a micro-institutional inquiry into resistance to ... · institutional theorizing as well as the...

r Academy of Management Journal2018, Vol. 61, No. 4, 1431–1466.https://doi.org/10.5465/amj.2016.0762

A MICRO-INSTITUTIONAL INQUIRY INTO RESISTANCE TOENVIRONMENTAL PRESSURES

OLIVER SCHILKEThe University of Arizona

This article contributes to the emerging stream of micro-institutional research, whichzooms in on the internal organizational processes that are responsible for organizations’differential responses to the external environment. Specifically, the investigation offersnew knowledge of how organizational identity processes can shape whether decision-makers will resist versus give in to environmental pressures. Building on the notion thatorganizational identity acts as a filter through which decision-makers relate to the ex-ternal environment, I develop the theoretical argument that strong organizationalidentification increases resistance to environmental pressures due to two mechanisms:(1) it bolsters the decision-maker’s certainty and (2) it deflects the decision-maker’sattention from the environment. A series of laboratory experiments not only test themediated relationship between organizational identification and resistance to envi-ronmental pressures but also contrast different types of organizational identity. Theempirical results support the hypothesized positive link between organizational iden-tification and resistance, which becomes particularly strong when the organizationalidentity is normative (vs. utilitarian). The findings reported here enrich institutionaltheory by adding microfoundations to organizational practice adoption decisionsand shedding new light on relevant enabling conditions for agency and within-fieldheterogeneity.

Institutional theory is currently undergoing a fun-damental shift from a formerly purely macro-levelapproach toward a comprehensive multilevel par-adigm that explicitly incorporates individuals(Bechky, 2011; Fine & Hallett, 2014; Thornton,Ocasio, & Lounsbury, 2012). This transformation isdriven by the recognition that “much analyticalpurchase can be gained by developing a micro-levelcomponent of institutional analysis” (Powell &

Colyvas, 2008: 276) to complement earliermacro-level accounts, with the goal of making thetheory both more precise and more general asa result (Thornton & Ocasio, 2008). After decadesof neglect, the motivations, cognitions, back-grounds, and behaviors of individual actors havethus moved to the foreground of institutionalinquiry (Boxenbaum, 2014; Lawrence, Suddaby,& Leca, 2011; Zilber, 2016).

The author is thankful for the insightful commentsprovided by associate editor Pursey Heugens and by threeanonymous reviewers. The author also gratefully ac-knowledges the helpful comments on earlier versions ofthis paper provided by Jeannette Colyvas, Karen Cook,Joseph Galaskiewicz, Jeffrey Lucas, Carrie Oelberger,Nathan Podsakoff, Davide Ravasi, Pat Reilly, MartinReimann, Gabriel Rossman, Leigh Plunkett Tost, EdwardWalker, and, last but most certainly not least, LynneZucker. The paper also benefitted from discussions withparticipants of the University of Arizona SociologyBrownbag, the University of Arizona Macro ReadingGroup, the Stanford University OB Macro Lunch, theUCLA Knowledge and Cognitive Systems WorkingGroup, the Harvard Business School Speaker Series inOrganizational Behavior, the Alberta School of Busi-ness McCalla Seminar Series, the Tuck School of Busi-ness Strategy and Management Seminar Series, the

American Sociological Association 2015 Annual Meet-ing, the 2015 Academy ofManagementMeeting, and the2016 Group Processes Conference. An earlier version ofthis paper was a finalist for the 2015 Pondy Best Dis-sertation Paper Award, and an extended abstract wasincluded in the 2015 Academy of Management AnnualMeeting Proceedings (OMT Division). The author isthankful for able research assistance by Jonathan BlairEvans, Mariam Hambarchyan, Jessi Mariah Rivin, andKatelyn Wirtz.

Funding: The author conducted this research in partwhile supported by a UCLA Dissertation Year Fellowship.This material is based upon work supported by the Na-tional Science Foundation under Grant Number 1302306.Any opinions, findings, and conclusions or recommen-dations expressed in this material are those of the authorand do not necessarily reflect the views of the NationalScience Foundation.

1431

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

https://doi.org/10.5465/amj.2016.0762

However, several important questions related tothe role of individuals in institutional theory haveremained largely unaddressed so far. The currentinvestigation tackles one such theoretical puzzle thatis particularly pressing: Facing the same envi-ronmental pressures to adopt certain organiza-tional practices, why is it that some organizationaldecision-makers resist those pressures while othersgive in to themandconform to isomorphic templates?Answering this question is of critical importance forat least two interrelated reasons, the first pertainingto its distinctive focus on individual decision-makers as the unit of analysis and the second to itsexplanandum of resistance to adopting organiza-tional practices.

First, in contrast to the conventional macro-institutional emphasis on field-level homogeneity,an analytical focus on decision-makers has sig-nificant potential for explaining within-fieldheterogeneity—that is, why not all organizations ina field are the same (Tolbert, David, & Sine, 2011;Zucker, 1991). While there is no doubt that envi-ronmental pressures can be very powerful, they arerarely so strong that all organizational decision-makers have no choice but to succumb to themunreflectedly (Binder, 2007; Fiol & O’Connor, 2003).Indeed, an increasing consensus is emerging thatdecision-makers can exercise considerable discre-tion in deciding to what extent their organizationbecomes isomorphic with the environment. There-fore, a revised account that does away with the long-standing “institutional dope” assumption (DiMaggio&Powell, 1991), and that instead endows decision-makers with the ability to subjectively make senseof the environment, has great potential to signifi-cantly broadenourunderstanding of the adoption ofinstitutionalized prescriptions. Unfortunately, rel-evant micro-sociological processes, which “forma necessary link in the causal chain connecting ab-stract isomorphic pressures to concrete organiza-tional actions, are all too often merely assumed oreven ignored by institutional theorists” (Heugens &Lander, 2009: 76). Consequently, greater attentionto the local conditions and how they shape the waydecision-makers perceive, interpret, and act on theenvironment are indispensable for institutionalinquiry to move forward (Bechky, 2011; Creed,Hudson, Okhuysen, & Smith-Crowe, 2014; Hallett,2010b; Powell & Bromley, 2015). It is this call todevelop themicrofoundations of institutions and tobring “individuals back into institutional theory”(Lawrence et al., 2011: 53) that provides a key mo-tivation for the current investigation.

Second, a reinvigorated commitment to explain-ing variations in resistance to adopting organiza-tional practices (e.g., Dhalla &Oliver, 2013; Pache &Santos, 2010) is warranted for both practical andtheoretical reasons. From a practical perspective,whether isomorphic templates for organizing areadopted by an organization can have major impli-cations not only for that organization’s social eval-uation but also for its technical efficacy anddifferentiation from competition, with deviatingorganizations often outperforming conformers interms of profitability (Barreto &Baden-Fuller, 2006;Meyer & Rowan, 1977; Zucker, 1987). This in-vestigation uncovers an important micro-levelpredictor of such deviance and thus perhaps of anorganization’s competitive advantage. From a the-oretical perspective, knowledge about the condi-tions under which actors resist environmentalpractices directly speaks to the enduring agencydebate (e.g., Heugens & Lander, 2009; Zilber,2002)—that is, the question of whether people aremere institutional carriers who passively re-produce their external world versus active humanagents who purposefully interpret and resist theirenvironment. Developing a middle-ground per-spective that transcends the conventional di-chotomy of structure versus voluntarism andunpacks the conditions under which people willfall closer to one or other end of this continuum is animportant element in building a richer notion ofagency and contingent actorhood in institutionaltheorizing (Battilana, 2006; Battilana & D’Aunno,2009), which has the potential to take institution-alism beyond a restricted model of constraint.

Despite the great relevance of understanding howdecision-makers experience and react to environ-mental pressures to adopt institutionalized pre-scriptions, prior investigations into this issue areseverely limited (cf. Suddaby, 2010; Zilber, 2008).We know very little about whether and how thecharacteristics of decision-makers and their imme-diate context shape organizational resistance. Tra-ditionally, scholars studying responses to adoptionpressures have used a macro lens (see Scott, 2014,for an overview), whereas recent micro-level in-vestigations evoking the notions of institutionalentrepreneurship (Battilana, Leca, & Boxenbaum,2009), institutional work (Lawrence et al., 2011), orinhabited institutions (Hallett & Ventresca, 2006)have so far largely bracketed organizational template-adoption decisions. As Chandler and Hwang (2015)emphasized, the institutional practice adoption andmicrofoundations literatures have yet to be integrated

1432 AugustAcademy of Management Journal

systematically in order to fully grasp the reasons fororganizational heterogeneity.

My particular approach to redressing this over-sight and identifying an important source ofdecision-makers’ heterogeneous responses to in-stitutional pressures is motivated by the need toavoid both the over-socialized view of earlier neo-institutional theorizing as well as the reductionisttrap of overly utilitarian, atomistic accounts (Powell& Colyvas, 2008). In most cases, organizationaldecision-makers do not respond to institutionalpressures in isolation but in the context of theirparticular social setting (Hallett, 2010a). Thus, inorder to capture human beings in their collectivecharacter, it is important to explicitly account for theeffect of their immediate social context on cognitionand behavior (Battilana & D’Aunno, 2009; Fine &Hallett, 2014).

In terms of decision-making related to practiceadoption, an especially salient immediate socialcontext is undoubtedly that of the organization(Suddaby, Elsbach, Greenwood, Meyer, & Zilber,2010; Zilber, 2002). Therefore, I build on recent ef-forts to infuse institutional theory with organiza-tional identity processes (Dejordy & Creed, 2016;Greenwood, Raynard, Kodeih,Micelotta, & Lounsbury,2011; Hatch & Zilber, 2012; Kraatz & Block, 2008;Thornton et al., 2012) to argue that actors’ sensemakingand adoption decisions are substantially influenced bytheir identification with and identity of the organi-zation. The concept of organizational identity rootsthe individual in the organizational context(Ashforth, Rogers, & Corley, 2011), structures on-going organizational sensemaking activity (Fine &Hallett, 2014), and provides an ideal window forinstitutionalists to examine micro-level dynamicsin organizational decision-making because it re-flects the taken-for-granted collective character oforganizational members and avoids the imagery ofmuscular, heroic individuals (Scott, 2014). Orga-nizational identity forms a critical link between theinstitutional environment and the cognitionand behavior of organizational decision-makers(Glynn, 2008; Thornton & Ocasio, 2008), but it hasyet to be systematically examined in conjunctionwith organizational responses to environmentalpressures.

Building on the notion that organizational identityrepresents an important filter through which theenvironment is perceived, interpreted, and actedupon (e.g., Greenwood et al., 2011; Raffaelli &Glynn,2014), this article develops a novel theoretical argu-ment regarding the role of organizational identity

processes in decision-makers’ responses to envi-ronmental pressures. I argue, and empirically dem-onstrate, that strong organizational identificationincreases organizational decision-makers’ resistanceto environmental pressures, and that this effect isdue to twocognitivemechanisms related tohow theyperceive themselves and the environment. Specifi-cally, strong organizational identification is shownto (a) bolster the decision-maker’s certainty and (b)deflect the decision-maker’s attention from the en-vironment. Moreover, this research moves beyondstudying identification to compare the consequencesof different types of organizational identity, specifi-cally contrasting a normative and a utilitarian idealtype. Overall, the article addresses the followingthree research questions: (1) Does organizationalidentification affect decision-makers’ resistance toenvironmental pressures?; (2) If so, why?; and (3)What particular attributes of an organization’s iden-tity modulate this resistance?

This investigation makes several important con-tributions. First, it adds to recent micro-institutionalinquiry by shedding new light on the role ofdecision-makers’ cognition inorganizationalpracticeadoption. While in no way devaluing the impact ofthe external environment on organizations, micro-institutionalists have called for greater emphasis oninterpreting macro-level environments through thelens of micro-level perceptions (e.g., Bitektine &Haack, 2015; Greenwood et al., 2011; Powell &Colyvas, 2008; Suddaby, 2010; Thornton et al.,2012), and they have identified the need for moretheorizing dedicated to the individual-level pro-cesses through which environments affect organi-zations. Because organizational decision-makersmediate the interface between organizational envi-ronment and organizational action, we can fullygrasp the nature and effect of external institutionsonly by understanding the cognitive processes in-volved in dealing with those institutions (Barley,2008; George, Chattopadhyay, Sitkin, & Barden,2006; Glaser, Kroezen, & Thornton, 2015). In partic-ular, identifying the local contexts and cognitiveprocesses that shield decision-makers from externalinstitutions is critical for understanding relevantenabling conditions of actorhood, an issue that re-searchers have identified as particularly critical toadvancing the agency debate in institutional theory(Battilana, 2006; Powell & Bromley, 2015). Specify-ing such enabling conditions will “help ease theunfortunate dichotomy between heroic actors and‘cultural or institutional dopes’” (Hwang & Colyvas,2011: 63). Contributing to this line of inquiry, the

2018 1433Schilke

current investigation highlights organizationalidentity as a central local condition promotingdecision-makers’ resistance to environmental pres-sures. It provides support for the notion that an in-dividual’s agency is strongly guided by his or hersocial identification and identity. Further, it spec-ifies two concrete cognitive mechanisms—certaintyand attention—that explain actors’ variation in re-sistance, thus directly addressing repeated calls forgreater insight into the cognitive processes throughwhich people experience and make sense of in-stitutional pressures (Battilana & D’Aunno, 2009;Bechky, 2011;Thornton&Ocasio, 2008;Zilber, 2002).

In addition to these theoretical advancements, thisarticle makes a methodological contribution by de-veloping a novel experimental setup with manifoldpotential uses for future institutional inquiry. Fora long time, experiments have been virtually un-heard of in institutional research, but recentlyscholars have begun to follow up on Zucker’s (1977)seminal work and leverage the ability of experi-mental methods to isolate theoretical factors of in-terest and clarify the mechanisms underlying causaleffects. Bringing quantitative micro work to in-stitutional theory, experiments represent a neglectedbut much-needed tool for uncovering micro-levelprocesses and explaining how individuals experi-ence, make sense of, and react to institutions. Ex-periments thus allow for building new causal theoryby addressing key questions at the heart of in-stitutional theory that would be difficult to studywith other methods (Bitektine, Lucas, & Schilke,2018; Haack, McKinley, Schilke, & Zucker, 2016).The procedures and findings reported in the currentresearch should encourage future scholars to employexperimental methods when studying institutionalphenomena, thus helping to expand the “resourcespace” of potential contributions available to theinstitutional research community (Bitektine, 2009;Bitektine & Miller, 2015).

THEORETICAL BACKGROUND

Toward a Micro-level Component ofInstitutional Theory

Why do organizations adopt practices whose ma-terial benefits are difficult to assess even in retro-spect? This important question is at the core ofinstitutional theory. A long line of institutional re-search suggests that variousmimetic, normative, andcoercive pressures cause organizations to adopttemplates that make them isomorphic with their

environment (Scott, 2014). Following the founda-tional works of Meyer and Rowan (1977), Zucker(1977), and DiMaggio and Powell (1983), institu-tionalismhas developed into a leading—perhaps thedominant—perspective in organizational analysisthese days (Greenwood, Hinings, & Whetten, 2014;Powell & Bromley, 2015).

Despite its significant contributions, institutionaltheory has become the subject of increasing criticismfor its overly strong emphasis on macro-level issues(Bitektine, 2011; Hallett, 2010b; Powell & Colyvas,2008). Traditionally, the theory focused heavily onenvironmental drivers of organizational behaviorwhile neglecting the processes by which organiza-tional actors cope with the environment and whichmay thus explain why organizations differ in theiracceptance of versus resistance to external pressures(cf. Felin, Foss, & Ployhart, 2015; George et al., 2006).That is, by viewing the source of organizational be-havior as exogenous, institutionalists have largely“black boxed” the organization (Gavetti, Levinthal, &Ocasio, 2007; King, Felin, & Whetten, 2010; Zilber,2016; Zucker, 1983).

This line of criticism is fueled by the observationthat, even in the long run, many organizations willnot conform to institutional pressures (Kraatz &Zajac, 1996; Pedersen &Dobbin, 2006). Heugens andLander’s (2009) comprehensivemeta-analysis of 144institutional studies analyzing the effect of environ-mental pressures on isomorphism found that, al-though this effect is significant, reported effect sizesare relatively small, and using environmental in-stitutions to explain a few percentage points of thevariance does not an iron cage make.

In response to such criticism, institutionalistshave increasingly come to accept the notion that or-ganizations do not simply react to environmentaldemands, and this opens up the way for in-vestigations into factors that help to explain whyresistance rather than conformity may characterizeanorganization’s response (Oliver, 1991;Park, Sine,&Tolbert, 2011). In particular, institutionalists haverecently become interested in investigating factorsrelevant to the micro level, which can provide con-siderable additional leverage for understanding thereasons for variability in organizations’ responses tocomparable institutional environments (DiMaggio &Markus, 2010; Zucker, 1991). Although organizationswithin a given institutional field are typically ex-posed to similar environmental pressures, theseorganizations’ decision-makers may not experi-ence, and consequently may not respond to, suchpressures in the same way (Raffaelli & Glynn, 2014;


Suddaby, Viale, & Gendron, 2016; Tolbert et al.,2011). A sharper focus on cognitive processes, whichfor too long have been “obscured by the macro-gazecommon in contemporary neo-institutionalism”

(Hallett, 2010b: 53), thus seems indispensable toexplain the feasibility of varying organizationalresponses within the same macro-institutionalenvironment.

In particular, the research stream on inhabitedinstitutionalism (Hallett & Ventresca, 2006) has re-cently made significant progress in illuminating therole of cognitive processes in how people engagewith external institutions. Binder (2007) and Hallett(2010b), for example, demonstrated that institutionsare “inhabited” by individuals who actively makesense of institutional rules through the lens of theirunique experiences and local context. Similarly,Fine and Hallett (2014) argued that actors’ cognitiveprocesses play a key role in interpreting externalinstitutions and ultimately guiding organizationalaction. In short, a micro-level understanding of in-stitutional processes goes hand in hand with a cog-nitive lens.

