a burden of responsibility: the role of...

Q Academy of Management Review2015, Vol. 40, No. 3, 345–369.http://dx.doi.org/10.5465/amr.2013.0027

A BURDEN OF RESPONSIBILITY: THE ROLE OF SOCIALAPPROVAL AT THE ONSET OF A CRISIS

JONATHAN BUNDYThe Pennsylvania State University

MICHAEL D. PFARRERUniversity of Georgia

We extend research on social evaluations and crisis management by explicating thesociocognitive mechanisms that influence how an organization and its external eval-uators perceive and respond to the onset of a crisis. Specifically, we highlight the role ofsocial approval—evaluators’ general affinity toward an organization—not only as a crit-ical outcome of crisis management but also as an important antecedent. We first identifythe distinct aspects of social approval and explain why it is an important perception ina crisis context. We then detail how managers attempt to limit the probability andmagnitude of social approval loss when responding to a crisis, and how an organization’sexisting endowment of social approval affects this decision. We theorize that social ap-proval will serve as either a buffer or a burden in influencing evaluators’ crisis sense-making and attributions. As a result, we argue, organizations endowed with higher andlower levels of social approval may bemotivated to take less responsibility at the onset ofa crisis than has been previously theorized. We conclude with a discussion of the broadermanagerial and social implications of our theory and how it expands our understandingof the crisis management process.

Management scholars have become increasinglyinterested in understanding the sociocognitiveprocesses associated with how external evaluatorsperceive and make sense of an organization’sbehavior (e.g., Bitektine, 2011; Devers, Dewett,Mishina, & Belsito, 2009; Lange & Washburn, 2012;Mishina, Block, & Mannor, 2012). Much of theresearch in this area has focused on the more de-liberate and analytical sociocognitive processesthat often serve as the foundation for a number ofsocial evaluations, including organizational repu-tation and legitimacy (e.g., Deephouse & Suchman,2008; Jensen, Kim, & Kim, 2012; Pfarrer, Pollock, &Rindova, 2010). However, researchers have alsobegun to explore evaluators’ more intuitive andaffective sociocognitive processes that serve as thefoundation for an organization’s “social approval,”which we define as evaluators’ general affinitytoward an organization (Zavyalova, Pfarrer, Reger,

& Shapiro, 2012: 1079). Hence, management re-search is beginning to show that the degree towhich evaluators like or how they feel about anorganization can be leveraged to build and main-tain relationships, engender higher performance,and enhance an organization’s chances of survival(e.g., Vergne, 2012; Zavyalova et al., 2012).Despite these recent efforts, scholars have yet to

fully explore the underlying sociocognitive pro-cesses associated with social approval and howsocial approval influences organizational out-comes. Additionally, scholars have yet to fullyunderstand the role of social approval when anorganization is associated with negative behav-iors, such as when experiencing a crisis—an un-expected, publicly known, and harmful event thathas high levels of initial uncertainty, interfereswith the normal operations of an organization, andgenerates widespread, intuitive, and negativeperceptions among evaluators (cf. Coombs, 2007b;Fink, 1986; Roberts, Madsen, & Desai, 2007). Recentexamples of organizational crises include BP’sDeepwater Horizon explosion and oil spill, Apple’slabor and human rights issues in China, andTarget’s consumer data breach. Given the intuitiveand affective nature of social approval, it seemsparticularly important to examine its role at theonset of a crisis, where similarly heuristic and

We thank former associate editor Cindy Devers and thethree anonymous reviewers for their exceptional guidanceduring the review process. We also thank Katy DeCelles,Ashley Gangloff, Amy Guerber, Patrick Haack, Jason Kiley,Alan Muller, Christine Shropshire, and Anastasiya Zavyalovafor their helpful comments on previous versions of the man-uscript. Finally, we thank participants at the 2012 Academy ofManagement annual meeting, especially those who partici-pated in the Pecha Kucha organized by David Deephouse.

345Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyrightholder’s express written permission. Users may print, download, or email articles for individual use only.

http://dx.doi.org/10.5465/amr.2013.0027

affective information processing dominates (cf.Billings, Milburn, & Schaalman, 1980; van derMeer & Verhoeven, 2014; Weick, 1988).

In this article we seek to advance research onsocial evaluations and crisis management byexplicating the role of social approval in thesociocognitive processes that shape managers’and evaluators’ perceptions of and responses tothe onset of a crisis. Recent commentary suggeststhat scholars often rely too heavily on anecdotal,case-based evidence instead of theory-basedarguments to systematically identify and modelthe key sociocognitive properties of crisis man-agement (Coombs, 2007a). Additionally, scholarshave largely ignored the role of an organization’sendowment of social approval in the crisissensemaking process. Given that evaluators relyon their prior social evaluations as a “cognitiveshorthand” to “gauge the probable outcomes ofinteracting with [an organization]” (Mishina et al.,2012: 460), endowed social approval likely playsan important role in how evaluators initially in-terpret and react to a crisis. We therefore drawfrom research on social judgment formation, per-ception management, and decision making toexamine not only how an organization’s socialapproval acts as a key outcome of crisis man-agement but also how it acts as an importantantecedent. In so doing, we contribute to multi-disciplinary organizational research in four pri-mary ways.

First, we define social approval and identify itssociocognitive properties. We highlight the in-tuitive and affective basis of social approval, aswell as several additional characteristics, in-cluding its ability to be collectively understoodand its individuating nature. We also comparesocial approval with two related evaluations: or-ganizational legitimacy and reputation. Finally,we position social approval as a key evaluation atthe onset of an organizational crisis, matching itsdistinct properties to those that similarly definea crisis.

Second, in focusing on social approval as anoutcome, we apply a sociocognitive framework tounderstand the onset of a crisis in terms of theexpected probability and magnitude of socialapproval loss. We argue that evaluators will usetwo primary sources of information to makesense of a crisis and formulate their attributions:(1) the perceived degree of an organization’s re-sponsibility based on the characteristics of thecrisis, which we term situational attributions, and

(2) an organization’s response strategy, whichis the set of coordinated communication andactions used to influence evaluators’ crisis per-ceptions (Barton, 2001; Coombs, 2007b). The de-gree to which these two sources of informationare matched will facilitate evaluators’ initialcrisis sensemaking and attributions. For exam-ple, for a crisis that triggers greater perceptionsof responsibility (e.g., financial fraud or environ-mental malfeasance), a response strategy thataccepts more responsibility will match eval-uators’ situational attributions and will simplifysensemaking. We theorize that a matched strat-egy will result in a more normalized loss of socialapproval such that the expected mean and vari-ance of social approval loss will be reduced. Incontrast, a mismatched strategy will complicateevaluators’ sensemaking and attribution efforts,which can increase the mean and variance ofsocial approval loss.Third, because prior social evaluations of an

organization influence “what observers expectand notice, as well as how actions and state-ments will be interpreted” (Mishina et al., 2012:462), we theorize that an organization’s endow-ment of social approval is an important yet oftenoverlooked antecedent for evaluators as theymake sense of a crisis at its onset. We extendorganizational research exploring the “doubleedge” of social evaluations (cf. Ashforth & Gibbs,1990: 177; Brooks, Highhouse, Russell, & Mohr,2003; Rhee & Haunschild, 2006) by arguing that anorganization’s endowment of social approval canserve as either a buffer or a burden in influencingevaluators’ perceptions of a crisis. We explainthe sociocognitive mechanisms behind thisdouble edge and show how an organization’sresponse strategy is an important factor in de-termining whether endowed social approval willact as a buffer or a burden. We ultimately arguethat higher and lower levels of social approvalare likely to be a burden when an organizationaccepts more crisis responsibility, whereas theyare likely to be a buffer when an organizationaccepts less responsibility.Fourth, we theorize that an organization’s en-

dowment of social approval will also influenceits choice of response strategy. We utilize argu-ments frommanagerial decisionmaking and risktaking (e.g., March & Shapira, 1987) to suggestthat—because of the burden of responsibilityassociated with social approval—organizationsendowed with higher and lower levels of social

346 JulyAcademy of Management Review

approval may be motivated to take less re-sponsibility for a crisis. In doing so we contributeto organizational research by specifying a moti-vation that goes beyond the typical legal and fi-nancial reasons an organization may choose oneresponse strategy over another.

SOCIAL APPROVAL ANDORGANIZATIONAL CRISES

There are several sociocognitive properties thatmake social approval distinctly suited for exami-nation at the onset of a crisis. These propertiesinclude (1) social approval’s more intuitive and af-fective cognitive basis and (2) its collective andindividuating nature. Throughout this section wehighlight how social approval compares with tworelated evaluations, organizational legitimacy andreputation. Table 1 summarizes all three evalua-tions along the dimensions detailed below.

Social Approval As the Result of a More Intuitiveand Affective Process

Research in social psychology has shown thatevaluators engage in two types of cognitive pro-cessing tomake sense of the social world (Chaiken& Trope, 1999; Kahneman, 2011; Kahneman &Frederick, 2002; Stanovich & West, 2000). The first,deliberate processing, is slow and effortful, in-volvingmore analytical reasoning and a controlledapplication of decision rules to reach optimal con-clusions (Bazerman, 2006; Kahneman & Frederick,2002). The second, intuitive processing, is fast andeffortless, involving more heuristic and affectivereasoning and a loose application of decisionrules to reach satisfactory conclusions. Both

processes have been shown to influence a varietyof social judgments, ranging from statistical rea-soning to medical diagnoses to military strategy(Kahneman, 2011).Drawing on this “dual-process” perspective

(Kahneman & Frederick, 2002: 51), researchersinvestigating how evaluators perceive and un-derstand an organization have often emphasizedthe specific outcomes of more deliberate socio-cognitive processes. For example, in research onorganizational legitimacy, scholars have oftenfocused on the result of evaluators’ more analyt-ical assessments of an organization’s conformityto a set of social norms and values (Bitektine,2011; Deephouse & Suchman, 2008; Tost, 2011).Thus, legitimacy is frequently characterized asa more deliberate judgment as to “whether theorganization, its form, its processes, its outcomes,or its other features are socially acceptable”(Bitektine, 2011: 157). Similarly, researchers typi-cally have characterized organizational reputa-tion as the outcome of a more deliberate processbased on an organization’s ability to delivervalue according to evaluators’ idiosyncraticexpectations (Jensen et al., 2012; Rindova,Petkova, & Kotha, 2007). Thus, reputation is oftendescribed as a more “analytical interpretiveframe” used by evaluators to understand anorganization’s ability to deliver value (Pfarreret al., 2010: 1134).In addition to these more deliberate and ana-

lytical assessments, some scholars have begunto investigate evaluators’ more intuitive andaffective perceptions of an organization (e.g.,Deephouse, 2000; Highhouse, Broadfoot, Yugo, &Devendorf, 2009; Love & Kraatz, 2009; Raithel &Schwaiger, in press). For example, Lange, Lee,

TABLE 1A Comparison of Social Evaluations

SociocognitiveProperty Social Approval Legitimacy Reputation

Definition Perception of generalaffinity toward an organization

Assessment of anorganization’sappropriateness

Assessment of anorganization’s abilityto deliver value

Cognitivereference point

No explicit reference Social norms and values Expectations based onidiosyncratic relationship

Emphasizedcognitive basis

Intuitive and affective Deliberate and analytical Deliberate and analytical

Collective understanding? Yes Yes NoIndividuating?

(Range)Yes(Continuous)

No(Categorical)

Yes(Continuous)

2015 347Bundy and Pfarrer

and Dai recognize a “generalized favorability”dimension of reputation, described as “an over-all, generalized assessment of the organiza-tion . . . asmore or less goodandattractive” (2011: 159).Similarly, the concept of cognitive legitimacydescribes an organization’s intuitive taken-for-grantedness as an appropriate entity (cf. Scott,1995; Suchman, 1995), and research has begun toexplore how more affective perceptions in-fluence legitimacy judgments (e.g., Haack, Pfarrer,& Scherer, 2014).

