (26.0%) 131 (23.4%) Rebuilding*** 61 (17.0%) 62 (31.0%) 123 (22.0%) Bolstering 5 (1.4%) 1 (0.5%) 6 (1.1%) Others 21 (5.8%) 13 (6.5%) 34 (6.1%) Total 359 (100.0%) 200 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 48

PAGE 49

Table 4-5. List of cross tabul ation of local administrative system crisis and communication strategies Local Total Absence Presence Attacking the accuser 118 (22.1%) 2 (8.3%) 120 (21.5%) Denial** 144 (26.9%) 0 (.0%) 144 (25.8%) Scapegoating 1 (.2%) 0 (.0%) 1 (.2%) Excusing 33 (6.2%) 1 (4.2%) 34 (6.1%) Justification 96 (17.9%) 1 (4.2%) 97 (17.4%) Compensation*** 40 (7.5%) 15 (62.5%) 55 (9.8%) Apology 65 (12.1%) 3 (12.5%) 68 (12.2%) Reminding 1 (.2%) 0 (.0%) 1 (.2%) Ingratiation 4 (.7%) 0 (.0%) 4 (.7%) Victimage 1 (.2%) 0 (.0%) 1 (.2%) Others 32 (6.0%) 2 (8.3%) 34 (6.1%) Total 535 (100.0%) 24 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 Table 4-6. List of cross ta bulation of national administ rative system crisis and communication strategies National Total Absence Presence Attacking the accuser*** 26 (37.1%) 94 (19.2%) 120 (21.5%) Denial*** 3 (4.3%) 141 (28.8%) 144 (25.8%) Scapegoating 0 (.0%) 1 (.2%) 1 (.2%) Excusing 3 (4.3%) 31 (6.3%) 34 (6.1%) Justification 7 (10.0%) 90 (18.4%) 97 (17.4%) Compensation*** 20 (28.6%) 35 (7.2%) 55 (9.8%) Apology 3 (4.3%) 65 (13.3%) 68 (12.2%) Reminding 0 (.0%) 1 (.2%) 1 (.2%) Ingratiation*** 2 (2.9%) 2 (.4%) 4 (.7%) Victimage 0 (.0%) 1 (.2%) 1 (.2%) Others 6 (8.6%) 28 (5.7%) 34 (6.1%) Total 70 (100.0%) 489 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 49

PAGE 50

Table 4-7. List of cross tabul ation of transnational administrative system crisis and communication strategies Transnational Total Absence Presence Attacking the accuser*** 96 (18.7%) 24 (52.2%) 120 (21.5%) Denial** 141 (27.5%) 3 (6.5%) 144 (25.8%) Scapegoating 1 (.2%) 0 (.0%) 1 (.2%) Excusing 32 (6.2%) 2 (4.3%) 34 (6.1%) Justification 91 (17.7%) 6 (13.0%) 97 (17.4%) Compensation 50 (9.7%) 5 (10.9%) 55 (9.8%) Apology* 68 (13.3%) 0 (.0%) 68 (12.2%) Reminding 1 (.2%) 0 (.0%) 1 (.2%) Ingratiation*** 2 (.4%) 2 (4.3%) 4 (.7%) Victimage 1 (.2%) 0 (.0%) 1 (.2%) Others 30 (5.8%) 4 (8.7%) 34 (6.1%) Total 513 (100.0%) 46 (100.0%) 559 (100.0%) p < .05, ** p < .01, *** p <.001 50

PAGE 51

CHAPTER 5 DISCUSSION This research explored the types of crises and the crisis communication strategies used during Rohs administration in Korea. The st udy also examined the Korean governments response to each crisis accordi ng to the level of responsibility associated with the crisis. Hypotheses and research questions assumed that the Korean government would follow Coombs suggestions as to how best to act during crises. Research question 1 attempted to determine the type of crises to which the Korean government was vulnerable during the Roh administ ration (2003-2008). At that time, the Korean government confronted diverse challenges from discontented stakeholde rs (26.5%). Protesting and criticism from stakeholders, the media, a nd opposing parties were the main challenges. The next most frequently occurring crisis involved organizational misdeeds (23.4%). Governmental officials breaking the law, moral hazards, and d eceptive behavior were typical of organizational misdeeds. Malevolence (12.1%) was an attack on the government from an opponent or a third party, including kidnapping, terro rism, and invasion by North Kor ea of the territory of South Korea. Human-error accidents (10.7%) came through blunder or negligence, and natural disasters (8.7%) were the results of typhoons, floods, and bird flu. Among Coombs 10 types of crisis typology, na tural disasters, rumors, malevolence, challenges, and human-error accidents were the most frequently occurring crises. Therefore, the government should give more attention to these t ypes of crises. In contrast, human-error product harm (1.5%), technical-error product harm (1.3 %), technical accident (1.1%), and workplace violence (1.0%) were rare. In the public s ector, human-error product harm, technical-error product harm, and technical accident were uncomm on because the public sector rarely produces 51

