Seeing a crisis through colored glasses: How partisan media leads to attribution of crisis responsibility and government trust

Seeing a crisis through colored glasses: How partisan media leads to attribution of crisis responsibility and government trust
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   ACADEMIC PAPER  Seeing a crisis through colored glasses: How partisan medialeads to attribution of crisis responsibility and government trust Myoung ‐ Gi Chon 1 |  Elisabeth Fondren 2 1 School of Communication and Journalism,Auburn University, Auburn, Alabama 2 Manship School of Mass Communication,Louisiana State University, Baton Rouge,Louisiana Correspondence Myoung ‐ Gi Chon, School of Communicationand Journalism, Auburn University, 237Tichenor Hall, Auburn, AL 36830.Email: This study uses a real national crisis, South Korea's 2014  Sewol  ferry disaster, toexamine how publics exposed to partisan media perceive the attribution of crisisresponsibility and government trust differently. The study also investigates the medi-ating role of the attribution of crisis responsibility on the relationship between parti-san media and government trust. The results demonstrate that citizens' partisanselective exposure influence their polarized perceptions of crisis responsibility andtheir trust in government. The attribution of crisis responsibility partially mediatedthe effects of partisan media on government trust. This study suggests the impor-tance for government public relations to understand partisan media users so thatpublic relations managers can engage and communicate effectively with all citizensduring a national crisis. 1  |  INTRODUCTION When a national crisis or catastrophe occurs, governments struggle toprotect their reputation and maintain trust. This process is often apublic one because citizens pay attention and respond to how theirgovernments navigate through a crisis (Atkeson & Maestas, 2012). In judging governmental crisis management, people form their opinionsthrough information presented in the news media. From the perspec-tive of government public relations (PR) managers, it is essential tohave and solidify a high level of government trust from citizens. Peo-ple in a multiple channel environment, however, may see a givennational crisis through differently colored glasses, meaning their parti-san and political positions. This paper offers new insights in how gov-ernment PR can overcome a national crisis by examining therelationship between partisan media outlets and the attribution ofresponsibility. In order to adjust their crisis management strategies,governments need to pay attention to the relationship betweenpeople'sopinions,mediaoutlets,andthelevelsoftrustinorganizations.Scholars have long indicated that partisan media, which provide aslanted presentation of the news, have become a culprit in that theylimit the diversity of information available to citizens and polarize theirindividual ‐ level attitudes about social issues (Iyengar & Hahn, 2009;Stroud, 2010; Sunstein, 2009). In moments of crisis, news media areespecially important because they tend to investigate and interpretthe unfolding events (Atkeson & Maestas, 2012) and often holdauthorities accountable. More importantly, news media are the finalarbitrator of framing a crisis (Coombs, 2007).Although scholars have studied government PR, the literature hasnot yet illuminated how partisan media may influence the perceptionof publics and their ability to see a crisis and safeguard trust in govern-ment (Liu, Horsley, & Levenshus, 2010; Napawan, 2018; Waymer,2013). In the context of an organizational crisis, the attribution of cri-sis responsibility plays a vital role in suggesting crisis response strate-gies such as situational crisis communication theory (SCCT; Coombs,2012). The issue of government trust, moreover, has been consideredas a critical factor because it leads to individuals' attitudinal andbehavioral consequences between the public and organizations(Huang, 2001; Ki & Hon, 2007). Consequently, trust results in anincrease in citizens' corporative behavior and their compliance withgovernment decisions (Kim, 2005) and to follow government policies(Tsang, Burnett, Hills, & Welford, 2009).However, how can government agencies improve their crisis man-agement and enhance public discourse in a fragmented media environ-ment? Using the issue of a national marine disaster in South Korea — the 2014 sinking of the  Sewol  ferry, which resulted in over 304 casu-alties — this study attempts to examine how audiences' informationselectivity in South Korea's media environment is related to the attri-bution of crisis responsibility to government authorities and govern-ment trust. Furthermore, this study tests how the attribution ofcrisis responsibility appears to mediate the effect of partisan media Received: 14 March 2019 Accepted: 31 March 2019DOI: 10.1002/pa.1950  J Public Affairs . 