Arguing the Value of Virtual Worlds: Patterns of ... · virtual worlds, this paper reports on a study of the 288 argu-ments used by 59 professionals in assessing the organizational

SPECIAL ISSUE

ARGUING THE VALUE OF VIRTUAL WORLDS: PATTERNS OF DISCURSIVE SENSEMAKING

OF AN INNOVATIVE TECHNOLOGY1

Nicholas BerenteTerry College of Business, University of Georgia, 312 Brooks Hall,

Athens, GA 30602 U.S.A. {[email protected]}

Sean HansenSaunders College of Business, Rochester Institute of Technology,

105 Lomb Memorial Drive, Rochester, NY 14623-5608 U.S.A. {[email protected]}

Jacqueline C. PikePalumbo-Donahue School of Business, Duquesne University,

600 Forbes Avenue, Pittsburgh, PA 15282 U.S.A. {[email protected]}

Patrick J. BatemanWilliamson College of Business Administration, Youngstown State University,

One University Plaza, Youngstown, OH 44515 U.S.A. {[email protected]}

With the rapid pace of technological development, individuals are frequently challenged to make sense ofequivocal innovative technology while being given limited information. Virtual worlds are a prime exampleof such an equivocal innovative technology, and this affords researchers an opportunity to study sensemakingand the construction of perspectives about the organizational value of virtual worlds. This study reports onan analysis of the written assessments of 59 business professionals who spent an extended period of time inSecond Life, a popular virtual world, and discursively made sense of the organizational value of virtual worlds. Through a Toulminian analysis of the claims, grounds, and warrants used in the texts they generated, weidentify 12 common patterns of sensemaking and indicate that themes of confirmation, open-ended rhetoric,demographics, and control are evident in the different types of claims that were addressed. Further, we assertthat the Toulminian approach we employ is a useful methodology for the study of sensemaking and one thatis not bound to any particular theoretical perspective.

Keywords: Virtual worlds, Second Life, sensemaking, discourse, argument, Toulmin, organizational value

1

1Robin Teigland was the accepting senior editor for this paper. Cathy Urquhart served as the associate editor.

MIS Quarterly Vol. 35 No. 3 pp. 685-709/September 2011 685

Berente et al./Patterns of Discursive Sensemaking

Introduction

Developed initially as social places for personal enjoymentand relaxation, virtual worlds are now being “colonized” byorganizations looking to understand the value these environ-ments may offer (Ives and Junglas 2008). Given the rapidlychanging technological landscape driven by continuousincreases in computing power and unprecedented Internet-based innovation (Lyytinen and Rose 2003), organizations areeager to capitalize on emerging technologies, but are cautiousof falling prey to unfounded hype (Ramiller 2001). At thisearly juncture, the way these organizations, and the indi-viduals that comprise them, make sense of virtual worlds willdirectly impact the evolution of these novel digital environ-ments, because virtual worlds are being co-constructed bytheir creators and residents (Whitton 2003). Virtual worldshave been described as a “blank slate” within which indi-viduals and organizations can bring about novel, customsituations (Davis et al. 2009). Thus the study of sensemakingis important to the study of virtual worlds, particularly in thisformative period.

A distinguishing feature of the particular technologicalinnovations commonly referred to as new media is that theycan be described as equivocal, since they are accompanied bymultiple, conflicting interpretations that “require hunches,discussion and social support” for individuals to deal withthem (Daft et al. 1987 p. 357). Such equivocal technologiesare marked by information that is incomplete, hyperbolic, orhighly ambiguous (Rosenberg 1994; Swanson and Ramiller1997). Even where it is anticipated that equivocal IT inno-vations will make new organizational activities possible, thespecific applications are not well articulated or understood(Swanson and Ramiller 1997), since these technologies oftenallow for a range of possibilities (Weick 1990). In makingsense of innovations, individuals discursively present,negotiate, and argue for a range of perspectives on the valueof the emergent technology (Weick 1995); thus they initiallyunderstand new technologies through sensemaking processes(Griffith 1999). It is these individually constructed argumentsabout the organizational potential of emergent technologiesthat organizations draw upon to guide strategic action (Maitlis2005; Weick 1979; Weick et al. 2005), ultimately influencingthe adoption and evolution of a technology. Therefore,individuals construct the future of equivocal technologiesrather than simply apply them (Daft et al. 1987; Weick 1990).

To understand the sensemaking of individuals about thepotential organizational value of the equivocal technology ofvirtual worlds, this paper reports on a study of the 288 argu-ments used by 59 professionals in assessing the organizationalvalue of virtual worlds. These professionals spent an average

of over 13 hours in-world and wrote essays that detailed theirexperiences and evaluated virtual worlds with respect to thepotential these technologies may offer “real-world” organiza-tions. In analyzing the data, we ask the following questions:

1. How do individuals discursively make sense of thepotential organizational value of Second Life?

2. What forms of arguments might they advance to justifytheir evaluation of that potential organizational value?

3. Finally, what patterns exist in those arguments for, oragainst, the potential organizational value?

We adapt Toulmin’s (2003) framework for deconstructingpractical reasoning to capture, analyze, and elicit patternswithin the arguments these professionals made about theorganizational value of virtual worlds. Using this framework,12 patterns of argumentation underlying claims on the organi-zational value of virtual worlds emerge from the analysis ofthe data. Further, we identify five broad types of sense-making about virtual worlds, and link these sensemakingtypes with major, often conflicting, traditions in organiza-tional theory. Based on these findings, we assert that theToulminian lens is a useful and theoretically neutral tool forrigorously examining sensemaking activity.

The paper is organized as follows. First we briefly establishvirtual worlds as equivocal technology and describe howindividuals respond to such new technologies discursively. Then we present a Toulminian lens for studying sensemakingand describe our research and findings. We conclude with adiscussion of these findings, focusing on the implications forpractice and for the study of virtual worlds, as well as forsensemaking research in general.

The Equivocality of Virtual Worlds

Many organizational actors are intrigued by the burgeoningphenomenon of virtual worlds, yet they desire guidance on thepossible applications, business value, and implications of usethat accompany these environments (Ives and Junglas 2008). One source of insight on these issues is the business press,which abounds with expositions on the untapped value ofvirtual worlds (e.g., Anderson 2006; Hemp 2006; Hof 2006),but also with stern warnings against the hype of an untestedmarketplace (Rose 2007). Some leading opinion-makers urgeorganizations to take virtual worlds seriously (Richards 2008),while Gartner, Inc. (2008) and others emphasize the high rateof failure for business ventures in virtual worlds. If the

686 MIS Quarterly Vol. 35 No. 3/September 2011


ambivalence of the press proves unsatisfying, one might turnto the academic research that focuses on virtual worlds. Inthis effort, one would find relevant research to be in a nascentstage and spread across many academic disciplines (Saira-mesh et al. 2004). What research there is can be characterizedas cautiously optimistic with respect to the practical value ofvirtual environments to organizations (e.g., Davis et al. 2009;Mennecke et al. 2007), but also discussing a wide array ofcaveats relating to issues of trust, privacy, security, reputation,and the quality of content and applications (Boulos et al.2007; Hansen et al. 2009; Sairamesh et al. 2004). Indeed,Second Life and other virtual worlds have largely beendesigned as open systems in which many modes of experienceare possible, rendering it impossible to foresee all applicationsand situations that will emerge as users pursue a variety oftasks (Galimberti et al. 2001).

Given the impossibility of predicting the future of thisevolving technology, one might turn to the broader IS litera-ture. This literature has long acknowledged the importance ofsensemaking in the evolution of novel technologies. The waysocial actors perceive new technologies is directly related tothe outcomes those technologies engender (Kling 1980).After adoption, individuals dynamically appropriate, change,and reproduce technologies through their use (Orlikowski1992). Individuals and organizations must build up therequisite sets of complementary shared cognitive schemas,skills, and practices to begin to make sense of IT innovations(Attewell 1992), particularly those open-ended technologiesthat have been described as base innovations that enable thegeneration of subsequent forms of innovation (Lyytinen andRose 2003; Swanson 1994 ).

A key component of these shared cognitive schemas thatrender new IT innovations more tractable has been called an“organizing vision” (Swanson and Ramiller 1997). Orga-nizing visions are articulated representations of IT innova-tions through which communities of individuals form perspec-tives, express these perspectives, and guide action. An orga-nizing vision is a reciprocal dynamic that forms a centralmechanism of sensemaking around inchoate information tech-nologies (Swanson and Ramiller 2004). While collectivelyshared, organizing visions are in their essence produced byindividuals as they seek to guide the perspectives of othersand, in turn, afford a ready-made interpretation of IT inno-vations upon which organizational actors can draw. Core tothis sensemaking process is that individuals respond to, andmake sense of, new information technologies through someform of oral or written communication (Griffith 1999;Orlikowski and Gash 1994; Ramiller 2001; Swanson andRamiller 1997, 2004). In other words, individuals make senseof new information technologies through discourse. Ac-

cordingly, the study of sensemaking is particularly relevant atthis point in the emergence of virtual worlds, as the founda-tion of these technologies is now being constructed by thecreators of these environments, their residents, and by com-mentators in organizations, academia, and media. Virtualworlds are an exemplar of equivocality, and the ultimate valueof the innovation remains uncertain, yet the basis for anypotential long-term value is being established through an on-going sensemaking process.

Sensemaking, Discourse,and Argument

Equivocal new phenomena engender a process of sense-making whereby humans meet ambiguous situations withaction, such as speech, and then retrospectively construct theirunderstandings of these phenomena (Weick 1979). While theconcept defies simple definition, sensemaking is a socialprocess where individuals and groups fashion an under-standing of a new phenomena through the iterative testing ofplausible explanations (Weick et al. 2005). Weick (1995)articulates six key characteristics of the sensemaking process:(1) the recognition of a discrepant set of cues in the ongoingflow of events; (2) the retrospective consideration of experi-ences; (3) the generation of plausible explanatory specula-tions; (4) enactment through written and oral communication;(5) social contact with other individuals and their ideas; and6) the involvement of issues about identity and reputation. Through this sensemaking process, novel meanings come intobeing:

When we say that meanings materialize, we meanthat sensemaking is, importantly, an issue of lan-guage, talk, and communication. Situations, organi-zations, and environments are talked into existence.(Weick et al. 2005, p. 409).

While Weick is perhaps the leading articulator of the sense-making concept, the perspective has been fruitfully appliedacross a wide range of organizational contexts, includingexecutive decision making (Starbuck and Milliken 1988),strategy formulation (Westley 1990), employee socialization(Louis 1980), IT innovation (Griffith 1999), organizationalcreativity (Drazin et al. 1999), and changes in higher educa-tion (Gioia and Chittipeddi 1991; Gioia and Thomas 1996). Across these distinct domains, sensemaking has been treatedas both an individual (e.g., Gioia and Chittipeddi 1991; Louis1980) and an organization-level (e.g., Westley 1990)phenomenon.