Despite these developments, however, the in-tuitive versus analytical nature of social evalua-tions remains in debate, as does the role of affectamong related constructs (cf. Bitektine, 2011;Deephouse & Suchman, 2008; Fombrun, 2012;Jensen et al., 2012). For example, Bitektine (2011)suggests that the intuitive taken-for-grantednessinherent in cognitive legitimacy is still the endresult of a more deliberate categorization processgrounded in widely held beliefs and norms. Sim-ilarly, Jensen et al. argue that conceptualizingreputation as an intuitive and affective constructhas resulted in “definitions that are vague andindistinguishable from other related theoreticalconstructs” (2012: 143). Furthermore, research hasrecognized that assessments of reputation andlegitimacy—either as more deliberate or moreintuitive judgments—do not necessarily involveperceptions of affinity (Haack et al., 2014; Pfarreret al., 2010). That is, an organization can be judgedas legitimate and reputable without evaluatorsnecessarily attaching affective value to the judg-ments. Given this debate, Lange et al. (2011: 154)argue that we are still in a “formative phase” ofresearch on the subject, characterized by un-certainty surrounding definitions, dimensionality,processes, and interrelationships.

In response to the “confusions and conflations”(Deephouse & Suchman, 2008: 62) associated witha conceptual thicket of related labels and defi-nitions, researchers have begun using the termsocial approval as an overarching construct todescribe the more intuitive and affective per-ceptions inherent in the social evaluations of anorganization (cf. Vergne, 2012; Zavyalova et al.,2012). As Jensen et al. (2012: 144) argue, dis-tinguishing such an overarching construct torepresent the “diffuse general impressions orsentiments among diverse audiences” is neededfor conceptual clarity. Research in social psy-chology also provides support for the use ofoverarching constructs to attenuate the potential

confusion related to more intuitive and affectiveperceptions. For example, the dual-process per-ception of cognition recognizes that affectivereactions are often separated from recallablecontent knowledge. In other words, “one mayfeel strongly about . . . [something] without re-membering all the reasons why” (Fiske & Taylor,1991: 452). Consistent with this logic, we arguethat it is difficult for scholars and evaluatorsto distinguish between the more affective andintuitive perceptions that may be inherent indifferent social evaluations.1 Instead, intuitiveperceptions of an organization’s affinity are oftenindiscriminate and overlapping.We therefore posit that the more affective and

intuitive perceptions inherent in social evalua-tions can be holistically summarized as socialapproval. As such, social approval reflects eval-uators’ general affinity toward an organization,including perceptions of its inherent goodnessor badness, attractiveness, or likability. Unlikemore deliberate and analytical assessments,social approval perceptions operate “automati-cally and quickly, with little or no effort and nosense of voluntary control” (Kahneman, 2011: 20).As the result of a “feeling state” (Slovic, Finucane,Peters, & MacGregor, 2002: 397), the evaluativequestion underlying social approval does notnecessarily make an explicit reference to a socialnorm (as with legitimacy) or specific value ex-pectation (as with reputation); it is simply, “Do Ilike this organization?” In this way social ap-proval is a nonspecific perception of affinity,whereas more intuitive forms of legitimacy andreputation have a cognitive reference groundedin norms or expectations.2 Additionally, while werecognize that more analytical andmore intuitiveforms of information processing are not mutuallyexclusive, we also recognize that certain evalu-ations tend to emphasize one form of processing

1 Evidence of this difficulty is shown in the academic lit-erature through the use of similar concepts and measures tocapture reputation (e.g., Deephouse, 2000), legitimacy (e.g.,Lamin & Zaheer, 2012), and social approval (e.g., Zavyalovaet al., 2012).

2 Because of their inherent overlap, an organization’s socialapproval likely increases as it gains legitimacy and reputa-tion, and vice versa. While the nuances of this relationshipare beyond the scope of this article and remain under-developed (cf. Deephouse& Suchman, 2008; Lange et al., 2011),the key difference in the context of our theorizing is theemphasized cognitive basis and reference point for eachassessment.


over the other (cf. Pfarrer et al., 2010; Slovic,Finucane, Peters, & MacGregor, 2004). Thus, asTable 1 depicts, we use “social approval” tocapture evaluators’ more nonspecific, intuitive,and affective perceptions toward an organiza-tion, while we use “reputation” and “legitimacy”to reference more deliberate and analyticalevaluations, consistent with their dominantcharacterizations in the literature.

An organization accrues social approval byregularly engaging in positively perceived behav-ior (Vergne, 2012; Zavyalova et al., 2012). Higherlevels of social approval are reflected in externalaccounts of praise, such as positive media cover-age, institutional endorsements, and placement onpopular social lists or rankings (Vergne, 2012). Incontrast, lower levels of social approval arereflected in external accounts that “challenge,criticize, or condemn [an organization’s] activities,behaviors, or values” (Vergne, 2012: 1027), such asnegative media coverage, institutional scrutiny,and placement on unpopular social lists or rank-ings. An organization’s level of social approval isalso important for building and maintaining rela-tionships with evaluators. For example, individ-uals have a natural tendency to direct theirbehaviors toward positive stimuli and away fromnegative stimuli (Elliot, 2006; Haidt & Bjorklund,2008; Lange et al., 2011). As such, evaluators willbe more inclined to interact with an organizationthey perceivemore positively and less inclined tointeract with an organization they perceive morenegatively. Thus, an organization with highersocial approval may benefit from better perfor-mance and more diverse and longer-lastingrelationships. In contrast, while an organizationwith lower approval may find ways to remainprofitable, it also may have difficulty creatingand maintaining relationships and may behampered with excessive regulation and scru-tiny, which may threaten its long-term viabilityand performance.

Social Approval As a Collective andIndividuating Perception

Scholars have asserted that individuals andorganizations are “immersed in cultural systemsfromwhich standards for judging favorability aresocially constructed” (Lange et al., 2011: 159-160).For example, moral intuition researchers haveargued that an individual’s innate sense of rightand wrong is the product of a group socialization

process (Haidt, 2001), and social psychologistshave long recognized that individuals validatetheir thoughts and opinions via social compari-son (Festinger, 1954). In the same way, eval-uators’ perceptions of social approval are shapedby a collective and intuitive social constructionprocess (Devers et al., 2009; Highhouse et al.,2009). Because of this, social approval perceptionstend to “transcend stakeholder group boundaries”(Rindova & Martins, 2012: 22), whether eval-uators are directly engaged with an organizationor are more diffuse observers, such as the gen-eral public or media. Thus, social approval canbe understood as a collective perception withina social system.However, it is important to recognize that

evaluators’ collective understanding of an orga-nization’s social approval does not require completeagreement among all evaluators. Instead, we ar-gue that an organization collectively gains higher orlower levels of social approval when evaluatorsreach “concurrence” (Pfarrer, DeCelles, Smith, &Taylor, 2008: 733), which reflects evaluators’ gener-ally shared perceptions or their dominant opinionof an organization (Devers et al., 2009; Highhouseet al., 2009; Pfarrer, DeCelles, Smith, & Taylor, 2008).Thus, while an organization with higher or lowerlevels of social approval may have defectors, theconcurring perception among most evaluators willbe consistent. For example, research has shownthat evaluators generally viewDisney and Amazonpositively and Goldman Sachs and Halliburtonnegatively (RepTrak, 2013), despite the fact that notall individual evaluators would agree.In contrast, when evaluators fail to reach con-

currence in their affinity toward an organization,its social approval fails to collectivize in a usefulmanner. For example, evaluators’ perceptions ofWalmart and McDonald’s are quite varied, bothwithin and across evaluators and evaluatorgroups (RepTrak, 2013). Because of this lack ofconcurrence, we argue that such organizationshave mixed levels of social approval. Giventhese conflicting perceptions of positive andnegative affinity, a mixed level of social ap-proval is ill-defined and therefore less useful ata collective level (cf. Bitektine, 2011). Thus, anorganization’s social approval is most salient athigher and lower levels, where evaluators havereached concurrence about the organization’sgoodness or badness and the perception iswidely shared. As Table 1 shows, concurrencealso makes social approval an individuating


evaluation—meaning that higher and lower lev-els of social approval can set the organizationapart such that it is perceived as a distinct entity(Devers et al., 2009).

Additionally, we limit our arguments to thoseorganizations that maintain a minimum level ofapproval among evaluators. For example, a cig-arette manufacturer’s negative media coverage,consistent regulatory challenges, and encoun-ters with advocacy groups suggest lower levelsof social approval. However, while facing col-lective perceptions of lower approval, the man-ufacturer still maintains standing as a credibleand legally bound organization; it is “approved”to exist and maintain operations, even if suchapproval is tenuous. In contrast, organizationsthat fail to maintain a minimum level of collec-tive social acceptance become “disapproved.”Such organizations are often made explicitlyillegal, are disparaged by a concurrence ofevaluators, and generally struggle to exist.Thus, we use the terms higher approval andlower approval to describe a social approvalcontinuum—with higher approval representingincreasingly positive perceptions and lower ap-proval representing increasingly negative per-ceptions. We use the term disapproved to describea point on the negative side of the continuum,beyond which an organization is no longer ap-proved to exist by a concurrence of evaluators.Finally, we use the terms mixed approval andaverage approval to describe the middle of thecontinuum. As with mixed approval, we arguethat average approval is ill-defined as a collectiveand individuating construct. Average approvalcaptures those organizations that have neitherpositive nor negative affinity, resulting in neutralperceptions. Examples include Safeway andDollar General (RepTrak, 2013).3

Scholars have also considered the collectiveand individuating nature of organizational legiti-macy and reputation. As with social approval,legitimacy is often understood as a collectiveperception (Deephouse & Suchman, 2008). How-ever, unlike social approval, legitimacy generallyis not considered individuating. For example,Devers et al. argue that “rather than conveying theunique aspects of organizations, it [legitimacy]categorizes organizations into those either havinglegitimacy (legitimate) or not having legitimacy(nonlegitimate)” (2009: 156). Once legitimacy hasbeen conferred on an organization, evaluatorsconsider it socially acceptable, but the judgmentdoes not convey unique or evaluative attributes.Because of this, legitimacy is often considereda categorical construct (Deephouse & Suchman,2008; Devers et al., 2009).In contrast, organizational reputation is often

viewed as an individuating assessment convey-ing distinct aspects of an organization that canbe used to compare that organization againstothers (Devers et al., 2009). As such, reputationis often considered a continuous construct andfundamentally differentiating (Deephouse &Suchman, 2008). However, unlike social approvaland legitimacy, an organization’s reputation isgenerated from how well it meets different eval-uators’ specific and idiosyncratic expectations(Devers et al., 2009; Fombrun, 2012; Lange et al.,2011; Rindova et al., 2007). Different evaluatorgroups have different perceptions of an orga-nization’s ability to provide value, including viafinancial performance, innovation, or social re-sponsibility. Thus, many current conceptualiza-tions of reputation do not depict it as a collectiveconstruct (cf. Fombrun, 2012; Jensen et al., 2012).

The Role of Social Approval in a Crisis

An organizational crisis is an unexpected,publicly known, and harmful event that has highlevels of initial uncertainty, interferes with thenormal operations of an organization, and gen-erates widespread, intuitive, and negative per-ceptions among evaluators. We focus on how anorganization manages evaluators’ perceptions ofa crisis, as opposed to how it minimizes harmand controls physical damage. We are also par-ticularly interested in evaluators’ perceptionsand reactions at the onset of a crisis. This stageincludes the public disclosure of the crisis,evaluators’ initial attempts to make sense of the

3 As evaluators’ affinity toward an organization decreases,their negative perceptions may also increase. This suggeststhat higher (positive) and lower (negative) approval may existon two separate continua (cf. Brooks et al., 2003; Vergne, 2012).However, the possibility of two continua is less important forour theorizing, given our focus on higher and lower levels ofsocial approval. For example, higher-approval organizationscan be defined either by the positive end of one continuum orby the presence of high positive approval on one continuumand low negative approval on another continuum. An orga-nization simultaneously facing high positive approval andhigh negative approval would have mixed approval and beill-defined as a collective perception.


crisis and attribute responsibility, and the orga-nization’s initial response (Fink, 1986; Mitroff, 1994;Turner, 1976; Veil, 2011).