PAGE 52

commodities. Workplace violence committed by an employee or a former employee also was not common in the public sector. Technical accident and techni cal-error product harm can be combined into one category, technical accident, an d human-error accidents and human-error product harm can be combined into one category, human-error accident. One cate gory can be added in cr isis types and can be named citizen or other actors faults (2.7%). In the continuum of low to high responsibility for crisis types, it can be located next to natural disasters and included into the victim cluster. Research question 2 confirmed that the Kor ean government employed different crisis communication strategies according to the cluster of crisis types. In the vi ctim cluster, attacking the accuser was a commonly used communication strategy. The government resorted to law enforcement for the strike of railroad workers an d threatened them with a lawsuit for the spread of rumors about a high-level governmental offi cials resignation. The Korean government used justification most frequently in the accident al cluster and apology mo st frequently in the preventable cluster. The results showed that the Korean governme nt appropriately used crisis communication strategies according to the level of government responsibility. In the victim cluster, which assumed low government responsibility, the government used the attack -the-accuser strategy most often, a denial strategy that allowed the government to accept the least amount of responsibility for the crises. In the accidental cluster, wh ich assumed medium government responsibility, the government empl oyed the justificati on strategy, a diminishment strategy that allowed the government to accept medium responsibil ity for the crises. In th e preventable cluster, which assumed high government responsibility, the government used the apology strategy most, a rebuilding strategy that allowed the government to accept high responsibility for the crises. 52

PAGE 53

Among the 10 crisis communica tion strategies, the Kor ean government used the scapegoating strategy in only one case. This finding implies that the government seemed to see this strategy as egocentric and possibly damaging to its authority status. Based on these research results, the scapegoating category may be elimin ated from the typology of crisis communication strategies used in the public sector. Among the crisis communication strategies, the others category was used in 5.8% of the cases. Most involved the governme nt taking actions in the crises but without apology or compensation, so it was difficult to check the re building strategies. Afte r researching Mercks Vioxx recall case, Vald et al. (2006) suggested that Merck should use a new mortification strategy rectificati on without assuming responsibility by making rectification through its corrective action of recalling Vioxx but without ever admitting fault, apologizing, or asking forgiveness for causing grave injury and deat h to Vioxx consumers (p. 357). The Korean government also often used rectification w ithout assuming responsibility by taking action during crises, but not providing an apology or aski ng for forgiveness. The government used this strategy especially when the causes of the crises were stakeholders. Therefore, one category can be added into crisis communication st rategies and can be named it as corrective action. In the continuum of crisis communication strategies from low to high, it can located in the just before compensati on and apology strategies and can be clustered into rebuilding strategy. The proposed changes in Coombs crisis type s and crisis communica tion types can apply to the public sector crises encountered by any na tion. This is because those changes stem mainly from the characteristics of the administration ra ther than from the individual administration system or culture of an individual nation. The citizen or other actors faults and corrective 53

PAGE 54

action are related with citi zen. Technical-error product harm and human-error product harm were related with production of commodity, and that production is not the main purpose of governments. As expected for hypothesis 1, the Korean gove rnment used the rebuilding strategy for crises in the preventable crisis cluster. The government used apology when crises occurred in the preventable crisis cluster, such as providi ng an apology for unlawful inquiry for individual information. Therefore, the governments respon se strategy conformed to that which Coombs (2007) recommended: Organizations should use rebuilding strate gies during any preventable crisis. Coombs (2007) suggested that bolstering strategies can help build a positive relationship between an organization and its stakeholders. However, he recommended that bolstering should be used as a supplement to the other three posture s because of its trait of egocentrism. Research question 3 confirmed that the Korean government did not frequently employ bolstering strategies (5.8%). When they were used, they were used to supplement other strategies (85%). The research results showed that bolstering strategies cannot become a stand-alone crisis communication strategies as Coombs suggested; the Korean government used them alone in only 15% of crisis communication situations. When the oil spill and typhoon caused damage to civilians, the government used ingratiation to supplement compensation. Among bolster ing strategies, the Korean government employed ingratiation (50.0 %) the most frequently during crises. The government displayed ingratiation by appreciati ng public support and victims patience. The Korean government used different strate gies during recurring crises and non-recurring crises. The research results confirmed that the Korean government used more rebuilding strategies in the crises that had a history. The government used the denial strategy most in 54