MIS Quarterly Vol. 35 No. 3/September 2011 687


Regardless of the domain to which it is applied, sensemakingprocesses are understood to be embodied in discourse. Weick(1979) famously posed the question: “How can I know whatI think until I see what I say?” (p. 207), emphasizing thenotion that individuals make sense of the world discursively.His contention is that the individual and social facets ofsensemaking are effectively inextricable (Weick 1995).Drawing upon sources as diverse as the Greek rhetoricianIsocrates and the English philosopher Francis Bacon, Billig(1996) observes that thinking is in practice an internalizedform of argumentation: “we think to ourselves as if ad-dressing someone else” (p. 142). This insight is mirrored inMead’s concepts of role-taking and the “generalized other,”in which all thought takes the form of an internalized debatein which the thinker adopts the role of another (Dodds et al.1997). Indeed, several researchers have drawn upon thisthread to develop the field of discursive psychology as acounterpoint to a prevailing cognitive science approach tosocial psychology (Billig 1997; Edwards and Potter 1992;Potter and Edwards 2001). In this view, sensemaking isalways discursive, and therefore in some respect social, evenif occurring within a lone individual (Weick 1995).

The discursive emphasis in sensemaking is also consistentwith a wide range of social theory that argues for the ongoingdynamic construction of human understanding throughcommunication (Boden 1994; Boland et al. 1994; Czarniaw-ska 2004; Taylor and Van Every 2000). The study of dis-course typically involves detailed research into communi-cation and language known as discourse analysis (Grant et al.2004; Johnstone 2002). Discourse analysis has becomeincreasingly relevant with the recent “discursive turn” inmanagement research (Czarniawska 2004) and represents adiverse and emerging tradition2 (Fairclough 2003; Grant et al.2004). In this tradition, a discourse is generally understoodto be an aggregating term describing multiple related“instances of communication in the medium of language”

(Johnstone 2002 p. 2). Discourses are comprised of texts,each of which is an instance of discourse (Johnstone 2002). Texts can take many forms, common examples of whichinclude a book, an essay, a speech, or a conversation.

Individuals draw upon discourses to make sense of newphenomena, and then generate novel texts that, in turn,recreate the discourse (Heracleous 2004; Phillips et al. 2004). While the generation of texts is essential for sensemaking, itis important to note that the form in which this sensemakingoccurs has a direct relationship to the content of humaninterpretation (Maitlis 2005). This content of a given inter-pretation can be called an argument following the literatureon practical reasoning (Habermas 1981; Toulmin 2003;Weick 1995). An argument is a mode of communicationwhereby an individual makes an explicit claim and thensupports, or thematizes, this claim to persuade others to acceptit while anticipating criticism (Habermas 1981). In this view,an argument does not require disagreement, rather theassertion itself puts forth the requisite evidence and logic inan effort to withstand anticipated scrutiny. Proactively sup-ported thematized claims are particularly important to thestudy of sensemaking because they indicate a certain amountof reflection, anticipation, and interest critical to sensemaking. Figure 1 locates the study of arguments as a stream in thediscursive sensemaking literature, which, in turn, representsa perspective in the fields of social psychology andorganizational research.

Toulmin’s Structure of Arguments

One of the most extensive contemporary approaches to thestudy of argumentation is that developed by Toulmin (2003,original 1958). Toulmin developed his model of argumenta-tion in an effort to overcome the limitations of the predomi-nant structure of formal logic: the syllogism. While thesyllogism presents the application of deductive inference fromaxiomatic presumptions as the standard for the judgment ofclaim validity, Toulmin suggests that in everyday discoursethere are a wide variety of means by which individualsestablish the validity of their arguments. Further, Toulminpoints out that arguments that can be deductively validatedare mere “analytic” argument—trivial arguments that neces-sarily flow from premises. “Substantive” claims, such asthose assessing the organizational value of virtual worlds,cannot be validated deductively, but must be supportedthrough inductive justification intended to convince others. Highly disciplined explicit logic in communication is rare inpractical application. Rather, humans generally apply a rangeof common sense justifications to support their arguments,

2Note that we use the term tradition loosely, since there is no singleepistemological tradition of discourse analysis. Heracleous and Barrett(2001) indicate that there are three vastly different streams of discourseanalysis: functional, interpretive, and critical. Each stream brings a differentset of assumptions and focuses on a different level of analysis. The func-tional streams tend to take an atomic view of language and focus on a veryhigh level of granularity and precision in its work. The interpretive, orhermeneutic, tradition, on the other hand, requires that some minimal levelof context must be included in the analysis, which brings the unit of analysisup to the level of sentences or sets of sentences that allow for this requisitecontext (Ricoeur 1981). Finally, the critical tradition emphasizes broadsocietal sets of texts as discourse and emphasizes power relations (Fairclough2003). One wishing to categorize this work, given our goal of understandingindividual sensemaking and a focus on argumentation, would find that it fitssquarely in the interpretive, hermeneutic tradition (i.e., Ricoeur 1981;Heracleous 2004).



Figure 1. Argumentation Within the Domain of Social Psychology

and the structure of this practical reasoning is implicit. However, the structure of arguments is explicable through aconsistent model of argumentation.

Toulmin’s model of an argument has three key components:claim, grounds, and warrants (Fairclough 2003). The claimis the central assertion of the argument, the “conclusionwhose merits we are seeking to establish” (Toulmin 2003, p.90). The grounds, also known as data or evidence, are state-ments offered in support of that claim. The grounds areintended to answer the question: “What do you have to goon?” Warrants, in turn, reflect the principles or rules ofinference which suggest that the movement from the groundsto the claim is appropriate. Warrants answer the question:“How did you get there?”

While claims and grounds represent explicit statements,warrants are often implicit assumptions reflected in a line ofargumentation. Building upon Toulmin’s model, Brockriedeand Ehninger (1960) identify different types of warrants encountered in practical argumentation: cause, sign, generali-zation, analogy, parallel case, authority, and principle. Thetypes of warrants, and the inferential action that they entail,are summarized in Table 1.

In this work, we employ the concept of an argument as athematized claim and the Toulmin model of argumentation.3

We suggest that the sensemaking around the organizational

value of virtual worlds occurs through the discursive argu-ments that individuals develop. We contend that the analysisof warrants is particularly applicable to the evaluation ofsensemaking, because warrants establish the validity of claimsbased on practical reasoning and assume away the necessity ofholding the validity to the unrealistic standard of formal logic(Kock 2006). Further, it is through the warrants that one cancharacterize the form of a given argument (Toulmin 2003). Across contexts one can expect regularities to stem from theuse of certain forms of argument. Based on this insight, welook to elicit the type of claims, the types of warrants, and thecontent of the grounds supporting the argument in relation tothe organizational value of virtual worlds in our data.

Research Methodology

One of the most visible and widely noted examples of a con-temporary virtual world is the online platform Second Life. As of September 2009, Second Life boasts 126 millionresidents, or uniquely named avatars,4 with almost 600,000residents that log into the platform weekly.5 Second Life isnow a self-sustaining economy within which users can buyand sell goods and services—in many cases translating intothe generation of substantial real-world revenue (Hobson2006; Noam 2007). Second Life was chosen because it is anexemplar of the virtual worlds of which organizations areattempting to make sense.

3We focus on the core model of claims, grounds, and warrants. WhileToulmin describes a number of secondary elements of an argument (e.g.,qualifiers, rebuttals, backing), such structures would not add substantively tothe research objectives that we have outlined. This limited Toulminianapproach is consistent with recent analysis of argumentation in organizationalresearch (see Green et al. 2009), and addresses the important structuralaspects of argument (Fairclough 2003).

4According to Linden Labs, the creators of Second Life; see http://lindenlab.com/pressroom/releases/22_09_09, accessed on October 13, 2009.

5According to Linden Labs; see http://secondlife.com/whatis/economy_stats.php accessed on October 13, 2009.



Table 1. Warrant Types in the Toulmin Model (Adapted from Brockriede and Ehninger 1960)

Warrant Types(Arguments from) Actions of the Warrant

Cause Attributes a generative power to the grounds (i.e., the grounds cause the claim)

SignInterprets the meaning or significance of the facts provided in the grounds; inference based onsymptomatic indications

GeneralizationAssumes that what is true of the items in the sample will be true for related phenomena; theinference from a sample to a population

AnalogyGrounds assert a relationship between two familiar items or events and this relationship is assumedby the warrant to hold for the distinct items or events reflected in the claim

Parallel Case Assumes an essential similarity between an event or condition in the grounds and that of the claim

AuthorityAsserts the reliability or validity of a presumed expert source and their statements (i.e., grounds)expressed. Classification warrants are implicitly rooted in authority

Principle Inference based on values, ideals, or an assumed moral common ground

Data Collection

To explore the sensemaking process used to assess the poten-tial value of virtual worlds, we enlisted 59 executives andbusiness professionals to act as informants. These informantswere selected based on their enrollment in graduate-levelmanagement courses and were asked to provide writtenassessments. For the assessments, the informants wereinstructed to conduct background research on Second Life andcreate a Second Life avatar to explore the virtual world first-hand. To guide their in-world exploration and narrow thearea of inquiry, informants were asked to think about suchissues as the relevance of the platform to organizations, thepossibility of internal or external applications of Second Lifewithin firms, and the opportunities and challenges that theinnovation would pose to organizations. Informants wereasked to share their opinions and evaluations of the tech-nology after experiencing it. Informants were directed towrite about the technology based on their experiences andresearch, noting how they see this technology being used ornot used in organizations and why. Thus, the deliverable wasa reflective assessment through which informants articulatedtheir thinking about the technology and its application in real-world organizations. Furthermore, it was stressed to infor-mants, both in the written project guidelines and verbally bycourse instructors, that there are no right answers and thattheir honest assessment of the technology as a business pro-fessional was of utmost importance. The project representedbetween 10 and 15 percent of the participant’s overall courseassessment.

Descriptive Characteristics of Informants

Informants were enrolled in graduate-level managementcourses at one of two Midwestern universities. Although anumber of researchers have argued that students in a coursedo not represent practitioners (e.g., Huang et al. 2008),Gordon et al. (1986) recommend a technique for improvingthe external validity of student subjects that involves usingstudents with demographic and interest profiles similar to thenon-student subjects that ideally would be used. In thisrespect, the informants involved in this study were, in fact,also practitioners, working in various industries with a varietyof professional backgrounds and many with extensive pro-fessional experience. One group of informants came from anExecutive MBA course consisting of 25 managers employedfull-time professionally, and another from a part-time MBAcourse that had 34 full-time professionals enrolled. Theopportunity to have access to experienced informants fromtwo different management-oriented programs allowed us toobtain a pool of informants that worked in a wide range ofindustries (see Table 2), possessed diverse professionalbackgrounds (see Table 3), and were at various stages of theircareer (see Table 4). Students who had not worked or werecurrently not working in industry were not part of this study,although there were some enrolled in the courses.