There are three characteristics of the onset ofa crisis that make the role of social approval im-portant to our theorizing. First, the onset of a crisisis characterized by high levels of uncertainty bothwithin the organization and with external eval-uators. Crises often have multiple explanations,ambiguity regarding responsibility and potentialdamages, and several feasible solutions (Billingset al., 1980; Pearson &Clair, 1998; Scherer, Palazzo,& Seidl, 2013). Under conditions of high uncer-tainty, most evaluators—whether directly en-gaged with the organization or more diffuse—willrely on more heuristic and intuitive processes asthey attempt to quickly make sense of the situa-tion (Kahneman, 2011; Kahneman & Frederick,2002; Sinaceur, Heath, & Cole, 2005). These pro-cesses are also necessarily more affective, withevaluators reacting emotionally to the perceivednegative consequences of a crisis (Coombs, 2007b;Sinaceur et al., 2005; van der Meer & Verhoeven,2014). Similarly, managers also rely on moreheuristic decision making in their responses touncertainty, often making rapid and emotionaldecisions in the heat of the moment (Sayegh,Anthony, & Perrewe, 2004; Weick, 1988). Givenevaluators’ and managers’ greater reliance onintuition and affect to make sense of uncertainty,it is more likely their attention will be directedtoward social approval at the onset of a crisis—itself the result of a more intuitive and affec-tive process.

Second, a crisis has a widespread impact on arange of evaluators, from direct stakeholders—forexample, investors and customers—to more in-direct evaluators—for example, themedia and thegeneral public (Pfarrer, DeCelles, Smith, & Taylor,2008; Yu, Sengul, & Lester, 2008). In this way a cri-sis is a collectively perceived event. Similar toevaluators’ concurrence about an organization’ssocial approval, our definition of a crisis requiresthat evaluators within and across groups concurin their perceptions of a crisis as a negative event,even if some evaluators may diverge from thedominant opinion.

Third, a crisis can be understood as an in-dividuating event because it compels evaluatorsto make attributions of responsibility. Attributiontheorists argue that individuals have an “innateneed to understand and control their envi-ronments” and act as “naıve psychologists” to

develop “causal explanations for significantevents” (Martinko, 1995: 8; see alsoHeider, 1958). Inturn, these causal explanations influence indi-viduals’ perceptions and behaviors (Martinko,1995; Weiner, 1986). Researchers have appliedthe basic tenets of attribution theory to crisismanagement, arguing that determining crisis re-sponsibility assigns individuality to the event (cf.Coombs, 1995; Coombs & Holladay, 2002). As such,evaluators quickly begin to associate a crisis withan individual organization, such as Exxon’s oilspill or Enron’s scandal.Thus, both the cognitive processes related to

social approval assessments and those associ-ated with the onset of a crisis can be character-ized as more intuitive, affective, collective, andindividuating. We are therefore able to developtheory that pairs our key construct and contextalong similar cognitive properties and at thesame level of analysis. In summary, we arguethat an organizational crisis, which generateshigh uncertainty and negative perceptionsamong a wide range of evaluators, should bepaired with an equally broad construct, such asthe “generalized, non-attribute-specific positiveattitudes” inherent in an organization’s socialapproval (Rindova & Martins, 2012: 22). We nextconsider the sociocognitive processes associatedwith how evaluators formulate their crisis attri-butions and modify their perceptions of anorganization’s social approval as an outcome ofa crisis.

CRISIS ATTRIBUTIONS AND THECRISIS-RESPONSE MATCH

A key element in understanding how eval-uators make sense of a crisis lies in their attri-butions of crisis responsibility: as evaluatorsattribute more crisis responsibility to an organi-zation, the threat to the organization’s social ap-proval increases (Coombs, 1995). Two sources ofinformation are critical in influencing theseattributions. First, evaluators rely on the char-acteristics of a crisis—including its perceivedintentionality, controllability, and severity—toformulate initial situational attributions alonga continuum of responsibility (Coombs, 2007b).Situational attributions are the result of a heu-ristic simplification process in which evaluatorsintuitively combine past experiences and expec-tations to reduce the complex nature of a crisis intoeasier-to-understand cognitive schemas (Coombs


& Holladay, 2004; Mitroff, 1988). For example,evaluators will intuitively attribute more re-sponsibility to an organization for a crisis origi-nating within the organization (e.g., an executivescandal) or resulting from a more controllablesituation (e.g., an industrial accident). In contrast,evaluators will intuitively attribute less re-sponsibility to an organization for a crisis origi-nating outside the organization (e.g., a naturaldisaster) or resulting from a less controllablesituation (e.g., product tampering).

Situational attributions can serve as usefulceteris paribus heuristics for understanding howevaluators comprehend crisis responsibility, andresearch confirms that evaluators make suchsimplified judgments when initially perceivinga crisis (cf. Coombs & Holladay, 2004). However,evaluators’ situational attributions are alsoa negotiated feature of crisis management and,therefore, subject to social influence (cf. Fediuk,Coombs, & Botero, 2012; Gephart, 2007; Robertset al., 2007). Thus, the second important source ofinformation for evaluators as they make crisisattributions is an organization’s response strategy.An organization’s initial response is influential inanchoring evaluators’ first impressions, which“form quickly and color the remainder of . . . [their]reception of the crisis communication efforts”(Coombs, 2011: 139).

Similar to the continuum of situational attri-butions detailed above, crisis managementscholars have described a response strategycontinuum in terms of the amount of responsibilityan organization takes for a crisis (Coombs, 2007b;Coombs & Holladay, 2004; Elsbach, 2003; Marcus &Goodman, 1991). Response strategies that ac-cept less responsibility—generally labeleddefensive—attempt to avoid social approvalloss by eliminating an organization’s perceivedassociation with a crisis (Coombs & Holladay,2004; Elsbach, 2003; Pfarrer, DeCelles, Smith, &Taylor, 2008). Examples range from outright denialof responsibility to attacking accusers and shift-ing blame onto other entities (Coombs& Holladay,2004; Elsbach, 2003). In contrast, strategies thataccept more responsibility—generally labeledaccommodative—attempt to manage social ap-proval loss by acknowledging an organization’scausal role in a crisis and reducing evaluators’negative perceptions (Bottom, Gibson, Daniels, &Murnighan, 2002; Coombs & Holladay, 2004).Examples range from outright acceptance ofresponsibility to apologies and displays of

mortification (Coombs, 2011; Kim & Yang, 2009).Finally, response strategies in the center ofthe continuum attempt to reframe how eval-uators perceive a crisis, while not necessarilyaccepting or denying responsibility (Coombs &Holladay, 2004). Examples include providingexcuses (downplaying an organization’s respon-sibility), justifications (minimizing the negativityof the event), and partial deflections (Elsbach,2003).4

The Crisis-Response Match

Communication and public relations research-ers have argued that an effective responsestrategy should match evaluators’ situationalattributions of a crisis (Coombs, 1995, 2007b;Coombs & Holladay, 2004). A crisis with highersituational attributions of responsibility should bematched with a response strategy that acceptsmore responsibility, and a crisis with lower situ-ational attributions of responsibility should bematched with a response strategy that acceptsless responsibility (Coombs, 1995; Coombs &Holladay, 2004). Empirical research provides sup-port for this logic. For example, Zavyalova et al.(2012) found that a more accommodative strategyled to less negative media coverage than a moredefensive strategy after a product recall—a cri-sis with generally higher situational attribu-tions. Similarly, Coombs and Holladay (2004)found that a matched response that neither ac-cepted nor rejected crisis responsibility led toless social approval loss than a mismatchedresponse that denied responsibility after a tech-nical industrial accident—a crisis with gener-ally moderate situational attributions.However, despite the breadth of work on

evaluators’ situational attributions of a crisisand an organization’s response strategy, crisismanagement scholars have yet to fully considerthe sociocognitive mechanisms that generatea crisis-response match (or mismatch) and theresulting benefit (or burden) to an organization’ssocial approval.

4 In addition to response strategies used to manage eval-uators’ attributions, we make the assumption that an organi-zation provides instructing information to help evaluatorsavoid harm that may result from the crisis (cf. Coombs, 2011).


A Sociocognitive Perspective of the Crisis-Response Match

As mentioned above, evaluators’ situationalattributions are subject to a degree of social in-fluence from an organization’s response strategy.However, the extent of this influence varies. Forexample, an organizationwill have a difficult timedenying complete responsibility for an on-siteexplosion or a scandal emanating from the exec-utive office. Thus, the degree to which a responsestrategy is able to influence evaluators’ sense-making will be constrained by how well thestrategy conforms to the evaluators’ intuitive sit-uational attributions of the crisis. Figure 1 depictsthe matched and mismatched zones betweenevaluators’ situational attributions of a crisis andan organization’s response strategy. Matchedstrategies conform to evaluators’ situationalattributions, as shown in the center diagonal areaof Figure 1, which we label the “zone of confor-mity.” In contrast, mismatched strategies either

underconform by accepting less responsibilitythan evaluators expect, based on their situationalattributions (shown in the lower right “zone ofunderconformity” in Figure 1), or overconform byaccepting more responsibility than evaluatorsexpect (shown in the upper left “zone of over-conformity” in Figure 1).We argue that the crisis-response match can

best be understood in how it influences theprobable distribution of social approval loss fora given crisis. All crises are uncertain events thatgenerate initial negative reactions. An organi-zation therefore can expect a negative residualeffect on its social approval at the onset of anycrisis, regardless of its response and the level ofsituational attributions. All else being equal,a crisis that generates higher situational at-tributions of responsibility should result ina greater average magnitude loss to an orga-nization’s social approval than a crisis thatgenerates lower attributions of responsibility.However, given the high uncertainty at the onset

FIGURE 1The Crisis-Response Match


of a crisis, the expected probability distributionsurrounding social approval loss for any givencrisis is less clear. We argue that a matchedresponse strategy that conforms to evaluators’situational attributions normalizes the proba-bility distribution of social approval loss fora given crisis such that the uncertainty sur-rounding potential loss is reduced. Stated moreformally, “normalize” describes a distribution ofexpected outcomes that is centered around themean, thus reducing the variance of social ap-proval loss.

Figure 2 depicts a set of crises with lower,moderate, and higher situational attributions. Thex-axis depicts a hypothetical scale representingthe potential magnitude of social approval loss.The y-axis represents the hypothetical probabilityof experiencing a particular magnitude of loss(scaled as percentages, with the total area un-derneath each distribution equal to 100 percent).Each distribution can be thought of as the aggre-gate of anyN number of organizations respondingto a similar yet independent crisis with a matchedstrategy. The three distributions in Figure 2 showthat the average magnitude of expected loss

to social approval increases as situationalattributions increase. Figure 2 also highlights thenormalized probability distributions resultingfrom the use of a matched strategy for a givenattribution level (as identified in Figure 1). Forexample, the hypothetical scales suggest that thelarge majority of organizations responding to amoderate attribution crisis with a matched strat-egy would experience a midrange magnitudesocial approval loss.In contrast to a matched strategy, we argue

that a mismatched strategy that either overcon-forms or underconforms to evaluators’ situationalattributions not only increases the averagemagnitude of loss by shifting the peak of thedistribution to the right but also broadens thecurve such that the variance of potential out-comes increases. Figure 3 depicts the distribu-tion of social approval loss expected fora matched strategy compared to the distributionfor a mismatched strategy, assuming a moderateattribution crisis. We explain each distribution inmore detail below.Matched strategy. The solid curves in Figures 2

and 3 depict a normalized distribution of social

FIGURE 2Expected Probability and Magnitude of Social Approval Loss for a Matched Response Strategy and

a Lower, Moderate, and Higher Attribution Crisisa

aThe probability and magnitude scales represent hypothetical unit increases.


approval loss associated with amatched responsestrategy from Figure 1. There are two primarysociocognitive explanations for this distribution.First, as mentioned above, a matched strategyprovides information to evaluators about anorganization’s role in a crisis that conforms totheir situational attributions. Scholars have longrecognized that evaluators display reluctance torevise already formed judgments and attributions(e.g., Bottom et al., 2002; Nickerson, 1998; Traut-Mattausch, Schulz-Hardt, Greitemeyer, & Frey,2004; Tversky & Kahneman, 1974). This confirma-tion bias restricts evaluators’ attention to con-firming information, causing them to ignore orresist disconfirming information (Traut-Mattauschet al., 2004). As such, a matched strategy createsa state of cognitive consonance in which eval-uators are likely to agree with an organization’sposture. A matched strategy therefore facilitatesevaluators’ sensemaking and can normalize theloss to social approval for a given level of situa-tional attributions.