Indeed, a cognitive micro-level approach to in-stitutional theory is highly consistent with some ofthe theory’s intellectual roots (see also Barley, 2008;DiMaggio & Powell, 1991; Hallett, 2010b; Powell &Colyvas, 2008; Zucker & Darby, 2005)—especiallySelznick (1957) and Berger and Luckmann (1966). Inhis seminal study of how institutionswork, Selznick(1957: 4) insisted that “no social process can be un-derstood save as it is located in the behavior of in-dividuals, and especially in their perceptions ofthemselves and each other.” Similarly, shared ob-jectified schemas of “how things are done” are key toBerger and Luckmann’s (1966) view of how in-dividuals make sense of everyday social reality,accentuating the subjective interpretations of in-dividuals that manifest in social groups. Notably,they highlighted that “identity is, of course, a keyelement of subjective reality” (Berger & Luckmann,1966: 194). Although social identities can stem froma variety of contexts, institutional scholars have de-voted special attention to organizational identity.

Organizational Identity and Institutional Theory

Organizational identity has a long history in in-stitutional inquiry.1Sixtyyearsago,Selznick (1957:40)

emphasized that an “infusion [of values] producesadistinct identity for theorganization.”Nonetheless,it has only been relatively recently that in-stitutionalists have rediscovered the importance oforganizational identity and have started to in-creasingly incorporate the concept into their theo-rizing (Glynn, 2008; Pedersen & Dobbin, 2006). Thisrenewed interest in organizational identity can be atleast partially explained by heightened efforts tobetter explain agency in the institutional theoryframework. To that end, scholars have started tostudy the conditions under which organizations aremore likely to possess agency (Battilana, 2006;Heugens & Lander, 2009) and, in doing so, havesingled out organizational identity as a potentiallycritical source of such actorhood (Dejordy & Creed,2016; Greenwood et al., 2011; Kraatz & Block, 2008).Because organizational identity shapes decision-makers’ views about what constitutes appropriatebehavior, it forms a fundamental basis for inten-tionality (King et al., 2010).2 As such, organizationalidentity is central to the very constitution of orga-nizations as social actors who are enabled to in-terpret and perform their own role within themacro-institutional order. Organizational identityis thus often viewed as a much-needed correctiveto the oversocialized view of traditional neo-institutionalism (Scott, 2014).

In my research model, I build on these ideas toadvance a view of organizational identity as integralto addressing this article’s puzzle of why decision-makers vary in their resistance to environmentalpressures. I focus on organizational identity because(a) it roots the individual in the organizational con-text (Ashforth et al., 2011; Fine & Hallett, 2014), andthus provides the linkage that is required for delvinginto the cognitive processes that are involved whenindividuals make adoption decisions on behalf oftheir organizations; (b) organizational identity isa key source of actorhood (King et al., 2010), and isaccordingly likely to be a relevant predictor of het-erogeneity and resistance; and (c) organizational

1 Indeed, institutional and identity theories share vari-ous common themes and assumptions (Kroezen&Heugens,2012: Table 1).

2 It is worth noting that an approach to organizationalidentity that emphasizes the perceptions of individualswithin organizations differs considerably from how pop-ulation ecologists have predominantly conceptualizedorganizational identity in terms of market categories (cf.Gioia, Price, Hamilton, & Thomas, 2010). Whether andhow these two conceptualizations of identity interact isbeyond the scope of this article (but see Gioia, Patvardhan,Hamilton, & Corley, 2013, for a relevant discussion), andthe focus here is on the former, internal perspective.

2018 1435Schilke

identity has the quality of an objective reality fororganizationalmembers (Dejordy&Creed, 2016) andis therefore highly consistent with the taken-for-grantedness aspect of institutional theorizing(Glynn, 2008; Pedersen &Dobbin, 2006). Rather thanconceiving of identity as a conceptually autonomousconcept external to institutional theory, I consider ita central element of microinstitutionalism, in linewith several other recent investigations (Dejordy &Creed, 2016; Greenwood et al., 2011; Hatch & Zilber,2012; Kraatz & Block, 2008; Thornton et al., 2012). Ielaborate this view by showing how organizationalidentity enables decision-makers to resist environ-mental pressures.

Conceptualizing Organizational Identity

Two key subdimensions of organizational identitythat need to be distinguished for analytic purposesare “organizational identification” and “organiza-tional identity content” (e.g., Haslam, Postmes, &Ellemers, 2003; Sluss & Ashforth, 2007). Whereasorganizational identification denotes the extent towhich members perceive themselves as subordinateparts of the organization, organizational identity re-fers to the particular characteristics that are (eitherobjectively or subjectively) shared among thosemembers (Gioia & Thomas, 1996; Whetten, 2006).The emphasis of organizational identification is thuson the act of classifying oneself as an organizationalmember (Ashforth &Mael, 1989; Thornton &Ocasio,2008). It reflects an individual’s perception that heor she belongs to a larger aggregate of humansdelineated by an organizational boundary. Orga-nizational identification can be thought of asa continuous concept ranging from low to high.

Organizational identity content, on the otherhand, pertains to the central, enduring, and distinctattributes of the organization (Albert & Whetten,1985). These organizational attributes offer shared3

social meanings that individuals attribute to them-selves in their roles as organizationalmembers (Scott,2014). Organizational identity content is a categoricalconcept that can be usefully represented by relevant

ideal types (Thornton & Ocasio, 2008), such asa normative and a utilitarian organizational iden-tity (Albert & Whetten, 1985; Gioia & Thomas,1996).

HYPOTHESES

Based on these considerations, I develop a re-search model that links organizational identity pro-cesses to resistance to environmental pressures.4

Given their distinct and complementary character(Ellemers, Spears, & Doosje, 2002; Haslam et al.,2003), I consider the roles of both organizationalidentification and organizational identity contentand develop distinct hypotheses for these two iden-tity subdimensions. Moving beyond main effects, Iderive relevant cognitive mechanisms that help toexplain the proposed identity–resistance relation-ship. While cognitive underpinnings of institutionscan be manifold (George et al., 2006), I focus on twoprocesses that are particularly pertinent for mymodel because they simultaneously operate as keyconsequences of organizational identity and as rel-evant sources of variation in resistance (as developedbelow)—namely, “certainty” and “attention.” Fi-nally, I contrast different types of identity and theirrespective effects on resistance. In constructing myconceptual argument, I draw from extant micro-institutional theorizing where possible, but I alsofollow Boxenbaum’s (2014) recommendation tocomplement and deepen these relatively nascentideas by borrowing from relevant social psycholog-ical research, particularly as it speaks to the conse-quences of identity processes.

The Effect of Organizational Identification onResistance to Environmental Pressures

The concept of identification is based on the no-tion that people have a tendency to classify them-selves and others into social categories, includingorganizations (as well as religious denominations,gender, age groups, etc.) (Thornton et al., 2012).Organizational identification thus pertains to self-categorization as a member of an organization(Thornton & Ocasio, 2008). Importantly, strongly

3 The degree to which social meanings are shared bymembers can of course vary from organization to organi-zation. This idea is captured by the notion of “organiza-tional identity strength”—the extent to which identitybeliefs are widely shared and densely articulated amongorganizational members (Ashforth, Harrison, & Corley,2008; Besharov & Brickson, 2016)—which is beyond thescope of the current investigation.

4 On a theoretical level, environmental pressures can beconceptualized as being of a mimetic, normative, or co-ercive nature (DiMaggio & Powell, 1983), although myempirical tests later will focus on the first two types ofpressures and any generalizations to coercive pressuresnecessarily have to remain speculative at this point.


identifying decision-makers not only locate them-selves in the social category of the organization butalso interact with their institutional environment interms of this category, allowing themselves to beguided by their role as a representative of the orga-nization (Scott, 2014). As a consequence, organiza-tional identification has profound implications fororganizational decision-making as it pertains to en-vironmental pressures.

In a way, identification transforms organizationsinto “institutions in their own right,” immunizingthem against external pressures for compliance(Greenwood et al., 2011; Zucker, 1983). In her land-mark investigation into institutional persistence,Zucker (1977: Experiment 3) compared participantsacting individually to those placed in an organiza-tional context, and she found the latter to be sub-stantially more resistant to third-party influenceattempts than the former. Extrapolating from thisfinding, I expect such a resistance effect stemmingfrom the organizational context to be significantlystronger for decision-makers who highly identifywith their organization than for those who identifyonly weakly. Given that strongly identifying actorstend to approach decision situations from the per-spective of the organization (rather than from that ofindividuals making decisions on their own account)(Scott, 2014), previously made organizational de-cisions are perceived as higher on institutionaliza-tion and thus as less situation dependent andsusceptible for change (Zucker, 1977). As a result,strong identification has the potential to outweighisomorphic pulls of the field (Battilana & Dorado,2010). In other words, the feeling of belonging to andrepresenting the organizational entity legitimizesdistinctiveness and, to a certain extent, overridesconcerns about field-level pressures (Binder, 2007;Greenwood et al., 2011), which ultimately explainsnonconforming action. I thus expect strongly iden-tifying decision-makers to display greater commit-ment to their organizations’ practices and greaterresistance to environmental pressures that challengeor offer alternatives to those practices, as per thefollowing baseline hypothesis:

Hypothesis 1. There will be a positive effect of orga-nizational identification on resistance to environ-mental pressures.

Mediating Processes

So far, I have argued that feelings of belonging toand representing the organizational entity create

changes in decision-makers’ tendencies to resistenvironmental pressures. But what specific cogni-tive attributes stemming from organizationalidentification are responsible for these behavioraldifferences? Zucker (1991) insisted that unpackingthe cognitive processes involved in themaintenanceand rejection of external institutions is a prime taskfor the micro-level approach to institutionalism.Notably, these cognitive processes constitute a criti-cal missing part of an understanding of agency in theinstitutional theory framework (Zilber, 2002).Building on extant micro-institutional theorizingand enriching it with social psychological insights, Ipropose that organizational identification increasesresistance to environmental pressures for two rea-sons: organizational identification (1) increasesthe decision-maker’s certainty and (2) shapes thedecision-maker’s attention to the environment (asI elaborate in greater detail below). Considerationof the decision-maker’s certainty and attention tothe environment is consistent with the sequentialdecision-making model outlined by Tolbert andZucker (1996), whereby organizational actors willfirst assess their own subjective situation as well asinformation from external sources, before makingtheir choice. Moreover, social identity scholarspropose that individuals’ identification with an in-group fundamentally shapes their perceived cer-tainty and focus of attention (Turner, 1982, 1987).Taken together, certainty and attention serve as keysocial mechanisms linking individuals with theirsocial systems and creating the fine-grained cou-pling between macro cause and micro effect thatBechky (2011) and Weber and Glynn (2006) havecalled for.

Certainty. A consistent finding in identity re-search is that decision-makers’ subjective certaintyis a positive function of social identification(e.g., Fransen, Haslam, Steffens, Vanbeselaere, DeCuyper, & Boen, 2015; Mullin & Hogg, 1998). Mostnotably, uncertainty identity theory (Hogg, 2000,2007) suggests that identifying with a larger entityincreases people’s situational certainty in regard towho they are, what they should think, and how theyshould feel and behave. This is not only becauseidentificationwith similar in-groupmembers createsa strong sense of belonging and connection that rai-ses people’s self-esteem (Ashforth et al., 2008;Bergami&Bagozzi,2000;Fine&Hallett, 2014;Pache&Santos, 2013), but also because identification in-creases perceived coherence with the in-group andthus furnishes consensual validation of self (Fan &Zietsma,2017;George&Chattopadhyay, 2005;Hogg&

2018 1437Schilke

Terry, 2000). Through categorizing themselves asprototypical group members, strongly identifyingindividuals perceive their own attitudes andjudgments to be aligned with those of other groupmembers, which bolsters their subjective cer-tainty. Interestingly, this consensual validationeffect through self-categorization operates even insettings where relatively little information aboutthe group prototype is available; through self-anchoring, strongly identifying individuals willgeneralize attributes of self to the prototype based onthe assumption of shared attributes due to commoncategory membership (Cadinu & Rothbart, 1996;Otten & Epstude, 2006). In sum, identifying witha larger social entity can significantly increase cer-tainty in decision-making.

Enhanced certainty, in turn, should increasedecision-makers’ resistance to environmental pres-sures. One of the central behavioral assumptions ofinstitutional theory is that organizational decision-makers have a strong preference for certainty(DiMaggio, 1988; Greenwood, Suddaby, & Hinings,2002; Oliver, 1991). Following a psychodynamicapproach, institutional theory is attuned to decision-makers’worries about failure and social disapprovaland their corresponding need to create behavioralclarity and thus a comforting sense of security(Powell, 1991; Sturdy, 2004). Indeed, it has beensuggested that a main reason why decision-makersgive in to institutional pressures is because doing somay decrease their subjective uncertainty (Berger &Luckmann, 1966). Conforming to institutionalizedscripts that appear to reflect the perceived collectivewisdom of the environment represents a commonresponse to a state of uncertainty (DiMaggio &Powell, 1983).

What is often made less explicit by in-stitutionalists is the flipside of this argument. Asdecision-makers’ subjective uncertainty decreases,so does their perceived need to follow external logics(Oliver, 1991, 1992). With growing confidence5 intheir own judgments and greater clarity regardinghow to behave in a certain situation, nonconformitywith environmental institutions becomes an in-creasingly tenable approach for decision-makers toachieve organizational goals, thus increasing thedegree of discretion assumed by the organization(Greenwood et al., 2011). As such, decision-makers’

subjective certainty is a key mechanism explainingresistance to environmental pressures.

Together, then, the above arguments suggest that,to the extent that organizational identification cre-ates certainty, it will also result in greater resistanceto environmental pressures. In other words, becauseincreased identificationmakes organizational decision-makers more confident, these decision-makers will beless likely to resort to environmental conformity asa certainty enhancement strategy.

Hypothesis 2. Organizational identification hasa positive effect on certainty, and certainty in turn hasa positive effect on resistance to environmentalpressures, such that certainty mediates the positiveeffect of organizational identification on resistance toenvironmental pressures.

Attention. Next, I offer the position that organiza-tional identification affects the extent to which at-tention is oriented toward environmental stimuli.According to social identity theory, self-categorizationas amember of a groupmodulates attention “in that itspecifies who should and should not be attended tofor appropriate information” (Abrams & Hogg, 1990:190). Specifically, strongly identifying individualstend to reject information that threatens the localreality accepted by the in-group, particularly whenthat information originates from external sourcesoutside of that in-group (Turner, 1982, 1987). Thereason is that the more an individual identifies witha social entity, the less valid and relevant she willperceive information from outsiders to be. Stronglyidentifying individuals use the categorical member-ship of the information source as a heuristic cue forinformation appropriateness, such that the validityand relevance of out-group information are dis-counted (Mackie, Gastardo-Conaco, & Skelly, 1992;Turner, 1987). As a result, out-group informationwill likely be dismissed and receive only minimalattention (Stapel, Reicher, & Spears, 1994). Mackieand colleagues found empirical support for thenotion that information from out-group memberselicits relatively little attention and systematicprocessing (Mackie et al., 1992; Mackie, Worth, &Asuncion, 1990). Consistent with this argument, Iexpect that decision-makers who strongly iden-tify with their organization (i.e., the in-group)will pay less attention to cues from other actors inthe organizational environment (i.e., the out-group).

Just as environmental stimuli vary in the attentionthey receive from different decision-makers, so dotheir availability, accessibility, and salience during

5 According to Sniezek (1992), “confidence” can be un-derstood as quasi-synonymous with subjective certaintyand can be formally defined as the strength of belief in thequality of one’s judgment.


decision deliberation, and thus their ability to shapeorganizational behavior (Glaser, Fast, Harmon, &Green, 2016; Nigam & Ocasio, 2010; Thornton et al.,2012). Environmental pressures that are largely ig-nored have little potential to evoke meaning anddefinewhat is anappropriate course of action.On theother hand, environmental pressures that receivesubstantial attention from decision-makers are morelikely to encounter critical reflection, influencesensemaking, and ultimately inform organizationalaction. For instance, if decision-makers devotesignificant resources to scanning and interpretingpractice adoption decisions by other organizationsin their field, these decisions will be more likely toshape those actors’ own adoption choices, dimin-ishing their likely resistance to mimetic pressures(Chandler & Hwang, 2015; Fiol & O’Connor, 2003).In short, responses to environmental pressures aresubstantially affected by the attention of organiza-tional decision-makers to their institutional envi-ronment. Consequently, Thornton et al. (2012)viewed decision-makers’ focus of attention asa crucial micro-level mechanism shaping organi-zational decisions vis-a-vis their institutionalenvironment.

Taken together, these arguments suggest that or-ganizational identification will lead decision-makersto devote less attention to environmental institutions,and reduced attention will, in turn, inhibit action to-ward adopting practices implied by the environment.Because attention to the external environment con-stitutes a necessary step toward conformity, reducedattention will buffer the organizational decision-maker from environmental pressures.

Hypothesis 3. Organizational identification has anegative effect on attention, and attention in turn hasa negative effect on resistance to environmentalpressures, such that attention mediates the positiveeffect of organizational identification on resistance toenvironmental pressures.6

The Role of Identity Content

Beyond organizational identification, I suggestthat the content of the organizational identity willshape decision-makers’ focus of attention and,thus, the degree to which they resist environmental

pressures. Organizational identities vary in termsof the key attributes that members believe makethe organization unique (Dutton & Penner, 1993;Selznick, 1957). The content of the organizationalidentity provides guidance on what the organiza-tion is and how it and its members should behave(Glynn, 2000; Scott, 2014). The central, enduring,and distinct attributes of the organization thusfunction as referents for members when they act orspeak on behalf of the organization (Dutton &Dukerich, 1991; Fox-Wolfgramm, Boal, & Hunt,1998; Whetten, 2006). Further, fellow members ofthe organization are also expected to hold viewsthat are in line with its identity (Fine & Hallett,2014). To achieve in-group consistency, thedecision-maker will thus tend to adopt the un-derlying norms herself in order to avoid potentialdifferences (Ridgeway, 2006; Thornton et al., 2012).Accordingly, the norms underlying organizationalidentity should lead the decision-maker to adopt theprototypical position.