It has also been argued that students can be used as surrogatesfor businesspeople when there is evidence that the studentshave adequate background for the research task (Khera 1970). For this study, professional experience (as discussed), no or



Table 2. Industries

IndustryNo. of

Informants

Construction 2

Consumer Products 7

Education 4

Information Technology 9

Environmental 1

Financial Services 8

Healthcare 8

Industrial Products 11

Manufacturing 1

Military 1

Public Relations 1

Retail 1

Transportation/Distribution 3

Utilities 2

Table 3. Professional Backgrounds

BackgroundNo. of

Informants

Accounting 5

Business Development 3

Engineering 6

Finance 2

General Management 8

Human Resources 3

Information Technology 9

Marketing 7

Medical 6

Non-disclosed 3

Operations 4

Physical Science 1

Product Management 2

Table 4. Organizational Position Held by Informants

Code DescriptionNo. of

Informants

Executives C-level, president, VP, director 18

Managers Middle managers 33

Professional Non-managerial professional 8

limited knowledge of Second Life, and the ability to useSecond Life were determined to be the prerequisite back-ground for the research task. As presented above, theinformants used in this study have sufficient professionalexperience. Informants were also asked to report informationabout their background research and time spent in-worldusing Second Life. Informants had access to publicly avail-able data sources (almost all of it electronic), and many drewupon the same or similar resources. Second Life was entirelynew to all but one of the informants.6 Each informantexperienced the virtual world first-hand. The average timespent in-world by the informants was 13.64 hours (range: 1 to100 hours), and the average number of resources consultedduring the background research was 4.35 sources (range: 0 to19 references; 49 assessments had at least one reference). The informants’ assessments averaged 2,230 words (range:

901 to 3,508 words), which equates to about 4.5 pages ofsingle-spaced text

Coding and Analysis Procedure

Grounded theory methodology is increasingly used, anddebated, in studies of information systems (e.g., Bryant et al.2004; Howcroft and Hughes 1999; Urquhart 2001) andinvolves an intensely data-driven process in the pursuit oftheoretical findings (Strauss and Corbin 1998). In this sense,our methodology reflected a grounded approach. However,strict interpretations of grounded theory insist that theoreticalsampling be a key component of any true application of themethodology (e.g., Morse 2007), and by this interpretation,our research may not strictly be considered grounded theoryby some. Nevertheless, Strauss and Corbin (1998) indicatethat their guidance for qualitative data analysis can be usefulfor a variety of research activities and can readily be adapted

6This informant visited the world two years previously and only for a shortperiod of time.



Figure 2. Process of Coding of Claims

to answering specific questions. Toward this end, weengaged in three cycles of coding following Strauss andCorbin’s process of (1) open coding, (2) axial coding, and(3) selective coding (see Figure 2).

The first cycle was open coding, involving notetaking asso-ciated with a microanalytical unit of analysis includingsentences and groups of sentences from the informants’

assessments (Corbin and Strauss 1990; Glazer and Strauss1967). First, each coder, individually, immersed themself inthe data, circling, highlighting, bolding, or underlining, signi-ficant participant quotes (Layder 1998). To become moreintimate with the data, each identified what Boyatzis (1998)refers to as “codable moments.” Here coders coded the databy identifying sentences where the assessments discussed thepotential value of virtual worlds.



In the second cycle,7 coders axially coded the data (Straussand Corbin 1998), and sought to identify assertions of poten-tial value, support for the assertions, and relationshipsbetween pairs of value and support. It has been advocatedthat for research projects involving multiple researchers,coding can, and should, be a collaborative effort (Ericksonand Stull 1998; Guest and MacQueen 2008). As such, to startthe second cycle, the coding procedure was defined amongthe four coders. Next, five assessments that were codedseparately by the four coders were compared and discussedcollectively to refine the process. Following this refinement,an additional four assessments were coded together to ensureunderstanding of the process and coding consistency. Incontrast to quantitative measures of intercoder reliability, theprimary objective of these steps was intensive group dis-cussion and consensus (Harry et al. 2005 p. 6; Saldana 2009).

Upon completion of this collective step, the remaining 50assessments were divided into two groups and assigned topairs of coders. Then each coder in each pair individuallycoded the assessments, employing the Toulmin (2003)framework (claim–ground–warrant). Coding was focused onthe argument as the level of analysis (operationalized speci-fically as thematized claims). Arguments on the value ofvirtual worlds that contained explicit evidence for the argu-ment were coded as claims, building on the perspective thatonly claims supported by evidence, or thematized, areintended to serve as arguments (Habermas 1981). For eachclaim, the grounds were coded and the type of warrant basedon Brockriede and Ehninger’s (1960) categorization (seeTable 1) was noted. Claims and grounds represent explicitstatements in the data. Most warrants were not explicitlydetailed in the data; rather, warrants are the coders’ inter-pretation based on assumptions reflected in a line of argumen-tation presented in the data. This is consistent with Toulmin’scharacterization of practical reasoning. Following is anexample (also diagramed in Figure 2) of a situation thatrequired an explicitly constructed warrant. Note that whilethe coders varied in the way they constructed implicitwarrants, the category of warrant was consistently coded.

Claim “[Virtual worlds] could be extremely valuablefor corporations or businesses that want toreach a wide demographic ...”

Ground “Wells Fargo Bank, Sun Microsystems, Coca-Cola, and Toyota have all started building stuffand doing stuff in Second Life as a method formarketing themselves online.”

Warrant [sign] Reputable companies marketing inSecond Life indicates that Second Life couldbe valuable for reaching a wide demographic.

After each coder pair coded their assessments individually,each pair discussed their coding. Instances where the codersdid not agree were negotiated and resolved. The purpose ofthis step was not to establish intercoder reliability measures,but rather to provide what Saldana (2009, p. 27) refers to asa “reality check” for coders involved in a collaborativeproject.

Once these arguments were amassed, the third cycle ofcoding, rooted in the paradigm of selective coding (Straussand Corbin 1998) commenced. In this cycle, the type of claimand the content of the ground were coded. The types ofclaims were not predetermined, but coded based on a givenargument’s core assertion regarding the value of Second Life.These assertions indicated that virtual worlds offer currentvalue, future value, contingent value, or no value. The cate-gories used to code the content emerged from the data andincluded organizational experience, personal experience,virtual worlds experience, convention, expert opinion, digiti-zation, games, Internet experience, demographic, desirabilityof control, and appropriateness. The type of claim, type ofwarrant, and the content of the grounds were coded to discernpatterns and themes across the arguments employed by theinformants. The assessments, types of claims, types of war-rants, and content of the grounds were loaded into ATLAS.ti,which facilitated the identification of patterns. A pattern is afrequent combination of three things: (1) a particular type ofwarrant, (2) a type of ground, and (3) a particular type ofclaim as the result of the first two.

The study of sensemaking involves the way in which indi-viduals make meaning, and tactics for studying sensemakingshould focus more on such patterns of meaning-making ratherthan “frequency counts” (Weick 1995, p. 173) or upon anyquestionable assumptions that argue for human consistency(James 1987; Simon 1956; Weick 1979). As a result, theprocedure did not focus on merely counting the number ofarguments and presenting the empirical results. Instead, thecodes and arguments were examined in search of meaningfulpatterns.

7The process through which qualitative researchers seek to understand therichness of their data can be rather complex. The data analysis procedureactually undertaken in qualitative research is frequently messier thandescribed in finished manuscripts, often for the benefit of the reader (Saldana2009). In the spirit of sharing a more realistic and accurate depiction ofactual coding procedures, an additional cycle of axial coding preceded oursecond cycle described in this manuscript. In this cycle, relationshipsbetween the coding and a set of tensions (i.e., pros and cons related to a claimof value) that appeared inherent in the assessment of the value of virtualworlds were identified. However, strong patterns between the tensions andclaims did not emerge, and the notion of central claims was perceived as aweakness.



Findings

The analysis of the data revealed that claims about the organi-zational value of virtual worlds clustered around one of fourgeneral value categories: (1) current value—claims thatvirtual worlds are currently valuable to organizations;(2) future value—claims that there is potential for value toorganizations; (3) contingent value—claims that any reali-zation of organizational value is contingent on certain factors;and (4) no value—claims that virtual worlds do not offerorganizational value, either due to their very nature or thatany conceivable value would never be able to be realized dueto factors that prohibit adoption within organizations.

The analysis of the data identified 12 patterns8 about theorganizational value of virtual worlds (summarized in Table5). What is referred to as a pattern is a combination of threethings: (1) a particular type of warrant9 and (2) a general typeof ground resulting in (3) a particular type of claim. In theremainder of this section, descriptions and illustrativeexamples of each of the patterns of argumentation, groupedaccording to their claims of organizational value, arepresented.

Current Value

With respect to claims categorized as current value, fourbroad patterns of argumentation claiming that virtual worldsare currently valuable to organizations emerged from the data. These arguments leveraged several warrants types and weresupported by various grounds, including analogy, generali-zation (two different types of grounds), and sign.

Analogy–Conventional: One pattern of argumentationaround claims of current value was based on grounds sug-gesting that aspects of Second Life were similar to conven-tional physical objects and environments, such as billboards,conventions, and retail stores. For example, an informant’s

claim argued that the virtual environment was quite similar tothe physical world in that there were varied options foradvertising media [Informant 9].

Claim [Current value] “One important aspect of SL isthe ability to advertise.”

Ground “Just like in the non-digital counterpart, thiscan be done in various ways: Classic bill-boards…as well sponsored areas, vendingmachines and especially virtual stores.”

Warrant [Analogy] Advertising in Second Life workslike in real life

This pattern argued that because Second Life shared attributeswith conventional offline objects (e.g., billboards), theseobjects would behave like those conventional phenomena in-world. In addition to offline analogies, other arguments werebased on grounds stating that aspects of Second Life weresimilar to other established online technologies. For example,one informant argued for instant messaging (IM) as a parallelcase to virtual worlds, and that it would be of value[Informant 17].

Claim [Current value] “Second Life provides an out-let for members of organizations to communi-cate without having to meet face to face.”

Ground “In today’s economy, especially with risinggas prices, members of businesses need tofrequently interact while keeping costs down. This means communicating without having tomeet face to face. Technologies such as web-cams, instant messaging, and email haveenabled businesses to thrive in the globalenvironment.”

Warrant [Parallel case] Second Life enables communi-cation like webcams, IM, and email.

Overall, claims based on an analogy–conventional argumen-tation pattern were generally positive in their assessment ofthe current organizational value of virtual worlds. This wasparticularly true when the object on which the analogy wasbased successfully demonstrated value to organizations, suchas various conventional advertising media, e-mail, websites,or social networking sites.

Generalization–Organizational Experience: A number ofclaims of current value were based upon generalization typesof warrant. These generalization warrants made use ofexamples related to organizations, their efforts, and relatedoutcomes as the grounds of the arguments. These claimsoftentimes were based on stories of how specific organi-zations received value from Second Life. These storiesleveraged organizations as the key actors (first claim below

8We make no claims as to the exhaustive, or exclusive, nature of thesepatterns; rather, these patterns were recurring, consistent patterns identifiedin the data.

9Claims and grounds represent explicit statements in the data. Warrants arethe interpretation of the coders based on assumptions reflected in a line ofargumentation presented in the data. Note that we are careful not to assertthat the combination of warrants and grounds actually drive the claim, for wedo not have this data. While it may stand to reason, the warrants and groundsmight also be retrospective constructions of a claim that one might wish tomake (Cohen et al. 1972; Weick 1979). However, we argue that theresulting discursive practices can be expected to affect organizationalactivity.