2019;e1950.© 2019 John Wiley & Sons,  1 of 9  on government trust. The results of this study offer new, importantinsights for government PR and mass communication theory, showinghow people in partisan and selective exposure environments perceiveattribution of crisis responsibility and government trust. 2  |  LITERATURE REVIEW2.1  |  Partisan media and selective exposure in SouthKorea The theoryof selectiveexposureexplainswhyviewerschoosepartisanmedia. Since the 1940s, political scientists and mass communicationscholars have recognized that people selectively expose themselves tomedia messages that are consistent with their predispositions andexistingbelief systems(Hovland,1954; Lazarsfeld,Berelson, & Gaudet,1944). For example, Klapper (1960) noted that partisan predispositionslead people to selectively expose themselves in accordance with theirexistingopinions.Inhispioneeringworksonhowmediacanaffectpub-lic opinion, Lippmann (1922) explained predispositions (stereotypes)through which people see the world. People's public opinions, Lipp-mann wrote, are formed through the lens of existing knowledge. Inthe theory of cognitive dissonance, Festinger (1957, 1964) stated thatpeople avoid incongruent information that contradicts their existingbelief.Recently,scholarshavesuggestedthatpartisans'politicalpredis-positions lead them to selectively seeking information consistent withtheir political views (Iyengar & Hahn, 2009; Stroud, 2010). In otherwords,peoplearemorelikelytoselectamediasourcethatreflectstheirpolitical predisposition (Stroud, 2008, 2010). For instance, mediascholars premise studies on the assumption that conservatives andRepublicans prefer Fox News and avoid news from CNN and NPR,whereasliberalsandDemocratsshowtheoppositemediaconsumptionbehaviors (Iyengar & Hahn, 2009). Accordingly, selective exposureoccurswhenpeoplewanttoseekinformationthatmatchestheirpredis-positions and avoid incongruent media sources.In addition, people who watch partisan media such as Fox Newsand MSNBC tend to have more extreme and polarized attitudes whenthey are exposed to one ‐ sided views consistent with their own. Themore individuals talk about political issues with like ‐ minded others,the more their political attitudes become polarized (Jones, 2002;Stroud, 2007). This political reinforcement behavior is strongly associ-ated with an  “ echo chamber ”  effect, which occurs when peopleencounter viewpoints they agree with and amplify their preexistingviews (Sunstein, 2009). Sunstein (2009) also predicts that the internaldiversity of a group is reduced as people talk to like ‐ minded others,intensifying their previous views. He supported his argument by usingseveral examples such as corroboration, social networks, reputation,and solidarity ties (Sunstein, 2009). Indeed, selective exposure andgroup extremes leading to polarization regarding social issues exacer-bate the difficulty of addressing national problems that need socialcompromise and consensus.In South Korea, partisan news are prevalent in both televisionbroadcast and newspaper industries. Among the three majorbroadcasting companies (KBS, MBC, and SBS), two media outlets(KBS and MBC) are government funded and the incumbent govern-ment likely influences the two companies' editorial tone (Kang & Yang,2003). It is known that the three mainstream newspapers (The ChosunIlbo, Joongang Ilbo, and Dong ‐ A Ilbo) with the highest national circula-tion rates are conservative. In contrast, the newspapers  HankyoreShinmoon  and the  Kyunghyang Shinmoon  are known to have a liberalperspective (Hahn, Ryu, & Park, 2015). When reporting governmentaffairs, the difference in editorial positions between conservativeand progressive media in South Korea is clearly evident. For example,a comparative content analysis of conservative and progressive framesof Korean journalism showed the ideological polarization of theKorean newspapers. Although 96% of the editorials in  Chosun Ilbo ,which is a representative of the conservative media, were positivetoward the Korean government issues, 60% of the editorials in Hankyore Sinmoon , a liberal paper, were negative toward the nationalpolicy (Shim, 2013). Accordingly, in this polarized media environment,individuals are more likely to aggravate partisan selective exposure. 