Relevant identity claims may be expressed asmission statements, policies, and routines. Theseoperate as the organization’s enduring signature,providing members with a shared set of phe-nomenological points of reference that guideconsequential deliberation and organizationaldecision-making (King et al., 2010; Whetten,2006). Configurations of such identity claimscan be usefully examined in terms of ideal types,with deliberate simplification affording comparativeanalysis to suggest testable hypotheses (Thornton &Ocasio, 2008).

Building on the works of Parsons (1960), Etzioni(1961), and Albert and Whetten (1985), Gioia andThomas (1996) identified two ideal-typical identityconfigurations, which I use in this article to derivea hypothesis regarding organizational identity con-tent; these are “utilitarian identity” and “normativeidentity.”Autilitarian identity is governed by valuesrelated to the maximization of profit and economicrationality. Obligations to the organization are de-fined in terms of self-interest,with remuneration andother extrinsic motivators as the major means ofcontrol over employees (Etzioni, 1961). The conceptof normative identity, on the other hand, is typifiedby Parsons’s (1960) pattern maintenance organi-zation with primarily cultural, educational, andexpressive functions. This church- or family-likeidentity underscores traditions, symbols, and valuesof altruism (Foreman&Whetten, 2002), andmemberstend to emphasize the authenticity and expressive

6 Note that, in a mediation model, the combination ofa negative effect of the independent variable on the medi-ator and a negative effect of themediator on the dependentvariable means that the indirect effect of the independentvariable through the mediator is positive.

2018 1439Schilke

value of their outputs (Lamertz, Heugens, &Calmet, 2005).

Comparing these two types, I propose membersof an organization with a utilitarian (as opposedto a normative) identity will pay relatively greaterattention to their external institutional environ-ment. Thornton and Ocasio (1999), for example,documented how publishing houses increasinglyfollowing a logic of capitalism and profit maximi-zation (cornerstones of a utilitarian identity) haveshifted their focus to the organization’s compara-tive market position. Perceiving their own legiti-macy to be based on the organization’s competitiveposition, decision-makers in these publishinghouses increasingly directed the focus of their at-tention to external market needs (rather than justproduce “good” books that would “sell themselves”).Along similar lines, the case study by Brown,Humphreys, and Gurney (2005) described a UK-based tour operator’s utilitarian identity as evokingthe need to devote constant attention to the marketin order to remain competitive and achieve orga-nizational goals. Likewise, Gioia and Thomas(1996: 381) cited the president of a university witha strongly utilitarian orientation as being “a greatbeliever in peer comparisons,” because such com-parisonswere seen as a key source of organizationallearning and profitability. In their quantitative an-alyses, these authors found an association betweena utilitarian (vs. normative) organizational identityand decision-makers favoring an offensive (vs. de-fensive) strategy that emphasizes diligent monitor-ing of external changes and an urge to be wellinformed about new programs. Therefore, I expecta utilitarian identity, with its strong focus on theorganization’s competitive position in its field, tolead decision-makers to pay greater attention totheir external environment, which in turn increasesthe potential for pressures from this environment toshape behavior.

A normative identity, on the other hand, de-emphasizes an external market orientation and di-rects attentionaway fromenvironmental issues (Gioia&Thomas, 1996). A normative identity has been found tofoster commitment to the organization’s ideals andcreateakindofaestheticautonomy(Moss,Short,Payne,& Lumpkin, 2011). Members of organizations charac-terized by this type of identity tend to focus on in-tangible organization-internal capabilities as definingfeatures and insulate themselves from organizationaloutsiders (Glynn, 2000). As such, individuals embed-ded in a normative identity will likely devote compar-atively fewer cognitive resources to their environment,

making them less prone to environmental pressures.7

Therefore:

Hypothesis 4. Resistance to environmental pressureswill be stronger when the organizational identity isnormative (vs. utilitarian), and this effect will be me-diated by attention to environmental stimuli.

METHODS

Study Overview

This article uses three experimental studies to testthe four hypotheses (see Table 1 for an overview).The first two studies build on a well-established ex-perimental task (introducedbyBerger&Fisek, 1970).In each of several trials, participants have the op-portunity to adjust their initial choice after learningabout responses from competing participants. Fo-cusingon those trials inwhich self andalter disagree,self staying with his/her previous choice is used asa measure of resistance to environmental pressures,the dependent variable in this research. The thirdstudy mirrors the structure of the first two but usesa slightly different task that introduces normative(rather than mimetic) pressures.

Methodological Considerations

As noted above, this article uses a set of experi-ments. Notwithstanding their limitations, experi-ments have the important advantage of controllingfor extraneous factors that would be difficult to iso-late in a field setting. By using random assignmentand systematically varying theoretically relevantinformation, one can be confident that observeddifferences in the dependent variable are due to themanipulated features rather than to other factors,thus allowing for strong internal causal inference(Brewer, 1985; Thau, Pitesa, & Pillutla, 2014;Webster & Sell, 2007). No other methodology canestablish causality—the gold standard of science—tothe extent that experiments can. What is more,experiments can significantly enhance our un-derstanding of underlyingmechanisms that are oftendifficult to detect and isolate in contextually richnon-experimental field studies. It is not surprising,

7 Note that I have no conceptual reason to believe, norwas I able to find any suggestive evidence pertaining tothe notion, that, all other things being equal, a utilitar-ian identity ought to engender greater or lesser cer-tainty (the second proposed mechanism underlying theidentification–resistance effect).


therefore, that experiments are increasingly becom-ing the go-to method for micro-institutional inquiry(see also Bitektine et al., 2018).

A possible source of concern regarding experi-mental methods may be the question of whethera laboratory setting can resemble a real-life organi-zation. In considering this issue, it is important tobear inmind that experiments generalize tonaturallyoccurring situations not directly but only throughtheory (Kanazawa, 1999; Lucas, 2003b; Martin &Sell, 1979; Stolte, Fine, & Cook, 2001; Zelditch,1980). Therefore, an adequate experimental settingneeds to ensure that the theoretical principles can betested so that the results inform the underlying the-ory, which bridges the experimental study and thereal world (Bitektine et al., 2018). Although in-stitutional theory is often thought of as an innatelymacro-level approach, researchers have argued that,through sociological miniaturism (Stolte et al.,2001), predictions of institutionalism can also beapplied to and tested at the level of smaller labora-tory organizations (Zucker, 1991). In fact, previouswork has successfully implemented experimentaldesigns to study key aspects of institutional theory,including the institutionalization of abstract groupstandards (Zucker, 1977), the institutionalized per-ception of female leadership in organizations (Lucas,2003a), and institutional processes leading to pricebubbles (Levine,Apfelbaum,Bernard, Bartelt, Zajac, &Stark, 2014; Levine & Zajac, 2008), as well as conse-quences of institutional complexity (Raaijmakers,

Vermeulen, Meeus, & Zietsma, 2015), different typesof institutional logics (Glaser et al., 2016), and in-stitutionalized belief systems (Hafenbradl & Waeger,2017). Consistent with this line of research, I developan experimental setting that fits the specific needs ofthe current investigation.

Study 1

Main task. The research problem required asituation in which participants are placed in“organizations” (defined here as “goal-directed,boundary-maintaining, and socially constructedsystems of human activity”; Aldrich&Rueff, 2006: 4)and make ambiguous decisions on behalf of theirorganization; moreover, it also required that they beconfronted with information on others’ behavior, inresponse to which participants could adjust theirinitial choice and emulate others’ behavior (whichallows for capturing variations in resistance to en-vironmental pressures). Finally, participants shouldbe task-oriented and should expect to be evaluated,so that they are motivated to perform the task well.

To establish such a setting, I used a variant of thestandardized experimental situation first introducedby Berger and Fisek (1970). This approach is wellestablished and has been shown to be applicable tothe study of various sociological problems (Berger,2007). Importantly, the standardized experimentalsituation is very similar to the designs used in earlierexperimental research in institutional theory studying

TABLE 1Study Overview

Studyno.

Hypothesis/-esaddressed Experimental conditions

Focal type ofenvironmental pressures Mediators Key findings

1 H1, H2, H3 (1) Weak organizationalidentification

Mimetic Certainty;attention

Organizational identificationincreases resistance toenvironmental pressures, andthis effect can be explained bysubjective certainty andattentionto environmental stimuli.

(2) Strong organizationalidentification

2 H4 (1) Utilitarian identity Mimetic Attention Resistance to environmentalpressures is stronger when theorganizational identity isnormative versus utilitarian, andthis effect can be explained byattention to environmentalstimuli.

(2) Normative identity

3 H4 (1) Utilitarian identity Normative Attention Results of Study 2 generalize tonormative pressures anda nonstudent sample.

(2) Normative identity(3) Control

2018 1441Schilke

actors’ responses to institutional pressures (Lucas,2003a; Zucker, 1977) but has the advantage of beingamenable to computerization. Computerization offersseveral well-known benefits, including the reductionof experimenter effects and greater control over timing(Molm, 2007).

More specifically, the task employedhere requiredparticipants to undergo a series of ambiguous de-cision trials, in which they made binary choices onvarious organizational problems. Participants weretold that their groupwas running a large flower shopand that they would need to make several strategicdecisions for the shop on behalf of their group (moreinformation on the group assignment appears be-low). For example, one of the questions read “Yourgroup’s flower shop needs to decide on the locationfor a new branch. Which of the following solutionsdo you choose?” The two options given were “openbranch in a shopping mall” and “open branchdowntown.” I extensively pretested and refinedthese scenarios before arriving at a final list of 25questions (see Appendix A for more details). Thesequestions represent common strategic problemsfaced by real-life organizations, thus ensuring highlevels of ecological validity (Lant & Montgomery,1992). In addition, the questions concern situationsin which the material benefits associated witha structure are ambiguous andnot readily calculable,consistent with Tolbert and Zucker’s (1996) recom-mendations regarding adequate empirical contextsfor institutional analysis. Finally, the questionswerenot overly complex, so as to ensure that even par-ticipants with limited in-depth knowledge aboutstrategic management would be able to comprehendthem (Wason, Polonsky, & Hyman, 2002).

In each decision round, participants were firstasked for their group’s initial response. Sub-sequently, they were shown a response that theywere told had been provided by a competing group.In reality, the competing group’s responses werecomputer-generated to disagree with the partici-pant’s initial choice in themajority (i.e., 20) of the 25trials,8 consistentwith Foschi (1996). Environmentalpressures were thus operationalized by confrontingparticipants with evidence of peer groups havingchosen the alternative response, which is consistentwith the core idea behind mimetic isomorphism—

that organizations feel pressured to adopt and repli-cate the solutions that others in their field have

chosen (DiMaggio & Powell, 1983)—as well as withearlier measures of environmental pressures used inarchival research (Mizruchi & Fein, 1999).9

After learning about the competing group’s choice,participants had the opportunity to adjust their owngroup’s initial response to provide a final choice forthe respective trial. That is, participants could eitherresist the influence attempt (15 stay with the initialchoice) or conform to the influence attempt and im-itate others (05 change the initial choice). Focusingon the 20 trials in which self and alter initially dis-agreed, self’s behavioral response of staying withhis or her original choice was used as a measure ofresistance (Berger & Fisek, 1970), the study’s de-pendent variable. After reading the instructions,participants went through a practice trial that fa-miliarized them with the structure of the task.

Participants and procedures. A total of 187 par-ticipants were recruited for a study on “decision-making” through the subject pool of the behaviorallab of a large public university. Participants werescheduled in groups (Msession size 5 10.2 partici-pants). To qualify for participation in the study,participants were required to have sufficiently highlevels of English proficiency (as self-reported in thepost-task questionnaire) and to demonstrate thatthey had paid sufficient attention to the study ma-terials (as assessed based on the response toa screener question hidden in the post-task ques-tionnaire). On the basis of these criteria, 18 peoplewere excluded from the study, yielding a usablesample size of 169 participants (72% were female,Mage 5 21 years).10 Participants were randomlyassigned to one of the two experimental conditions:

8 The following 20 trials were randomly chosen to havecompetitor disagreement: rounds 1, 3, 4, 5, 6, 7, 8, 9, 10, 12,13, 16, 17, 18, 19, 20, 21, 22, 24, and 25.

9 The conceptualization of resistance to environmentalpressureshas certain similarities to thenotionof escalationof commitment (Staw, 1976), in that both ideas have to dowith organizational persistence (or the lack thereof). Whatmakes the institutional theory-grounded concept of re-sistance different is its distinct focus on the organization’sposition vis-a-vis external forces, such as competitors’behaviors (Schreyogg & Sydow, 2011).

10 Specifically, two of the participants indicated theirlevel of English language proficiency to be novice or less,while 16 participants responded incorrectly to thescreener item “To show that you have read this question,please choose 0 (the very left button) as your response”(this item was adapted from Berinsky, Margolis, & Sances,2014, and was hidden within a number of other question-naire items). I also ran the analyses including these 18participants (i.e., with n 5 187), and the results were sub-stantively similar.


weak organizational identification (n5 84) or strongorganizational identification (n 5 85).

Uponarriving at the lab, participantswereusheredinto separate cubicles equipped with computers. Aresearch assistant (who was blind to the study’s hy-potheses) obtained the participants’ informed con-sent and told them they had to wait until all sessionparticipants had arrived before the experiment couldstart. Once the signal to begin was given, all furtherinstructions were displayed on the computer screenin each participant’s cubicle. Participants first un-derwent the identification manipulation beforecompleting the study’s main task and subsequentlyfilling out a post-task questionnaire that includedmanipulation checks and other survey items. Fi-nally, participants were debriefed and receivedU.S. $6.50 as remuneration for their participation.The procedure took about 35 minutes to complete.

Manipulation. To vary organizational identifica-tion, I employed a variant of the well-establishedminimal group paradigm (introduced by Tajfel,Billig, Bundy, & Flament, 1971), which uses catego-rization into a superordinate organization as theprimary means for manipulating identification.11

The minimal group paradigm has the advantages ofbeing compatible with ad hoc (as opposed to preex-isting) groups and of not requiring potentially con-founding face-to-face interaction between the groupmembers. Even under these relatively stripped-down conditions, prior research has shown thisconservative approach to be very effective in in-ducing identification (e.g., Brewer, 1979; Grieve &Hogg, 1999). In particular, I adopted the organiza-tional identification manipulation procedure de-vised by Doosje, Spears, and Koomen (1995) andEllemers, Spears, and Doosje (1997), which has beenrecommended by Haslam (2004: Appendix 2) spe-cifically for lab studies with ad hoc organizations.12

This manipulation procedure is described next.The first phase of the procedure required partici-

pants to engage in a word and number associationtest, which ostensibly served the purpose of dividing

people into one of two groups that would latercompete. The test presented participants with a se-ries of key words (e.g., water) and key numbers(e.g., 1111), after which they were asked to chooseone of four alternative responses they associatedmostwith the keyword (e.g., rain, fire, drink, orwell)or with the key number (e.g., 1110, 1112, 111, or 4;see Appendix B for the full list of items used in theassociation test). Based on their responses, partici-pants were led to believe that a computer algorithmwould be able to assess their dominant thinking styleand assign them to either the group of “inductivethinkers” or the group of “deductive thinkers” (inreality, all participants in all study conditions wereassigned to the group of “inductive thinkers”).13

Theywere also told that each group consisted of fourmembers who would compete against other groupson several tasks.

I then varied identification between conditions asfollows. In the strong organizational identificationcondition, participants were told that the researchteam would like them to stay in the group to whichthey had been assigned for the rest of the study, andtheywere askedwhether ornot theywould agreewiththis (which all but one participant did). This questionwas included because prior research has shown thatvoluntarygroupcommitmentenhances identificationwithagroup (Turner,Hogg,Turner,&Smith, 1984). Inthe weak identification condition, participants weresimply informed that they would stay in theirassigned group for the rest of the study, without beingasked to voluntarily commit to doing so.

Afterward, participants in the strong organiza-tional identification condition performed an in-tergroup reward allocation task, in which theyallocated points using “Tajfel matrices” (Tajfel et al.,1971). These matrices provided 14 different payofftuples for two individuals, with payoffs on one sideof the spectrum benefitting the first person whilehurting the second and payoffs on the other sidebenefitting the second personwhile hurting the first.In four of the sixmatrices, participants were asked to

11 The minimal group paradigm is consistent withThornton et al.’s (2012: 79) view that social identification“can arise even in the absence of networks of interpersonalrelations and interactions.”

12 The identification manipulation procedures adoptedhere feature several similarities to those used in the famousRobbers Cave study (Sherif, Harvey,White,Hood, &Sherif,1961), including group formation based on ostensiblesimilarities and intergroup competition that increasesgroup salience.

13 To explore whether people have differential associ-ations with the two group labels, I had 40MTurk workersrate their sentiments toward a group of deductive thinkers(n520) or a group of inductive thinkers (n520). Based onresponses to an answer scale ranging from 1 (stronglynegative) to 5 (strongly positive), I did not find a signifi-cant difference in sentiments toward deductive thinkers(Mdeductive thinkers 5 3.65) versus inductive thinkers(Minductive thinkers 5 3.60), t 5 0.25, df 5 38, d 5 0.08,p . .10.

2018 1443Schilke

divide points between members of their own group(the inductive thinkers) and members of the othergroup (the deductive thinkers), whereas, in anothermatrix, the points were divided between twoin-group members and, in another still, betweentwo out-group members.14 Leyens, Yzerbyt, andSchadron (1994: 68) suggested that performing thisintergroup reward allocation task increases categorysalience and thus identification. Participants in theweakorganizational identificationconditionperformedthe same task but only after (rather than before) thestudy’s main task.