Table 5. Patterns of Argumentation about the Organizational Value of Virtual Worlds

Claim Type of Warrant Type of Ground Example of Organizational Value

Current Value(n = 122)

Analogy Conventional MediaLike retail stores, billboards, etc., Second Lifehas value.

Generalization Organizational ExperienceOrganization successfully meeting or training in-world.

Generalization Personal Experience Organization ads were effective and presencehad impact.

Sign Organizational Exemplars Business press identifying business participation.

Future Value(n = 23)

Authority Expert OpinionOrganizations should evaluate the value of virtualworlds based on authoritative experts.

Causal DigitalizationSynchronous collaboration and training; broadmarketing

Parallel case InternetSecond Life will mirror the good and bad ofInternet, IM, etc.

Contingent Value(n = 39)

Causal Demographic Access to specific populations; narrow marketing

No Value(n = 104)

Analogy GamesJust as games are an unproductive distraction,so is Second Life.

Causal Control Control of activity and technology; pornography;employee time

Generalization Personal Experience Technological problems; pornography

Principle AppropriatenessOrganizations should avoid virtual worlds basedon standards.

[Informant 2]), but also occasionally depicted stories of otherindividuals (second claim below [Informant 11]).

Claim [Current value] “business [value] of SecondLife…a company to hold meetings on theplatform.”

Ground “ArcelorMittal, which is the largest steel com-pany in the world…recently had an investormeeting, and held it concurrently live, inLuxembourg, as well as virtual, on SecondLife.”

Warrant [Generalization] If ArcelorMittal gets valuefrom virtual meetings, other organizations willas well.

Claim [Current value] “[Second Life can] greatlyenhanced distance learning.”

Ground “Ms. Smith (a pseudonym), a Ph.D. candidateof the Harvard Extension School who broughther computer science class to Second Life lastsemester. ‘Things pop up in a less linearfashion than they do in a regular classroom.

Still, even when 10 students chime in, thethreads of a discussion are easy to follow.’”

Warrant [Generalization] Smith’s experience withdistance learning in Second Life is similar toexperiences of others.

Arguments that leveraged secondhand narratives from eitherthe organizational or individual perspectives were over-whelmingly positive about the organizational value of virtualworlds. The core of the warrants suggested that if the subjecthighlighted in the example (e.g., company, individual) cancurrently realize value, then the value realization cangeneralize to other entities as well.

Generalization–Personal Experience: Another commonargumentation pattern involving the generalization warrantrelied upon examples drawn from an informant’s personalexperience with Second Life as the grounds. Worthy of noteis that these experiences were typically viewed from amarketing perspective that offered potential organizationalvalue [Informant 24].



Claim [Current value] “Second Life also offers agreat deal of opportunities for [organizations].”

Ground “I went to the Gotham City Merchants whereI could have become The Joker from The DarkKnight if I wanted, which was a simple yeteffective way to draw attention to the movie. It’s late…and all of a sudden The Joker runsby you go hey, that new Batman movie iscoming out, wonder when, or ‘Hey? Where’dyou get that?’”

Warrant [Generalization] What interests me willinterest others and indicate organizationalopportunity.

With this argumentation pattern, informants describedencounters with in-world advertising or an organization’svirtual location. The detail of the arguments ranged fromstories being quite involved, such as when one informantdescribed his/her “amazing experience” with a virtual car testdrive [Informant 43], to informants simply indicating thatsince they noticed certain ads others are likely to as well. Essentially, informants generalized that if they noticed certainthings or had pleasurable experiences in-world, others userswould also, which would ultimately equate to there beingvalue in these worlds for organizations.

Sign–Organizational Exemplars: Informants dedicated aconsiderable amount of space in their assessments listing thevarious exemplar organizations that have a presence inSecond Life, without necessarily detailing their stories. Rather, the participation of certain companies was seen asevidence in itself of the organizational value of virtual worlds. Technology organizations such as Dell [Informant 41] andIBM were particularly common exemplars, but a variety ofother organizations were also used as evidence, ranging fromBarack Obama’s campaign [Informant 13] to religiousorganizations [Informant 4] to CNN [Informants 38 and 52]. Often the informants just listed organizations as evidence[Informant 28].

Claim [Current value] “Organizations…can leverageSecond Life in a number of ways.”

Ground “An example of this would be how Toyotaallowed residents to test drive its Scion auto-mobile within Second Life…Cisco, AmericanApparel, and Adidas use product placement inthe virtual world to gain more market recog-nition for specific products.”

Warrant [Sign] If example organizations leverageSecond Life in a variety of ways, then organi-zations in general can also do so.

The general sentiment of participants utilizing a sign–organi-zational exemplar pattern of argumentation was that sincereputable companies are investing in Second Life, it must bea signal that virtual worlds are valuable. Unlike when thegeneralization warrant type was employed, these sign–organizational exemplar arguments did not discuss outcomesof the exemplar organization’s efforts; the mere fact of theirparticipation was deemed evidence enough.

Future Value

With respect to claims categorized as future value, three broadpatterns of argumentation claiming there is potential for futureorganizational value emerged from the data. These argumentsleveraged several warrants types, and were supported byvarious grounds, including causal, authority, and parallel.

Authority–Expert Opinion: Resulting largely from theirInternet-based research used to support their argument,informants made claims that built upon the claims of expertsand opinion leaders, predominantly for asserting the futureorganizational value of virtual worlds. These opinion leadersranged Linden Labs (the creator of Second Life) to industryanalysts [Informant 57].

Claim [Future value] “Clearly there is some benefitfrom a business perspective to…evaluateeconomic [viability] within the virtual setting.”

Ground “In a recent Businessweek.com article(November 2006 by Reena Jana) the articlestated that “big brands are increasingly turningto a new demographic to market their goods”[and that] as reported in an article titled “TheComing Second Life Business Cycle” [byMatthew Beller, August 2, 2007 on Mises.org],“Second Life has attracted attention fromWired Magazine, The Economist, and othermedia with stories of a burgeoning economyand entrepreneurs earning their sole incomesby selling virtual goods and services. Ac-cordingly, real-world economists and SecondLife’s residents alike could benefit from acloser look into the actual workings of itseconomy.”

Warrant [Authority] The business press offers soundadvice.

Most commonly, this form of claim was grounded in thebusiness press and argued for the emerging, or future, valueof virtual worlds. In one case, an argument cited business



press that, in turn, cited business press as evidence. What wasevident in the data was that for claims that utilized this formof argumentation, there was not a question of whether therewill be organizational value, but rather when. Additionally,the argumentation for these claims tended to be tentative,suggesting critical evaluation and an experimental attitude,rather than outright advocacy.

Causal–Digitization: Claims utilizing this warrant-groundpattern involved the causal warrant type, with informantsarguing that team meetings, classroom learning, marketresearch, and many other common physical world activitiescan be simply moved over to a virtual world (i.e., digiti-zation). Furthermore, organizational value would take theform of monetary savings. Most typically it was argued thatsavings would occur around activities associated with com-munication or social interaction [Informant 4].

Claim [Future value] “Second Life technology canbe used effectively for training purposes bycorporations who have multiple locations....Efficiencies are gained by the savings in traveltime and travel expense.”

Ground “By using Second life technology, all em-ployees can effectively meet online, providedthat they have an appropriate internet connec-tion and computer.”

Warrant [Cause] If all employees can meet online, thatwill save the organization expense.

While argument for the value of digitizing social interactionand thus saving money was quite common in the data, savingmoney was not the only way informants argued for futurevalue brought about by digitization in the virtual world. Informants also argued that virtual worlds would one dayoffer value to organizations through the digitization ofexperiential marketing initiatives. This marketing focusassociated with causal–digitization arguments includedthemes such as ease of product testing, surveying, andawareness. The concept of awareness was a particularlycommon way to characterize future value [Informant 37].

Claim [Future value] “I think virtual worlds…can bevery beneficial for corporate organizations,nonprofits and educational institutions…brandawareness.”

Ground “A person can experience the brand [campus/training] without having to go to a brick andmortar store.”

Warrant [Cause] If people can experience a brandvirtually, this will increase the organization’sbrand awareness.

In both synchronous collaboration and marketing-orientedarguments that focused on digitization, there was an impliedemphasis on abstract possibilities and the future. Claimssuggested that synchronous collaboration would help, but atsome point in the future as organizations became morefamiliar with virtual worlds. Further it was argued that broad-based marketing would indeed be valuable one day, but atsome point in the future. These claims emphasized futurepotential value, not necessarily current value.

Parallel Case–Internet: The most common form of a parallelcase (i.e., anchoring) argument involved comparing virtualworlds to other Internet-based technologies, or to the Internetitself. Virtual worlds were described as a parallel to theInternet, both being large-scale technological environments. In addition, argument focused on virtual worlds havingadditional functionality that the Internet did not, so virtualworlds were seen essentially as an upgrade of the Internet. Assuch, it was argued that virtual worlds would likely followdiffusion patterns similar to what the Internet experienced. For example, it was suggested that real estate in Second Lifeis parallel to domain names, but richer and more immersive. One informant summed up this perspective in his advocativeclaim encouraging organizations to pursue virtual worlds[Informant 51].

Claim [Future value] “In the longer term no enter-prise could afford to ignore this exciting newdevelopment.”

Ground “It is possible to see parallels with the earlydevelopment of the World Wide Web in themid-1990s.”

Warrant [Parallel case] The development of virtualworlds will be similar to the development ofthe Internet.

When arguments utilized the parallel case–Internet argumen-tation pattern, the claims were advocative (i.e., suggestingwhat should be), encouraging companies to pursue thetechnology of virtual worlds or else be left behind and missout on the organizational value they will provide in the future.

Contingent Value

With respect to claims categorized as contingent value, onebroad pattern of argumentation claiming that any realizationof organizational value is contingent on certain factors waspresent in the data. This argumentation pattern leveraged onetype of warrant: causal.



Causal–Demographic: In this argumentation pattern, it wasargued that whether an organization can gain value fromvirtual worlds was contingent upon whether the organi-zation’s desired audience fits specific demographic criteria. Informants based many of their claims on their assumptionsof the demographic characteristics of participants in virtualworlds. Informants tended to believe, but not necessarilysupport with statistics, that the demographic of users ofSecond Life were younger and often highly technical[Informant 51].

Claim [Contingent value] “Getting access to a smallgroup of creators…is worth the investment forleading-edge companies, large and small.”

Ground “Virtual worlds today are almost totally popu-lated by creators….Many active residents inSecond Life are content creators and gamechangers that have creative ideas to offercompanies.”

Warrant [Cause] Organizational participation in virtualworlds provides valuable access to creativetypes.

Informants perceived this demographic as an exciting oppor-tunity, as it represents a somewhat elusive demographic formarketers to reach effectively. What was particularly ap-parent in the data, as illustrated in this claim, is the argumentthat some (e.g., leading-edge companies) but not all com-panies can benefit from access to the people within SecondLife. For example, organizations that primarily serve marketsthat are characterized as older or more serious—manufac-turers [Informant 2], business-to-business companies [Infor-mant 9], and local companies [Informants 9, 18]—were seenas not desiring to reach this demographic contingency. Otherinformants [e.g., Informant 2] went further, arguing thatvirtual worlds were only useful for those organizationswishing to access specific demographic markets, such asgamers or teens.