Second, there is also a normative componentto the crisis-response match along the entire

zone of conformity in Figure 1. For example, anorganization’s acceptance of responsibility fora crisis with higher situational attributions sat-isfies social expectations of justice, sincerity, andfairness (Coombs, 2007b; Dean, 2004; Pfarrer,DeCelles, Smith, & Taylor, 2008; Tyler, 1997).Evaluators may interpret an organization’s ac-commodativeness as a sign of goodwill (Benoit,1995; Pfarrer, DeCelles, Smith, & Taylor, 2008;Sutton & Callahan, 1987). Evaluators may alsogrant an organization a pardon, in which case itis essentially rewarded—or penalized less—forcoming forward and taking responsibility (Pfarrer,Smith, Bartol, Khanin, & Zhang, 2008; Tyler, 1997).Thus, accepting increasing responsibility as eval-uators’ situational attributions increase shouldcontribute to a more predictable and normal dis-tribution of social approval loss.For crises generating moderate situational

attributions of responsibility, evaluators’ norma-tive expectations may be focused more on “set-ting matters aright going forward” than onhaving an organization take responsibility for thecrisis (Koehn, 2013: 249). Thus, a response in the

FIGURE 3The Trade-off Between a Matched and Mismatched Response Strategya

aThe probability and magnitude scales represent hypothetical unit increases. The matched strategy for the moderateattribution crisis from Figure 2 is used for clarity.


center of the response strategy continuum thatneither accepts nor rejects responsibility is likelyan effective match. For example, in response toa product recall that resulted when rogue sup-pliers added lead paint to toys, Mattel took effortsto prevent a similar situation from happeningin the future but did not accept outright re-sponsibility for its suppliers’ actions (Koehn,2013). Mattel’s response was consistent withevaluators’ normative expectations, since fewexpected the company to have “omniscient”control over unprecedented supplier deviance(Koehn, 2013: 247).

Finally, evaluators typically have lower nor-mative expectations of an organization for criseswith lower situational attributions. Thus, a moredefensive response strategy where an organiza-tion does not accept responsibility is not likely toviolate evaluators’ normative expectations and,therefore, is the matched response for such cri-ses. For example, while a general expression ofsorrow may be appropriate in the wake of a nat-ural disaster or an act of terrorism, evaluatorslikely do not have expectations for an organiza-tion to accept responsibility for the event itself(Koehn, 2013).

Returning to Figure 1, we expect the responsestrategies in the matched zone of conformity tonormalize the distribution of social approval lossbecause of their conformity to evaluators’ initialsituational attributions of responsibility andnormative expectations. This normalized distri-bution is depicted in Figure 2 for a range ofsituational attribution crises. In contrast, wetheorize that the mismatched strategies in thezones of underconformity and overconformityfrom Figure 1 will generate a distribution of so-cial approval loss more reflective of the mis-matched (dashed) curve in Figure 3.

Mismatched strategy. Instead of conformingto evaluators’ situational attributions, a mis-matched strategy challenges evaluators’ initialperceptions of a crisis. Such a challenge isa “sensebreaking” event that triggers cognitivedissonance among evaluators (Festinger, 1957;Pratt, 2000: 464) and results in a higher averagemagnitude of loss distributed over a widerrange for a given crisis. As with the matchedcurve, there are several explanations for thisdistribution.

First, we argue that cognitive dissonance cre-ated by the sensebreaking response is a strongmotivator for evaluators to increase information

search (Ashforth, Harrison, & Corley, 2008; Elliot& Devine, 1994). This search can lead to con-clusions that are contrary to the organization’smismatched response. In such instances eval-uators may deduce that an organization’smismatched response was careless at best, orinaccurate, wrong, or even deceitful at worst(Benoit, 1995; Dean, 2004; Elsbach, 2003; Pfarrer,Smith, Bartol, Khanin, & Zhang, 2008). For exam-ple, an organization that is overconforming bybeing accommodative during a crisis with lowersituational attributions risks being perceived asinsincere or overreaching, leading evaluators toreduce their perceptions of social approval asthey attempt to rectify their dissonance (Ashforth& Gibbs, 1990). Similarly, an organization that isunderconforming by being defensive in responseto a crisis with higher situational attributionsrisks being perceived as unethical and manipu-lative. In such cases amismatched responsemaylead to a state of “perpetual discordance,” pre-venting crisis resolution and likely increasing anorganization’s social approval loss (Pfarrer,DeCelles, Smith, & Taylor, 2008: 737). This in-crease in expected loss is depicted by the mis-matched (dashed) curve in Figure 3, which showsa higher average magnitude loss compared tothe matched curve.Second, evaluators’ attribution biases can

lead them to resist an underconforming or over-conforming mismatched strategy. As mentionedabove, the confirmation bias leads evaluators toprefer their initial attributions, either ignoring orchallenging inconsistent information. Similarly,the anchoring bias suggests that even if eval-uators do make subsequent adjustments, theywill be hesitant to stray too far from their initialattributions (Tversky & Kahneman, 1974). Thus,the more a response strategy is mismatched toevaluators’ situational attributions, the lesslikely evaluators are to view it as appropriate. Forexample, evaluators will likely respond nega-tively to a defensive response from an executivecaught misappropriating organizational funds. Asa result, the distribution of social approval lossshifts to the right since evaluators are inclined toreject a nonconforming response.Finally, we also recognize that the distribution of

potential outcomes is wider and more uncertainwhen using a mismatched strategy. The dis-confirming information in an organization’s mis-matched strategy can hinder evaluators frommaking definitive attributions regarding the nature


of a crisis and an organization’s responsibility.This may cause some evaluators to alter theirperceptions of the response or the crisis. For ex-ample, evaluators may perceive an organization’soverconformity in taking more responsibility fora moderate attribution crisis as an additional signof goodwill rather than as a sign of disingenuous-ness. Similarly, some evaluators may resist theirbiases and begin to doubt their original intuitiveattributions. Others will be reluctant to expend thetime and effort required to gathermore informationand, thus, will be pacified by an organization’smismatched posture (Connelly, Certo, Ireland, &Reutzel, 2011; Sanders & Carpenter, 2003). Stillothers will continue their search but may not findthe diagnostic information needed to challenge themismatched response. For instance, an under-conforming defensive response to a crisis withhigher situational attributions may sow seeds ofdoubt among evaluators. Thus, uncertainty fromthe disconfirming response—and the effect of un-certainty on evaluators’ sensemaking and attribu-tions—can increase the expected distribution ofsocial approval loss.

In sum, a mismatched strategy, whether lyingin the zones of overconformity or underconformityin Figure 1, is likely to increase the averagemagnitude of social approval loss for a givencrisis. However, because it is difficult to predicthow evaluators will collectively perceive anorganization’s nonconformity, especially amidhigh uncertainty (cf. Heckert & Heckert, 2002), weargue that the variance of the distribution willalso increase. Figure 3 shows that the peak of themismatched curve is to the right of the matchedcurve, indicating an increased probability fora higher-magnitude loss. Figure 3 also showsthat the mismatched distribution is wider andmore uncertain, making it more difficult for man-agers and evaluators to understand and predictthe consequences for an organization’s socialapproval. Combining these arguments with ourarguments related to a matched strategy, wepropose the following.

Proposition 1: The more an organiza-tion’s response strategy matches eval-uators’ situational attributions of crisisresponsibility, the lower the mean andvariance of social approval loss.

In this section we focused on an organization’ssocial approval as an outcome of a crisis. Belowwe examine how an organization’s endowment of

social approval acts as an important antecedentfor evaluators’ perceptions of a crisis.

THE EFFECT OF AN ORGANIZATION’SENDOWMENT OF SOCIAL APPROVAL ON

EVALUATORS’ CRISIS PERCEPTIONS

In addition to situational attributions and anorganization’s response strategy, one important—yet often neglected—source of information forevaluators as they make sense of a crisis is anorganization’s endowment of social approval.Evaluators rely on social evaluations as a “cog-nitive shorthand” to help them make sense of anorganization’s actions (Mishina et al., 2012: 460).Consequently, an organization’s endowment ofsocial approval will likely influence the wayevaluators perceive and react to a crisis and,thus, the probable social approval loss associ-ated with an organization’s response strategy.In this section we theorize that the probability

distributions associated with the matched andmismatched strategies described above are al-tered as an organization’s endowed social ap-proval increases or decreases. These altereddistributions are depicted in Figure 4. We arguethat endowed levels of social approval can serveas a buffer to decrease the magnitude of socialapproval loss (depicted as the darkly shadedmatched [solid] curve on the far left side ofFigure 4) or as a burden to increase the magni-tude of social approval loss (depicted as themismatched [dashed] curve on the far right sideof Figure 4), relative to the average-approvalorganization (depicted as the lightly shadedmatched [solid] curve in the center of Figure 4).

Crisis Perceptions of an Organization withHigher Social Approval

Research investigating the influence of endowedsocial approval following a negative event sug-gests a conundrum. On the one hand, scholarshave argued and found that evaluators givehigher-approval organizations the benefit of thedoubt when disconfirming information arises(cf. Burgoon, 1993; Coombs & Holladay, 2006;Hollander, 1958; Pfarrer et al., 2010; Schnietz &Epstein, 2005). Because of evaluators’ affinity to-ward a higher-approval organization, they maybe hesitant to conclude that such an organiza-tion could be as responsible for a negative eventas other organizations. Thus, an endowment of


higher social approval can serve as a bufferagainst loss as evaluators either ignore ordownplay negative information that is incon-gruent with their prior positive social judgments(Burgoon, 1993). In this way evaluators may beinclined to attribute less responsibility for a cri-sis to an organization with higher levels of so-cial approval, even if the crisis characteristicsmight normally signal higher situational at-tributions. Empirical findings provide supportfor this buffer. For example, Coombs andHolladay (2001) found a significant negative cor-relation between evaluators’ perceptions of crisisresponsibility and an organization’s socialapproval.

On the other hand, a separate stream of re-search suggests that higher levels of social ap-proval can serve as a burden that exacerbatesloss following a negative event (cf. Ashforth &Gibbs, 1990; Brooks et al., 2003; Graffin, Bundy,Porac,Wade, & Quinn, 2013; Merton, 1968; Mishinaet al., 2012; Rhee & Haunschild, 2006; Sutton &Galunic, 1996). Evaluators may have heightenedstandards and expectations for organizations they

view positively (Burgoon, 1993; Pfarrer et al., 2010).When a higher-approval organization is associ-ated with a negative event, it can serve as a starkviolation of these standards and expectations,which can lead to dissonance and feelings ofbetrayal (e.g., Brehm & Brehm, 1981; Decker, 2012;Mishina et al., 2012; Rhee & Haunschild, 2006;Sutton & Galunic, 1996). Evaluators may alsobegin to suspect that “ulterior motives” werebehind a higher-approval organization’s priorbehavior, causing them to question the orga-nization’s values and motivations (Mishina et al.,2012: 466). For example, describing a product re-call as a breach of an “implicit promise,” Rheeand Haunschild (2006: 103) found that more highlyapproved companies suffered a greater loss ofmarket share when issuing recalls than did lessapproved competitors.Taking these past theoretical arguments and

empirical findings together, it remains unclearhow higher levels of social approval might affectevaluators’ perceptions of and reactions to theonset of a crisis. We theorize that this conundrumis best understood by considering the effects of

FIGURE 4The Trade-off Between aMatched and Mismatched Response Strategy for Higher- and Lower-Approval

Organizationsa

aThe probability and magnitude scales represent hypothetical unit increases. The matched strategy for the moderateattribution crisis from Figure 2 is used for clarity.


an organization’s endowment of social approvalon evaluators’ initial situational attributions ofresponsibility combined with an organization’sresponse strategy.