PAGE 55

nonrecurring crises. However, when the crises ha d a history, the government used compensation and apology more than the other strategies. The government apologized for recurring crises, such as the abuse of human rights in the army and jail s, and used denial stra tegies in non-recurring crises, such as the acceptance of wart ime operational control from the U.S. According to Coombs theory, when a crisis has a history, stakeholders attach more responsibility to the organization. The organization shoul d, therefore, adopt more accommodative strategies. From that perspec tive, the Korean government responded well to recurring crises. From the study, it is difficult to identify whether the Korean government employed the two-step process of Coombs (initial assessment of responsibility and modification of the initial assessment according to the history of the crisis), but this research revealed that the Korean government did use more accommodativ e strategies in recurring crises. The Korean government faced crises that mos tly (87.5%) originated w ithin the control of the national administrative system. Crises that originated within the control of a local administrative system were not many (4.3%). The government used different strategies according to crises that occurred in different le vels of the administration systems. The results showed that the government took moderate responsi bility on national-level crises by using denial strategies (28.8%), a lower level of responsibility in international crises by using the strategy of attacking the accuser (52.2%), and a higher level of responsibility during local-level crises by using the strategy of compensation (62.5%). For exam ple, in the case of an outbreak of bird flu in a certain region under the control of local government, the central government provided immunization services and compensation with money, and when the international reporters association criticized the Ko rean government for oppressing freedom of the press, the government attacked the associa tion for displaying malevolence. 55

PAGE 56

Implications This study suggests a summary of how Korean government reacts to crises. By analyzing the entire crises and crisis responses for one administration, it helps th e understanding of the Korean governments crisis communication. Many researchers focus on the individual crisis and provide detailed information abou t it. However, case studies cannot describe whole trends of phenomena, and this research gives public sector crisis managers a nd scholars a glimpse of crises in one countrys public sector. The research can offer public sector crisis managers more time to respond to crises. Because crises are unpredictable and sudden events that put an organization into uncertain and time-sensitive situations, the research can provi de several benefits to public sector crisis managers. First, the public sector usually prepares for crises by developing crisis response plans and creating crisis response teams, then perf orming drills designed to respond to hypothetical crises. However, the public sector cannot prepare for all the crises it is expected to confront. It can just prepare for some of the crises that are assumed to be the most likely to happen. Therefore, most crisis preparati on is apt to concentrate on the natu ral disaster. However, as seen in the study results, natural disast er is part of only 8.7% of all crises; challenges, organizational misdeeds, and malevolence are more common causes for crises in the Kor ean public sector. The research can present the crises that the governme nt is most likely to face and can help public sector crisis managers cope w ith crises more effectively. Second, the research can provide st andard criteria to wh ich public crisis managers can refer. Public sector crisis managers can use diverse communication plans, and their choices sometimes depend on the results of previous case studies, their experience, or their instinct. They use diverse communication strategies without knowing whether they are appropriate or not. The research can provide public crisis managers with the ability to easily de tect what would be a 56

PAGE 57

good or bad crisis response by comparing Coom bs study with their co mmunication strategies. Crisis managers can use the most appropriate cr isis communication strategy based on each crisis situation by considering both the cr isis history of an organizati on and the different levels of responsibility of th e specific administrative system. The study also suggests categories that shoul d be added and categor ies that should be deleted when applying Coombs crisis typology a nd communication strategy to the public sector. In contrast to the private sect or, the public sector does not produce commodities, and crisis typology related to production should be diminished. On the contrary, the crises types and crises communication strategies that originated from the private sectorthe at tribution can be on the stakeholders from the private sectorshould be added to the crisis typology. The category rectification without assuming responsibility can be a dded in the crisis communication strategies (Vald et al., 2006, p. 357). Limitations and Future Research The study applies Coombs theory to the Korean governments response to various crises. It analyzed crises that were pr esented in press releases, but st akeholders perceptions can differ from the facts presented in the releases. Stak eholders can become mo re focused on certain aspects of a crisis than on the ac tual characteristics of the crisis They also can perceive crisis communication strategies differently when the government admits responsibility. The research did not measure perceptions. Future research sh ould analyze the medias or the publics opinion to learn how stakeholders perceive crises. Certain methodological limitations constrained this research. Wh en selecting crises, this researcher read only the first paragraph of news releases because of a la rge number of releases. However, if the entire news releases had been examined thoroughly, the number of crises might have risen to more than 619. The research al so excluded overlapping co mmunication strategies 57