2.2  |  The attribution of crisis responsibility andpublic trust in government The public's perception has a measurable effect on how much an orga-nization takes on the responsibility for a crisis (Coombs, 2007). Forthis reason, PR practitioners dealing with a crisis have been interestedin monitoring or influencing news media for crisis management basedon the belief that the public experiences a particular crisis via theframes chosen by news media (Coombs, 2012). Weiner's (1985) attri-bution theory provides a theoretical framework for a relationalapproach to crisis management (Coombs, 2000). More specifically,the attribution of the responsibility frame is defined as  “ a way ofattributing responsibility for a cause or solution to either the govern-ment or to an individual or group ”  (Semetko & Valkenburg, 2000).According to Coombs (2000), organizational crisis responsibility isequated to the degree to which key publics attribute blame for a crisisto an organization. Attributions of organizational crisis responsibilityare influenced by perceptions of a crisis situation in terms of causalfactors, and these attributions, in turn, affect people's feelings andbehaviors (Weiner, Perry, & Magnusson, 1988).When it comes to government PR, trust has been considered as thecornerstone to showing greater commitment, cooperation, and alsocompliance from the public toward the government (Ki & Hon, 2007;Kim, 2005; Tsang et al., 2009). Scholars have also found that trust inan organization increases individuals' attitudinal (e.g., satisfaction andshared values) and behavioral responses (e.g., cooperation and compli-ance) to the organization (Huang, 2001; Ki & Hon, 2007). In PR, then,trust is defined as  “ a belief by publics that an organization is reliable,honest, and stands by its words as well as accomplishes its promisedobligations ”  (Ki & Hon, 2007, p. 42). Meanwhile, the definition of trustin political science and public administration refers to the degree towhich an individual perceives the government as  “ do[ing] the rightthing ”  (Wang & Wart, 2007, p. 266). Furthermore, trust is one of many 2 of 9  CHON AND FONDREN  dimensions of organizational PR. Trust may lead to signaling commit-ment,to “ theextenttowhicheachpartybelievesandfeelsthattherela-tionship is worth spending energy to maintain and promote ”  (Hon &Grunig, 1999, p. 3). Consequently, public trust in government authori-ties is very important because trust is an essential component of effec-tive governance that increases publics' compliance with government(Tsang et al., 2009; Tyler, 2003) and maintain robust relations with thepublic.Crisis communicators and PR scholars have focused on how peopleperceive, interpret, and respond to a crisis depending on which framesnews media use (e.g., image repair theory and attribution theory andSCCT). Depending on an organization's locus of control, SCCT explainsthat PR practitioners implement either accommodative strategies (e.g.,a full apology) or defensive strategies (e.g., attack and denial). Argu-ably, news media are the final arbitrator to frame crises (Coombs,2007). Furthermore, individuals who expose themselves to onlycertain media outlets are more likely to be influenced by media frames,and then, they may have stronger opinions in a given situation (e.g.,echo chamber effects). In a national crisis, we assume that the phe-nomenon of selective exposure may amplify publics' perceptions of agiven national crisis. Regarding the framing of the  Sewol  ferry tragedy,recent research revealed that there were considerable differences incoverage on the responsibility framing between conservative newspa-pers and liberal newspapers in South Korea (Fong, Jeon, & Wan,2016). Although conservative newspapers had a tendency to protectgovernment positions, liberal newspapers tended to criticize thegovernment, emphasizing the lack of responsibility by overlookingsafety and regulations (Fong et al., 2016). Based on the abovediscussion, we propose the following hypotheses: H1.  (a) Individuals who were exposed to conservativenews media are less likely to attribute crisis responsibility to government in a given crisis issue. (b) Individuals whowere exposed to liberal news media are more likely toattribute crisis responsibility to government in a given cri-sis issue. H2.  (a) Individuals who were exposed to conservativenews media are more likely to trust government in a givencrisis issue. (b) Individuals who were exposed to liberalnews media are less likely to trust government in a givencrisis issue. 