In taking the form of causal-demographic arguments, many ofthe claims emphasize there was value to organizations, butthis value was contingent on factors related to the demo-graphics of virtual world users. To a large degree, it wasbelieved the users currently participating in virtual worlds fita particular profile (e.g., younger and technologicallyinclined). While the demographics of virtual worlds representa market that is valuable, the value is only valuable to organi-zations looking to access certain populations.

No Value

The final claim classification identified was no value. Fourbroad patterns of argumentation claiming that virtual worldsdo not offer organizations any value, due to their inherentnature or issues that would lead to non-adoption, emergedfrom the data for this classification. These arguments lever-aged several warrants types, which were supported on variousgrounds, including analogy, causal, generalization, andprinciple.

Analogy–Games: As perhaps might be expected due toSecond Life’s immersive user interface, many of the infor-mants made analogous arguments likening Second Life to agame and classifying Second Life as a form of computer game[Informant 20].

Claim [No value] “I’m not so sure that is the appro-priate environment to bring into a real worldbusiness.”

Ground “To me Second Life is an online, interactivevideo game. It appears to have been designedby gamers for gamers. It has all of the markingof a game in that there are hidden things to do(such as find the kissing parrot) and tasks thatmust be performed (such as dancing the hula infront of an idol).”

Warrant [Parallel case] If Second Life is a game, itdoes not have business value.

Claims of this warrant-ground argumentation pattern largelydismissed the organizational value of virtual worlds from theonset based on the argument that games offered no value toorganizations. Further, in the assessments of some infor-mants, games simply had no place in organizations. BecauseSecond Life was seen as being analogous to a game, it wasstated that it should not be brought into a real world organi-zation. While there were some informants that utilized theanalogy–game argument, but acknowledged that Second Lifewas not solely a game and did pose some potential fororganizational value, they too ultimately claimed value wouldbe minimal, or non-achievable, due to issues related to games. For example, one claim focused on the tendency for game-like technologies to be faddish [Informant 3].

Claim [No value] “Programs like Second Life tend tobe fads.”

Ground “I draw a parallel to online video games, suchas Warcraft, that may see incredible popularityfor a short period of time, say several years,



but traffic will tail off as new technologiesemerge and people lose interest in the specificprogram.”

Warrant [Analogy] Games come and go in popularity,so will SL.

Whether the ground took the stronger analogous form andclassified Second Life as a game, or the weaker form thatdrew analogies to games, the anchoring of virtual worlds tothe phenomenon of any sort of game was simply viewed asdetrimental for organizational value. The core argument ofthose imploring the analogy–game argument type was thatorganizations simply should not waste time in Second Life ifas it is just a game.

Causal–Control: The issue of control was a frequently citedground in the arguments of informants. When issues ofcontrol served as the grounds of the argument, claimsgenerally stated there was no (or minimal) organizationalvalue. The issue of control manifested in two broad ways: (1) control of organizational activity with respect to the virtualworld and (2) control of in-world phenomena. For example,a number of claims focused on an organization’s inability tocontrol its image or to control activity at an in-world function[Informant 11].

Claim [No value] “The anarchistic society of usersmaking content and social structure has nega-tive effects on marketing campaigns and userenjoyment.”

Ground “The atmosphere has an inherently underworldfeeling and could easily be mixed with yourcompany’s culture”; “There are thieves amongthe ‘lifers’ and the anonymity of your second-self makes bad behavior more prevalent”;“Commercial pranks ranging from thrownvirtual feces to a rogue helicopter crashing intothe digital Nissan building have cost avatarlives and made certain brands look twice attheir involvement with Second Life.”

Warrant [Cause] The anarchistic society of usersmaking content and social structure has nega-tive effects on marketing campaigns and userenjoyment.

Further, rampant pornography was often cited as grounds forarguments against the organizational value of virtual worlds.In one situation, a claim focused on liability issues associatedwith a hypothetical causal chain [Informant 3].

Claim [No value] “I do not see a lasting place forSecond Life in mainstream business.”

Ground “The material in the world is far too risqué…the potential for sexual harassment lawsuitsopens on several fronts. If colleagues areoffended by the material, and repeatedly seeimages of smut on a fellow coworker’sscreen....The potential liability of the programis too great to risk.”

Warrant [Cause] If the material is highly risqué, therisk will outweigh any lasting organizationalvalue.

Claims utilizing the causal-control argumentation patternlargely argued that the anonymity and relative freedomafforded to users of Second Life would provide nihilistpranksters with the ability to disrupt events, paint graffiti onwalls, and generally destroy the environment. Due to thislack of control, and the problems it would then cause fororganizations, virtual worlds were claimed to be of no valueto organizations.

Generalization–Personal Experience: Many of the argu-ments involving first-person narratives were quite negativeabout the organizational value of virtual worlds. Thesenegative claims were often supported by narratives describingproblematic interactions with the Second Life technologyitself, including slow response time, crashes, and problemsnavigating, but also described issues with Second Liferesidents. Not only were some avatars reportedly unfriendly,many were quite offensive and led informants to question anyreal value [Informant 47].

Claim [No value] “The security and ethical chal-lenges [that organizations face]…are con-siderable.”

Ground “During my SL journeys, performance crashesplagued my computer and the reality of theharassment crimes that appear on SL Policeblotter became clear when I was harassed by amale avatar who teleported on top of me andviolated me with gestures left best for theimagination. I was offended and taken back bythe experience, feeling violated.”

Warrant [Generalization] My personal experience willbe shared by others and will reflect the chal-lenges associated with the organizational useof virtual worlds.

Essentially, informants using the generalization–personalexperience argumentation form based their arguments on thebelief that their negative personal experience with SecondLife would generalize to other users, and as such virtualworlds are of no value to organizations.



Principle–Appropriateness: Informants occasionally madeclaims that were supported by a moral standard, or imperative,rather than any external evidence. Most often these claimsindicated that virtual worlds should not be seriously enter-tained on grounds of ethics, appropriateness, or practicality. For example, one claim about the future organizational valuewas based on what employees could do in Second Life that isnot appropriate in organizational contexts [Informant 15].

Claim [No value] “There are some hazards asso-ciated with Second Life that may deterbusinesses from using the technology.”

Ground “In the virtual world that Second life offers,people may feel more free to do things orexpress themselves more inappropriately in thevirtual world than they would at work.”

Warrant [Principle] Inappropriate behavior shouldinhibit business adoption.

Claims based on the principle–appropriateness argumentationmade the case for organizational avoidance of Second Life inorder to steer clear of the potentially objectionable aspects,including bad behavior, pornography, and idle activityassociated with Second Life.

Discussion

In the very early stages of a technological innovation, theorganizational value exists nowhere except in rhetoric (Latour1996). Like other early nascent technologies, virtual worldsdo not come ready-made with the necessary institutionalresources for organizations to draw upon (Attewell 1992;Ramiller 2001; Swanson and Ramiller 1997). Further, theseenvironments represent technology that is arguably moreequivocal than other early-stage innovations, since they werenot created for organizational purposes at all, and any busi-ness applications were little more than an afterthought (Ivesand Junglas 2008). Virtual worlds do not present themselvesto organizations complete with well-packaged value propo-sitions and vendors seeking to justify their relevance. Instead,a handful of organizational actors proactively explore suchworlds to determine how they might appropriate this tech-nology for organizational gain.

Weick (1995) contends that all sensemaking is individual, butit is socially applied in that individuals make projectionsabout their identities in the context of social groups ororganizations. Sensemaking—even if it is the work of asingle person reflecting alone—is necessarily social (Billig1996; Dodds et al. 1997). As individuals make sense of any

novel situation, such as that posed by an innovative tech-nology, they draw upon preexisting organizational andinstitutional resources to form the relevant discursive prac-tices. These discursive practices, in turn, shape the contoursof emergent organizations (Boden 1994; Maitlis 2005; Taylorand Van Every 2000), and eventually become reproducedacross fields, thereby reproducing the discourse concerningthose innovative technologies (Boczkowski and Orlikowski2004; Heracleous and Barrett 2001). Individuals draw uponand elaborate these discourses, thus continually reinforcingthem while simultaneously engendering new institutionalstructures and practices (Phillips et al. 2004). Institutionali-zation is not an outcome, but a process (Powell and DiMaggio1991), and IT innovations are continuously undergoing aprocess of institutionalization throughout their adaptation inorganizational contexts (Ramiller 2001). However, as thepatterns of action become institutionalized, the initialarguments justifying them disappear as the claims becometaken-for-granted (Green et al. 2009).

The study of virtual worlds at this juncture provides a uniqueopportunity to study the very early stages of sensemaking,organizing, and institutionalization with respect to this noveland sweeping technological genre, and affords an analysis ofthe reasoning that will contribute to subsequent path–dependent outcomes. In this study we looked at the patternsof sensemaking individuals exhibit when reflecting upon theorganizational value of virtual worlds, identifying 12 patternsin the arguments put forth by our informants and eliciting fourbroad themes of arguments. Next we will present thesethemes, followed by a reflection on the methodological impli-cations of using Toulmin’s model as a methodological device.

Themes Across Patterns of Sensemaking

Informants in this study were engaged in perspective making(Boland and Tenkasi 1995) through the production of texts inan effort to arrive at explanations of the organizational valueof virtual worlds. The assessments they generated docu-mented sensemaking by individuals as they made ideasexplicit in text, and developed their own understandings incombination with this text (Weick 1979). By focusing ononly thematized claims (Habermas 1981), or those claims thatinformants explicitly supported with grounds (i.e., evidence),we were able to isolate and address only those claims thatinformants considered in a fundamental way (i.e., not thoseaddressed superficially) and that were likely to be importantand relevant to the individuals.

The 12 patterns of argument associated with these thematizedclaims represent a primary contribution of this research. These



Table 6. Themes of Argumentation on Organizational Value of Virtual Worlds

Key Themes Claims Warrants Grounds

Confirmation

Current value Generalization Single organizational experience

Current value Generalization Personal experience with organization

Current value Sign Organizational exemplars

Demographics Contingent value Causal Demographic

Control

No value Causal Control

No value Generalization Personal Experience with Second Life

No value Principle Appropriateness

Open Ended

Current value Analogy Conventional media

Future value Causal Digitalization

Future value Authority Opinion leader

Future value Parallel case Internet

No value Analogy Games

patterns offer insight not only for the champions (andopponents) of virtual world technology in organizations asthey look to influence others, but also for those tasked withassessing the rhetorical statements of others. This researchidentifies patterns that illustrate the way that certain forms ofargument are brought to bear in arriving at specific con-clusions. The ability to identify and interpret these patternscan provide managers with additional guidance whenassessing new technologies. Furthermore, our patterns ofargumentation represent a set of potentially testable relation-ships that can guide future research in the area of virtualworlds, as well as the exploration of organizational value forother equivocal technologies.