The higher-approval buffer.As detailed above,different biases and expectations may inclineevaluators to attribute less crisis responsibilityto a higher-approval organization, regardless ofthe situational crisis characteristics. However,evaluators can only attribute less responsibilityas long as the higher-approval organization’sresponse strategy matches the evaluators’ al-tered perceptions of the crisis. For example,Figure 4, like Figure 3, illustrates a crisis thathas moderate situational attributions for anaverage-approval organization. As Figure 1 and2 show, an average-approval organization’smatched strategy would be in the center of theresponse strategy continuum, and the organi-zation would expect a moderate-magnitude so-cial approval loss. Figure 4 depicts this matchas the lightly shaded matched (solid) curve. Incontrast, a higher-approval organization facingthe same crisis would likely be buffered suchthat evaluators would attribute less respon-sibility to the organization—perceiving themoderate attribution crisis as a lower attribu-tion crisis. The effect of the buffer makes a moredefensive strategy a more suitable match fora higher-approval organization, as depicted inthe lower left of Figure 1. Combined with eval-uators’ reduced attributions, a more defensivestrategy should decrease the expected magni-tude of social approval loss when compared toan average-approval organization facing thesame crisis. This buffering effect of highersocial approval is shown on the far left side ofFigure 4 by the darkly shaded matched (solid)curve.

By matching its response to evaluators’ re-duced attributions, a higher-approval organiza-tion allows evaluators to rely on their positiveprior judgments to downplay divergent crisisinformation. For example, when referring toApple’s defensive strategy related to its iPhone 4antenna issues, an industry analyst noted, “Ap-ple has a history of making ridiculous claims andhaving them accepted by an adoring fan baseand worshipful press” (Lyons, 2010: 1). Apple’shigher social approval—combined with its moredefensive response—likely reduced evaluators’situational attributions of responsibility. Whatcould have been perceived as a crisis with

moderate situational attributions (a technicalproduct error) for an average-approval organi-zation was instead largely perceived as a crisiswith lower situational attributions (a rumor) forApple, a higher-approval organization.The overall implication of the buffer is that, for

a given crisis, a higher-approval organizationhas the opportunity to accept less crisis re-sponsibility than its average-approval coun-terpart. In contrast, when a higher-approvalorganization accepts the same or more crisisresponsibility, we expect a burdening effect toprevail.The higher-approval burden. The heuristics

and biases associated with the social approvalbuffer are likely only effectual when an orga-nization’s response strategy matches evaluators’reduced attributions of crisis responsibility.Thus, a response strategy that accepts more cri-sis responsibility than indicated by evaluators’altered situational attributions is mismatchedand creates a sensebreaking situation of non-conformity. In such situations evaluators arelikely compelled to question their prior judg-ments and may be forced to face the conclusionthat a higher-approval organization is respon-sible for a crisis. To rectify their cognitive dis-sonance, evaluators likely will reduce theirapproval of the organization. As mentionedabove, such dissonance is also likely to triggerfeelings of betrayal (Koehler & Gershoff, 2003;Mishina et al., 2012), further exacerbating loss asevaluators intuitively react to a higher-approvalorganization’s failure to meet their heightenedstandards.Returning to Figure 4, we argue that a higher-

approval organization’s movement away froma more defensive strategy and toward a moreaccommodative strategy will trigger evaluators’perceptions of mismatched overconformity. Theorganization’s acceptance of more crisis re-sponsibility does not conform to evaluators’ priorpositive perceptions and their reduced situa-tional attributions. The effect of this burden ofresponsibility is depicted on the far right side ofFigure 4 by the mismatched (dashed) curve. Themismatched curve shows a higher probabilityfor a higher-magnitude social approval loss.Importantly, a mismatched strategy for a higher-approval organization may be a matched strat-egy for an average-approval organization. Forexample, an average-approval organizationwould match a higher attribution crisis with


a more accommodative response, while eval-uators would likely respond to the same re-sponse from a higher-approval organization asa mismatch.

We also recognize that the distribution ofpotential outcomes is increased when a higher-approval organization uses a mismatched re-sponse strategy. Again, this is due to theheightened uncertainty and sensebreaking thataccompany a mismatched response. It may bedifficult for evaluators to make sense of the non-conformity, which increases the distribution of po-tential outcomes. For example, a higher-approvalorganization providing a comprehensive accom-modative explanation of a crisis—including offer-ing compelling logic for its responsibility andspecific details of its proactive efforts to preventreoccurrence—may be spared excessive socialapproval damage, even when the response isperceived as a mismatch (Coombs, 2007a). How-ever, such a nuanced response may be difficult todeliver at the onset of a crisis, when uncertainty isat its peak.

In summary, we expect a burdening effect toprevail when a higher-approval organizationaccepts the same or more crisis responsibility fora given crisis, relative to an average-approvalcounterpart. As a result, the distribution of socialapproval loss identified by the mismatched(dashed) curve in Figure 4 is more likely to occur.In contrast, we expect a buffering effect to prevailwhen a higher-approval organization acceptsless crisis responsibility, relative to an average-approval counterpart. As a result, the distributionof social approval loss identified by the darklyshaded matched (solid) curve in Figure 4 is morelikely to occur.

Proposition 2: For an organization withhigher social approval, a responsestrategy that accepts less crisis re-sponsibility, relative to an average-approval organization, will generatea lower mean and variance of socialapproval loss than a response strategythat accepts the same or more crisisresponsibility.

Crisis Perceptions of an Organization with LowerSocial Approval

Scholars have done little research on theeffects of negative social evaluations in the

context of a negative event. We develop theory inthis domain by arguing that a lower-approvalorganization in a crisis also faces a potentialbuffer or burden. Unlike for a higher-approvalorganization, however, the buffering effect fora lower-approval organization does not resultfrom evaluators’ altered attributions of crisis re-sponsibility (cf. Coombs & Holladay, 2001). Rather,it results from evaluators’ reduced standards forthe lower-approval organization.An organization accrues lower approval by

consistently invoking negative feelings andaffinity. Over time, evaluators may becomedesensitized to a lower-approval organization’sbehaviors, giving each successive disappoint-ment less attention and concern (e.g., Ahmadjian& Robinson, 2001). Because evaluators’ standardsof conduct are intuitively reduced for a lower-approval organization, it is harder for such anorganization to violate evaluators’ expectations(Burgoon, 1993). Additionally, evaluators arelikely to be less motivated to engage in extensivesensemaking when an organization they per-ceive negatively is associated with a crisis. Suchan organization is effectively “screened out ofconsideration” in the first place (Phillips &Zuckerman, 2001: 385). Thus, for a lower-approvalorganization, evaluators’ response to a crisis—interms of the expected magnitude of social ap-proval loss—may be muted.However, as with a higher-approval organiza-

tion, evaluators’ reactions to a lower-approvalorganization may also be more extreme, sug-gesting a potential burdening effect. For exam-ple, in research on reciprocity and recidivism,scholars argue that a crisis may serve asa breaking point, compelling evaluators to enact“durable changes to the rules” and to punisha lower-approval organization for its repeatedfailures (Ballinger & Rockmann, 2010: 374;Braithwaite, 1989; Davidson, Worrell, & Lee, 1994;Fehr & Gachter, 2000; Pfarrer, DeCelles, Smith, &Taylor, 2008). Thus, rather than facing the samemeasurable magnitude of social approval lossthat a higher-approval organization faces,a lower-approval organization risks evokingcollective disapproval, which may threaten itssurvival.As with higher approval, we argue that this

conundrum is best understood by consideringthe influence of evaluators’ situational at-tributions and an organization’s responsestrategy.


The lower-approval buffer. We argue that a re-sponse strategy that rejects or is more equivocalabout crisis responsibility is likely to buffera lower-approval organization. As mentionedabove, evaluators already have reduced expec-tations and reduced standards for a lower-approval organization and likely will not besurprised when such an organization is associ-ated with a crisis (Burgoon, 1993). A more de-fensive strategy does little to alter these beliefsand, instead, can generate additional uncertainty.As evaluators’ dissonance and sensemaking arereduced in response to this uncertainty, so, too, isthe potential magnitude of damage to socialapproval. For example, in responding to chal-lenges regarding the addictive nature of nicotine,many cigarette manufacturers denied the accu-sations in an attempt to “place an element ofdoubt or confusion in the minds of consumers orpotential consumers” (Ulmer & Sellnow, 1997:231). When confusion and doubt grow, a lower-approval organization can be spared further lossas many evaluators become desensitized anddisengaged. Thus, a lower-approval organiza-tion’s response strategy that accepts less re-sponsibility can be understood as a conformingmatched response and can trigger a bufferingeffect.

The lower-approval burden. In contrast,a lower-approval organization that acceptsmore responsibility for a crisis is likely to trig-ger evaluators’ extreme negative reactions.Accepting more crisis responsibility risks rein-forcing evaluators’ negative affinity and trig-gering a tipping point, beyond which evaluatorscan no longer accept the organization as anapproved entity. That is, responsibility for a cri-sis is likely to compel the active punishment ofa lower-approval organization as evaluatorsenforce notions of reciprocity and signal theimportance of positive behavior (e.g., Abbink,Irlenbusch, & Renner, 2000; Davidson et al., 1994;Fehr & Gachter, 2000). Thus, we argue thata response strategy that accepts the sameor more crisis responsibility relative to anaverage-approval organization is mismatchedand will trigger a burdening effect for a lower-approval organization. This effect of theburden of responsibility is represented by themismatched (dashed) curve on the far right ofFigure 4, which shows a greater probability ofa higher-magnitude social approval loss fora lower-approval organization, relative to an

average-approval organization. As with higherapproval, what is considered a mismatchedstrategy for a lower-approval organization canbe considered a matched strategy for an average-approval organization.We also recognize that, as with a higher-

approval organization, the variance of outcomesis increased when a lower-approval organizationuses a mismatched response strategy. For ex-ample, it is possible that evaluators may per-ceive a lower-approval organization’s moreaccommodative response as a positive expec-tancy violation (Burgoon, 1993) and a signal thatthe organization is motivated to change its ways.This could trigger evaluators to reward the lower-approval organization for its acceptance of re-sponsibility, so the magnitude of social approvalloss would be reduced. However, proactive ac-commodation can also confirm evaluators’ gen-eral perceptions of lower approval, which islikely to reinforce evaluators’ perceptions thatthe organization deserves conferral of negativeaffinity. Thus, the resulting probability distri-bution would widen because of the increaseduncertainty regarding how evaluators might re-spond to the nonconformity of a mismatchedresponse.In summary, we expect the burdening effect to

prevail when a lower-approval organizationaccepts the same or more crisis responsibility fora given crisis, relative to an average-approvalcounterpart. As a result, the distribution of socialapproval loss identified by the mismatched(dashed) curve in Figure 4 is more likely to occur.In contrast, we expect the buffering effect toprevail when a lower-approval organizationaccepts less crisis responsibility, relative to anaverage-approval counterpart. The resultingdistribution of social approval loss will resemblethe darkly shaded matched (solid) curve on thefar left of Figure 4, which shows a greater proba-bility of a lower-magnitude loss relative to anaverage-approval organization.

Proposition 3: For an organization withlower social approval, a responsestrategy that accepts less crisis re-sponsibility, relative to an average-approval organization, will generatea lower mean and variance of socialapproval loss than a response strategythat accepts the same or more crisisresponsibility.


PREDICTING RESPONSE STRATEGIES FORORGANIZATIONS ENDOWED WITH HIGHER

OR LOWER APPROVAL

Given our arguments above, managers ata higher- or lower-approval organization may betempted to accept less crisis responsibility atthe onset of a crisis. Managers at a higher-approval organization will want to protecta valuable asset, whereas managers at a lower-approval organization will want to avoid dis-approval and potential failure. Spurred by thesemotivations, a higher-approval organizationmay act more defensively, knowing that itsendowed approval can provide a buffer frommore unfavorable attributions of crisis re-sponsibility. A lower-approval organization isalso likely to respond more defensively, know-ing that it has little to lose in the way of goodwilland hoping that evaluators perceive the crisisas just another example of why the organizationdeserves lower approval—as opposed to out-right disapproval.