Next, participants were asked to indicate, on a 9-point scale, the extent to which they agreed with fivegeneral questions that directly or indirectly re-lated to group membership or social contact(e.g., “Relationships with other people are very im-portant to me”).15 After they had provided their re-sponses, they were told that these five questionswere “group involvement questions” that would,together with the participant’s choices on the asso-ciation test, allow for computing a personalizedgroup involvement score—that is, a measure of theextent to which the participants felt involved withtheir respective group. It was deliberately left am-biguous how exactly this involvement score wouldbe computed. Participants in the strong identificationcondition were then informed that their group in-volvement scorewas 53 points,which purportedlywassignificantly above the average score of 40, whereasparticipants in the weak identification condition weretold that their group involvement score was 27 pointsand thus significantly below the average score of 40.

To analyze the effectiveness of themanipulation, Imeasured perceived identification using four itemspertaining to the inductive thinkers in-group (“Iidentifywith this group,” “I donot fit inwellwith theother members of this group” (reverse coded), “Iwould like to get to know the other groupmembers,”and “I feel like I belong to this group”), anchoredon a 9-point answer scale ranging from 1 (“stronglydisagree”) to 9 (“strongly agree”). The coefficientalpha (a 5 0.79) demonstrated sufficient reliabilityof the 4-item identificationmeasure. Results of a one-way analysis of variance (ANOVA) showed that themean of this measure among participants in thestrong identification condition (Mstrong identification 56.29) was significantly greater than the mean among

participants in the weak identification condition(Mweak identification55.60),F(1, 167)511.69,d50.53,p # .001. As a supplementary behavioral manipu-lation check, I also compared the number of pointsallocated to in-group members in the reward allo-cation task (focusing on the four trials of the rewardallocation task in which participants were asked todivide points between members of their own groupand members of the other group) and found thatsignificantly more points were allocated to in-groupmembers in the strong identification condition(Mstrong identification 5 38.55) than in the weakidentification condition (Mweak identification5 29.80),F(1, 167) 5 7.61, d 5 0.42, p # .01. This result pro-vided further behavioral support for a successfulmanipulation. Appendix E offers details on the com-prehensionandsuspicionchecks thatwereconducted.

Main effect. Because participants completed 20relevant individual rounds of the decision task, Icomputed an average proportion of stay decisionsacross these trials to capture the dependent variable ofresistance.16Aone-wayANOVArevealed significantdifferences in resistance across conditions,F(1, 167)510.44, d 5 0.50, p # .01.17 Participants in the strongidentification condition were more resistant (Mstrong

identification 5 0.89) compared to participants in theweak identification condition (Mweak identification 50.81), providing empirical support for Hypothesis 1.

Mediational analyses. Hypotheses 2 and 3 statedthat the effect of organizational identification on re-sistance to environmental pressures is mediated bythe decision-maker’s certainty (Hypothesis 2) andattention to environmental stimuli (Hypothesis 3),such that organizational identification increasescertainty but decreases attention to environmentalstimuli, both of which, in turn, affect resistance.“Subjective certainty” can be formally defined as thestrength of belief in the quality of one’s judgments(e.g., Sniezek, 1992). It was measured at the partici-pant level in the post-task questionnaire using thefollowing three items (with an answer scale rangingfrom 1, “strongly disagree,” to 5, “strongly agree”): “Ifeel confident about my responses,” “I feel confidentabout being a participant in the experiment,” and “Ifound it easy to provide appropriate responses” (see

14 See Appendix C for a list of the Tajfel matrices used inthe intergroup reward allocation task.

15 See Appendix D for all five “group involvementquestions.”

16 As an alternative to averaging the responses acrosstrials, I also conducted random intercept logistic re-gressions with participant as clustering variable (Rabe-Hesketh & Skrondal, 2012). The results of these analysesare consistent with those reported in the article.

17 In the article, I report the results of ANOVAs, but allfindings are robust to using Tobit regressions instead.


Sniezek, 1992, and Zucker, 1977, for similar items).The measure demonstrated good psychometricproperties (a 5 0.82).

Attention can be captured by considering the timeallocated to processing information from a source(e.g., Dutton, 1986: 4). Consistentwith this approach,attention was proxied at the trial level using thebehavioral measure of reaction time (Prinzmetal,McCool, & Park, 2005) for the second-stage decisionin the individual trials of the main task, indicatingthe degree to which the participant paid attention tothe competing group’s choices (with greater atten-tion being reflected by longer reaction times).While Iacknowledge that reaction time is a composite mea-sure of a number of unobserved mental processes, itis well aligned with the (similarly multifaceted)concept of attention, and specifically with its sub-process of executive attention, which is central toorganizational decision-making and involves theallocation of cognitive resources inworkingmemoryto incoming schema-inconsistent stimuli (Ocasio,2011). Although the duration of the second-stagedecision presumably also encompasses the processof indicating the group’s final choice, in addition toprocessing the competitor’s response, it is reason-able to assume that participants who completedStage 2 quickly have not paid much attention towhatever their competitor’s choice was, whereasthose that spent considerable time on this screenhave tried to make sense of the response from thecompetitor. Since reaction times (measured in sec-onds) were highly skewed, I redefined the variableusing a logarithmic transformation. I then created anaverage across the 20 trials for each participant.

I began testing Hypotheses 2 and 3 by running twosimple mediation bootstrapping tests (Preacher &Hayes, 2004; the mediation tests were run with5,000 bootstrap samples using version 2.16 of thePROCESS macro in SPSS; Hayes, 2013), in whicheach of the two potential mediators was analyzedindependently (Hayes, 2013:Model 4). The indirecteffect of identification on resistance was positiveand significant for certainty (ab 5 0.019; 95% con-fidence interval (CI) [0.005, 0.042]), as was the in-direct effect in the mediation model for attention(ab 5 0.029; 95% CI [0.008, 0.057]). FollowingPreacher and Kelley (2011), I also report the com-pletely standardized indirect effects as measures ofeffect size; these were abcs5 0.065 for certainty andab cs 5 0.101 for attention.

I then examined both factors simultaneously ina single multiple mediators model (Hayes, 2013:Model 6). Estimation results for this model showed

that the twoparallel indirect effectswere statisticallysignificant (via certainty: ab5 0.017; 95% CI [0.005,0.036]; ab cs 5 0.059; via attention: ab 5 0.027; 95%CI [0.005, 0.055]; abcs 5 0.094). However, the serialindirect effect through both certainty and attentionwas not found to be significant (ab 5 0.002; 95% CI[20.005, 0.013]; abcs 5 0.006). Finally, the total in-direct effect (i.e., the sum of the three specific in-direct effects described above) was statisticallysignificant in this model (ab 5 0.046; 95% CI [0.020,0.079]; ab cs 5 0.159). Figure 1 shows the estimatesfor the individual direct effects in the multiple me-diators model. It is worth noting that the direct effectof identification on resistance was positive but nolonger significant once certainty and attention wereincluded as covariates, indicating full (Baron &Kenny, 1986) or indirect only (Zhao, Lynch, &Chen, 2010) mediation. A correlation table for thisand the other studies reported in this paper is pro-vided in Appendix F (Tables F1–F3). Three supple-mentary post hoc analyses exploring gender effects,temporal tendencies, and measure validity aresummarized in Appendix G.

Discussion. The results of Study 1 provided sup-port for Hypotheses 1 to 3. As predicted, I founda significant main effect of organizational identifi-cation on resistance to environmental pressures,with greater resistance in the strong than in the weakidentification condition. Further, Study 1 providedimportant insights into the theoretical processes re-sponsible for the observed main effect. Specifically,in linewithHypotheses 2 and3, bothdecision-makercertainty and attention to environmental stimulimediated the identification–resistance relation-ship. In sum, the results of Study 1 contribute tounderstanding how organizational identificationinfluences cognition and affects the way decision-makers respond to pressures from the environment.

Study 2

Designed to test Hypothesis 4, Study 2 uses theexperimental procedures of the strong identificationcondition inStudy1while adding identity content tothe manipulation by requiring participants to con-tribute to a mission statement for their group. Orga-nizations frequently codify their identity claims intheir mission statements, which gives employeesa sense of meaning and guides the actions of the or-ganization (Leuthesser & Kohli, 1997; Pearce &David, 1987). In this experiment, mission state-ments serve to activate and increase the salience ofpreexistingunderstandings of normative andutilitarian

2018 1445Schilke

logics stored in individuals’ memories (Higgins,1996; Thornton et al., 2012). Using mission state-ments to operationalize identity type is particularlyconsistent with the social actor view on identity(Whetten, 2006; Whetten & Mackey, 2002),18 whichemphasizes that organizations make overt “claims”about who they are in society and that such claimshave a profound influence not only on external butalso on internal audiences.

Participants and procedures. The participantsfor this study included 180 business major un-dergraduate students at a large public university.Students participated in exchange for U.S. $6.50 andcourse credit. As in Study 1, self-reported levels ofEnglish proficiency and an attention screener ques-tion were used to exclude, in this case, seven peoplefrom the study, resulting in a usable sample size of173 participants (51%were female,Mage5 21 years).Participants were randomly assigned to either theutilitarian identity condition (n592)or thenormativeidentity condition (n5 81). Except for the manipula-tion, the procedures and materials were identical tothose of the strong identity condition in Study 1.

Manipulation. After finishing the practice trial ofthe flower shop task, participants were informedthat, like any other organization, their flower shopneeded a “mission statement,”whichwas defined asa coherent business purpose that is both central anddistinctive. To craft such a mission statement, each

participant was told that all group members wouldneed to contribute one sentence and that he or shewould be the last member of the group to do so. Be-fore adding their own sentence, participants in thenormative identity condition were shown the fol-lowing statements, ostensibly contributed by theother members of their team (these statements buildon the work of Gioia & Thomas, 1996): “Superiorquality (rather than competitive performance) iswhat makes us unique,” “Our strategy is based onwhat we ourselves consider to be the right way ofdoing business (rather than what the competitivemarket might reward the most),” and “What moti-vates our business is that we do what we believe isright for us (even if it’s not always themost profitablecourse of action).” Conversely, participants in theutilitarian identity condition were exposed to thefollowing team member statements: “Superior com-petitive performance (rather than quality, etc.) iswhat makes us unique,” “Our strategy is based onwhat the competitive market values and rewards themost (rather than what we ourselves might considerto be the right way of doing business),” and “Whatmotivates our business is that we are highly profit-able (even if this means changing our ways).” Sub-sequently, participants in both conditions wererequired to complete the sentence “When we makedecisions, we primarily take into consideration. . .”by choosing between two alternatives: “. . . our ownassumptions, traditions, and ideology (rather thancompetitive information)” or “. . . the objective in-formation that is available from the competitive

FIGURE 1Study 1: Results of the Mediation Analyses

R2 = 0.13

R2 = 0.04

Attention toEnvironmental

stimuli

R2 = 0.33

Certainty

Identification Resistance

b = 0.03*, SE = 0.01,t = 2.42b = –0.02,

SE = 0.05,t = 0.42

b = 0.56***, SE = 0.11,t = 4.93

b = –0.16*,SE = 0.07, t = 2.28

b = 0.02, SE =0.02, t = 1.21

b = –0.17***,SE = 0.02, t = 7.79

Notes: n5 169. Strong identification was coded as 1, and weak identification was coded as 0. The statistics indicate direct effects from themultiple mediators model. Asterisks indicate significant coefficients.

* p# .05*** p# .001

18 For more details on the prevalent views on organiza-tional identity, see Gioia et al. (2013).


environment (rather than specific assumptions, tra-ditions, or ideologies).”19 They were encouraged toselect the sentence that would be consistent withwhat the other team members had entered beforethem. Appendix H summarizes the manipulationchecks for Study 2.

Main effect. Results of a one-way ANOVAshowed that resistance averaged across trials wassignificantly greater in the normative identity con-dition (Mnormative identity 5 0.92) than in the utilitar-ian identity condition (Mutilitarian identity5 0.87), F(1,171) 5 4.69, d 5 0.33, p # .05. These results pro-vided empirical support for Hypothesis 4.

Mediational analyses. I then tested whether theobserved identity type effect was mediated by at-tention to environmental stimuli, also including thedecision-maker’s certainty as a second potentialmediator in the analysis. I used the same 3-itemmeasure for certainty that I employed in Study 1(a5 0.84) and again used the national logarithm ofreaction time to gauge attention.

Amultiplemediatorsmodel (Hayes, 2013: Model 6)showed that the indirect effect of identity type on re-sistance via attentionwas statistically significant (ab 50.007; 95% CI [0.001, 0.021]; abcs 5 0.029). However,the indirect effect via certainty was not significant(ab50.000; 95%CI [20.010, 0.008]; abcs520.002),nor was the serial indirect effect through both cer-tainty and attention (ab 5 0.000; 95% CI [20.001,0.001]; ab cs5 0.000). Finally, the total indirect effectlacked statistical significance (ab 5 0.007; 95% CI[20.005, 0.021]; abcs 5 0.027). Figure 2 summarizesthe estimates for the individual direct effects in themultiple mediators model, and Appendix I describesa post hoc analysis pertaining to gender effects.

Discussion.The results of Study2extended thoseofthe previous study by showing that the type of identitymatters, as participants in the normative identitycondition were significantly more resistant to en-vironmental pressures thanwere participants in theutilitarian identity condition. Mediation analyses

indicated that this was due to differences in the atten-tion that participants paid to environmental stimuli.

Study 3

The environmental pressures on which Studies 1and 2 focused emphasized mimetic processes lead-ing to conformity. Another important way in whichinstitutions can exert themselves on organizationaldecision-makers is through normative pressures(DiMaggio & Powell, 1983). To investigatewhether thefindings related to identity type generalize to thesesorts of pressures, I conducted an additional study.Study 3 builds on the structure of Study 2 but employsa different task that introduces normative pressures.Moreover, Study 3 introduces a control condition inaddition to the normative and utilitarian identity con-ditions. Finally, Study 3 uses a nonstudent sample.

Main task. Normative pressures are imposedthrough field norms that specify how things are done(Scott, 2014). One particularly common and importantsource of normative pressures are industry standardsestablished by professional organizations that defineappropriate behavior (Berrone, Fosfuri, Gelabert, &Gomez-Mejia, 2013;Greenwoodet al., 2002;Rossman&Schilke, 2014). Study 3’s main task focused on oneparticular standard established by the InternationalOrganization for Standardization (ISO), the world’slargestdeveloperof industrystandards (Helms,Oliver,&Webb, 2012). This particular standard—ISO 31000—lays out certain codified procedures for organizationalriskmanagement. In the task, participantswere asked toread somebasic background information aswell as a listofprosandconsrelatedtothestandard(for thefull textofthe vignette, please seeAppendix J) and then to respondto two items,whichasked themto indicate their supportfor adopting ISO 31000 for their flower shop (“Ourflower shop is going to use ISO 31000 as a guide to in-tegrate risk management in our company” and “Ourflower shop is going to invest in ISO 31000 adoptiontraining,” anchored on a 9-point scale ranging from 1,“totally disagree,” to 9, “totally agree”). The two-itemswere reverse-coded and averaged to capture partici-pants’ resistance to normative pressures.

Participants and procedures. The participantsincluded 531 adults recruited via Amazon Mechan-ical Turk (MTurk), an online crowdsourcing serviceoffering large volumes of small web-based tasks toanonymous online workers for monetary compen-sation (for further details about MTurk and analysesthat confirm the quality of MTurk responses, seeBuhrmester, Kwang, & Gosling, 2011, andWeinberg,Freese, & McElhattan, 2014). Participants were paid

19 To validate the eight statements used for theutilitarian/normative identity manipulation, I provideddefinitions of the two identity types to 40 MTurk workersand had them rate each statement on a bipolar scale rang-ing from 1 (utilitarian) to 9 (normative). Results confirmedmy a priori classification of the items. A paired sample ttest showed that the mean across the four utilitarianidentity statements was significantly smaller (M 5 2.73,SD 5 1.79) than the mean across the four normative iden-tity statements (M5 6.52,SD5 1.51),meandiff.5 3.79, t58.13, d 5 1.29, p# .001.

2018 1447Schilke

U.S. $1 for an online study that took about 10 min-utes to complete. An attention screener questionwasused to exclude 25 people from the study, yielding ausable sample size of n5 506.20 Within that sample,48.0% of participants were female, and they were36.6 years old on average (SD 5 11.8), with an aver-age of 14.6 years of full-time work experience (SD511.1). To assess the degree to which participants hadrelevant first-hand experience with organizationaldecision-making, I included the following question:“Haveyouever run anorganizationor organizationalunit—no matter how large or small—such that youpersonally had a significant say in its key strategicdecisions?”Of theparticipants, 56%answered “yes”to this question.21 Participants were randomly

assigned to either a control condition (n 5 170),a utilitarian identity condition (n 5 172), or a nor-mative identity condition (n 5 164). Participantsunderwent the manipulation, completed the maintask, and responded to a brief post-task survey; theywere then given a code to be entered in MTurk inorder to receive their compensation.

Manipulation. Study 3 used the manipulation oforganizational identity type employed in Study 2while adding a third control condition in which noidentitymanipulation took place. To keep the lengthof the study approximately constant, participants inthis control condition still completed the associationtest and the reward allocation task, but without anyreference to group formation (specifically, in thiscondition, the association test ostensibly served thepurpose of evaluating participants’ personal style ofthinking, the reward allocation task only referred toanonymous study participants rather than in-group/out-group members, and participants did not con-tribute to or learn about a mission statement). Par-ticipants in the control condition were onlyinformed right before the main task that they werebeing assigned to a group with three other randomlychosen participants that would need to make orga-nizational decisions together. Appendix K demon-strates the effectiveness of the manipulation.