In addition to identifying the patterns of argumentation withinour data, we elicited four broad themes that ran acrossmultiple patterns (see Table 6).

1. Positive assessments of the current value of virtualworlds contained elements of confirmation bias.

2. Contingent value assessments consistently involvedissues associated with demographic assumptions.

3. For no value assessments, the salient themes focused onissues of control, both in-world and in real-worldcontexts.

4. The rhetorical choices associated with the future valueassessments of virtual worlds were open-ended.

Next we discuss each of these in turn.

Confirmation

Based on the analysis, claims of current value of virtualworlds were often made through some form of modeling uponexisting exemplars. Modeling is a common mechanism bywhich individuals make sense of uncertain situations (Cyertand March 1963), such as equivocal technologies. Thismodeling takes the form of success stories or simply uses thepresence of well-known organizations as success signals(Feldman and March 1981). Modeling is the root of thenotion of mimetic isomorphism, or the tendency of individualsto imitate the practices of other individuals withoutnecessarily subjecting these practices to rational scrutiny(DiMaggio and Powell 1983), which is foundational to thecontemporary institutional tradition.

A noteworthy observation associated with modeling, par-ticularly as it relates to the claims of current value of virtualworlds, involves the possibility of confirmation bias.Confirmation bias describes a non-intentional, “unwittingselectivity” of only positive evidence as grounds for a certainclaim (Nickerson 1998). It is one of the most commonlydocumented issues associated with the human condition andholds a long-lasting tradition in both psychological andphilosophical traditions (Nickerson 1998). Therefore, itshould not be surprising to find evidence of confirmation biasin our data.

However, we must be careful to avoid regarding the selectionof only positive exemplars as an unintended bias. Rather thancausing informant interpretations, the positive models couldrepresent ex post justification or rationalizations of opinions



about the current value of virtual worlds. Such retrospectivesensemaking is altogether consistent with Weickian socialpsychology (Weick 1979). Either way, all of the patterns weidentified claiming current value in virtual worlds emphasizedpositive cases as grounds. For instance, claims that used the“sign” warrant listed companies in Second Life and thenclaimed value based on the presence of these exemplars. Following the same reasoning, one could list many moreorganizations not in Second Life and therefore make anequally compelling, or perhaps more compelling, claim usingthe same logic. Similarly those who generalized from theexperience of a single organization, or their own personalexperience with an organization within Second Life, oftenargued that as the experience was positive, there must bevalue. This is particularly interesting, since many of theinformants offered quite negative experiences, but these wereoften as caveats and qualifiers to their argument. Inter-estingly, these negative models or examples were rarely usedfor dismissing the current organizational value of virtualworlds. Oftentimes, however, they did result in a contingentassessment of the value of virtual worlds—based on certaindemographic characteristics.

Demographics

When informants chose to focus on their assumptions aboutthe demographic characteristics of participants (i.e., avatars)in Second Life, they tended to conclude with some form ofcontingent explanation of the organizational value of virtualworlds. The informants suggested that virtual worlds will bevaluable for companies seeking to market to a specific type ofindividual or to recruit a specific type of individual. The typeof individual that informants described largely reflected thestereotypes of young, anti-social “techies.” Thus, it wassuggested that organizations could benefit if they were high-tech firms, and the value was limited for firms with olderworkforces.

While many of these claims are certainly defensible, a focuson demographic concerns draws attention away from moregeneral potential organizational applications. When infor-mants focus on marketing demographics, they may beinadvertently discounting the potential for virtual teamcollaboration (Davis et al. 2009). Similarly, arguments thatfocus on the “old industry” context of a particular organiza-tion may ignore possible opportunities to innovate andreinvent aspects of their organizations. Demographic-basedarguments focused the informants’ attention on what is,according to their assumptions, and not on the possibilities ofwhat can be. Many of the negative assessments of organi-zational value also emphasized present limitations rather than

long-term opportunities, and a theme that ran through manyof these critical arguments involved issues of control.

Control

While the warrants employed vary, the concept of controlunderlies several of the discursive patterns that we have out-lined. Indeed, informants perceived the absence of sufficientcontrol over virtual worlds to be a primary impediment toorganizational adoption. This is quite consistent with findingsfrom other areas of IS research where the management oftraditional organization-based systems (i.e., those that havebeen planned, implemented, and maintained within organi-zations) has been aimed at the ability to clearly define systemcapabilities and control users (Kirsch 1997). However, virtualworlds are unique systems, as they have been designed toallow users the freedom to express themselves, which makesthe limitation of capabilities, uses, and issues related to theseequivocal technologies challenging. Further, issues withthrown feces and crashing helicopters do not involve thecontrol of employees by organizational systems. Rather, theyare simply inappropriate for business applications—suchbehavior is not “natural, rightful, expected, and legitimate”(March and Olsen 2004, p. 3) in an organizational context. The logic of appropriateness that businesspeople draw uponto express issues with control reflect the relatively stable setof rules and resources that they collectively draw upon topattern their behavior, and these actors will habitually workto reinforce this logic (Giddens 1984; March and Olsen 2004).Therefore, it is clear that for organizations to see significantbenefit in virtual worlds, it will likely make more sense tocontinue to portion off and segment organizational elementsof virtual worlds from the general population. Although issuesof control were described by virtually all of the informants,many chose not to root their assessments in these issues, butin others.

Open-Ended

Individuals make choices about the way they frame andanchor their arguments. The issue of framing was particularlyevident in the arguments that employ warrants of analogy andparallel case, “anchoring” (Moscovici 1984) the novel envi-ronment of virtual worlds to an existing phenomenon. Rhetorical choices of the informant were particularly salientto claims about the future value of virtual worlds, but notrestricted to that perspective. For instance, when the analogyto a game was employed, claims of value are consistentlynegative. The informants in this study clearly conceive ofgames as trivial, unproductive, and not deserving of serious



business consideration. However, not all anchoring effortsrender a negative result. When the virtual world is viewed asanalogous to a virtual version of a prevailing business prac-tice, perceptions of value are positive. Similarly, informantshave a favorable perception of virtual worlds when theyunderstand them as similar to the World Wide Web, in termsof their potential for adoption and growth.

It is important to note that this anchoring is not a randomchoice, but is inexorably linked to the informant’s view ofself, history, and personal and social experiences (Moscovici1984). Sensemaking in general is largely about continuity; itis a process aimed at creating and maintaining a consistent,positive sense of self vis-à-vis the phenomenon or cuesencountered (Weick 1995). Therefore the way in which anindividual anchors a new technology not only offers insightinto the referent history and social context, but may alsobecome rigid, or objectified, and thus impact the futureappropriation significantly (Gal and Berente 2008). Forexample, if social actors are uncomfortable with video gamesand three-dimensional graphics label virtual worlds as “videogames,” this would have significantly different implicationsthan if they labeled them as electronic “billboards.”

Arguing Across Sensemaking Traditions

Humans do not act like the omniscient rational actors inclassical models of homo economicus (Henrich et al. 2001). Rather, individual rationality is necessarily bounded by thelimits on cognitive ability, access to information, attention,and varying preferences (Simon 1956). Various modes ofsensemaking have been discussed at length in a variety ofcontexts across multiple domains and disciplines, and in thisstudy we provide findings that offer insight for the study ofmodes of rationality involved with sensemaking about theorganizational value of virtual worlds. Functional, rationalexplanations dominate much of the discussion around tech-nological dynamics in organizations (e.g., Chandler 1977;Rogers 1995; Yates 1989), and we found such logic in causalarguments of the organizational value of virtual worlds (e.g.,demographics will drive the applicability of virtual worlds;lack of control will limit an organization’s willingness topursue virtual worlds). However, there are a variety of othertraditions in social psychology and organizational researchthat characterize the process of human sensemaking. It wasnot our intention to tackle this topic from the social represen-tations perspective, but this perspective was apparent throughthe warrants of analogy and parallel case used in argumentsby our informants. We also identified a number of othertheoretical traditions that could be implied by other warrants,and came to the conclusion that our methodology and use of

Toulmin’s model enabled us to get at sensemaking whileaccommodating a diverse set of sensemaking perspectives.

One set of such traditions includes socio-cognitive perspec-tives that focus on particular notions of cognitive frames orsocial representations (Moscovici 1984; Orlikowski and Gash1994; Pinch and Bijker 1987; Ramiller 2001). From theseperspectives, individuals come to terms with novel phenom-ena by relating, or anchoring, them to familiar phenomena. In our study, this anchoring process was evident in analogy-based modes of argumentation (e.g., virtual worlds willdevelop like the Internet, or like video games). Anotherperspective on human sensemaking can be described as anarrative lens that highlights the power of stories in makingsense of ambiguous situations (Boje 1991; Boland 2001;Bruner 1990; Pentland 1999). This narrative approach isdirectly related to generalization arguments in our study. Bygeneralizing from particular experiences, informants narrateexamples of organizational value (e.g., virtual worlds arevaluable because I had a really great experience with an in-world organization). Finally, institutional theory is often heldup as an alternative to the rational perspective (Powell andDiMaggio 1991) as it emphasizes taken-for-granted assump-tions and highlights forces associated with legitimacy and therelevance of cultural embeddedness in sensemaking(DiMaggio and Powell 1983; Granovetter 1985; Uzzi 1997).In their seminal work, DiMaggio and Powell (1983) indicatethat there are three types of legitimizing institutional forcesthat act in lieu of purely “rational” action, which they describeas normative, coercive, and mimetic. Normative and coerciveforces are essentially referring to propriety, that is, appro-priate behaviors often encouraged by authorities in a givensocial context. Mimetic forces encourage organizationalactors to follow exemplars as a way of taking legitimateaction (DiMaggio and Powell 1983).

Toulmin (2003) indicated that the type of argument is deter-mined fundamentally by the warrant brought to bear. Aspresented in Table 7, the forms of sensemaking-orientedtraditions match well with the warrants we identified in thereflections of our informants. While we make no claims aboutthe exhaustiveness of our study, it is apparent that differentmodes of cognition can be observed within a single assess-ment and across assessments.

Based on our account, we assert that the Toulminian lensoffers a methodological tool that is theoretically neutral, or“agnostic” (Nord and Connell 1993), in that it avoids com-mitting to a particular tradition about human sensemaking exante. Since, as we assert, different combinations of warrantsand grounds have a relationship to the nature of the claimsthemselves, this leads to an important implication to thepractice for researchers: if an investigator were to limit the



Table 7. Types of Arguments and Sensemaking Traditions

Warrant/Type ofArgument Sensemaking Traditions Theoretical Foundations Selected Citations

CausalRational: sensemakingthrough mechanisticprocesses

• Functionalist theory• Information processing• Contingency theory

Chandler 1977; Galbraith 1977;Lawrence and Lorsch 1967; Rogers1995; Yates 1989

Analogy/ParallelCase

Anchored: sensemakingrelation to other experiencedphenomena

• Frames of reference• Social representations

Gal and Berente 2008; Moscovici1984; Pinch and Bijker 1987;Orlikowski and Gash 1994; Starbuckand Millikin 1988

GeneralizationNarrative: sensemakingthrough storytelling

• Narrative cognition• Storytelling

Boje 1991; Boland 2001; Boland andTenkasi 1995; Bruner 1990;Pentland 1999

SignMimetic: sensemakingthrough imitation • Institutionally embedded

perspectives• Cultural norms

Cyert and March 1963; DiMaggioand Powell 1983; Feldman andMarch 1981; Grannovetter 1985;Greenwood et al. 2008; March andOlsen 2004; Meyer and Rowan1977; Uzzi 1996;

Authority/PrincipleNormative: sensemakingbased on standards

mode of inquiry to a particular set of assumptions abouthuman understanding, and use a particular lens, that investi-gator may only capture a portion of the forms of sensemaking. For example, a focus on narrative (i.e., generalizationwarrant) as a form of sensemaking would focus on theexperiences of organizations, and likely yield some interestinginsight; but this would likely result in only positive,confirming conclusions about the organizational value ofvirtual worlds. The Toulmin model allows researchers toaddress multiple sensemaking perspectives.