PAGE 58

when analyzing some hypotheses and questions because of methodol ogical difficulties. Therefore, some crises were not examined. Also, the study does not determine why the government used differe nt strategies for different administrative systems. It only concludes that the government responded differently to crises originating within the control of local, na tional, and transnational administrative systems. The next step in this research should be to determine why the differences were introduced. The research into the relationship between administra tive system level and political nature, such as liberal or conservative, and relationship betw een administrative system level and degree of centralization, such as ce ntralization or decentralization, can serv e as topics for future research. This research only analyzed the written aspect of government crisis responses. However, people receive information not only from what is written but also th rough the actions of government representatives. The gestures and the way statements are presented can be the cues on which the stakeholders build their perceptions When combining these aspects of a crisis response, the crisis response can be more accurately analyzed. The research tested characteristics of crisis types and crisis communication strategies in Rohs regime in Korea. The research results may not apply to other countries because of differences in culture and administrative syst ems, except for the revision of Coombs typology. The U.S. has a different culture and administrative system from Korea; thus, the characteristics of the U.S. government crisis and crisis response may differ from those of the Korean government. Likewise, even within Korea, Roh s administration and the current administration might possibly show different results in crisis types and crisis responses. More study should be conducted across culture and across administrations to identify more accurate characteristics and trends of governmental crises. If private sector crises are also analyzed in the long term, research 58

PAGE 59

59 can compare government crises with private sector crises. This will help to clarify the nature of governmental crises. Future research can analyze whether public sector crisis managers execute crisis communication strategies according to pre-pla nned crisis communications guidelines. By analyzing gaps between actually used communication strategies a nd pre-planned plans, managers can understand what causes the gaps, and more r ealistic communication plans can be developed.

PAGE 60

APPENDIX A CODING SHEET FOR CONTENT ANALYSIS Coder Name: __________ Coding Period: __________ 1. Administrative System 1-1. Check the presence or absence of government system. 1-1-1. Local Administration: (1) Presence__ (2) Absence __ 1-1-2. National Administration: (1) Pr esence__ (2) Absence __ 1-1-3. Transnational Administration: (1) Presence__ (2) Absence __ 1-1-4. Others (specify) ________________________ : (1) Presence__ (2) Absence __ 2. Crisis Typology 2-1. Check the presence or absence of crisis typology. 2-1-1. Victim Cluster 2-1-1-1. Natural disasters: (1 ) Presence__ (2) Absence __ 2-1-1-2. Workplace violence: (1) Pr esence__ (2) Absence __ 2-1-1-3. Rumors: (1) Presence__ (2) Absence __ 2-1-1-4. Malevolence: (1) Presence__ (2) Absence __ 2-1-2. Accidental Crisis Cluster 2-1-2-1. Challenges: (1) Presence__ (2) Absence __ 2-1-2-2. Technical-error accidents: (1) Presence__ (2) Absence __ 2-1-2-3. Technical-error product harm: (1) Presence__ (2) Absence __ 2-1-3. Preventable Crisis Cluster 2-1-3-1. Human-error product harm : (1) Presence__ (2) Absence __ 2-1-3-2. Human-error accidents: (1) Presence__ (2) Absence __ 2-1-3-3. Organizational misdeeds: (1) Presence__ (2) Absence __ 2-1-4. Others (specify) __________________________ : (1) Presence__ (2) Absence __ 3. Crisis Communication Strategy 3-1. Check the presence or ab sence of crisis comm unication strategy (check all that apply). 3-1-1. Denial Posture 3-1-1-1. Attacking the accuser: (1) Presence__ (2) Absence __ 3-1-1-2. Denial: (1) Presence__ (2) Absence __ 3-1-1-3. Scapegoating: (1) Presence__ (2) Absence __ 3-1-2. Diminishment Posture 3-1-2-1. Excusing: (1) Presence__ (2) Absence __ 3-1-2-2. Justification: (1 ) Presence__ (2) Absence __ 3-1-3. Rebuilding Posture 60

PAGE 61

3-1-3-1. Compensation: (1) Presence__ (2) Absence __ 3-1-3-2. Apology: (1) Presence__ (2) Absence __ 3-1-4. Bolstering Posture 3-1-4-1.Reminding: (1) Pr esence__ (2) Absence __ 3-1-4-2. Ingratiation: (1) Presence__ (2) Absence __ 3-1-4-3. Victimage: (1) Presence__ (2) Absence __ 3-1-5. Others (specify) __________________________ : (1) Presence__ (2) Absence __ 61