Main effect. Results of an ANOVA showed thatresistance (a 5 0.96) was significantly differentacross the three conditions, F(2, 503) 5 11.07, h2 50.04,p# .001. Resistancewas significantly greater in


R2 = 0.00

Certainty

R2 = 0.03

Identity Type Resistance

b = 0.02,SE = 0.04,

t = 0.64

b = –0.01, SE = 0.11,t = 0.11

b = 0.03*, SE = 0.01,t = 2.72

b = –0.14*,SE = 0.06, t = 2.36

b = –0.05*,SE = 0.03, t = 2.17

b = 0.03, SE =0.02, t = 1.80


stimuli

R2 = 0.09

Notes: n 5 173. Normative identity was coded as 1, and utilitarian identity was coded as 0. The statistics indicate direct effects from themultiple mediators model. Asterisks indicate significant coefficients.

* p# .05

20 In Study 3, the attention screener included in the post-task questionnaire asked participantswhat type of productthe company they were representing in the main task wasprimarily offering (with answer choices being food, med-ical devices, and flowers).

21 Those participants who indicated they did haveorganizational decision-making experience were sub-sequently asked to provide more details about the organi-zation and their job. The median number of employees inthese organizations was 10, suggesting that most organiza-tions that participants were in charge of were rather small.Inspection of freely entered text describing the main in-dustry showed that many of these organizationswere in thefields of retail, information technology, education, andmiscellaneous services. Self-reported job titles included co-owners, managers, supervisors, and consultants.


the normative identity condition (Mnormative identity 55.67) than in theutilitarian identitycondition (Mutilitarian

identity5 4.52), F(1, 334)5 18.47, d5 0.47, p# .001, infurther support of Hypothesis 4. Additionally, in thecontrol condition, resistance (Mcontrol 5 4.71) wassignificantly smaller than in the normative iden-tity condition, F(1, 332) 5 13.94, d 5 0.41, p #.001, but not significantly different than in theutilitarian identity condition, F(1, 340)50.55,d50.08, p . .10.

Mediational analyses. Certainty was measuredusing the 3-item scale employed previously (a 50.90). Similar to the previous studies, attention wascaptured by the natural logarithm of the time spentreading about the ISO standard. I first conductedmediation analyses using a multicategorical in-dependent variable (control, utilitarian identity,normative identity) and the unweighted contrastcoding approach described by Hayes and Preacher(2014) and implemented in the PROCESS script(Hayes, 2013: Model 4).22 The model with certaintyas mediator showed that the control condition rela-tive to the two identity conditions combined did nothave a significant indirect effect on resistancethrough certainty (ab 5 20.025; 95% CI [20.113,0.028]), nor was this indirect effect through cer-tainty significant for the contrast comparing the twoidentity conditions (ab 5 0.011; 95% CI [20.010,0.078]).23 Moreover, in the model with attention asmediator, the control condition relative to the twoidentity conditions combined did not have a signifi-cant indirect effect on resistance through attention(ab 5 20.036; 95% CI [20.153, 0.061]). However,this indirect effect through attentionwas statisticallysignificant for the utilitarian versus normative iden-tity contrast (ab 5 0.127; 95% CI [0.028, 0.263]).

Further, a multiple mediator model using a binaryindependent variable contrasting the utilitarian andnormative identity conditions (Hayes, 2013:Model 6)showed that the indirect effect of identity type on re-sistance via attention was statistically significant(ab 5 0.139; 95% CI [0.035, 0.310]; abcs 5 0.027).However, the indirect effect via certainty was notsignificant (ab 5 0.003; 95% CI [20.031, 0.067];

abcs 5 0.001), and nor was the serial indirect effectthrough both certainty and attention (ab 5 20.004;95% CI [20.026, 0.002]; abcs 520.001). Finally, thetotal indirect effect was statistically significant (ab 50.138; 95% CI [0.024, 0.320]; abcs 5 0.027).

Figure 3 summarizes the estimates for the directeffects in the multiple mediators model. The maineffect of identity type on resistance was statisticallysignificant in this model, suggesting partial media-tion. Appendix L reports the results of post hoc an-alyses exploring the roles of gender and taskexperience.

GENERAL DISCUSSION

This article suggests that organizational identityprocesses can significantly affect institutional ac-tion, providing anovel impetus for adding identity tothe list of key concepts that help explain variations inresponses to environmental pressures and within-field heterogeneity. The findings have broad theo-retical significance because they offer new supportfor the claim that micro-level cognition plays a criti-cal role in how decision-makers respond to macro-institutional pressures, thus giving new insight intothe questions of why some organizational decision-makers may resist institutional pressures whileothers do not and, as a result, why not all organiza-tions are the same. The article’s findings contributeto the literature in several ways.

First and foremost, this research promotes the de-velopment of the microfoundations of institutionaltheory called for by Zucker (1991), Powell andColyvas (2008), and many others (e.g., Bitektine &Haack, 2015; Greenwood et al., 2011; Thornton et al.,2012). This emerging literature asks the importantquestion of how individuals’ location in social re-lations affects the way in which they interpret andrespond to their institutional context (Glaseret al., 2016; Powell & Colyvas, 2008). Deviatingfrom traditional neo-institutional theory, micro-institutionalism thus places great emphasis onintra-organizational processes to better understanddifferences in the behavior of organizations (Gavettiet al., 2007; King et al., 2010; Zucker, 1983). Myfindings augment this work and indicate the impor-tance of organizational identification as a key sourceof variation in how organizational decision-makersinteract with their external context. I offer initialexperimental support for the notion that organiza-tional identification has a significant influence ona key institutional outcome variable—resistance toenvironmental pressures. Adding greater nuance to

22 For multicategorical independent variables, thePROCESS script currently doesnot accommodatemultiplemediator models (Hayes, 2013: Model 6). To probe serialmediation, I separately estimated a multiple mediatormodel for the two identity conditions only (see below).

23 The effect size option is currently not available inthe PROCESS script with multicategorical independentvariables.

2018 1449Schilke

our understanding of the cognitive processesthrough which the local context shapes organiza-tional decision-making, the article recognizes twomechanisms—certainty and attention to environ-mental stimuli—that explain the identification–resistance effect. Identifying these mechanismsequips institutional theory to move beyond the no-tion of blind imitation to capture more nuancedadoption processes that explain why some decision-makers give in to environmental forces while othersare able to resist them. The article thus contributesboth novel empirical evidence and additional theo-retical richness to the micro-institutional researchagenda. In particular, it provides support for a situ-ated stance (Boxenbaum, 2014), highlighting that therepertoire of practices an organization comes toadopt ismaterially affected by the nature of decision-makers’ relationships with their organization. Orga-nizations’ tendencies to succumb to or resist toinstitutional pressures may thus arise not only fromfeatures of the environment but also from keydecision-makers’ sense of connection to the orga-nization, aswell as from fundamental characteristicsthat are shared among organizational members.From this perspective, organizational identity con-stitutes an important enabling condition for organi-zational agency.Conceptualizing thedegreeof agencyas a continuous (rather than binary) concept, thispaper advocates that organizational actors are likelyto fall somewhere in between the two extremes ofheroic actors and institutional dopes, depending on

their degree of identification and the type of identity,thus advancing a much-needed, contingent middleground concerning agency within institutional theory(Battilana, 2006; Battilana & D’Aunno, 2009; Hwang &Colyvas, 2011; Powell & Bromley, 2015).

Further, the article makes a methodological con-tribution by re-incorporating experimental tech-niques into institutional analysis and helping torevive the experimental approach to institutionaltheory initiated 40 years ago by Zucker (1977). Theunfortunate lack of methodological diversity in theinstitutional line of research has restricted it to spe-cific kinds of research questionswhile leaving othersunanswered. For too long, a rigid dichotomy of“macro 5 quantitative/explanatory” and “micro 5qualitative/interpretative” has persisted and con-fined institutional research to an unnecessarily lim-ited interpretation of institutions. Experimentalmethods help to break away from this false di-chotomy and bring useful new explanatory per-spectives to themicro-level study of institutions andtheir effects, thus opening up a whole new array oftopics accessible to institutionalists. For example,the experimental approach has the potential tomakesignificant contributions to explaining how groupsof individuals organize and carry out their goals(Heath & Sitkin, 2001) and to clarifying causal effectsof micro-level process (Bitektine, 2011; Brewer,1985; Raaijmakers et al., 2015; Webster & Sell,2007), thus allowing for an approach to institution-alism from the bottom up and complementing


R2 = 0.00

Certainty

Identity Type Resistance

b = –0.05,SE = 0.05,

t = 0.98

b = –0.11, SE = 0.10,t = 1.18

b = –0.20*,SE = 0.08, t = 2.40

b = –0.03, SE = 0.15,t = 0.17

b = –0.68***,SE = 0.17, t = 3.98

b = 1.02***,SE = 0.27,t = 3.82


StimuliR2 = 0.02

R2 = 0.10

Notes: n 5 336. Normative identity was coded as 1, and utilitarian identity was coded as 0. The statistics indicate direct effects from themultiple mediators model. Asterisks indicate significant coefficients.

* p# .05*** p# .001


existing insights. In my particular investigation, us-ing an experimental design enabledme to study howinstitutions specifically impact ambiguous organi-zational decisions, which then allowed me to ad-vancemy broader theoretical argument for agency asa contingent continuum. Further, experiments likethe ones reported here can give unique insights intorelevant theoretical mechanisms (such as cer-tainty and attention) that help to explain how or-ganizational decision-makers perceive and reactto institutional pressures. Such mechanisms arefundamental to theory building and enrichment(Stinchcombe, 1991) but are often difficult to detectand isolate in contextually rich field studies, whichmay be why they have so far gone untested in suchstudies (David & Bitektine, 2009; Kennedy & Fiss,2009; Thornton et al., 2012). Moving from inside theorganization to external audiences, future experi-ments also have huge potential to enrich the studyof legitimacy perceptions, a key construct in in-stitutional theory that too often is merely assumedrather than directly measured and linked to relevantcognitive processes (Haack et al., 2016). Experimentscan thus add considerable richness to the in-stitutional research agenda, following a sociologicalminiaturism approach to institutionalism, wherebylarge-scale social issues are investigated by meansof small-scale social situations (see Stolte et al.,2001, for more details on sociological miniaturism).In sum, experimental methods should help to ex-pand the “resource space” of potential contributionsavailable to the institutional research communityand help the institutional paradigm to remain vi-brant (Bitektine, 2009; Bitektine & Miller, 2015).Specifically, the current article’s findings show thatan experimental manipulation can successfully in-duce different levels of identification and, in turn,certainty, attention, and resistance to environmentalpressures. It is my hope that these findings, alongwith theprocedures andmeasures employed, help topave the way for future micro-quantitative work ininstitutional theory.

Finally, the article also makes an integrative con-tribution (Okhuysen & Bonardi, 2011) by furthersynthesizing the highly complementary but thusfar too-often separated streams of institutional andidentity research. Despite their common foundationin social constructionism and Durkheim’s theory ofmeaning, these two lines of research have evolvedlargely independently (Pedersen & Dobbin, 2006),with relatively little cross-fertilization (but seeGlynn, 2008, and Kroezen & Heugens, 2012). An in-tegrative approach provides significant theoretical

leverage by allowing for the construction ofmeaningto take place both within fields and within organi-zations. It thus enables a richer and more completeapproach to understanding organizational decision-making—one that may help to reconcile the seem-ingly contradictoryobservationsof homogeneity anddistinctiveness that are central to the two literatures.While recent progress has been made in better un-derstanding how the institutional environment mayshape an organization’s identity, the current workcomplements these efforts by suggesting how an or-ganization’s present identity affects its responses toenvironmental pressures (see Besharov & Brickson,2016). In this way, the article underlines how iden-tity theory can add to institutional theory the in-sightful notion of organizations as filters of externalinstitutional pressures. It is through this filtering rolethat organizations themselves can open up room foractorhood in a way that is orthogonal to the choicesof decision-makers. Such a position is highly com-patible with an organizations-as-institutions per-spective (Greenwood et al., 2011; Zucker, 1983),whereby the complexity created through the in-terplay between environmental and organizationalpractices provides greater potential for discretionand flexibility. All this implies that previous dis-cussions of enabling conditions of actorhood, whichwere usually located in either the individualdecision-maker or the organizational environ-ment, must be augmented to accommodate theorganization itself.24

In addition to its scientific merit, this researchspeaks to an issue of high practical importance. Es-pecially in times of great environmental uncertainty(e.g., during an economic crisis),many organizationstend to give up some of their agentic qualities andalmost automatically adopt certain institutionalpractices in order to “play it safe.” However, as pre-vious research has shown, conformity with the en-vironment is not always exclusively beneficial;while it may improve symbolic performance byincreasing legitimacy, it can also hurt technicalefficacy and differentiation-based competitive ad-vantage (Barreto & Baden-Fuller, 2006; Meyer &Rowan, 1977; Zucker, 1987). Thus, organizationsmay benefit from proactive management of envi-ronmental conformity versus resistance. My in-vestigation suggests (but certainly does notconclusively prove) that one way to do so may be tosteer identity salience and content. While an

24 I am grateful to one of the anonymous reviewers forgenerously sharing this thought.

2018 1451Schilke

organization’s identity has traditionally beenthought to be rather enduring (Albert & Whetten,1985), others have argued that it may be moremalleable (Gioia, Schultz, & Corley, 2000) andhave pointed to several communication- andsocialization-based approaches to identity adap-tation (Battilana & Dorado, 2010; Harquail, 2006).Moreover, whenever the organizational goal is toimplement practices that challenge traditionalmodels that are well established in an organiza-tional field (Battilana, 2006), selecting leaderswho strongly identify with their organization maybe particularly advisable.

Although this research sheds new light on the linkbetween organizational identity processes and re-sistance to environmental pressures, it is limited,like prior studies, by its conceptual focus and themethods employed. These limitations open up im-portant avenues for future research. For example, thearticle clearly focuses on the consequences of orga-nizational identity while largely bracketing thequestion of how identity is constructed. In doing so,it adopts an embedded agency perspective whileneglecting reproduction and transformation from thebottom up (Steele & King, 2011; Thornton et al.,2012). Clearly, further research is needed to elu-cidate the duality of organizational identity andinstitutionalized decision-making (Kroezen &Heugens, 2012).

Moreover, this investigation takes the commonbutsomewhat simplistic theoretical approach of indi-vidual decision-makers representing organizationalactors (King et al., 2010), thus implicitly framing theorganizationas a reflectionof its individualmembers(Hambrick & Mason, 1984: 193). In many cases, theprocess of aggregation can be considerably morecomplicated than a linear pattern, and further insightis needed to understand how mechanisms such associal interaction and interdependencies affectemergent organization-level outcomes. Qualitativemethods and agent-based modeling seem particu-larly promising for extending the current in-vestigation to study both how institutional pressuresare enacted through ongoing interactions with otherorganizational members and how collective organi-zational actions involve complex social aggregationprocesses (Felin et al., 2015; Smith & Rand, 2017).

Finally, it is worth repeating that empirical find-ings from experimental research, like those reportedhere, are generalizable only via theory rather thanthrough one-to-one application (Kanazawa, 1999;Lucas, 2003b; Martin & Sell, 1979; Zelditch, 1980).Clearly, simplifying assumptions and a focus on the

features of particular theoretical interest are neces-sary to bring organizational research questions intothe laboratory, trading off enhanced internal validityfor a certain degree of external validity. Therefore,experimental studies in institutional theory shouldspark further research that uses complementarymethods (Levine, 2003), most notably qualitativework (Zilber, 2016), to shed more light on the dy-namic process by which organizational identity de-velops and is negotiated in organizations. Suchstudies will also be able to scrutinize the extent towhich the findings reported here generalize to pop-ulations of practicingmanagers in large corporationsoutside the laboratory. As such, the current in-vestigation provides plenty of opportunity for futureresearch (e.g., lab studies involving practitioners,field replications, mixed methods work) to furthervalidate the findings reported here. Nonetheless, it isimportant to note that the usage of realistic andplausible scenarios, such as in this study, can help toreduce concerns about experiments’ external val-idity (Finch, 1987). As institutions are otherwisedifficult to research in real time, the use of scenariotechniques allows for “introducing” environmentalpressures while assessing the effect of focal in-dependent variables in situ and not in a vacuum.

Overall, it is clear that microinstitutionalism isstill in its infancy and that most of the work in thisfield lies ahead (Powell & Colyvas, 2008; Zucker,1991). My research adds to this emerging researchstream by building a strong argument for micro-institutionalism that goes beyond exploring microlinks for their own sake and instead showinghow thedevelopment of microfoundations can further ourunderstanding of organizational theory addressingorganizational practice adoption and deviance.Extending the current investigation, future researchis needed to develop a more comprehensive un-derstanding of the various conditions and mecha-nisms relevant to decision-makers conforming withversus resisting environmental pressures. Moreover,environmental pressures often contradict each otherin reality,making it important to studyhowdecision-makers process and prioritize competing institu-tional logics (Glaser et al., 2016). It is also clearthat responses to institutional pressures do not al-ways follow a binary conformance-versus-resistancechoice; thus, qualitatively different types of re-sponses should be investigated aswell (Oliver, 1991;Yoo, Bachmann, & Schilke, 2016), most notablystrategies that are at the heart of the institutionalwork literature, such as challenging, attacking, andmanipulating institutions (Lawrence, Leca, & Zilber,


2013). Another significant limitation in extant workis that institutionalization has rarely been studied asa dynamic process, so our current understanding ofthe individual stages through which institutionali-zation occurs is at best incomplete (Schilke & Cook,2013; Tolbert & Zucker, 1996). Finally, a compre-hensive approach tomicroinstitutionalismwill needto expand its focus to analyze not only decision-makers in organizations but members of externalaudiences as well. Only rarely have institutionalistsexplicitly studied the antecedents and cognitiveconsequences of audiences’ legitimacy judgments(Haack et al., 2016), despite the centrality of thisconcept in institutional theory (Tost, 2011). In sum, Iexpect microinstitutionalism to provide a particu-larly fertile ground for inquiry in the years to come.

CONCLUSIONS

In conclusion, this research provides valuablenovel findings on a key source of resistance to envi-ronmental pressures. It goes significantly beyond thetraditional neo-institutional macro-level approachto delve into cognitive processes that help to explainorganizational choices. The organizational actor andhis/her relationships to the organization and its en-vironment assume center stage in this line of inquirythat the current investigation aims to bolster, thuslaying further groundwork for a micro-institutionalagenda of research aimed at opening the black box oforganizational decision-making.