Research investigating managerial risk pref-erences and decision making provides supportfor how endowed social approval influencesmanagers’ choice of response strategies (e.g.,Argote & Greve, 2007; Cyert & March 1963; March& Shapira, 1987). For example, in situations ofhigh uncertainty, such as the onset of a crisis,managers will look for strategic opportunities tocontrol loss (Cyert & March 1963; March &Shapira, 1987, 1992). Managers at a higher- orlower-approval organization are likely to per-ceive that a more accommodative strategy willpresent an almost certain social approval loss,whereas a less accommodative strategy willpresent an opportunity to reduce loss. Addition-ally, when making risky decisions, such asresponding to a crisis, managers are often moresensitive to the potential magnitude of lossrather than to the probability of loss (March &Shapira, 1987). Given that managers are likelyto perceive that a more accommodative re-sponse will increase the magnitude of loss,managers of a higher- or lower-approval or-ganization may be motivated to avoid suchoutcomes.

Proposition 4: Managers of a higher- orlower-approval organization will bemore likely to accept less crisis re-sponsibility, relative to managers ofan average-approval organization.

DISCUSSION

In this article we have applied theories fromsocial judgment formation, perception manage-ment, and decision making to examine the role ofan organization’s social approval as both a criti-cal outcome of and important antecedent toevaluators’ perceptions of a crisis. In doing sowe offered four primary contributions to re-search. First, we specified the sociocognitivemechanisms that make social approval a distinctsocial evaluation ideally suited for research oncrises. Second, we advanced understanding ofcrisis management by identifying the socio-cognitive processes underlying the crisis-response match. Third, we detailed how anorganization’s endowment of social approvalinfluences this match, arguing that higher andlower levels of social approval act as eithera buffer or a burden to modify evaluators’ crisisperceptions and attributions. Finally, given theburden of responsibility associatedwith endowedsocial approval, we theorized that managers athigher- and lower-approval organizations will bemore likely than managers at other organizationsto use a response strategy that accepts less re-sponsibility for a crisis. Below we discuss thesocial and theoretical implications of our theory,as well as limitations and directions for futureresearch.

Social Implications

We recognize that our theory may raise nor-mative and ethical concerns for scholars andmanagers. Our intention has been to theorize anddescribe how evaluators’ sociocognitive biasesand reactions related to social approval in-fluence an organization’s crisis managementstrategies. Importantly, we do not prescribe thathigher- and lower-approval organizations shouldaccept less responsibility for a crisis. Rather, wesuggest that they could be buffered from socialapproval loss, at least at the onset of a crisis.Who is responsible for a crisis is often a questionof perception rather than fact, especially at theonset (Coombs, 2007b; Gephart, 2007; Sellnow &Seeger, 2013). Such equivocality serves asa foundation for investigating the sociocognitiveeffects we describe in this article. Additionally,because of this ambiguity, many scholars havechallenged whether it is appropriate and ethicalfor organizations to always be accommodative in


response to a crisis (cf. Coombs & Holladay,2008; Koehn, 2013). Indeed, some have suggestedthat, in extreme cases of uncertainty, the mostethical response is to “remain strategically am-biguous” so as not to unnecessarily fan theflames (Sellnow & Seeger, 2013: 227).

This tension highlights an important boundarycondition for our theory, which only considers theeffects of social approval under the conditionsof high uncertainty prevailing at the onset ofa crisis. Of course, as a crisis evolves, more in-formation may come to light, and uncertaintyabout an organization’s responsibility may bereduced. Thus, in the latter stages of a crisis,a truly misleading defensive strategy may offerfew benefits for a higher- or lower-approval or-ganization, and the burden of being overly de-fensive may lead to greater social approval lossthan being more accommodative in the firstplace (cf. Kim, Ferrin, Cooper, & Dirks, 2004;Pfarrer, DeCelles, Smith, & Taylor, 2008; Pfarrer,Smith, Bartol, Khanin, & Zhang, 2008). This isalso true for crises in which managers un-ambiguously know that their organization is re-sponsible. Thus, like Benoit, we advocate that“those who are truly at fault should admit it im-mediately and take appropriate corrective ac-tion” (2005: 409). Of course, we also contend thatsuch unequivocal knowledge of responsibility israre at the onset of a crisis, and it may take muchtime before an organization and evaluatorsagree on the facts, rather than their perceptionsof what happened and why (cf. Gephart, 2007;Pfarrer, DeCelles, Smith, & Taylor, 2008).

Given these social implications, it is our hopethat our theory can have important societal andorganizational applications. From a societal per-spective, an understanding of the role of an orga-nization’s social approval should allowevaluators torecognize their potential biases when making crisisattributions. This may help them be wary of overlydefensive strategies or more accepting of accom-modative strategies, especially from higher- andlower-approval organizations and particularly atthe onset of a crisis. It should also allow evaluatorsto have a deeper consideration of the crisis situa-tion, reducing their tendencies toward intuitive andvisceral reactions. Such consideration may changethe dynamics of the social approval buffer andburden and, ultimately, may encourage higher-and lower-approval organizations to take more re-sponsibility for a crisis than theory, investors, andtheir legal counsel might suggest.

From an organizational perspective, an en-hanced awareness of the trade-offs associatedwith a given endowment of social approval, thedifferent response strategies, and the dynamicsof the crisis-response match should reduce thecommon tendency for organizations to offer aninitial, automatic, and often defensive responsethat limits attributions of responsibility. Suchawareness should also enhance the ability oforganizations to focus on resolving crisis sit-uations and to move more quickly toward re-integration with evaluators (Pfarrer, DeCelles,Smith, & Taylor, 2008). The result would be moreeffective organizational communication, more ef-fective management, and, ultimately, increasedbenefits for society.

Implications for Theory and Future Research

Research in risk management and managerialdecision making has highlighted the trade-offsamong strategic choices in terms of probabilitydistributions and expected outcomes (e.g., Gephart,Van Maanen, & Oberlechner, 2009; March &Shapira, 1987; Tversky & Kahneman, 1974). Usingthis logic, we identified the sociocognitive mech-anisms that drive the probable distributions ofsocial approval loss in a crisis. In doing so weadded refinement to traditional research on thecrisis-response match, including highlightingzones of overconformity and underconformity andthe trade-offs associated with matched and mis-matched strategies.Our sociocognitive focus on the crisis-response

match was centered on evaluators’ situationalattributions of responsibility relative to theamount of responsibility an organization takesfor a crisis. Underlying our argument was anassumption that an organization is satisfyingevaluators’ “basic information needs,” such asproviding emergency information to protectevaluators from harm (Coombs, 2007b; Coombs &Holladay, 2004: 99), and that the organization istaking appropriate actions to contain the crisis(cf. Pfarrer, DeCelles, Smith, & Taylor, 2008). Al-though these assumptions are inherent in crisiscommunication research (cf. Coombs, 2011), fu-ture research could investigate the moderatingeffects of substantive organizational actionsother than response strategies that influenceevaluators’ attributions (e.g., Pfarrer, DeCelles,Smith, & Taylor, 2008; Zavyalova et al., 2012).Additionally, some organizations are more


prepared to handle a crisis than others (cf. Turner,1976). For example, high-reliability organiza-tions may emphasize substantive actions tocontrol the crisis over more symbolic actions tomanage social approval (Roberts, 1990; Weick &Sutcliffe, 2001). An organization’s ability toassert control over the crisis, especially at itsonset, may influence the dynamics of our theoryand could dampen the effect of the respon-sibility burden.

Our theory focused on crisis-response strate-gies in terms of an organization’s level of ac-ceptance of responsibility. However, we recognizethat, in practice, organizational responses aresometimes more nuanced (e.g., Coombs, 2007a;Lamin & Zaheer, 2012). For instance, we suggestedthat an organization with higher or lower socialapproval may benefit from using a more de-fensive strategy at the onset of a crisis. Futureresearch could investigate how that benefit wouldbe moderated by more or less extreme defen-siveness (e.g., an outright denial versus a de-flection strategy), more or less extreme crises (e.g.,those with particularly high or low situationalattributions of responsibility), or during the latterstages of a crisis. Additionally, we did not con-sider an organization’s reticence at the onset ofa crisis. However, researchers have found evi-dence that reticence is often suboptimal to otherresponse strategies (e.g., Decker, 2012; Ferrin, Kim,Cooper, & Dirks, 2007). Because theory and find-ings suggest the need for an organization to pro-vide a consistent message at the onset of a crisis(Massey, 2001), we assumed that if an organizationwants to protect its social approval, it not only willprovide an initial response but will provide a re-sponse that speaks to both direct andmore diffuseevaluators.

We also recognize the special case of a mis-matched strategy for higher- and lower-approvalorganizations facing crises that would normallytrigger higher situational attributions of re-sponsibility. In such crises the burdening effectmay be possible when a higher- or lower-approval organization goes too far by giving anoverly defensive response. Such a response risksincurring evaluators’ anger and resentmentshould it be found to be untrue, particularly whendiagnostic information suggests that the orga-nization is indeed responsible for the crisis. Forexample, once a member of the highly ap-proved Big Five group of accounting firms,Arthur Andersen found its higher social

approval threatened as its role in the Enron crisiswas revealed. Arthur Andersen executives im-mediately denied all responsibility for Enron’sfailures, while secretly destroying evidence(McLean & Elkind, 2003). Pundits claimed thatwhen this came to light, it “tar[red] the name ofthe venerable Arthur Andersen” (Kadlec, 2002),and several scholars have argued that ArthurAndersen’s subsequent loss of social approvalled to its eventual downfall (e.g., Chaney &Philipich, 2002; Krishnamurthy, Zhou, & Zhou,2006; Linthicum, Reitenga, & Sanchez, 2010).Additionally, research on product recalls hasshown that a lower-approval organization imple-menting an aggressive defensive responserisks being perceived as overly self-interestedand lacking credibility, which can increaseevaluators’ negative perceptions and threatenthe organization’s survival (Siomkos & Shri-vastava, 1993). This suggests a fine line for higher-and lower-approval organizations, which may beafforded the opportunity to accept less crisis re-sponsibility, but only to a certain point, beyondwhich an overly defensive strategy could backfire.Our interest in the sociocognitive mechanisms

of crisis management made focusing on the onsetof a crisis, when uncertainty and sensemakingare high and, thus, when a response strategy canhave its greatest effect on evaluators’ percep-tions, a natural boundary for our theorizing. Yet,as mentioned above, uncertainty may decreaseand sensemaking may be reduced as a crisisevolves. An organization therefore may switch itsresponse strategy based on new information andfeedback from evaluators. How an organization’sendowment of social approval would influencethe likelihood and effectiveness of switching isan interesting question. For example, switchingis often triggered by the release of authoritativeinformation that conflicts with an organization’sinitial response (e.g., the U.S. government’schallenges to BP’s initial estimates of oil leakingfrom the Deepwater Horizon rig). Because suchinformation is difficult to contest, it is likely thatany organization would alter its message to beconsistent with the message evaluators willperceive as more credible.Nevertheless, a higher- or lower-approval or-

ganization may face more or less extreme reac-tions to its switching response strategies than theaverage organization. For example, the bufferingeffect of endowed approval may continue to in-fluence evaluators’ reactions through the latter


stages of a crisis, giving a higher- or lower-approval organization some leeway in eithermaintaining its initial response or switchingto a new one. In contrast, the burdening effectof social approval may trigger enhanced nega-tive reactions. Evaluators’ heightened expec-tations for a higher-approval organization mayinclude that it provide a consistent responsethroughout the crisis. Alternatively, evaluatorsmay view a lower-approval organization’s lackof persistence as further evidence that it shouldbe disapproved. Ultimately, the role of socialapproval on an organization’s switching be-havior may be another theoretical conundrum,or it may be an empirical question worthy offuture study.