PAGE 62

APPENDIX B THE CODE BOOK FO R CONTENT ANAYSIS Instructions: After reading each press release, please chec k the items that are applicable to the press release. If you have any que stions regarding the coding pro cess, feel free to contact me. Date of Press Release : Write the date when the press release was issued. Coding Period : Indicate the period in the news release that you are coding 1. Administrative System 1-1. Local Administration: The crisis originally occurred within the control of a Metropolitan City, Province (Do), City (Si), County (Gun), District (Gu), eup, myeon, or dong. 2-1. National Administration: The cr isis originally occurred within the control of the central government. 3-1. Transnational Administration: The crisis orig inally occurred outside the territory of Korea or within the control of other nations government. 2. Crisis Typology Read the press release an d check either 1 for presence or 2 for absence. 2-1. Victim Cluster Feature Item Description Natural disasters Natural disasters are crises resulti ng from acts of God, for example, earthquakes, hurricanes, tornados, floods, mud slides, and blizzards e.g., typhoon swamps the East coast of the country. Workplace violence Workplace violence is violence committed by an employee or a former employee of an organization to other employees. e.g., a shooting incident occurs among coworkers. Rumors Rumors are false information inte ntionally distributed about an organization and its products to damage the organization. e.g., a rumor about high ranking government officials forced them to resign. Malevolence Malevolence is an o pponent or a third partys attack on the organization, such as tampering, kidnapping, terrorism, or computer hacking. e.g., commercial ship kidnapped by Somalian pirates. 62

PAGE 63

2-2. Accidental Crisis Cluster Feature Item Description Challenges Challenges are caused by confr ontations with discontented stakeholders who claim that the organization operate inappropriately. e.g., labor union strikes in viol ation of government labor policy. Technical-error accidents Technical-error accidents are industrial accidents caused by the failure of technology or equipm ent utilized or supplied by the organization. e.g., air Force fighter plane cras hes due to defect in engine. Technical-error product harm Technical-error product harm is w hen the technology utilized or supplied by the organization fails and results in a defect of a potentially harmful product. e.g., several government websites crash due technical error. 2-3. Preventable Crisis Cluster Feature Item Description Human-error accidents Human-error accidents are caused by human errors, such as failed judgment, negligence, blunders, or sabotage e.g., government officer used wron g prescription for oil spill and thus damage increased. Human-error product harm accidents Human-error product harm is a product defect or harm to consumers caused by human error. e.g., government electronic passport not operating properly due to government workers accidentally eradicating one element of passport. Organizational misdeeds Organizational misdeeds occur when management takes actions it knows may place stakeholders at ri sk or knowingly violates the law. e. g., top government officers ac cused of inappropriate sexual harassment of female employees. 63

PAGE 64

3. Crisis Communication Strategies Read the press release and mark either 1 for pres ence or 2 for absence. Coder can check all that apply. 3-1. Denial Posture Feature Item Description Attacking the accuser The organization confronts the pers on or organizations arguing for the existence of a crisis. The organization can threaten the a ccuser to use force, such as a lawsuit. e.g., government criticized labor union for violating the law. Denial strategy The organization denies the existence of the crisis. The organization states that there is no crisis and sometimes discusses the rationale why the crisis does not exist. e.g., denial of any misdeed, or late response to a misdeed. Scapegoating The organization tries to shift th e blame from the organization to other person or organizations e.g., blaming the company for causing labor union strike and blocking public transportation. 3-2. Diminishment Posture Feature Item Description Excusing The organization tries to diminish its responsibility for the crisis. This strategy denies that the organization had any intention to do harm or has not the power to control the crisis. e.g., making excuses with saying that the government has no intention of revealing pr ivate citizen information. Justification The organization tries to mi nimize the perceived damage associated with the crisis. The response can include stating that there was no severe damage or injuries or stating that the people affected by the crisis deserved what they received. e.g., emphasizing benefits of cons tructing a new road by building a tunnel outweigh any damages. 64

PAGE 65

65 3-3. Rebuilding Posture Feature Item Description Compensation The organization provides various bene fits, such as money, gifts, or services, to reduce or assuage the negative emotions of the victims toward the organization. e.g., governmental support for spendi ng seven billion dollars in a disaster area. Apology The organization admits full responsibility for the crisis and requests forgiveness publicly. e.g., making an apology for failure s with regulation of university entrance examination. 3-4. Bolstering Posture Feature Item Description Reminding The organization reminds stakeholders about the organizations good deeds in the past. e.g., reminding positive accomplishments of Free Trade Agreement (FTA) with other nations when co nfronted with opposition to the FTA agreement with the U.S. Ingratiation The organization praises the stakehol ders as a most important asset to the organization. e.g., offering thanks to the fa mily of released hostage. Victimage The organization insists that they are also a victim of the crisis e.g., argument that the government is also a victim of sudden localized heavy rain.