REFERENCES

Abrams, D., & Hogg, M. A. 1990. Social identification, self-categorization and social influence. European Re-view of Social Psychology, 1: 195–228.

Albert, S., & Whetten, D. A. 1985. Organizational identity.Research in Organizational Behavior, 7: 263–295.

Aldrich, H. E., & Rueff, M. 2006. Organizations evolving.Thousand Oaks, CA: SAGE.

Ashforth, B. E., Harrison, S. H., & Corley, K. G. 2008.Identification in organizations: An examination offour fundamental questions. Journal ofManagement,34: 325–374.

Ashforth, B. E., &Mael, F. 1989. Social identity theory andthe organization. Academy of Management Review,14: 20–39.

Ashforth, B. E., Rogers, K.M., &Corley, K. G. 2011. Identityin organizations: Exploring cross-level dynamics.Organization Science, 22: 1144–1156.

Barley, S. R. 2008. Coalface institutionalism. InR. Greenwood, C. Oliver, K. Sahlin & R. Suddaby

(Eds.), The SAGE handbook of organizational insti-tutionalism: 491–518. London, England: SAGE.

Baron, R. M., & Kenny, D. A. 1986. The moderator–mediator variable distinction in social psychologicalresearch: Conceptual, strategic, and statistical con-siderations. Journal of Personality and Social Psy-chology, 51: 1173–1182.

Barreto, I., & Baden-Fuller, C. 2006. To conform or to per-form? Mimetic behaviour, legitimacy-based groupsand performance consequences. Journal of Manage-ment Studies, 43: 1559–1581.

Battilana, J. 2006. Agency and institutions: The enablingrole of individuals’ social position.Organization, 13:653–676.

Battilana, J., & D’Aunno, T. 2009. Institutional work andthe paradox of embedded agency. In T. Lawrence,R. Suddaby & B. Leca (Eds.), Institutional work:Actors and agency in institutional studies of organi-zations: 31–58. Cambridge, England: Cambridge Uni-versity Press.

Battilana, J., & Dorado, S. 2010. Building sustainable hy-brid organizations: The case of commercial micro-finance organizations. Academy of ManagementJournal, 53: 1419–1440.

Battilana, J., Leca, B., & Boxenbaum, E. 2009. How actorschange institutions: Towards a theory of institutionalentrepreneurship. The Academy of ManagementAnnals, 3: 65–107.

Bechky, B. A. 2011. Making organizational theory work:Institutions, occupations, and negotiated orders. Or-ganization Science, 22: 1157–1167.

Bergami, M., & Bagozzi, R. P. 2000. Self-categorization,affective commitment and group self-esteem as dis-tinct aspects of social identity in the organization.British Journal of Social Psychology, 39: 555–577.

Berger, J. 2007. The standardized experimental situation inexpectation states research: Notes on history, uses,and special features. In M. Webster & J. Sell (Eds.),Laboratory experiments in the social sciences:353–378. Amsterdam, Netherlands: Academic Press.

Berger, J., & Fisek, M. H. 1970. Consistent and inconsistentstatus characteristics and the determination of powerand prestige orders. Sociometry, 33: 287–304.

Berger, P. L., & Luckmann, T. 1966. The social construc-tion of reality: A treatise in the sociology of knowl-edge. Garden City, NY: Doubleday.

Berinsky, A. J., Margolis, M. F., & Sances, M. W. 2014.Separating the shirkers from theworkers?Making surerespondents pay attention on self-administered sur-veys. American Journal of Political Science, 58:739–753.

Berrone, P., Fosfuri, A., Gelabert, L., & Gomez-Mejia, L. R.2013. Necessity as the mother of “green” inventions:

2018 1453Schilke

Institutional pressures and environmental innovations.Strategic Management Journal, 34: 891–909.

Besharov, M. L., & Brickson, S. L. 2016. Organizationalidentity and institutional forces: Toward an in-tegrative framework. In M. Pratt, M. Schultz,B. Ashforth & D. Ravasi (Eds.), The Oxford hand-book of organizational identity: 396–416. Oxford,England: Oxford University Press.

Binder, A. 2007. For love and money: Organizations’ cre-ative responses to multiple environmental logics.Theory and Society, 36: 547–571.

Bitektine,A. 2009.Whatmakesus faddish?Resource spaceconstraints and the “garbage can” model of socialscience research. Scandinavian Journal of Manage-ment, 25: 217–220.

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

Bitektine, A., & Haack, P. 2015. The “macro” and the“micro” of legitimacy: Toward a multilevel theory ofthe legitimacy process. Academy of ManagementReview, 40: 49–75.

Bitektine, A., Lucas, J., & Schilke, O. 2018. Institutionsunder a microscope: Experimental methods in in-stitutional theory. In A. Bryman & D. A. Buchanan(Eds.), Unconventional methodology in organiza-tion and management research: 147–167. Oxford,England: Oxford University Press.

Bitektine, A., & Miller, D. 2015. Methods, theories, data,and the social dynamics of organizational research.Journal of Management Inquiry, 24: 115–130.

Boxenbaum, E. 2014. Toward a situated stance in organi-zational institutionalism: Contributions from Frenchpragmatist sociology theory. Journal of ManagementInquiry, 23: 319–323.

Brewer, M. B. 1979. In-group bias in the minimal in-tergroup situation: A cognitive-motivational analysis.Psychological Bulletin, 86: 307–324.

Brewer, M. B. 1985. Experimental research and socialpolicy: Must it be rigor versus relevance? The Journalof Social Issues, 41: 159–176.

Brown, A. D., Humphreys, M., & Gurney, P. M. 2005.Narrative, identity and change: A case study ofLaskarina Holidays. Journal of OrganizationalChange Management, 18: 312–326.

Buhrmester,M.,Kwang, T., &Gosling, S.D. 2011.Amazon’sMechanical Turk: A new source of inexpensive, yethigh-quality, data? Perspectives on PsychologicalScience, 6: 3–5.

Cadinu, M. R., & Rothbart, M. 1996. Self-anchoring anddifferentiation processes in theminimal group setting.Journal of Personality and Social Psychology, 70:661–677.

Chandler, D., & Hwang, H. 2015. Learning from learningtheory: A model of organizational adoption strategiesat the microfoundations of institutional theory. Jour-nal of Management, 41: 1446–1476.

Creed, D. W. E., Hudson, B. A., Okhuysen, G. A., & Smith-Crowe, K. 2014. Swimming in a sea of shame: In-corporating emotion into explanations of institutionalreproduction and change. Academy of ManagementReview, 39: 275–301.

David, R. J., & Bitektine, A. B. 2009. The de-institutionalization of institutional theory? Exploringdivergent agendas in institutional research. InA. Bryman & D. Buchanan (Eds.), The SAGE hand-book of organizational research methods: 160–175.London, England: SAGE.

Dejordy, R., &Creed,W. E. D. 2016. Institutional pluralism,inhabitants, and the construction of organizationaland personal identities. In M. Pratt, M. Schultz,B. Ashforth & D. Ravasi (Eds.), The Oxford handbookof organizational identity: 374–395. Oxford, En-gland: Oxford University Press.

Dhalla, R., & Oliver, C. 2013. Industry identity in an oli-gopolistic market and firms’ responses to institutionalpressures. Organization Studies, 34: 1803–1834.

DiMaggio, P. 1988. Interest and agency in institutionaltheory. In L. G. Zucker (Ed.), Institutional patternsand organizations: Culture and environment: 3–21.Cambridge, MA: Ballinger.

DiMaggio, P., & Markus, H. R. 2010. Culture and socialpsychology: Converging perspectives. Social Psy-chology Quarterly, 73: 347–352.

DiMaggio, P., & Powell, W. W. 1983. The iron cagerevisited: Institutional isomorphism and collectiverationality in organizational fields. American So-ciological Review, 48: 147–160.

DiMaggio, P., & Powell, W. W. 1991. Introduction. InW. W. Powell & P. DiMaggio (Eds.), The new institu-tionalism inorganizationanalysis: 1–38. Chicago, IL:University of Chicago Press.

Doosje, B., Spears, R., & Koomen, W. 1995. When bad isn’tall bad: Strategic use of sample information in gener-alization and stereotyping. Journal of Personalityand Social Psychology, 69: 642–655.

Dutton, J. E. 1986.Understanding strategic agendabuildingand its implications for managing change. Scandina-vian Journal of Management Studies, 3: 3–24.

Dutton, J. E., &Dukerich, J.M. 1991.Keeping an eye on themirror: Image and identity in organizational adap-tation. Academy of Management Journal, 34:517–554.

Dutton, J. E., & Penner, W. J. 1993. The importance of or-ganizational identity for strategic agenda building. InJ. Hendry & G. Johnson (Eds.), Strategic thinking:


Leadershipand themanagement of change: 89–113.Chichester, England: Wiley.

Ellemers, N., Spears, R., & Doosje, B. 1997. Sticking to-gether or falling apart: In-group identification asa psychological determinant of group commitmentversus individual mobility. Journal of Personalityand Social Psychology, 72: 617–626.

Ellemers, N., Spears, R., & Doosje, B. 2002. Self and socialidentity.Annual Review of Psychology, 53: 161–186.

Etzioni, A. 1961. A comparative analysis of complexorganizations. New York, NY: Free Press.

Fan, G. H., & Zietsma, C. 2017. Constructing a sharedgovernance logic: The role of emotions in enablingdually embedded agency. Academy of ManagementJournal, 60: 2321–2351.

Felin, T., Foss, N. J., & Ployhart, R. E. 2015. The micro-foundations movement in strategy and organizationtheory. The Academy of Management Annals, 9:575–632.

Finch, J. 1987. The vignette technique in survey research.Sociology, 21: 105–114.

Fine, G. A., & Hallett, T. 2014. Group cultures and the ev-eryday life of organizations: Interaction orders andmeso-analysis.OrganizationStudies, 35: 1773–1792.

Fiol, C.M.,&O’Connor, E. J. 2003.Wakingup!Mindfulnessin the face of bandwagons.Academy ofManagementReview, 28: 54–70.

Foreman, P., & Whetten, D. A. 2002. Members’ identifica-tion with multiple-identity organizations. Organiza-tion Science, 13: 618–635.

Foschi, M. 1996. Double standards in the evaluation ofmenand women. Social Psychology Quarterly, 59: 237–254.

Fox-Wolfgramm, S. J., Boal, K. B., & Hunt, J. G. 1998. Or-ganizational adaptation to institutional change: Acomparative study of first-order change in prospectorand defender banks. Administrative Science Quar-terly, 43: 87–126.

Fransen, K., Haslam, S. A., Steffens, N. K., Vanbeselaere,N., De Cuyper, B., & Boen, F. 2015. Believing in “us”:Exploring leaders’ capacity to enhance team confi-dence and performance by building a sense of sharedsocial identity. Journal of Experimental Psychology:Applied, 21: 89–100.

Gavetti, G., Levinthal,D., &Ocasio,W.2007.Neo-Carnegie:The Carnegie School’s past, present, and reconstruct-ing for the future.OrganizationScience, 18: 523–536.

George, E., & Chattopadhyay, P. 2005. One foot in eachcamp: The dual identification of contract workers.Administrative Science Quarterly, 50: 68–99.

George, E., Chattopadhyay, P., Sitkin, S. B., & Barden, J.2006. Cognitive underpinnings of institutional

persistence and change: A framing perspective.Academy of Management Review, 31: 347–365.

Gioia, D. A., Patvardhan, S. D., Hamilton, A. L., & Corley,K. G. 2013. Organizational identity formation andchange. The Academy of Management Annals, 7:123–193.

Gioia, D. A., Price, K. N., Hamilton, A. L., & Thomas, J. B.2010. Forging an identity:An insider-outsider studyofprocesses involved in the formation of organizationalidentity. Administrative Science Quarterly, 55:1–46.

Gioia, D. A., Schultz, M., & Corley, K. G. 2000. Organiza-tional identity, image, and adaptive instability.Academy of Management Review, 25: 63–81.

Gioia, D. A., & Thomas, J. B. 1996. Identity, image, andissue interpretation: Sensemaking during strategicchange in academia. Administrative Science Quar-terly, 41: 370–403.

Glaser, V. L., Fast, N. J., Harmon, D. J., & Green, S. 2016.Institutional frame switching:How institutional logicsshape individual action.Research in the Sociology ofOrganizations, 48A: 35–69.

Glaser, V., Kroezen, J., & Thornton, P. H. 2015. Learninginstitutional logics. Paper presented at the FourthTriennial Alberta Institutions Conference, Banff,Alberta, Canada.

Glynn, M. A. 2000. When cymbals become symbols: Con-flict over organizational identity within a symphonyorchestra. Organization Science, 11: 285–298.

Glynn, M. A. 2008. Beyond constraint: How institutionsenable identities. In R. Greenwood, C.Oliver, K. Sahlin& R. Suddaby (Eds.), The SAGE handbook of organi-zational institutionalism: 413–430. Thousand Oaks,CA: SAGE.

Greenwood, R., Hinings, C. R., & Whetten, D. 2014. Re-thinking institutions and organizations. Journal ofManagement Studies, 51: 1206–1220.

Greenwood, R., Raynard,M., Kodeih, F., Micelotta, E. R., &Lounsbury, M. 2011. Institutional complexity and or-ganizational responses. The Academy of Manage-ment Annals, 5: 317–371.

Greenwood, R., Suddaby, R., & Hinings, C. R. 2002. Theo-rizing change: The role of professional associations inthe transformation of institutionalized fields. Acad-emy of Management Journal, 45: 58–80.

Grieve, P. G., & Hogg, M. A. 1999. Subjective uncertaintyand intergroup discrimination in the minimal groupsituation. Personality and Social Psychology Bulle-tin, 25: 926–940.

Haack, P., McKinley,W., Schilke, O., & Zucker, L. G. 2016.The theory–method gap in organizational legitimacyresearch: A critical review, synthesis, and directionsfor future research. Paper presented at the American

2018 1455Schilke

Sociological Association Annual Meeting, Seattle,WA.

Hafenbradl, S., & Waeger, D. 2017. Ideology and themicrofoundations of CSR: Why executives believe inthe business case for CSR and how this affects theirCSR engagements. Academy of Management Jour-nal, 27: 502–520.

Hallett, T. 2010a. From inhabited institutions to aninhabited institutionalism? Exploring the theoreticalposition of inhabited institutionalism. Paper pre-sented at the 2010 EGOS Colloquium: Micro-foundations of Institutions, Lisbon, Portugal.

Hallett, T. 2010b. The myth incarnate: Recoupling pro-cesses, turmoil, and inhabited institutions in an urbanelementary school. American Sociological Review,75: 52–74.

Hallett, T., & Ventresca, M. J. 2006. Inhabited institutions:Social interactions and organizational forms inGouldner’s patterns of industrial bureaucracy.Theoryand Society, 35: 213–236.

Hambrick, D. C., &Mason, P. A. 1984. Upper echelons: Theorganization as a reflection of its top managers.Academy of Management Review, 9: 193–206.

Harquail, C. V. 2006. Making use of organizationalidentity: Icons as symbolic identity proxies (Workingpaper). Stockholm, Sweden: Stockholm School ofEconomics.

Haslam, S. A. 2004. Psychology in organizations: Thesocial identity approach (2nd ed.). Thousand Oaks,CA: SAGE.

Haslam, S. A., Postmes, T., & Ellemers, N. 2003. More thana metaphor: Organizational identity makes organiza-tional life possible. British Journal of Management,14: 357–369.

Hatch, M. J., & Zilber, T. 2012. Conversation at the borderbetween organizational culture theory and institut-ional theory. Journal of Management Inquiry, 21:94–97.

Hayes,A. F. 2013. Introduction tomediation,moderation,and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

Hayes, A. F., & Preacher, K. J. 2014. Statistical mediationanalysis with a multicategorical independent vari-able. British Journal of Mathematical & StatisticalPsychology, 67: 451–470.

Heath, C., & Sitkin, S. B. 2001. Big-B versus Big-O: What isorganizational about organizational behavior? Jour-nal of Organizational Behavior, 22: 43–58.

Helms,W. S., Oliver, C., &Webb, K. 2012. Antecedents ofsettlement on a new institutional practice: Negotia-tion of the ISO 26000 standard on social re-sponsibility.Academy of Management Journal, 55:1120–1145.

Heugens, P. P. M. A. R., & Lander, M. W. 2009. Structure!Agency! (And other quarrels): A meta-analysis of in-stitutional theories of organization. Academy ofManagement Journal, 52: 61–85.

Higgins, E. T. 1996. Knowledge activation: Accessibility,applicability, and salience. In E. T. Higgins &A. W. Kruglanski (Eds.), Social psychology: Hand-book of basic principles: 133–168. New York, NY:Guilford Press.

Hogg, M. A. 2000. Subjective uncertainty reductionthrough self-categorization: A motivational theory ofsocial identity processes. European Review of SocialPsychology, 11: 223–255.

Hogg, M. A. 2007. Uncertainty–identity theory. Advancesin Experimental Social Psychology, 39: 69–126.

Hogg, M. A., & Terry, D. J. 2000. Social identity and self-categorization processes in organizational con-texts. Academy of Management Review, 25:121–140.

Hwang, H., & Colyvas, J. A. 2011. Problematizing actorsand institutions in institutional work. Journal ofManagement Inquiry, 20: 62–66.

Kanazawa, S. 1999. Using laboratory experiments to testtheories of corporate behavior. Rationality and Soci-ety, 11: 443–461.

Kennedy, M. T., & Fiss, P. C. 2009. Institutionalization,framing, and diffusion: The logic of TQM adoptionand implementation decisions among U.S. hospitals.Academy of Management Journal, 52: 897–918.

King, B. G., Felin, T., & Whetten, D. A. 2010. Finding theorganization in organizational theory: A meta-theoryof the organization as a social actor. OrganizationScience, 21: 290–305.