We also extended organizational research byspecifying the sociocognitive foundations of so-cial approval and its role as an intuitive per-ception ideal for a crisis context. We treatedsocial approval as a distinct construct, but wealso acknowledged that a crisis could havenegative effects on legitimacy and reputation.While social evaluations scholars continue todebate definitions, dimensionality, and properlevels of analysis (e.g., Deephouse & Suchman,2008; Lange et al., 2011; Rindova & Martins, 2012),future work should consider how more deliberateand more intuitive social evaluations may in-teract to influence the crisis management pro-cess. Incorporating social approval and crisisresponses into the processes related to reputa-tion formation (cf. Rindova et al., 2007) and legit-imation (cf. Haack et al., 2014) may also providefruitful future research opportunities. Addition-ally, beyond the onset of a crisis, different eval-uators may become more or less salient, and anorganization’s response may become more nu-anced and targeted (Mitchell, Agle, &Wood, 1997;Pfarrer, DeCelles, Smith, & Taylor, 2008). Wetherefore encourage future research to explorethe multiple interactions among the nature of thecrisis, the response strategy, and the range ofstakeholders and social evaluations.

Finally, we recognize that our theory could beenhanced by a number of additional moder-ators. For example, we suggested that diagnosticinformation contrary to an organization’s under-conforming or overconforming mismatched re-sponse strategy could lead to more severe socialapproval loss. Thus, the presence of a “smokinggun” could be an important moderator to en-hance our understanding of the trade-offs

among response strategies. Additionally, wesuggested above that Arthur Andersen’s overlydefensive response likely contributed to itsdownfall. However, the Enron crisis was excep-tional, being one of the largest corporate fraudsin U.S. history. Thus, the magnitude or salience ofthe crisis could represent an important modera-tor altering the effect of different response strat-egies on social approval. Similarly, the conceptof moral intensity—the perceived moral impera-tive in a crisis situation—could influence howevaluators respond (Jones, 1991).In conclusion, organizational scholars are

continuing to add to their understanding of the“microfoundations of strategic managementconcepts” (Mishina et al., 2012: 460). Extendingthis line of inquiry, we theorized that social ap-proval can act as both a buffer and a burden ina crisis context, and we identified the socio-cognitive mechanisms behind this seeming con-tradiction. We believe that future research cancontinue to untangle the mechanisms and pro-cesses that generate social approval and crisisperceptions, as well as other factors that maycontribute to the double edge of social approvalin crisis and noncrisis situations.

REFERENCES

Abbink, K., Irlenbusch, B., & Renner, E. 2000. The moonlightinggame: An experimental study on reciprocity and retribution.Journal of Economic Behavior & Organization, 42: 265–277.

Ahmadjian, C. L., & Robinson, P. 2001. Safety in numbers:Downsizing and the deinstitutionalization of permanentemployment in Japan. Administrative Science Quarterly,46: 622–654.

Argote, L., & Greve, H. R. 2007. A behavioral theory of thefirm—40 years and counting: Introduction and impact.Organization Science, 18: 337–349.

Ashforth, B., & Gibbs, B. 1990. The double-edge of organiza-tional legitimation. Organization Science, 1: 177–194.

Ashforth, B. E., Harrison, S. H., & Corley, K. G. 2008. Identifi-cation in organizations: An examination of four funda-mental questions. Journal of Management, 34: 325–374.

Ballinger, G. A., & Rockmann, K. W. 2010. Chutes versus lad-ders: Anchoring events and a punctuated-equilibriumperspective on social exchange relationships. Academyof Management Review, 35: 373–391.

Barton, L. 2001. Crisis in organizations (2nd ed.). Cincinnati:South-Western College Publishing.

Bazerman, M. H. 2006. Judgment in managerial decisionmaking (6th ed.). New York: Wiley.

Benoit, W. 1995. Accounts, excuses, and apologies: A theory ofimage restoration strategies. Albany: State University ofNew York Press.


Benoit,W. L. 2005. Image restoration theory. In R. L. Heath (Ed.),Encyclopedia of public relations, vol. 2: 407–410. ThousandOaks, CA: Sage.

Billings, R. S., Milburn, T. W., & Schaalman, M. L. 1980. Amodel of crisis perception: A theoretical and empiricalanalysis. Administrative Science Quarterly, 25: 300–316.

Bitektine, A. 2011. Toward a theory of social judgments oforganizations: The case of legitimacy, reputation, andstatus. Academy of Management Review, 36: 151–179.

Bottom,W., Gibson, K., Daniels, S., & Murnighan, J. 2002.Whentalk is not cheap: Substantive penance and expressionsof intent in rebuilding cooperation.Organization Science,13: 497–513.

Braithwaite, J. 1989. Crime, shame and reintegration. Cam-bridge: Cambridge University Press.

Brehm, S. S., & Brehm, J. W. 1981. Psychological reactance: Atheory of freedom and control.New York: Academic Press.

Brooks, M., Highhouse, S., Russell, S., & Mohr, D. 2003. Famil-iarity, ambivalence, and firm reputation: Is corporatefame a double-edged sword? Journal of Applied Psy-chology, 88: 904–913.

Burgoon, J. K. 1993. Interpersonal expectations, expectancyviolations, and emotional communication. Journal ofLanguage and Social Psychology, 12: 30–48.

Chaiken, S., & Trope, Y. 1999. Dual-process theories in socialpsychology. New York: Guilford Press.

Chaney, P. K., & Philipich, K. L. 2002. Shredded reputation:The cost of audit failure. Journal of Accounting Research,40: 1221–1245.

Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. 2011.Signaling theory: A review and assessment. Journal ofManagement, 37: 39–67.

Coombs, W. T. 1995. Choosing the right words: The developmentof guidelines for the selection of the “appropriate” crisisresponse strategies. Management Communication Quar-terly, 8: 447–476.

Coombs, W. T. 2007a. Attribution theory as a guide for post-crisis communication research. Public Relations Review,33: 135–139.

Coombs, W. T. 2007b. Protecting organization reputationsduring a crisis: The development and application of situ-ational crisis communication theory. Corporate Repu-tation Review, 10: 163–176.

Coombs, W. T. 2011.Ongoing crisis communication: Planning,managing, and responding. Thousand Oaks, CA: Sage.

Coombs, W. T., & Holladay, S. J. 2001. An extended examina-tion of the crisis situations: A fusion of the relationalmanagement and symbolic approaches. Journal of PublicRelations Research, 13: 321–340.

Coombs, W. T., & Holladay, S. J. 2002. Helping crisis managersprotect reputational assets. Management Communica-tion Quarterly, 16: 165–186.

Coombs,W. T., & Holladay, S. J. 2004. Reasoned action in crisiscommunication: An attribution theory based approach tocrisis management. In D. P. Millar & R. L. Heath (Eds.),Responding to crises: A rhetorical approach to crisis

communication: 95–116. Mahwah, NJ: Lawrence ErlbaumAssociates.

Coombs, W. T., & Holladay, S. J. 2006. Unpacking the haloeffect: Reputation and crisis management. Journal ofCommunication Management, 10: 123–137.

Coombs, W. T., & Holladay, S. J. 2008. Comparing apology toequivalent crisis response strategies: Clarifying apology’srole and value in crisis communication. Public RelationsReview, 34: 252–257.

Cyert, R. M., & March, J. G. 1963. A behavioral theory of thefirm. Englewood Cliffs, NJ: Prentice-Hall.

Davidson, W. N., Worrell, D. L., & Lee, C. I. 1994. Stock marketreactions to announced corporate illegalities. Journal ofBusiness Ethics, 13: 979–987.

Dean, D. H. 2004. Consumer reaction to negative publicity:Effects of corporate reputation, response, and re-sponsibility for a crisis event. Journal of Business Com-munication, 41: 192–211.

Decker, W. H. 2012. A firm’s image following allegedwrongdoing: Effects of the firm’s prior reputation andresponse to the allegation. Corporate Reputation Re-view, 15: 20–34.

Deephouse, D. 2000. Media reputation as a strategic re-source: An integration of mass communication andresource-based theories. Journal of Management, 26:1091–1112.

Deephouse, D. L., & Suchman, M. C. 2008. Legitimacy in or-ganizational institutionalism. In R. Greenwood, C. Oliver,R. Suddaby, & K. Sahlin-Andersson (Eds.), The Sage hand-book of organizational institutionalism: 49–77. ThousandOaks, CA: Sage.

Devers, C. E., Dewett, T., Mishina, Y., & Belsito, C. A. 2009. Ageneral theory of organizational stigma. OrganizationScience, 20: 154–171.

Elliot, A. J. 2006. The hierarchical model of approach-avoidancemotivation. Motivation and Emotion, 30: 111–116.

Elliot, A. J., & Devine, P. G. 1994. On the motivational nature ofcognitive dissonance: Dissonance as psychological dis-comfort. Journal of Personality and Social Psychology, 67:382–394.

Elsbach, K. D. 2003. Organizational perception management.Research in Organizational Behavior, 25: 297–332.

Fediuk, T. A., Coombs, W. T., & Botero, I. C. 2012. Exploringcrisis from a receiver perspective: Understanding stake-holder reactions during crisis events. In W. T. Coombs &S. J. Holladay (Eds.), The handbook of crisis communica-tion: 635–656. Malden, MA: Wiley.

Fehr, E., & Gachter, S. 2000. Cooperation and punishment inpublic goods experiments. American Economic Review,90: 980–994.

Ferrin, D., Kim, P., Cooper, C., & Dirks, K. 2007. Silence speaksvolumes: The effectiveness of reticence in comparison toapology and denial for responding to integrity- andcompetence-based trust violations. Journal of AppliedPsychology, 92: 893–908.

Festinger, L. 1954. A theory of social comparison processes.Human Relations, 7: 117–140.


Festinger, L. 1957. A theory of cognitive dissonance. Stanford,CA: Stanford University Press.

Fink, S. 1986. Crisis management: Planning for the inevitable.New York: AMACOM.

Fiske, S., & Taylor, S. 1991. Social cognition. New York:McGraw-Hill.

Fombrun, C. J. 2012. The building blocks of corporate reputation:Definitions, antecedents, consequences In M. L. Barnett &T. G. Pollock (Eds.), The Oxford handbook of corporate rep-utation: 94–113. Oxford: Oxford University Press.

Gephart, R. P. 2007. Crisis sensemaking and the public in-quiry. In C. M. Pearson, C. Roux-Dufort, & J. A. Clair (Eds.),International handbook of organizational crisis manage-ment: 123–160. Thousand Oaks, CA: Sage.

Gephart, R. P., Van Maanen, J., & Oberlechner, T. 2009.Organizations and risk in late modernity. OrganizationStudies, 30: 141–155.

Graffin, S. D., Bundy, J., Porac, J. F., Wade, J. B., & Quinn, D. P.2013. Falls from grace and the hazards of high status: The2009 British MP expense scandal and its impact on par-liamentary elites. Administrative Science Quarterly, 58:313–345.

Haack, P., Pfarrer, M. D., & Scherer, A. G. 2014. Legitimacy-as-feeling: How affect leads to vertical legitimacy spilloversin transnational governance. Journal of ManagementStudies, 51: 634–666.

Haidt, J. 2001. The emotional dog and its rational tail: A socialintuitionist approach to moral judgment. PsychologicalReview, 108: 814–834.

Haidt, J., & Bjorklund, F. 2008. Social intuitionists answersix questions about moral psychology. In W. Sinnott-Armstrong (Ed.), Moral psychology: The cognitive scienceof morality: Intuition and diversity, vol. 2: 181–218. Cam-bridge, MA: MIT Press.

Heckert, A., & Heckert, D. M. 2002. A new typology of deviance:Integrating normative and reactivist definitions of de-viance. Deviant Behavior, 23: 449–479.

Heider, F. 1958. The psychology of interpersonal relations.NewYork: Wiley.

Highhouse, S., Broadfoot, A., Yugo, J. E., & Devendorf, S. A.2009. Examining corporate reputation judgments withgeneralizability theory. Journal of Applied Psychology, 94:782–789.

Hollander, E. P. 1958. Conformity, status, and idiosyncrasycredit. Psychological Review, 65: 117–127.

Jensen, M., Kim, H., & Kim, B. K. 2012. Meeting expectations:A role-theoretic perspective on reputation. In M. L. Barnett& T. G. Pollock (Eds.), The Oxford handbook of corporatereputation: 140–159. Oxford: Oxford University Press.

Jones, T. 1991. Ethical decision making by individuals inorganizations: An issue-contingent model. Academy ofManagement Review, 16: 366–395.