2.3  |  The attribution of crisis responsibility as amediator between partisan media uses andgovernment trust By examining the ink between partisan media and the attribution ofcrisis responsibility and government trust, (H1 to H2), the currentstudy explores possible mediation effects of the attribution of crisisresponsibility on government between partisan media uses andgovernment trust. When publics face an organizational crisis, theygenerally seek causes and make attributions (Coombs, 2007). Theprimary goal of crisis communication is to protect organization's repu-tation from the negative effects of a crisis situation. It is essential forPR practitioners to know how publics perceive crisis responsibility inan organizational crisis. Once a crisis occurs, publics perceive crisisresponsibility based upon  “ how the crisis is being framed ”  (Coombs,2007, p. 166). Publics then attribute crisis responsibility to the organi-zation, leading to reputational threat (Coombs, 2007). Stronger orincreased attribution of crisis responsibility produces greater reputa-tional damage and, in turn, negative behavioral consequences suchas negative word of mouth and decreased use of products or services(Coombs, 2007, 2012).Despite considerable differences in the coverage of framingresponsibility among newspapers in South Korea, the attribution ofcrisis responsibility — as an important concept of crisis communication — may be a mediating element between partisan media and govern-ment trust (Fong et al., 2016). According to SCCT, the attribution ofcrisis responsibility is a key driver to determine crisis responsibilitystrategies. Previous research has tested the role of crisis responsibilityto suggest appropriate crisis strategies to protect an organizationalreputation. Expanding on these premises, it is also important to testthe mediation effect of the attribution of crisis responsibility in thecontext of government PR. Hence, this study proposes the followingresearch question:RQ1: To what extent does the attribution of crisis responsibilitymediates the effects of partisan media use on government trust? 2.4  |  South Korea's  Sewol  ferry disaster case andgovernment crisis One of South Korea's most catastrophic maritime accidents took placeon April 16, 2014. The  Sewol , a South Korean ferry, sank, claiming 304lives. The mechanical reasons remain unresolved and the crisis remainsa controversial topic to this day. The impeachment of South Korea'sformer president, Park Geun Hye, for instance, was partially due toissues associated with the accident. This disaster has become a gov-ernmental crisis but has also sparked political and social reactionsacross South Korea. Several debates were brought up that questionedwho was responsible for the accident. Accusations were made towardthe government, operation regulators, the ferry operator and crew,and also the South Korean media. Such criticisms scrutinized thesegroups for how they responded to the disaster and their efforts tominimize government accountability. The mismanagement of the ini-tial response, inadequate safety procedures, and the unethical rela-tionship between authorities and the ship operator had a majorimpact on President Park's administration. If there had been a bettersystem in place for disaster response and stricter regulations on theshipping industry, critics believe that this accident may have beenavoided. Therefore, the public outrage of how this incident washandled shows that the public took notice of the government's roleand scrutinized the official response to the accident. CHON AND FONDREN  3 of 9  3  |  METHOD3.1  |  Participants Data for this study were collected in South Korea as one of panelsurvey results. This is based on second survey, which was collected inOctober 2014 after first survey in August, 2014. This study employedanonlinesurveybyusing 347participants livingin SouthKorea.Partic-ipants were recruited from the online panel of Macromill Embrain with1,239,610 panelists during the month of October 2014 (Embrain,2017). The survey yielded a total of 347 valid responses. The agesranged from 20 to 69 years old with the average age of 41.2. Althoughmales were 51.6% of respondents ( n  = 179), females were 48.4%( n  = 168). In terms of their education level, 15.9% of respondents( n  = 55) had a high school degree or less, 10.4% ( n  = 36) had a 2 ‐ yearassociate degree or less than associate level, 65.2% ( n  = 216) had abachelor's degree or less than 4 ‐ year university level, and 10.7%( n  = 37) had a postgraduate degree or some graduate education. 