Kraatz, M. S., & Block, E. S. 2008. Organizational impli-cations of institutional pluralism. In R. Greenwood,C. Oliver, K. Sahlin & R. Suddaby (Eds.), The SAGEhandbook of organizational institutionalism: 243–275. Thousand Oaks, CA: SAGE.

Kraatz,M. S., & Zajac, E. J. 1996. Exploring the limits of thenew institutionalism: The causes and consequences ofillegitimate organizational change. American Socio-logical Review, 61: 812–836.

Kroezen, J. J., & Heugens, P. P. M. A. R. 2012. Organiza-tional identity formation: Processes of identity im-printing and enactment in the Dutch microbrewinglandscape. In M. Schultz, S. Maguire, A. Langley &H. Tsoukas (Eds.), Constructing identity in andaround organizations, vol. 2: 89–127. Oxford, En-gland: Oxford University Press.

Lamertz, K., Heugens, P. P.M.A. R., &Calmet, L. 2005. Theconfiguration of organizational images among firms inthe Canadian beer brewing industry. Journal ofManagement Studies, 42: 817–843.


Lant, T. K., & Montgomery, D. B. 1992. Simulation gamesasa researchmethod for studying strategicdecision-making: The case of MARKSTRAT (Working paper).Stanford, CA: Stanford University.

Lawrence, T. B., Leca, B., & Zilber, T. B. 2013. Institutionalwork: Current research, new directions and over-looked issues. Organization Studies, 34: 1023–1033.

Lawrence, T., Suddaby, R., & Leca, B. 2011. Institutionalwork: Refocusing institutional studies of organization.Journal of Management Inquiry, 20: 52–58.

Leuthesser, L., & Kohli, C. 1997. Corporate identity: Therole of mission statements. Business Horizons, 40:59–66.

Levine, M. 2003. Times, theories and practices in socialpsychology. Theory & Psychology, 13: 53–72.

Levine, S. S., Apfelbaum, E. P., Bernard, M., Bartelt, V. L.,Zajac, E. J., & Stark, D. 2014. Ethnic diversity deflatesprice bubbles.Proceedings of theNational Academyof Sciences of the United States of America, 111:18524–18529.

Levine, S. S., & Zajac, E. J. 2008. Institutionalization inefficientmarkets: The case of price bubbles.Academyof Management Best Paper Proceedings, 2008: 1–6.

Leyens, J.-P., Yzerbyt, V. Y. A., & Schadron, G. 1994. Ste-reotypesand social cognition. London, England: SAGE.

Lucas, J. W. 2003a. Status processes and the institutional-ization of women as leaders. American SociologicalReview, 68: 464–480.

Lucas, J. W. 2003b. Theory testing, generalization, and theproblemof external validity.Sociological Theory, 21:236–253.

Mackie, D. M., Gastardo-Conaco, M. C., & Skelly, J. J. 1992.Knowledge of the advocated position and the pro-cessing of in-group and out-group persuasive mes-sages. Personality and Social Psychology Bulletin,18: 145–151.

Mackie, D. M., Worth, L. T., & Asuncion, A. G. 1990. Pro-cessing of persuasive in-group messages. Journal ofPersonality and Social Psychology, 58: 812–822.

Martin,M.W., & Sell, J. 1979. The role of the experiment inthe social sciences. The Sociological Quarterly, 20:581–590.

Meyer, J. W., & Rowan, B. 1977. Institutionalized organi-zations: Formal structure as myth and ceremony.American Journal of Sociology, 83: 340–363.

Mizruchi,M. S., & Fein, L. C. 1999. The social constructionof organizational knowledge: A study of the uses ofcoercive, mimetic, and normative isomorphism. Ad-ministrative Science Quarterly, 44: 653–683.

Molm, L. D. 2007. Experiments on exchange relations andexchange networks in sociology. In M. Webster &J. Sell (Eds.), Laboratory experiments in the social

sciences: 379–406. Amsterdam, Netherlands: Acade-mic Press.

Moss, T. W., Short, J. C., Payne, G. T., & Lumpkin, G. T.2011. Dual identities in social ventures: An explor-atory study. Entrepreneurship Theory and Practice,35: 805–830.

Mullin, B.-A., & Hogg, M. A. 1998. Dimensions of sub-jective uncertainty in social identification and mini-mal intergroup discrimination. British Journal ofSocial Psychology, 37: 345–365.

Nigam, A., & Ocasio, W. 2010. Event attention, environ-mental sensemaking, and change in institutionallogics: An inductive analysis of the effects of publicattention to Clinton’s health care reform initiative.Organization Science, 21: 823–841.

Ocasio, W. 2011. Attention to attention. OrganizationScience, 22: 1286–1296.

Okhuysen, G., & Bonardi, J.-P. 2011. The challenges ofbuilding theory by combining lenses. Academy ofManagement Review, 36: 6–11.

Oliver, C. 1991. Strategic responses to institutionalprocesses.Academy of Management Review, 16: 145–179.

Oliver, C. 1992. The antecedents of deinstitutionalization.Organization Studies, 13: 563–588.

Otten, S., & Epstude, K. 2006. Overlapping mental repre-sentations of self, ingroup, and outgroup: Unravelingself-stereotyping and self-anchoring. Personality andSocial Psychology Bulletin, 32: 957–969.

Pache, A.-C., & Santos, F. 2010. When worlds collide: Theinternal dynamics of organizational responses toconflicting institutional demands. Academy of Man-agement Review, 35: 455–476.

Pache, A.-C., & Santos, F. 2013. Embedded in hybrid con-texts: How individuals in organizations respond tocompeting institutional logics. Research in the Soci-ology of Organizations, 39(B): 3–35.

Park, S., Sine, W. D., & Tolbert, P. S. 2011. Professions, or-ganizations, and institutions: Tenure systems in collegesand universities.Work and Occupations, 38: 340–371.

Parsons, T. 1960. Structure and process in modern soci-eties. New York, NY: Free Press.

Pearce, J. A., & David, F. 1987. Corporate mission state-ments: The bottom line. The Academy of Manage-ment Executive, 1: 109–115.

Pedersen, J. S., & Dobbin, F. 2006. In search of identity andlegitimation: Bridging organizational culture andneoinstitutionalism. The American Behavioral Sci-entist, 49: 897–907.

Powell, W. W. 1991. Expanding the scope of institutionalanalysis. In W. W. Powell & P. J. DiMaggio (Eds.), Thenew institutionalism in organizational analysis:183–203. Chicago, IL: University of Chicago Press.

2018 1457Schilke

Powell,W.W.,&Bromley, P. 2015.New institutionalism inthe analysis of complex organizations. In J. D. Wright(Ed.), International encyclopedia of social and be-havioral sciences, vol. 16 (2nd ed.): 764–769. Oxford,England: Elsevier.

Powell, W. W., & Colyvas, J. A. 2008. Microfoundations ofinstitutional theory. In R. Greenwood, C. Oliver,K. Sahlin & R. Suddaby (Eds.), The SAGE handbookof organizational institutionalism: 276–298. Thou-sand Oaks, CA: SAGE.

Preacher, K. J., & Hayes, A. F. 2004. SPSS and SAS pro-cedures for estimating indirect effects in simplemediation models. Behavior Research Methods,Instruments, & Computers, 36: 717–731.

Preacher, K. J., & Kelley, K. 2011. Effect size measures formediation models: Quantitative strategies for com-municating indirect effects. Psychological Methods,16: 93–115.

Prinzmetal, W., McCool, C., & Park, S. 2005. Attention:Reaction time and accuracy reveal different mecha-nisms. Journal of Experimental Psychology: Gen-eral, 134: 73–92.

Raaijmakers, A., Vermeulen, P., Meeus, M., & Zietsma, C.2015. I need time! Exploring pathways to complianceunder institutional complexity. Academy of Man-agement Journal, 58: 85–110.

Rabe-Hesketh, S., & Skrondal, A. 2012. Multilevel andlongitudinal modeling using Stata (3rd ed.). CollegeStation, TX: Stata Press Publication.

Raffaelli, R., & Glynn, M. A. 2014. Turnkey or tailored?Relational pluralism, institutional complexity, andthe organizational adoption of more or less custom-ized practices. Academy of Management Journal,57: 541–562.

Ridgeway, C. L. 2006. Linking social structure and in-terpersonal behavior: A theoretical perspective oncultural schemas and social relations. Social Psy-chology Quarterly, 69: 5–16.

Rossman, G., & Schilke, O. 2014. Close, but no cigar: Thebimodal rewards to prize-seeking. American Socio-logical Review, 79: 86–108.

Schilke, O., & Cook, K. S. 2013. A cross-level process the-ory of trust development in interorganizational re-lationships. Strategic Organization, 11: 281–303.

Schreyogg, G., & Sydow, J. 2011. Organizational path de-pendence: A process view.Organization Studies, 32:321–335.

Scott, W. R. 2014. Institutions and organizations: Ideas,interests, and identities (4th ed.). Los Angeles, CA:SAGE.

Selznick, P. 1957. Leadership in administration: A so-ciological interpretation. Evanston, IL: Row.

Sherif, M., Harvey, O. J., White, B. J., Hood,W. R., & Sherif,C.W. 1961. Intergroup conflict and cooperation: TheRobbers Cave experiment. Norman, OK: UniversityBook Exchange.

Sluss, D. M., & Ashforth, B. E. 2007. Relational identity andidentification: Defining ourselves through work relation-ships.Academy ofManagement Review, 32: 9–32.

Smith, E. B., & Rand, W. 2017. Simulating macro-level ef-fects from micro-level observations. ManagementScience. Published online aheadof print. doi: 10.1287/mnsc.2017.2877

Sniezek, J. A. 1992. Groups under uncertainty: An exami-nation of confidence in group decision-making. Or-ganizational Behavior and Human DecisionProcesses, 52: 124–155.

Stapel, D. A., Reicher, S. D., & Spears, R. 1994. Socialidentity, availability and the perception of risk.SocialCognition, 12: 1–17.

Staw, B. M. 1976. Knee-deep in the big muddy: A study of es-calating commitment to a chosen course of action. Orga-nizationalBehaviorandHumanPerformance,16:27–44.

Steele, C. W., & King, B. G. 2011. Collective intentionalityin organizations: A meta-ethnography of identity andstrategizing. Advances in Group Processes, 28: 59–95.

Stinchcombe, A. L. 1991. The conditions of fruitfulness oftheorizing about mechanisms in social science. Phi-losophy of the Social Sciences, 21: 367–388.

Stolte, J. F., Fine, G. A., & Cook, K. S. 2001. Sociologicalminiaturism: Seeing the big through the small in socialpsychology.Annual Review of Sociology, 27: 387–413.

Sturdy, A. 2004. The adoption of management ideas andpractices: Theoretical perspectives and possibilities.Management Learning, 35: 155–179.

Suddaby, R. 2010. Challenges for institutional theory.Journal of Management Inquiry, 19: 14–20.

Suddaby, R., Elsbach, K. D., Greenwood, R.,Meyer, J.W., &Zilber, T. B. 2010. Organizations and their in-stitutional environments: Bringing meaning, values,and culture back in. Academy of ManagementJournal, 53: 1234–1240.

Suddaby, R., Viale, T., & Gendron, Y. 2016. Reflexivity:The role of embedded social position and entrepre-neurial social skill in processes of field level change.Research in Organizational Behavior, 36: 225–245.

Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. 1971.Social categorization and intergroup behaviour. Eu-ropean Journal of Social Psychology, 1: 149–178.

Thau, S., Pitesa, M., & Pillutla, M. 2014. Experiments inorganizational behavior. InM.Webster & J. Sell (Eds.),Laboratory experiments in the social sciences:433–447. San Diego, CA: Academic Press.


Thornton, P. H., &Ocasio,W. 1999. Institutional logics andthe historical contingency of power in organizations:Executive succession in the higher education pub-lishing industry, 1958–1990. American Journal ofSociology, 105: 801–843.

Thornton, P. H., & Ocasio, W. 2008. Institutional logics. InR. Greenwood, C. Oliver, K. Sahlin & R. Suddaby(Eds.), The SAGE handbook of organizational insti-tutionalism: 99–129. Los Angeles, CA: SAGE.

Thornton, P. H., Ocasio, W., & Lounsbury, M. 2012. Theinstitutional logics perspective: A new approach toculture, structure and process. Oxford, England:Oxford University Press.

Tolbert, P. S., David, R. J., & Sine, W. D. 2011. Studyingchoice and change: The intersection of institutionaltheory and entrepreneurship research. OrganizationScience, 22: 1332–1344.

Tolbert, P. S., &Zucker, L.G. 1996. The institutionalizationof institutional theory. In S. Clegg, C. Hardy &W. R. Nord (Eds.),Handbook of organization studies:175–190. London, England: SAGE.

Tost, L. P. 2011. An integrative model of legitimacy judg-ments.AcademyofManagementReview, 36: 686–710.

Turner, J. C. 1982. Towards a cognitive redefinition of thesocial group. In H. Tajfel (Ed.), Social identity andintergroup relations: 15–40. Cambridge, England:Cambridge University Press.

Turner, J. C. 1987. Rediscovering the social group: Self-categorization theory. Oxford, England: Blackwell.

Turner, J. C., Hogg, M. A., Turner, P. J., & Smith, P. M. 1984.Failure and defeat as determinants of group cohesive-ness. British Journal of Social Psychology, 23: 97–111.

Wason, K. D., Polonsky, M. J., & Hyman, M. R. 2002. De-signing vignette studies in marketing. AustralasianMarketing Journal, 10: 41–58.

Weber, K., & Glynn, M. A. 2006. Making sense with in-stitutions: Context, thought and action inKarlWeick’stheory. Organization Studies, 27: 1639–1660.

Webster, M., & Sell, J. 2007. Why do experiments? In M.Webster & J. Sell (Eds.), Laboratory experiments inthe social sciences: 5–23. Amsterdam, Netherlands:Academic Press.

Weinberg, J. D., Freese, J., & McElhattan, D. 2014. Com-paring data characteristics and results of an onlinefactorial survey between a population-based anda crowdsource-recruited sample. Sociological Sci-ence, 1: 292–310.

Whetten, D. A. 2006. Albert and Whetten revisited:Strengthening the concept of organizational identity.Journal of Management Inquiry, 15: 219–234.

Whetten, D. A., & Mackey, A. 2002. A social actor con-ception of organizational identity and its implications

for the study of organizational reputation. Business &Society, 41: 393–414.

Yoo, T., Bachmann, R., & Schilke, O. 2016. Acquiescentdefiance: Tuscan wineries’ partial reactivity to theItalian government’s quality regulation system. Paperpresented at the Annual Meeting of the Academy ofManagement, Anaheim, CA.

Zelditch, M. 1980. Can you really study an army in a labo-ratory? In A. Etzioni & E. W. Lehman (Eds.), A socio-logical reader on complex organizations (3rd ed.):528–539. NewYork, NY: Holt, Rinehart, andWinston.

Zhao, X., Lynch, J., & Chen, Q. 2010. Reconsidering Baronand Kenny: Myths and truths about mediation analy-sis.The Journal of Consumer Research, 37: 197–206.

Zilber, T. B. 2002. Institutionalization as an interplay be-tweenactions,meanings, andactors:The caseof a rapecrisis center in Israel. Academy of ManagementJournal, 45: 234–254.

Zilber, T. B. 2008. The work of meanings in institutionalprocesses. In R. Greenwood, C. Oliver, K. Sahlin &R. Suddaby (Eds.), The SAGE handbook of organi-zational institutionalism: 151–169. London, England:SAGE.

Zilber, T. B. 2016. How institutional logics matter: Abottom-up exploration. Research in the Sociology ofOrganizations, 48A: 137–155.

Zucker, L. G. 1977. The role of institutionalization in cul-tural persistence.American Sociological Review, 42:726–743.

Zucker, L.G. 1983.Organizations as institutions.Researchin the Sociology of Organizations, 2: 1–47.

Zucker, L. G. 1987. Institutional theories of organization.Annual Review of Sociology, 13: 443–464.

Zucker, L. G. 1991. Postscript: Microfoundations of in-stitutional thought. In W. W. Powell & P. DiMaggio(Eds.), The new institutionalism in organizationalanalysis: 103–106. Chicago, IL: University of ChicagoPress.

Zucker, L. G., & Darby, M. R. 2005. An evolutionary ap-proach to institutions and social construction: Processand structure. In K. G. Smith &M. A. Hitt (Eds.),Greatminds in management: The process of theory de-velopment: 547–571. Oxford, England: Oxford Uni-versity Press.

Oliver Schilke ([email protected]) is an assistant pro-fessor of management and organizations (tenure track) andan assistant professor of sociology (by courtesy) at theUniversity of Arizona. He received his PhD from UCLA.Oliver is primarily interested in organizational theory, andmuch of his extant research has examined issues related to

2018 1459Schilke

mailto:[email protected]

routinesand trust, often in thecontextof interorganizationalrelationships and applying an institutional theory lens. Hiscurrent research contributes to elaborating the micro-foundations of institutional theory.

APPENDIX A

Study 1, Main Task—OrganizationalProblems Used

Initially, I created a long list of 30 decision scenarios(each consisting of a question and two answer choices)based on a variety of strategy andmarketing textbooks andcase studies. I then ran a pretest via MTurk (n 5 41), inwhich I asked participants to select a response and thenindicate how certain they were about their choice (ona scale from 1 to 7) for each of the 30 questions. Focusingon those questions with relatively little behavioral re-sponse variance and high levels of perceived certainty,I dropped five questions and slightly reworded others tomake sure that questionswere sufficiently ambiguous andcertain answer choices didn’t appear objectively supe-rior. The full list of the 25 organizational problems used inStudy 1 appears below. The self-reported mean of per-ceived certainty across these 25 questions in the MTurkpretest was 4.61 (SD 5 0.35).