Toulmin Model as Methodological Tool

Individuals rarely argue from fully formed, valid, and inter-nally consistent positions, but rather from a smattering ofperceptions, experiences, and motivations. Such a pragmaticform of reasoning allows for a variety of methods in sup-porting claims. In the data, we found a set of claims based ona variety of arguments of different forms and different con-tent. The structure of practical reasoning (Toulmin’s claim–ground–warrant) provided us with a tool whereby we couldrigorously capture and code the sensemaking activity, withoutstripping it of its richness and without fitting the results intopreformed assumptions about human rationality. Avoidingsuch assumptions and characterizing different forms of argu-ments proved to be one of the key contributions of the study.

Various adaptations of the Toulmin model have been used inthe IS discipline, although not as an analytical device tounderstand the different forms of human sensemaking. Instead, the Toulmin model is often used to better understandthe impact of explanations on issues such as trust andacceptance technology (Kim and Benbasat 2006; Ye andJohnson 1995), and how to support the design of explanationsin various IS domains, including expert systems, knowledgemanagement, and e-commerce (Gregor and Benbasat 1999;Kim and Benbasat 2006; Ye and Johnson 1995). Further,various commentators have used the Toulmin model as a wayto frame IS-related arguments and navigate different IS-related perspectives (King and Lyytinen 2004; Klein andHirschheim 2001). In this research we go beyond prior workin the discipline and use Toulmin’s model as a methodo-logical device for assessing patterns in practical discourse(Fairclough 2003). In doing so, we find that this perspectiveenables us to capture dynamics of sensemaking across avariety of intellectual traditions.

It has been established elsewhere that discursive practicesignificantly impacts the adoption of information technologies(Boczkowski and Orlikowski 2004; Heracleous and Barrett2001; Swanson and Ramiller 1997). Toulmin’s model offersa way to capture IT-related discourses and link these discur-sive forms to claims about the technology. This is a powerfultool for use in the early stages of a technology where no



single discourse has yet been institutionalized (Green et al.2009) or “black boxed” (Latour 1996). The discourses remainfluid enough to be readily influenced.

Limitations and Future Research

Although we present implications for IS research and practice,these implications must be qualified by a number of limita-tions of this study. First, the source of our data can be viewedas a limitation because informants were students completinga course assignment. Student data is never ideal; however,there are some practices that help validate the use of studentinformants in certain situations, which we addressed in themethodology section. Further, the study was an assignmentwithin a course, and as such may not reflect the context thisstudy seeks to generalize (i.e., the assessments do not repre-sent sensemaking of the informants in the capacity of businessprofessionals). In order to address the future role of anemerging technology and how individuals go about makingsense of the technology and its potential for organizations, weneed to have access to a group of individuals with businessexperience that have engaged the technology before thetechnology becomes widespread and used by every firm. Fortunately, to gather appropriate data, we found a uniqueopportunity involving management classes that allowed us tofind informants that (1) are real businesspersons, (2) haveengaged the technology, and (3) do not work for organizationsthat have employed the technology previously. For thesereasons, we believe the informants’ assessments provideinsights into individual sensemaking of equivocal tech-nologies and their potential for organizations. However, theseassessments may not reflect the arguments the informantsmight make in organizational contexts where they are heldaccountable for the content. Future research should seek tounderstand similar dynamics within organizational environ-ments, and it would be interesting to incorporate socialnetworks as well.

A related limitation involves the framing of both the assign-ment and this manuscript. We asked students to assess theorganizational value of virtual worlds in general and requiredsome experience with Second Life as an example of a virtualworld. Instead of focusing solely upon virtual worlds ingeneral, on a number of occasions informants conflatedSecond Life and the broader domain of virtual worlds. Thisis perhaps an artifact of the novelty of the environment tomany of them, but still may materially affect their claims.Further, since there is an ever-increasing variety of virtualworlds continually emerging, and since we cannot possiblyforesee all future virtual worlds, future research must

acknowledge this variety and attend to the potential organiza-tional value of differentiated virtual environments.

An area where the data was quite rich was that of individualreflections upon their identities in relation to the virtual world. Identity formulation is fundamental to studies of individualsensemaking (Moscovici 1984; Weick 1995), and yet in thispaper we looked solely at the argument structure in relation tothe nature of the claim. Since informants made numerous,often inconsistent claims, we concluded that we could notproperly link a given assessment of identity with any or allarguments, since the reflections on identity were typicallyremoved from the documents on discussions of the organiza-tional value of virtual worlds. Attention to issues relating toidentity would be in order for future studies.

Conclusion

The emergence of virtual worlds offers an opportunity tostudy the sensemaking—and resulting appropriation andevolution—of an entirely open-ended technology. At thispoint, the organizational value of this technology resideslargely within the discourse, but this discourse will contributeto the shaping of future reality. Our findings suggest thatpatterns of argumentation can be consistent with particularperspectives of novel phenomena, while they also illustratethe diversity of such patterns that individuals bring to bear. From rational argumentation around conceptual capabilitiesto the metaphorical association with earlier waves of inno-vation, individuals draw upon a rich tapestry of sensemakingstrategies to confront the equivocality that they encounter.

Acknowledgments

The authors would like to thank Dorothy Leidner, Robin Teigland,Cathy Urquhart, and the anonymous reviewers for their invaluableguidance in crafting this paper. We are also grateful to Brian Butler,John King, and Kalle Lyytinen for their insight throughout theprocess.

References

Anderson, C. 2006. “Wired Travel Guide: Second Life,” WIREDMagazine (available online at http://www.wired.com/wired/archive/14.10/sloverview.html).

Attewell, P. 1992. “Technology Diffusion and OrganizationalLearning: The Case of Business Computing,” OrganizationScience (3:1), pp. 1-19.



Billig, M. 1996. Arguing and Thinking: A Rhetorical Approach toSocial Psychology, Cambridge, UK: Cambridge UniversityPress.

Billig, M. 1997. “The Dialogic Unconscious: Psychoanalysis,Discursive Psychology and the Nature of Repression,” BritishJournal of Social Psychology (36:2), pp. 139-159.

Boczkowski, P. J., and Orlikowski, W. J. 2004. “OrganizationalDiscourse and New Media: A Practice Perspective,” in The SageHandbook of Organizational Discourse, D. Grant, C. Hardy, C.Oswick, and L. L. Putnam (eds.). Thousand Oaks, CA: SagePublications, pp. 360-377.

Boden, D. 1994. The Business of Talk, Cambridge UK: PolityPress.

Boje, D. M. 1991. “The Storytelling Organization: A Study ofStory Performance in an Office- Supply Firm,” AdministrativeScience Quarterly (36:1), pp. 106-126.

Boland, R. J. 2001. “The Tyranny of Space in OrganizationalAnalysis,” Information and Organization (11:1), pp. 3-23.

Boland, R. J., and Tenkasi, R. V. 1995. “Perspective Making andPerspective Taking in Communities of Knowing,” OrganizationScience (6:4), pp. 350-372.

Boland, R. J., Tenkasi, R. V., and Te'eni, D. 1994. “DesigningInformation Technology to Support Distributed Cognition,”Organization Science (5:3), pp. 456-477.

Boulos, M. N. K., Hetherington, L., and Wheeler, S. 2007. “SecondLife: An Overview of the Potential of 3-D Virtual Worlds inMedical and Health Education,” Health Information & LibrariesJournal (24:4), pp. 233-245.

Boyatzis, R. E. 1998. Transforming Qualitative Information: Thematic Analysis and Code Development, Thousand Oaks, CA: Sage Publications.

Brockriede, W., and Ehninger, D. 1960. “Toulmin on Argument: An Interpretation and Application,” Quarterly Journal of Speech(46), pp. 44-53.

Bruner, J. S. 1990. Acts of Meaning, Cambridge, MA: HarvardUniversity Press.

Bryant, A., Hughes, J., Myers, M. D., Trauth, E. M., and Urquhart,C. 2004. “Twenty Years of Applying Grounded Theory in Infor-mation Systems: A Coding Method, Useful Theory GenerationMethod, or an Orthodox Positivist Method of Data Analysis?,”in Information Systems Research: Relevant Theory and In-formed Practice, B. Kaplan, D. P. Truex, D. Wastell, A. T.Wood-Harper and J. I. DeGross (eds.), Norwell, MA: KluwerAcademic Publishers, pp. 649-650.

Chandler, A. D. 1977. The Visible Hand: The Managerial Revolu-tion in American Business, Cambridge, MA: Belknap Press.

Cohen, M. D., March, J. G., and Olsen, J. P. 1972. “A Garbage CanModel of Organizational Choice,” Administrative ScienceQuarterly, (17:1), pp. 1-25.

Corbin, J. M., and Strauss, A. L. 1990. “Grounded TheoryResearch: Procedures, Canons, and Evaluative Criteria,” Quali-tative Sociology (13:1), pp. 3-21.

Cyert, R. M., and March, J. G. 1963. A Behavioral Theory of theFirm, Englewood Cliffs, NJ: Prentice Hall.

Czarniawska, B. 2004. “Turning to Discourse,” in The Sage Hand-book of Organizational Discourse, D. Grant, C. Hardy,C. Oswick, and L. L. Putnam (eds.). Thousand Oaks, CA: SagePublications, pp. 399-404.

Daft, R. L., Lengel, R. H., and Trevino, L. K. 1987. “MessageEquivocality, Media Selection, and Manager Performance: Implications for Information Systems,” MIS Quarterly (11:3), pp.355-366

Davis, A., Khazanchi, D., Murphy, J., Zigurs, I., and Owens, D.2009. “Avatars, People, and Virtual Worlds: Foundations forResearch in Metaverses,” Journal of the Association forInformation Systems (10:2), pp. 90-117.

DiMaggio, P. J., and Powell, W. W. 1983. “The Iron CageRevisited: Institutional Isomorphism and Collective Rationalityin Organizational Fields,” American Sociological Review (48:2),pp. 147-160.

Dodds, A. E., Lawrence, J. A., and Valsiner, J. 1997. “ThePersonal and the Social: Mead’s Theory of the ‘GeneralizedOther,’” Theory & Psychology (7:4), pp. 483-503.