3.2  |  Measures3.2.1  |  Independent variables: Partisan media Partisan media exposure Inordertomeasureexposuretopartisannewsmedia,weusedconserva-tive newspapers ( The Chosun Ilbo ,  Joongang Ilbo , and  Dong  ‐  A Ilbo ; M  = 2.31,  SD  = 1.16) and liberal newspapers ( Hankyore Shinmoon  andthe KyunghyangShinmoon ; M =1.99, SD =1.00).Respondentswereasked “ How often did you get selected information from the media during thelast week? ”  Respondents answered on a gradual scale (e.g., never, 1 pera week, 3 – 5 per a week, 1 per day, and above 3 per day; seeTable 1). 3.2.2  |  Control variables: Party identification, ideol-ogy, and demographics Party identification Respondents' political parties were measured through the followingquestions:  “ Which political party do you support? ”  The answer choiceswere as follows: (a) the ruling and conservative party (Saenuri), (b) theopposition and liberal parties (New Politics Alliance for Democracy),and (c) independence (no preference). Ideology  Respondents' ideology was measured through asking the question:  “ Ingeneral, how would you describe your political view? ”  The answerchoices were as follows: (a) very liberal, (b) liberal, (c) moderate, (d)conservative, and (e) very conservative ( M  = 3.03,  SD  = 0.73). Demographics For this study, respondents' demographic characteristics were basedon gender, level of education, and income. Gender (female = 1,male = 0) and partisanship (ruling party, opposition party, and indepen-dent party) were encoded as dichotomous variables. 3.2.3  |  Dependent variables: The attribution of crisisresponsibility and public trust on government As a dependent variable and a mediator variable, attribution of crisisresponsibility on the government was measured with one item,indicating that  “ the blame for the crisis lies with the government. ” The respondents indicated their answers on a 5 ‐ point Likert ‐ type scalethat ranged from 1 ( not at all ) to 5 ( very much ).Trust in PR is defined as  “ a belief by publics that an organization isreliable, honest, and stands by its words as well as accomplishes itspromised obligations ”  (Ki & Hon, 2007, p. 42). Public trust ingovernment – public relationship also reflects individuals' overall evalu-ation of the performance of political authorities and institutions (Miller& Listhaug, 1990). Accordingly, trust in this study is a concept thatinfluences the relationship between the citizenry and the politicalobjects in general (e.g., Chen, Hung ‐ Baesecke, & Kim, 2017; Huang,1997). Based on this rationale, this study evaluates individuals' trustin their government by adapting Hon and Grunig's (1999, p. 19)dimensions among six levels: integrity items (the belief that govern-ment performs fairly), dependability (the belief that government willdeliver its promises), and competence (the belief that the governmenthas the ability to manage controversies or problems affecting society). TABLE 1  Descriptive statistics  Variable (Cronbach's alpha) Mean  SD  Items Government trust ( α  = .91) Integrity 2.31 0.94 The government treats people like me fairly and justly.Integrity 2.21 0.97 Whenever the government makes an important decision,I know it will be concerned about people like me.Dependability 2.10 0.98 The government can be relied on to keep its promises.Dependability 2.15 0.92 I believe that the government takes the opinions of peoplelike me into account when making decisions.Competence 2.37 1.00 I feel very confident about the government's skills.Competence 2.64 1.12 The government has the ability to accomplish what it says it will do.Partisan newspapers Conservative newspapers 2.31 1.16 During the past week, how often did you use following media to getinformation (i.e.,  Chosun Ilbo ,  Joongang Ilbo , and  Dong  ‐  A Ilbo )?Liberal newspapers 1.99 1.00 During the past week, how often did you use following media toget information (i.e.,  Hankyore Shinmoon  and  Kyunghyang Shinmoon )?Attribution of crisis responsibility 3.37 3.04 The blame for the crisis lies with the government. 4 of 9  CHON AND FONDREN  These six items were combined into a sing index ( M  = 2.29,  SD  = 0.84, α  = .91). All of these items were measured on a 5 ‐ point Likert scaleranging from 1 ( strongly disagree ) to 5 ( strongly agree ; seeTable 1). 