Kadlec, D. 2002. Enron: Who’s accountable? Time Magazine,January 21: 28–35.

Kahneman, D. 2011. Thinking, fast and slow. New York: Farrar,Straus and Giroux.

Kahneman, D., & Frederick, S. 2002. Representativenessrevisited: Attribute substitution in intuitive judgment. InT. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristicsand biases: The psychology of intuitive judgment: 49–81.Cambridge: Cambridge University Press.

Kim, H. K., & Yang, S. U. 2009. Cognitive processing of crisiscommunication: Effects of CSR and crisis response strat-egies on stakeholder perceptions of a racial crisis dy-namics. Public Relations Journal, 3: 1–39.

Kim, P., Ferrin, D., Cooper, C., & Dirks, K. 2004. Removing theshadow of suspicion: The effects of apology versus denialfor repairing competence- versus integrity-based trustviolations. Journal of Applied Psychology, 89: 104–118.

Koehler, J. J., & Gershoff, A. D. 2003. Betrayal aversion: Whenagents of protection become agents of harm. Organiza-tional Behavior and Human Decision Processes, 90:244–261.

Koehn, D. 2013. Why saying “I’m sorry” isn’t good enough: Theethics of corporate apologies. Business Ethics Quarterly,23: 239–268.

Krishnamurthy, S., Zhou, J., & Zhou, N. 2006. Auditor reputa-tion, auditor independence, and the stock‐market impactof Andersen’s indictment on its client firms. Contempo-rary Accounting Research, 23: 465–490.

Lamin, A., & Zaheer, S. 2012. Wall Street vs. Main Street: Firmstrategies for defending legitimacy and their impact ondifferent stakeholders. Organization Science, 23: 47–66.

Lange, D., Lee, P. M., & Dai, Y. 2011. Organizational reputation:A review. Journal of Management, 37: 153–184.

Lange, D., &Washburn, N. T. 2012. Understanding attributionsof corporate social irresponsibility. Academy of Man-agement Review, 37: 300–326.

Linthicum, C., Reitenga, A. L., & Sanchez, J. M. 2010. Socialresponsibility and corporate reputation: The case of theArthur Andersen Enron audit failure. Journal of Account-ing and Public Policy, 29: 160–176.

Love, E., & Kraatz, M. 2009. Character, conformity, or thebottom line? How and why downsizing affected corpo-rate reputation. Academy of Management Journal, 52:314–335.

Lyons, D. 2010. Apple’s rotten response. Newsweek, July 16:http://www.thedailybeast.com/newsweek/2010/07/16/apple-s-rotten-reponse.html.

March, J., & Shapira, Z. 1987. Managerial perspectives on riskand risk taking. Management Science, 33: 1404–1418.

March, J., & Shapira, Z. 1992. Variable risk preferences and thefocus of attention. Psychological Review, 99: 172–183.

Marcus, A., & Goodman, R. 1991. Victims and shareholders:The dilemmas of presenting corporate policy duringa crisis. Academy of Management Journal, 34: 281–305.

Martinko, M. J. 1995. The nature and function of attributiontheory within the organizational sciences. In M. J.Martinko (Ed.), Attribution theory: An organizational per-spective: 7–16. Delray Beach, FL: St. Lucie Press.

Massey, J. E. 2001. Managing organizational legitimacy:Communication strategies for organizations in crisis.Journal of Business Communication, 38: 153–183.


http://www.thedailybeast.com/newsweek/2010/07/16/apple-s-rotten-reponse.html

http://www.thedailybeast.com/newsweek/2010/07/16/apple-s-rotten-reponse.html

McLean, B., & Elkind, P. 2003. The smartest guys in the room:The amazing rise and scandalous fall of Enron. New York:Penguin.

Merton, R. K. 1968. The Matthew effect in science. Science, 159:56–63.

Mishina, Y., Block, E. S., & Mannor, M. J. 2012. The path de-pendence of organizational reputation: How social judg-ment influences assessments of capability and character.Strategic Management Journal, 33: 459–477.

Mitchell, R. K., Agle, B. R., & Wood, D. J. 1997. Toward a theoryof stakeholder identification and salience: Defining theprinciple of who and what really counts. Academy ofManagement Review, 22: 853–886.

Mitroff, I. I. 1988. Crisis management: Cutting through theconfusion. Sloan Management Review, 29(2): 15–20.

Mitroff, I. I. 1994. Crisis management and environmentalism: Anatural fit. California Management Review, 36(2): 101–113.

Nickerson, R. S. 1998. Confirmation bias: A ubiquitous phe-nomenon in many guises. Review of General Psychology,2: 175–220.

Pearson, C. M., & Clair, J. A. 1998. Reframing crisis manage-ment. Academy of Management Review, 23: 59–76.

Pfarrer, M. D., DeCelles, K. A., Smith, K. G., & Taylor, M. S. 2008.After the fall: Reintegrating the corrupt organization.Academy of Management Review, 33: 730–749.

Pfarrer, M. D., Pollock, T., & Rindova, V. 2010. A tale of twoassets: The effects of firm reputation and celebrity onearnings surprises and investors’ reactions. Academy ofManagement Journal, 53: 1131–1152.

Pfarrer, M. D., Smith, K., Bartol, K., Khanin, D., & Zhang, X. 2008.Coming forward: The effects of social and regulatoryforces on the voluntary restatement of earnings sub-sequent to wrongdoing.Organization Science, 19: 386–403.

Phillips, D. J., & Zuckerman, E. W. 2001. Middle-status confor-mity: Theoretical restatement and empirical demonstra-tion in two markets. American Journal of Sociology, 107:379–429.

Pratt, M. G. 2000. The good, the bad, and the ambivalent:Managing identification among Amway distributors.Administrative Science Quarterly, 45: 456–493.

Raithel, S., & Schwaiger, M. In press. The effects of corporatereputation perceptions of the general public on share-holder value. Strategic Management Journal.

RepTrak. 2013. 2013 most reputable U.S. companies: Part of theworld’s largest study on corporate reputation. New York:Reputation Institute.

Rhee, M., & Haunschild, P. 2006. The liability of good reputa-tion: A study of product recalls in the U.S. automobileindustry. Organization Science, 17: 101–117.

Rindova, V. P., & Martins, L. L. 2012. A multidimensional per-spective on reputation as an intangible asset. In M. L.Barnett & T. G. Pollock (Eds.), The Oxford handbook of cor-porate reputation: 17–33. Oxford: Oxford University Press.

Rindova, V. P., Petkova, A. P., & Kotha, S. 2007. Standing out:How new firms in emerging markets build reputation.Strategic Organization, 5: 31–70.

Roberts, K. H. 1990. Some characteristics of one type of highreliability organization. Organization Science, 1: 160–176.

Roberts, K. H., Madsen, P., & Desai, V. 2007. Organizationalsensemaking during crisis. In C. M. Pearson, C. Roux-Dufort, & J. A. Clair (Eds.), International handbook oforganizational crisis management: 107–122. ThousandOaks, CA: Sage.

Sanders, W. G., & Carpenter, M. A. 2003. Strategic satisficing?A behavioral-agency theory perspective on stockrepurchase program announcements. Academy of Man-agement Journal, 46: 160–178.

Sayegh, L., Anthony, W. P., & Perrewe, P. L. 2004. Managerialdecision-making under crisis: The role of emotion in anintuitive decision process.Human Resource ManagementReview, 14: 179–199.

Scherer, A. G., Palazzo, G., & Seidl, D. 2013. Managing legiti-macy in complex and heterogeneous environments: Sus-tainable development in a globalized world. Journal ofManagement Studies, 50: 259–284.

Schnietz, K., & Epstein, M. 2005. Exploring the financial valueof a reputation for corporate social responsibility duringa crisis. Corporate Reputation Review, 7: 327–345.

Scott, W. R. 1995. Institutions and organizations. ThousandOaks, CA: Sage.

Sellnow, T. L., & Seeger, M. W. 2013. Theorizing crisis com-munication. Malden, MA: Wiley.

Sinaceur, M., Heath, C., & Cole, S. 2005. Emotional and de-liberative reactions to a public crisis: Mad cow disease inFrance. Psychological Science, 16: 247–254.

Siomkos, G., & Shrivastava, P. 1993. Responding to productliability crises. Long Range Planning, 26: 72–99.

Slovic, P., Finucane, M., Peters, E., & MacGregor, D. G. 2002.Rational actors or rational fools: Implications of the af-fect heuristic for behavioral economics. Journal of Socio-Economics, 31: 329–342.

Slovic, P., Finucane, M., Peters, E., & MacGregor, D. G. 2004.Risk as analysis and risk as feelings. Risk Analysis, 24:311–322.

Stanovich, K. E., & West, R. F. 2000. Advancing the rationalitydebate. Behavioral and Brain Sciences, 23: 701–717.

Suchman, M. C. 1995. Managing legitimacy: Strategic and in-stitutional approaches. Academy of Management Review,20: 571–610.

Sutton, R., & Callahan, A. 1987. The stigma of bankruptcy:Spoiled organizational image and its management.Academy of Management Journal, 30: 405–436.

Sutton, R. I., & Galunic, D. C. 1996. Consequences of publicscrutiny for leaders and their organizations. Research inOrganizational Behavior, 18: 201–259.

Tost, L. P. 2011. An integrative model of legitimacy judgments.Academy of Management Review, 36: 686–710.

Traut-Mattausch, E., Schulz-Hardt, S., Greitemeyer, T., & Frey,D. 2004. Expectancy confirmation in spite of disconfirmingevidence: The case of price increases due to the in-troduction of the euro. European Journal of Social Psy-chology, 34: 739–760.


Turner, B. 1976. The organizational and interorganizationaldevelopment of disasters. Administrative Science Quar-terly, 21: 378–397.

Tversky, A., & Kahneman, D. 1974. Judgment under uncertainty:Heuristics and biases. Science, 185: 1124–1131.

Tyler, L. 1997. Liability means never being able to say you’resorry: Corporate guilt, legal constraints, and de-fensiveness in corporate communication. ManagementCommunication Quarterly, 11: 51–73.

Ulmer, R. R., & Sellnow, T. L. 1997. Strategic ambiguity and theethic of significant choice in the tobacco industry’s crisiscommunication. Communication Studies, 48: 215–233.

van der Meer, T. G. L. A., & Verhoeven, J. W. M. 2014. Emotionalcrisis communication. Public Relations Review, 40: 526–536.

Veil, S. V. 2011. Mindful learning in crisis management. Jour-nal of Business Communication, 48: 116–147.

Vergne, J. P. 2012. Stigmatized categories and public disap-proval of organizations: A mixed-methods study of the

global arms industry, 1996–2007. Academy of Manage-ment Journal, 55: 1027–1052.

Weick, K. E. 1988. Enacted sensemaking in crisis situations.Journal of Management Studies, 25: 305–317.

Weick, K., & Sutcliffe, K. 2001. Managing the unexpected: As-suring high performance in an age of uncertainty. SanFrancisco: Wiley.

Weiner, B. 1986. An attributional theory of motivation andemotion. New York: Springer-Verlag.

Yu, T., Sengul, M., & Lester, R. H. 2008. Misery loves company:The spread of negative impacts resulting from an orga-nizational crisis. Academy of Management Review, 33:452–472.

Zavyalova, A. A., Pfarrer, M. D., Reger, R. K., & Shapiro, D. L.2012. Managing the message: The effects of firm actionsand industry spillovers on media coverage followingwrongdoing. Academy of Management Journal, 55:1079–1101.

Jonathan Bundy ([email protected]) is an assistant professor of management and orga-nization in the Smeal College of Business at The Pennsylvania State University. Hereceived his Ph.D. from the University of Georgia. His research investigates the socialand cognitive forces that shape organizational outcomes, with a focus on social eval-uations, crisis management, stakeholder management, and corporate governance.

Michael D. Pfarrer ([email protected]) is an associate professor in the Terry College ofBusiness at the University of Georgia. He received his Ph.D. from the University ofMaryland. His research focuses on social evaluations of the firm, including reputationand celebrity; impression and crisis management; media accounts; and the role ofbusiness in society.


mailto:[email protected]

mailto:[email protected]

a burden of responsibility: the role of...

Documents