PAGE 66

LIST OF REFERENCES Baker, B. (1997). Public relations in government, In C. L. Caywood (Eds .), The handbook of strategic public relations & integrated communication (pp. 453-480). New York: McGraw-Hill. Banks, K. F. (2005). Crisis Communications: A Casebook Approach. Mahwah, NJ: Lawrence Erlbaum. Barton, L. (1993). Crisis in organizations: Managing and communicating in the heat of chaos Cincinnati, OH: South-We stern Publishing Company. Benoit, W. L. (1995). Accounts, excuses, and apologies: A th eory of image restoration discourse Albany, NY: State University of New York Press. Benoit, W. L. (1997). Image repair discourse and crisis communication, Public Relations Review, 23 (2), 177-186. Birch, A. H. (1993). The British system of government (9th ed.) New York: Routledge. Boin, A., & Lagadec, P. (2000). Preparing for the future: critical ch allenges in crisis management, Journal of Contingenc ies and Crisis Management, 8(4), 185-191. Boyd, J. (2000). Actional legitimation: No crisis necessary. Journal of Public Relations Research, 12 (4), 341-353. Coombs, W. T. (1998). An analytic framework for crisis situations : Better responses from better understanding of the situation. Journal of Public Re lations Research, 10, 177. Coombs, W. T. (2004). Impact of past crises on current crisis communi cation: Insights from situational crisis communication theory. Journal of Business Communication, 41(3), 265289. Coombs, W. T. (2006). Crisis management: A co mmutative approach, In C. H. Botan, & V. Hazleton (Eds.), Public Relations Theory (pp.171-198). Mahwah, NJ: Lawrence Erlbaum. Coombs, W. T. (2007). Ongoing crisis communication: Planning, managing, and responding Thousand Oaks, CA: Sage. Coombs, W. T. (2007b). Protecti ng organization reputations duri ng a crisis: The development and application of situational crisis communication theory, Corporate Reputation Review 10(3), 163-176. Coombs, W. T., & Holladay, S. J. (1996). Comm unication and attributio ns in a crisis: An experimental study in crisis communication. Journal of Public Relations Research, 8 279295. 66

PAGE 67

Coombs, W. T., & Holladay, S. J. (2001). An exte nded examination of the crisis situations: a fusion of the relational management and symbolic approaches. Journal of Public Relations Research, 13 (4), 321-340. Coombs, W. T., & Holladay, S. J. (2002). Helping crisis managers protect reputational assets: Initial test of the situationa l crisis communication theory. Management Communication Quarterly, 16 (2), 165-186. Coombs, W. T., & Holladay, S. J. (2004). Reas oned action in crisis communication: An attribution theory-based approach to crisis management. In D. P. Miller, & R. L. Heath (Eds.), Responding to crisis: A rhetorical approach to crisis communication (pp. 95-115). Mahwah, NJ: Lawrence Erlbaum. Coombs W. T., & Holladay, S. J. (2005). Expl oratory study of stakehol der emotions: Affect and crisis. In N. M. Ashkanasy, W. J. Zerbe, & C. E. J. Hartel (Eds.), Research on emotion in organizations: volume 1: the effect of affect in organizational setting (pp. 271 288). New York: Elsevier. Coombs. W. T., Hazleton, V., Holladay, S. J., & Ch andler, R. C, (1995). The crisis grid: Theory and application in crisis mana gement. In L, Barton (Eds.), New avenues in risk and crisis management (Vol. 4, pp. 3-39). Las Vegas, NV: Univ ersity of Nevada Small Business Development Center. Crandall, W. R., & Mensah, E. C. (2008). Cris is management and sustainable development: A framework and proposed research agenda, International Journal of Sustainable Strategic Management,1 (1), 16-34. Englehardt, K. J., Sallot, L. M., & Springston, J. K. (2004). Compassion without blame: testing the accident decision flow chart with the crash of ValuJet Flight 592. Journal of Public Relations Research 16(2), 127-156. Fishman, D. A. (1999). ValuJet Flight 592: Cris is communication theory blended and extended. Communication Quarterly, 47 (4), 345-375. Fink, S. (1986). Crisis management: Planning for the inevitable. New York: American Management Association. Friesema, H. P., Caporaso, J. A., Goldstein, G ., Lineberry, R. L., & McCleary, R. (1979). Aftermath: Communities after natural disaster Beverly Hills, CA: Sage. Gundel, S. (2005). Towards a new typology of crises. Journal of Contingencies and Crisis Management, 13(3), 106-115. Heath, R. L., & Millar, D. P. (2004). A rhet orical approach to crisis communication: management, communication processes, and strate gic responses. In R. L. Heath, & D. P. Millar (Eds.), Responding to crisis: A rhetorical approach to Crisis Communication (pp. 118). Mahwah, NJ: Lawrence Erlbaum. 67