4  |  RESULTS4.1  |  Hypotheses testing To test H1 and H2, multiple ordinary least squares regression analyseswere conducted in SPSS statistical software program 24. Our firsthypothesis (H1) predicts individuals who were exposed to partisanmedia differently perceive the attribution of crisis responsibility togovernment in the context of  Sewol  ferry disaster (see Table 2). Forcontrol purposes, demographic variables (gender, age, education, andincome) were entered into the regression first, followed by ideologyand party identification. Finally, the measure of reading conservativenewspapers and reading liberal newspapers was entered into theregression. Controlling for effects of other independent variables inthe model, individuals who were exposed to conservative news mediawere less likely to attribute crisis responsibility to government( β  =  − .19,  p  < .001); individuals who were exposed to liberal newsmedia were more likely to attribute crisis responsibility to government( β  = .28,  p  < .01). Supporting H1, these findings suggest that partisanmedia exposure is a strong predictor of the attribution of crisis respon-sibility to government in the  Sewol  ferry disaster.Our second hypothesis (H2) examines whether partisan media useis associated with government trust. To answer this question, we useda multiple ordinary least squares regression in the second model (seeTable 2). Demographic variables (gender, age, education, and income)entered into the regression first, followed by ideology and party iden-tification. Finally, the measure of reading conservative newspapersand reading liberal newspapers entered the regression. The resultsindicated that individuals who were exposed to conservative newsmedia were more likely to trust government ( β  = .18,  p  < .001); individ-uals who were exposed to liberal news media were less likely to trustgovernment ( β  =  − .15,  p  < .01). Thus, H2 was supported in this study,and these results suggest that partisan media use is a predictor to gov-ernment trust in the context of  Sewol  ferry disaster. 4.2  |  Testing the mediation of the attribution ofcrisis responsibility  A path analysis using SPSS Amos was conducted to explore the medi-ating role of the attribution of crisis responsibility on governmenttrust. For this model to fit the evaluation of the path model, we usedthree model fit indices: comparative fit index (CFI), standardized rootmean square residual (SRMR), and root mean square effort of approx-imation (RMSEA) based on Hu and Bentler's joint ‐ criteria approach(Hu & Bentler, 1999; Kline, 2016). According to the joint ‐ criteriaapproach, an acceptable model fit is determined by a CFI value above.96 and an SRMR value less than .10; an RMSEA value less than .06and SRMR less than .10. Once the model achieved all the standardcriteria, we interpreted their paths to evaluate the mediating role ofthe attribution of crisis responsibly in the model:  x  2 df   (23) = 47.06,CFI = .97, SRMR = .02, RMSEA = .055.To confirm the mediation effects, we conducted a bootstrappinganalysis (Preacher & Hayes, 2008). Bootstrap estimates were calcu-lated on the basis of 5,000 bootstrap samples at a 95% bias ‐ correctedpercentile. The results indicated that the mediation effect of attribu-tion of crisis responsibility was significant (seeTable 3). The attributionof government crisis responsibility partially mediated the effect of par-tisan media on government trust. As shown inTable 3, partial media-tion effects of the attribution of crisis responsibility were foundbetween conservative newspaper use and government trust (i.e., con-servative newspaper use → attribution of crisis responsibility on gov-ernment → government trust):  β  = .14, SE = .03 (95% BC ‐ CI interval:.09, .20),  p  < .001; between liberal newspaper use and governmenttrust (i.e., liberal newspaper use → attribution of crisis responsibilityon government → government trust):  β  =  − .17, SE = .03 (95% BC ‐ CIinterval:  − .23,  − .11),  p  < .001. The results of this study, therefore,showed partial mediation effects of the attribution of crisis responsi-bility between partisan media and government trust. 5  |  DISCUSSION Using the 2014 national marine disaster in South Korea, this studyexplored audiences' information selectivity in partisan media and thepolarization about the attribution of crisis responsibility and govern-ment trust toward the South Korean government in a national crisis. TABLE 2  Regression predicting attribution of crisis responsibility ongovernment and government trust  Attribution of crisisresponsibility towardgovernmentGovernmenttrust DemographicsGender (female = 1) .08  − .05Age .02 .12*Education  − .05  − .04Income .01  − .01Political valueIdeology (conservative = high)  − .21*** .22***Party identification (rulingparty = high) − .18** .35***Partisan mediaConservative newspaperreading − .19*** .18***Liberal newspaper reading .28**  − .15** R 2 .25 .38 F   14.38 25.54 Note . Data:  N  = 347. Cell entries for all models are standardized regressioncoefficients.* p  < .05. ** p  < .01. *** p  < .001. CHON AND FONDREN  5 of 9
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!