(1) Your group’s flower shop needs to reduce per-sonnel costs. Which of the following solutionsdo you choose?• Discharge a small number of employees• Reduce the working hours of all employees

(2) Your group’s flower shop needs to decide on itshiring policy.Which of the following solutionsdo you choose?• Hire certified florists only• Hiremotivatedpeople nomatter their education

(3) Your group’s flower shop needs to decide on itsgeographic scope. Which of the following so-lutions do you choose?• Focus on local area• Spread over various regions

(4) Your group’s flower shop needs to decide on itsproduct breadth. Which of the following solu-tions do you choose?• Focus on flowers only• Offer flowers and other garden products

(5) Your group’s flower shop needs to decide on itspricing strategy. Which of the following solu-tions do you choose?• Low prices (meaning low profit margins but

higher volumes)

• High prices (meaning high profit margins butlower volumes)

(6) Your group’s flower shop needs to decide on itsreplenishment system. Which of the followingsolutions do you choose?• Reordering inventory at fixed points in time• Reordering inventory when running low

(7) Your group’s flower shop needs to decide onprice variations. Which of the following solu-tions do you choose?• Keep prices constant throughout the year (to

reduce consumer confusion)• Run frequent price promotions

(8) Your group’s flower shop needs to decide on itsflower portfolio. Which of the following solu-tions do you choose?• Wide variety of flowers from all over the

world• Specific types of flowers the shop is known

for(9) Your group’s flower shopneeds todecideon the

location for a new branch. Which of the fol-lowing solutions do you choose?• Open branch in a shopping mall• Open branch downtown

(10) Your group’s flower shop needs to decide onwho places orders. Which of the following so-lutions do you choose?• Orders are placed centrally by corporate

management• Orders are placed decentrally by every

branch manager(11) Your group’s flower shop needs to decide on its

ownership form. Which of the following solu-tions do you choose?• Privately held company• Publicly traded company

(12) Your group’s flower shopneeds todecideon thetarget segment of its next ad campaign. Whichof the following solutions do you choose?• Weddings and events• Individual customers

(13) Your group’s flower shop needs to decide on itssourcing strategy. Which of the following so-lutions do you choose?• Order from few exclusive flowerwholesalers• Order from many different flower

wholesalers(14) Your group’s flower shop needs to decide on

how to reward outstanding employees. Whichof the following solutions do you choose?• Personnel trainings• Financial bonuses


(15) Your group’s flower shop needs to decide ona way to improve the stores. Which of the fol-lowing solutions do you choose?• Improve store design• Improve store size

(16) Your group’s flower shop needs to increasecustomer retention. Which of the followingsolutions do you choose?• Loyalty card/bonus points• Volume discounts

(17) Your group’s flower shopneeds todetermine itskey financial goal. Which of the following so-lutions do you choose?• Increase sales growth• Increase profit margins

(18) Your group’s flower shop needs to decide onhow to obtainmarket intelligence.Whichof thefollowing solutions do you choose?• Set up an inhouse market research unit• Hire a market research consultancy

(19) Your group’s flower shop needs to decidewhich advertising channel to prioritize. Whichof the following solutions do you choose?• Advertising via Internet• Advertising via local media

(20) Your group’s flower shop needs to redesign itslogo. Which of the following solutions do youchoose?• Make the logo blue• Make the logo red

(21) Your group’s flower shop needs to decide onwhether or not to offer home delivery.Which ofthe following solutions do you choose?• Offerdeliveryservice—that’swherethemoneyis• Don’t offer delivery service—it’s not worth

the hefty investment(22) Your group’s flower shop needs to decide on its

return policy.Which of the following solutionsdo you choose?• “7 days fresh or your money back”• “If you don’t love it, we’ll take it back”

(23) Your group’s flower shop needs to decide on itsexpansion strategy. Which of the following so-lutions do you choose?• Organic growth (growth from existing business)• External growth (growth from acquiring

other flower shops)(24) Your group’s flower shop needs to decide on its

opening hours. Which of the following solu-tions do you choose?• 9 a.m.–9 p.m. for all branches• Store hours consistent with other stores in

the local neighborhood

(25) Your group’s flower shop needs to increaseits reach among commercial customers.Which of the following solutions do youchoose?• Offer discounted flower subscriptions• Hire a sales rep to go out to restaurants, fu-

neral homes, etc.

APPENDIX B

Study 1, Association Test—List of Items Used

(1) Which of the four words listed below do youassociate most with the keyword cow?• horse• farmer• grass• milk

(2) Which of the four words listed below do youassociate most with the keyword water?• rain• fire• drink• well

(3) Which of the four words listed below do youassociate most with the keyword house?• apartment• home• roof• school

(4) Which of the four words listed below do youassociate most with the keyword film?• director• cinema• camera• movie

(5) Which of the four words listed below do youassociate most with the keyword lamp?• shade• genie• street• desk

(6) Which of the four words listed below do youassociate most with the keyword cup?• tea• coffee• glass• mug

(7) Which of the four words listed below do youassociate most with the keyword phone?• dial• cell

2018 1461Schilke

• ring• call

(8) Which of the four words listed below do youassociate most with the keyword book?• page• reservation• story• library

(9) Which of the four words listed below do youassociate most with the keyword car?• engine• truck• vehicle• auto

(10) Which of the four words listed below do youassociate most with the keyword pool?• table• water• swim• cue

(11) Which of the four numbers below do you asso-ciate most with the keynumber 12?• 11• 6• 13• 24

(12) Which of the four numbers below do you asso-ciate most with the keynumber 1111?

• 1110• 1112• 111• 4


• 99• 12• 6• 3


• 111• 131• 11• 202


• 0• 2• 11• 10

APPENDIX C

Study 1, Intergroup Reward Allocation Task—Tajfel Matrices Used

(1) Please choose one of the following reward allocation structures:Line 1: Rewards/penalties for Member 2 of your group (“inductive thinkers”)Line 2: Rewards/penalties for Member 3 of the other group (“deductive thinkers”)


219 216 213 210 27 24 21 0 1 2 3 4 5 66 5 4 3 2 1 0 21 24 27 210 213 216 219

1 2 3 4 5 6 7 8 9 10 11 12 13 1414 13 12 11 10 9 8 7 6 5 4 3 2 1



(4) Please choose one of the following reward allocation structures:Line 1: Rewards/penalties for Member 3 of your group (“inductive thinkers”)Line 2: Rewards/penalties for Member 2 of your group (“inductive thinkers”)

(5) Please choose one of the following reward allocation structures:Line 1: Rewards/penalties for Member 4 of the other group (“deductive thinkers”)Line 2: Rewards/penalties for Member 3 of the other group (“deductive thinkers”)


APPENDIX D

Study 1, Group Involvement Questionsa

Next, please indicate the extent to which you agree withthe following statements:

(1) Relationships with other people are very im-portant to me.

(2) I basically never feel lonely.(3) Belonging to larger social entities and groups is

a crucial part of life.(4) Sometimes, the welfare of groups I belong

to can be as important as my own personalwelfare.

(5) I care about groups I belong to and want them tobe different/better than other groups.

(anchored on a 9-point answer scale ranging from 1,“not at all,” to 9, “very much”).

APPENDIX E

Study 1, Comprehension and Suspicion Checks

Post-task questionnaire items were used to assess self-reported comprehension and suspicion regarding the task.The mean of the item “I understood the instructions well”—with answer categories ranging from1, “strongly disagree,” to5, “strongly agree”—was 4.21, suggesting that study compre-hension was high. Further, themean of the item “There wereother participants involved in the task” was 4.33, indicatingthat participants were not suspicious about the ostensiblegroup setting. In addition, an open-ended question askedparticipants to list any aspects of the experiment that theyfound “weird” or “hard to believe” (if any).While some of theparticipants expressed doubt that the association test wasa valid instrument for discriminating between inductive anddeductive thinkers, or that the specific group involvementscore they were provided exactly matched their group in-volvement, the actual existence of other study participantsand the reality of their responseswere very rarely questioned.

214 212 210 28 26 24 22 21 3 7 11 15 19 2323 19 15 11 7 3 21 22 24 26 28 211 212 214

12 10 8 6 4 2 0 21 25 29 213 217 221 225225 221 217 213 29 25 21 0 2 4 6 8 10 12

18 17 16 15 14 13 12 11 10 9 8 7 6 55 6 7 8 9 10 11 12 13 14 15 16 17 18

17 14 11 8 5 2 21 22 23 24 25 26 27 2828 27 26 25 24 23 22 21 2 5 8 11 14 17

a Based on Ellemers, Spears, and Doosje (1997).

2018 1463Schilke

APPENDIX FCorrelation Tables

TABLE F1Study 1

Variable 1 2 3 4

1 Identification 12 Certainty 0.36*** 13 Attention 20.20** 20.10 14 Resistance 0.24*** 0.25*** 20.54*** 1

Note: n 5 169.**p# .01

***p# .001

TABLE F2Study 2

Variable 1 2 3 4

1 Identity type 12 Certainty 20.01 13 Attention 20.18* 0.05 14 Resistance 0.16* 0.19* 20.18* 1

Note: n 5 173.*p# .05

TABLE F3Study 3

Variable 1 2 3 4 5

1 Utilitarian identity 12 Normative identity 20.50*** 13 Certainty 0.11* 0.02 14 Attention 0.10* 20.07 20.03 15 Resistance 20.13** 0.20*** 20.04 20.23*** 1

Notes: n 5 506. Baseline: no identity/control condition.*p# .05

**p# .01***p# .001


APPENDIX G

Study 1, Post Hoc Analyses

In supplementary post hoc analyses, I first explored po-tential gender effects by estimating a one-way ANOVAcomparing resistance by gender. Results showed that genderhad no significant effect, F(1, 167)5 1.47, d5 0.21, p. .10.

Second, I explored temporal tendencies to determinewhether respondents would learn to resist or accept envi-ronmental pressures throughout the trials of the experiment.Running a regression with resistance as dependent variableand trial number (ranging from 1 to 20) as independent var-iableyieldedanonsignificantcoefficient (b50.01;SE50.01;z 5 1.38; p . .10). Manual inspection of the responses ag-gregated across participants at the trial level also failed toreveal any discernable cross-trial trends (see Figure G1).

Third, I assessed differences between the 20 focal trials inwhich the competitor’s response ostensibly differed and thefive trials in which there was agreement between the partic-ipant’s and the competitor’s initial choice. Running a pairedsamples t test, I found that participants stayed with theirprevious choice significantly more frequently in the latterfive trials (Mcontrol trials50.94) than in the20 focal trials (Mfocal

trials5 0.85), t5 5.15,df5 168,d5 0.60,p# .001, indicatingthat response changes were indeed a consequence of envi-ronmental pressures and lending further credibility to theoperationalization of the dependent variable.

APPENDIX H

Study 2, Manipulation Checks

After participants completed the main task, post-taskquestionnaire items were used to check the effectivenessof the manipulation. Specifically, I included four items inthe post-task questionnaire, based on the measure devisedby Gioia and Thomas (1996), that asked participants to

characterize their group. The four questionnaire itemswere “Our group really cares about high quality,” “Tradi-tions and symbols are fundamental to our group’s func-tioning,” “Economic performance is considered key tofulfilling our mission,” and “Financial returns are thecentral successmeasure for us.”The response scale rangedfrom 1 (“strongly disagree”) to 7 (“strongly agree”). Con-sistent with Gioia and Thomas (1996), I reverse-coded thelast two items of the scale so that higher scores indicateda more normative identity.

Themean of the identity typemeasure (a5 0.79) differedconsiderably between conditions; specifically, it was signif-icantly higher in the normative identity condition (Mnormative

identity 5 4.84) than in the utilitarian identity condition(Mutilitarian identity 5 3.05), F(1, 171) 5 115.06, d 5 1.64,p# .001, supporting the effectiveness of themanipulation.As a supplementary manipulation check, I included anitem asking participantswhere they saw their group on a 9-point continuum, with “Investment Bank on Wall Street”and “Catholic Church” as anchors at the low andhigh endsof the scale, respectively. The mean of this item differedsignificantly between conditions, F(1, 171) 5 46.49, d 51.03, p# .001, with participants in the utilitarian identitycondition providing responses closer to the investmentbank end of the continuum (Mutilitarian identity 5 2.91) andparticipants in the normative identity condition beingcloser to the church archetype (Mnormative identity 5 4.85).I also analyzed whether the identity type manipulationinadvertently led todifferences in identification and foundno significant difference in the perceived identificationmeasure employed earlier (a 5 0.82) between conditions,F(1, 171) 5 1.33, d 5 0.17, p . .10. Further, there was nosignificant difference between conditions in terms of thenumber of points allocated to in-group members in thereward allocation task, F(1, 171)5 0.58, d5 0.12, p. .10.These results suggested that the manipulation was ef-fective in inducing the two different types of identity

FIGURE G1Study 1: Average Resistance to Environmental Pressures (Participants Staying with Their Initial Response)

across Individual Trials

0.7

0.75

0.8

0.85

0.9

0.95

1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20Res

ista

nce

to

envi

ron

men

tal

pre

ssu

res

Trial

2018 1465Schilke

(utilitarian and normative) while holding identificationapproximately constant.

APPENDIX I


Interestingly, I found a significant negative main effectof gender in the data of Study 2 (b520.04; SE5 0.02; t52.17; p # .05), suggesting that women showed less re-sistance to environmental pressures than men did. Theinteraction effect between identity type and gender, how-ever,wasnotsignificant (b50.06;SE50.04; t51.45;p. .10).I reran all regressions for Study2with gender as a control, andthe results remained qualitatively unchanged.

APPENDIX J

Study 3, Main Task—Text

How to engage in riskmanagement is a complex strategicissue that the top management of virtually any company isfacing thesedays.Yourgroup’s flower shophas consistentlyinvested in risk management initiatives about as much asother companies in the industry and has never faced anyserious problems related to risk management issues.

Not too long ago, the ISO developed a voluntary riskmanagement standard for companies worldwide toconsider—the ISO 31000, which provides principles,framework, and a process for managing risk. Topics thatare covered by this particular standard range from imple-mentation guidelines to risk assessment techniques.

ISO 31000 has been praised by some for helping compa-nies to perform well in an environment full of uncertainty,but, at the same time, the standard is not without criticism.Opponents claim it permits only an overly narrow range ofmethods, has a narrow scope, cannot be universally in-tegrated, and is full of illogical definitions of key terms.

To learn a bit more about ISO 31000, please click here:http://www.iso.org/iso/home/standards/iso31000.htm

Now you are asked to decide whether or not your flowershop should adopt ISO 31000. To be clear, this is not a de-cision on whether to make greater investments in riskmanagement generally; it is specifically a decision onwhether to adopt this particular ISO 31000 standard.

APPENDIX K

Study 3, Manipulation Checks

An ANOVA revealed significant differences in the self-reported identity-type measure (see Study 2) across thethree conditions, F(2, 503)5 175.81,h25 0.41, p# .001. Insupport of the effectiveness of the manipulation, themean was significantly higher in the normative identitycondition (Mnormative identity 5 4.76) than in the utilitarian

identity condition (Mutilitarian identity 5 2.72), F(1, 334) 5266.69, d 5 1.78, p # .001, and the control condition(Mcontrol 5 3.71), F(1, 332) 5 112.33, d 5 1.15, p # .001.These latter two conditions were also significantly differ-ent from each other, F(1, 340) 5 98.22, d5 1.07, p# .001.Similarly, participants in the three conditions differed sig-nificantly in their assessment on the “Investment Bank onWall Street” versus “Catholic Church” scale previously usedin Study 2, F(2, 503) 5 87.64, h2 5 0.26, p # .001, with themean in the normative identity condition (Mnormative identity55.13) being significantly higher than those in both the utili-tarian identity condition (Mutilitarian identity5 2.55), F(1, 334)5169.12, d5 1.42, p# .001, and the control condition (Mcontrol

5 3.65), F(1, 332)5 55.20, d5 0.49, p# .001. The differencebetween theutilitarian identity and the control conditionson thismeasurewas also statistically significant,F(1, 340)5 34.47, d 5 0.63, p # .001.

Moreover, a comparison of the three conditions acrosslevels of self-reported identification (see the measure fromStudy1)wassignificant,F(2,503)531.36,h250.11,p# .001.There was a significantly lower identification in the controlcondition (Mcontrol5 5.15) than in both the utilitarian identity(Mutilitarian identity5 6.47,F(1, 340)5 52.94,d5 0.79,p# .001)and the normative identity conditions (Mnormative identity 56.20), F(1, 332)5 37.73, d5 0.67, p# .001. These latter twoconditions did not differ significantly from one another intermsof self-reported identification,F(1,334)52.42,d50.16,p. .10.

APPENDIX L


Similar to Study 2, results showed a negativemain effect ofgender on resistance that was significant at a 10% level (b 520.39;SE50.22; t51.80;p5 .07),but the interactiontermsofgender with the two condition dummies were not significant(ps . .10). Taken together, the gender-related post hoc ana-lyses reported here, although inconclusive, indicate that theremay be gender differences in people’s susceptibility to envi-ronmental pressures—an intriguing exploratory finding wor-thy of further study.

Additionally, I was interested in exploring whether theresults of Study 3 hold when restricting the sample to onlythose participants who had firsthand experience with or-ganizational decision-making. I thus ran subgroup ana-lyses among thosewho indicated theyhad suchexperience(n5 285). Consistent with the main results reported in themain body, one-way ANOVAs revealed greater resistancein the normative identity condition (Mnormative identity 55.71) than in the utilitarian identity condition (Mutilitarian

identity5 4.38), F(1, 181)5 13.10, d5 0.54, p# .001, and inthe control condition (Mcontrol5 5.05),F(1, 189)53.52,d50.27, p 5 .06. Moreover, the indirect effect of normativeidentity (vs. utilitarian identity) on resistance was signifi-cantly mediated by attention (95% CI [0.001, 0.386]) butnot certainty (95% CI [20.124, 0.026]).


http://www.iso.org/iso/home/standards/iso31000.htm

a micro-institutional inquiry into resistance to ... · institutional theorizing as well as the...

Documents