PAGE 68

Hearit, K. M. (1995). Mistakes were made: Or ganizations, apologia, and crises of social legitimacy. Communication Studies, 46, 1-17. Ho, F., & Hallahan, K. (2004). Post-earthquak e crisis communications in Taiwan: An examination of corporate adve rtising and strategy motives, Journal of Communication Management, 8 (3), 291-306. Horsley, J. S., & Barker, R. T. (2002). Toward a synthesis model for crisis communication in the public sector. Journal of Business and Technical Communication. 16 (4). 406-440. Hung, Y. (2008). Trust and relational commitment in corporate crises: The effects of crisis communicative strategy and fo rm of crisis response. Journal of Public Relations Research, 20, 297-327. Kersten, A. (2005). Crisis as usual: Organizational dysf unction and public relations. Public Relations Review, 31, 544. Kim, Y., Cha, H., & Kim, J. R. (2008). Developi ng a crisis management index: Applications in South Korea, Journal of Public Relations Research, 20, 328-355. Knight, J. (1982, October 11). Tylenols ma ker shows how to respond to a crisis, The Washington Post p. 1. Korean Culture and Information Service. (2008). Local Government. Retrieved December 19, 2008, from http://www.korea.net Kouzmin, Alexander, Jarman, & Alan, M.G. (1989). Crisis decision-making: Towards a contingent decision path perspec tive. In U. Rosenthal., M. Ch arles., & P. 't Hart (Eds.), Coping with crisis: The management of disasters, riots and terrorism (pp. 77-89). Spinfield, IL: Charles C. Thomas. Lee, B. K. (2004). Audience-oriented approach to crisis communication: As study of Hong Kong consumers evaluation of an organizational crisis, Communication Research 31(5), 600618. Lee, K. (2008). How the Hong Kong government lo st the public trust in SARS: Insights for government communication in a health crisis. Public Relations Review, 30, 1-3. Lerbinger, Otto. (1997). The crisis manager: Facing risk and responsibility Mahwah, NJ: Lawrence Erlbaum. Lijphart, A. (1999). Patterns of democracy: Government forms and performance in thirty-six countries, New Haven, CT: Yale University Press. Linke, C. G. (1989). Crisis-dealing with the une xpected. In B, Cantor, & C. Burger (Eds.), Experts in action: Inside public relations (pp. 166-178). New York: Longman. 68

PAGE 69

Liu, B. F., & Horsley, J. S. (2007). The Govern ment communication decision wheel: toward a public relations model for the public sector Journal of Public Relations Research 19 (4), 377-393. Marcus, A. A., & Goodman, R. S., (1991). Victim s and shareholders: the dilemma of presenting corporate policy during a crisis, Academy of Management Journal, 34 (2), 281-305. Marra, F. J. (1998). Crisis comm unication plans: poor predictors of excellent crisis public relations. Public Relations Review 24 (4), 461-474. McConnell, A. (2003). Overview: Crisis manage ment, influences, responses and evaluation. Parliamentary Affairs, 56 393-409. McAuley, E., Duncan, T. E., & Russell, D. W. (1992). Measuring causal attributions: The Revised Causal Dimension Scale (CDSII). Personality and Social Psychology Bulletin, 18, 566-573. Ministry of Government Affairs and Home Affairs (2008a). The national government organization act Seoul: Korea. Ministry of Government Affairs and Home Affairs (2008b). The local government act Seoul: Korea. Mitroff, I. I., & Anagnos, G. (2001). Managing crises before they happen: What every executive and manager needs to know about crisis management. New York: AMACOM. Mitroff, I. I., Pauchant, C. T., & Shrivastava, P. (1998). The structure of man-made organizational crises: Conceptual and empirical issues in the development of a general theory of crisis management. Technological Forecasti ng and Social Change, 33 83. Pearson, C. M., & Clair, J. A. (1998). Reframing crisis management, The Academy of Management Review 23 (1), 59-76. Pearson, C. M., & Mitroff, I. I. (1993). From crisis prone to crisis pr epared: A framework for crisis management, The Executive, 7 (1), 48-59. Perrow, C. (1984). Normal accident: Loving w ith high-risk technologies New York: Basic Books. Petak, W. J. (1985). Emergency managemen t: A challenge to public administration. Public Administration Review, 45, 3-7. Riffe, D., Lacy, S., & Fico, F. (2005). Reliabilit y. In D. Riffe, S. Lacy, & F. Fico (Eds.), Analyzing media messages: using quantitative content analysis in research (pp. 122-155). Mahwah, NJ: Erlbaum Associates. Regester M., & Larkin J. (2002). Risk issues and crisis management (2nd ed.). Philadelphia, PA: Kogan Page. 69

PAGE 70

Rosental, U., Boin, A., & Comfort, L. K. ( 2001). The changing world of crises and crisis management. In U. Rosental, A. Boin, & L. K. Comfort (Eds.), Managing crises: Threats, dilemmas, opportunities (pp.5-27). Springfield, IL: Charles C Thomas. Rosenthal, U., & Kouzmin, A. (1993). Globalizi ng an agenda for contingencies and crisis management: An editorial statement. Journal of Contingencie s and Crisis Management, 1(1), 1-21. Rosenthal, U., & Kouzmin, A. (1997). Crises and crisis management: Toward comprehensive government decision making. Journal of Public Administ ration Research and Theory, 2, 277-304. Rosental, U., t Hart, P., & Charle s, M. T. (1989). The world of crises and crisis management. In Rosental, U., t Hart, P., & Charles, M. T. (Eds.), Coping with crises: The management of disasters, riots and terrorism (pp. 3-33). Springfield, IL: Charles C Thomas. State Department's Bureau of International Information Programs & Braddock Communications. (2003). About America: How the United States governed Herndon, VA: Braddock Communications. Seeger, M. W., Sellnow, T. L., & Ulmer, R. R. (1998). Communication, orga nization, and crisis. In M. E. Roloff (Eds.), Communication yearbook 21 (pp. 231-276). Thousand Oaks, CA: Sage. Seeger, M. W., Sellnow, T. L., & Ulmer, R. R. (2001). Communication and organizational crisis London: Praeger publishers. Siomkos, G., & Shrivastava, P. (1993). Responding to product liability crises: Long Range Planning, 26 (5), 72-79. Stubbs, J. (2002). Forward. In R. Haywood (Eds.), Manage your reputation; How to plan public relations to build & protect the organization's most powerful asset (2nd ed.), London, UK: Kogan Page. Susskind, L., & Field, P. (1996). Dealing with an angry public: The mutual gain approach to resolving disputes. New York: Free Press. Hart, P., Rosenthal, U., & Kouzmin, A. (1993). Cr ises decision making: Th e centralization thesis revisited. Administration & Society, 25 (1), 12-45. Tucker, L., & Melewar, T. C. (2005). Corporate reputation and cr isis management: The threat and manageability of anti-corporatism, Corporate Reputation Review, 7 (4), 377-387. Ulmer, R. R. (2001), Effective crisis manageme nt through established stakeholder relationships: Malden Mills as a case study, Management Communication Quarterly, 14 (4), 590-615. 70

PAGE 71

71 Vlad, I., Sallot, L. M., & Reber, B. H. (2006) Rectification without assuming responsibility: Testing the transgression flowchart with the Vioxx Recal, Journal of Public Relations Research,18( 4), 357-379. Weiner, B. (1985). Human motivation New York: Springer-Verlag. Wimmer, R. D., & Dominick, J. R. (2006). Mass media research: An introduction (8th ed.). Belmont, CA: Thomson Higher Education. Wollmann, H. (2003). Coordination in the intergovernmental setting. In B. G. Peter & J. Pierre (Eds.), Handbook of public administration (pp. 594-605). London: Sage Publications.

PAGE 72

BIOGRAPHICAL SKETCH Jinsuk Lee received her B.A. in political science and international relations from Sookmyung Womens University, and a M.A. in In ternational relations fr om Korea University. She received her M.A.M.C. from the University of Florida in the summer of 2009. She has worked in the Korean Government at the Minist ry of Information and Government Information Agency since 1993. She will continue to work and contribute at the Ministry of Culture, Tourism, and Sports in Korea after graduation. 72