Drazin, R., Glynn, M. A., and Kazanjian, R. K. 1999. “MultilevelTheorizing About Creativity in Organizations: A SensemakingPerspective,” Academy of Management Review (24:2), pp.286-307.

Edwards, D., and Potter, J. 1992. Discursive Psychology, London:Sage Publications Ltd.

Erickson, K., and Stull, D. 1998. Doing Team Ethnography: Warnings and Advice, Thousand Oaks, CA: Sage Publications.

Fairclough, N. 2003. Analyzing Discourse: Textual Analysis forSocial Research, London: Routledge:

Feldman, M. S., and March, J. G. 1981. “Information in Organiza-tions as Signal and Symbol,” Administrative Science Quarterly(26:2), pp. 171-186.

Gal, U., and Berente, N. 2008. “A Social Representations Perspec-tive on Information Systems Implementation: Rethinking theConcept of ‘Frames,’” Information Technology and People(21:2), pp. 133-154.

Galbraith, J. R. 1977. Organization Design, Reading, MA: Addison Wesley.

Galimberti, C., Ignazi, S., Vercesi, P., and Riva, G. 2001. “Com-munication and Cooperation in Networked Environments: AnExperimental Analysis,” CyberPsychology & Behavior (4:1), pp.131-146.

Gartner. 2008. “Gartner Says 90 Per Cent of Corporate VirtualWorld Projects Fail Within 18 Months,” Gartner Group, Stamford, CT (available online at http://www.gartner.com/it/page.jsp?id=670507).

Giddens, A. 1984. The Constitution of Society: Outline of theTheory of Structuration, Cambridge, UK: Polity Press.

Gioia, D. A., and Chittipeddi, K. 1991. “Sensemaking and Sense-giving in Strategic Change Initiation,” Strategic ManagementJournal (12:6), pp. 433-448.

Gioia, D. A., and Thomas, J. B. 1996. “Identity, Image, and IssueInterpretation: Sensemaking During Strategic Change in Acade-mia,” Administrative Science Quarterly (41:3), pp. 370-403.



Glazer, B. G., and Strauss, A. L. 1967. The Discovery of GroundedTheory: Strategies for Qualitative Research, New York: Aldine.

Gordon, M. E., Slade, L. A., and Schmitt, N. 1986. “The ‘Scienceof the Sophomore’ Revisited: From Conjecture to Empiricism,”Academy of Management Review (11:1), pp. 191-207.

Granovetter, M. 1985. “Economic Action and Social Structure: The Problem of Embeddedness,” American Journal of Sociology(91:3), pp. 481-510.

Grant, D., Hardy, C., Oswick, C., and Putnam, L. L. 2004.“Organizational Discourse: Exploring the Field,” in The SageHandbook of Organizational Discourse, D. Grant, C. Hardy,C. Oswick, and L. L. Putnam (eds.), Thousand Oaks, CA: SagePublications, pp. 1-36.

Green, S. E., Li, Y., and Nohria, N. 2009. “Suspended in Self-SpunWebs of Significance: A Rhetorical Model of Institutionalizationand Institutionally Embedded Agency,” Academy of ManagementJournal (52:1), pp. 11-36.

Greenwood, R., Oliver, C., Shalin, K., and Suddaby, R. 2008. TheSage Handbook of Organizational Institutionalism, ThousandOaks, CA: Sage Publications.

Gregor, S., and Benbasat, I. 1999. “Explanations from IntelligentSystems: Theoretical Foundations and Implications for Practice,”MIS Quarterly (23:4), pp. 497-530.

Griffith, T. L. 1999. “Technology Features as Triggers for Sense-making,” Academy of Management Review (24:3), pp. 472-488.

Guest, G., and MacQueen, K. M. 2008. Handbook for Team-Based Qualitative Research, Lanham, MD: AltaMira Press.

Habermas, J. 1981. The Theory of Communicative Action, VolumeOne: Reason and the Rationalization of Society, Boston: BeaconPress.

Hansen, S., Berente, N., Pike, J., and Bateman, P. J. 2009.“Productivity and Play in Organizations: Executive Perspectiveson the Real-World Organizational Value of Immersive VirtualEnvironments,” Artifact (2:3), pp. 69-81.

Harry, B., Sturges, K. M., and Klingner, J. K. 2005. “Mapping theProcess: An Exemplar of Process and Challenge in GroundedTheory Analysis,” Educational Researcher (34:2), pp. 3-13.

Hemp, P. 2006. “Avatar-Based Marketing,” Harvard BusinessReview (84:6), pp. 48-57.

Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H.,and McElreath, R. 2001. “In Search of Homo Economicus: Behavioral Experiments in 15 Small-Scale Societies,” AmericanEconomic Review (91:2), pp. 73-78.

Heracleous, L. T. 2004. “Interpretivist Approaches to Organi-zational Discourse,” in The Sage Handbook of OrganizationalDiscourse, D. Grant, C. Hardy, C. Oswick, and L. L. Putnam(eds.), Thousand Oaks, CA: Sage Publications, pp. 175-192.

Heracleous, L. T., and Barrett, M. 2001. “Organizational Changeas Discourse: Communicative Actions and Deep Structures inthe Context of Information Technology Implementation,”Academy of Management Journal (44:4), pp. 755-778.

Hobson, N. 2006. “Is it Time You Got a Second Life?,” StrategicCommunication Management (11:1), p. 3.

Hof, R. D. 2006. “My Virtual Life,” Business Week, May 1(available online at http://www.businessweek.com/magazine/content/06_18/b3982001.htm).

Howcroft, D., and Hughes, J. 1999. “Grounded Theory: I Men-tioned it Once But I Think I Got Away with It,” in InformationSystems— The Next Generation: Proceedings of the 4th UKAcademy for Information Systems Conference, L. Brooks and C.Kimble (eds.), Maidenhead, UK: McGraw-Hill, pp. 129-141.

Huang, X., Gattiker, T. F., and Schwarz, J. L. 2008. “Can StudentsPass the Test: Should We Use Student Subjects in OperationsManagement Research,” in Proceedings of the 39th AnnualMeeting of the Decision Sciences Institute, Baltimore, MD,November 22-25, pp. 4931-4936 (available online at http://www.ecisionsciences.org/Proceedings/DSI2008/docs/493-3475.pdf)

Ives, B., and Junglas, I. 2008. “APC Forum: Business Implicationsof Virtual Worlds and Serious Gaming,” MIS QuarterlyExecutive (7:3), pp. 151-156.

James, W. 1987. Pragmatism, Albany, NY: Library of America.Johnstone, B. 2002. Discourse Analysis, Malden, MA: Blackwell

Publishing.Khera, I. P., and Benson, J. D. 1970. “Are Students Really Poor

Substitutes for Businessmen in Behavioral Research?,” Journalof Marketing Research (7:4), pp. 529-532.

Kim, D., and Benbasat, I. 2006. “The Effects of Trust-AssuringArguments on Consumer Trust in Internet Stores: Application ofToulmin’s Model of Argumentation,” Information SystemsResearch (17:3), p. 286.

King, J. L., and Lyytinen, K. 2004. “Reach and Grasp,” MISQuarterly (28:4), pp. 539-551.

Kirsch, L. S. 1997. “Portfolios of Control Modes and IS ProjectManagement,” Information Systems Research (8:3), pp. 215-239.

Klein, H. K., and Hirschheim, R. 2001. “Choosing between Com-peting Design Ideals in Information Systems Development,”Information Systems Frontiers (3:1), pp. 75-90.

Kling, R. 1980. “Social Analyses of Computing: TheoreticalPerspectives in Recent Empirical Research,” ACM ComputingSurveys (12:1), pp. 61-110.

Kock, C. 2006. “Multiple Warrants in Practical Reasoning,” inArguing on the Toulmin Model: New Essays in ArgumentAnalysis and Evaluation, D. Hitchcock and B. Verheij (eds.),Heidelberg: Springer, pp. 247-259.

Lawrence, P. R., and Lorsch, J. W. 1967. Organization andEnvironment: Managing Differentiation and Integration, Boston: Harvard Business School Press.

Latour, B. 1996. Aramis, or the Love of Technology, Cambridge,MA: Harvard University Press.

Layder, D. 1998. Sociological Practice: Linking Theory andResearch, London: Sage Publications Ltd.

Louis, M. R. 1980. “Surprise and Sense Making: WhatNewcomers Experience in Entering Unfamiliar OrganizationalSettings,” Administrative Science Quarterly (25:2), pp. 226-251.

Lyytinen, K., and Rose, G. M. 2003. “The Disruptive Nature ofInformation Technology Innovations: The Case of Internet Com-puting in Systems Development Organizations,” MIS Quarterly(27:4), pp. 557-596.

Maitlis, S. 2005. “The Social Processes of Organizational Sense-making,” Academy of Management Journal (48:1), pp. 21-49.

March, J. G., and Olsen, J. P. 2004. “The Logic of Appro-priateness,” ARENA Working Papers, WP 04/09, Centre for



European Studies, University of Oslo (available online athttp://www.arena.uio.no/publications/wp04_9.pdf).

Mennecke, B. E., Roche, E. M., Bray, D. A., Konsynski, B.,Lester, J., Rowe, M., and Townsend, A. M. 2007. “Second Lifeand Other Virtual Worlds: A Roadmap for Research,” inProceeedings of the 28th International Conference on InformationSystems, Montreal, Canada, December 9-12.

Meyer, J., and Rowan, B. 197. “Institutionalized Organizations: Formal Structure as Myth and Ceremony,” The American Journalof Sociology (83:2), pp. 340-363.

Morse, J. M. 2007. “Sampling in Grounded Theory,” in The SageHandbook of Grounded Theory, A. Bryant and K. Charmaz(eds.), London: Sage Publications, Ltd. pp. 229-244.

Moscovici, S. 1984. “The Phenomenon of Social Representations,”in Social Representations, R. M. Farr and S. Moscovici (eds.),Cambridge, UK: Cambridge University Press, pp. 3-69.

Nickerson, R. S. 1998. “Confirmation Bias: A UbiquitousPhenomenon in Many Guises,” Review of General Psychology(2:2), pp. 175-220.

Noam, E. M. 2007. “The Dismal Economics of Virtual Worlds,”The DATA BASE for Advances in Information Systems (38:4), pp.106-109.

Nord, W. R., and Connell, A. F. 1993. “From Quicksand toCrossroads: An Agnostic Perspective on Conversation,” Organi-zation Science (4:1), pp. 108-120.

Orlikowski, W. J. 1992. “The Duality of Technology: Rethinkingthe Concept of Technology in Organizations,” OrganizationScience (2:3), pp. 398-427.

Orlikowski, W. J., and Gash, D. 1994. “Technology Frames: Making Sense of Information Technology in Organizations,”ACM Transactions on Information Systems (12:2), pp. 174-207.

Pentland, B. T. 1999. “Building Process Theory with Narrative: From Description to Explanation,” Academy of ManagementReview (24:4), pp. 711-724.

Phillips, N., Lawrence, T. B., and Hardy, C. 2004. “Discourse andInstitutions,” Academy of Management Review (29:4), pp.635-652.

Pinch, T., and Bijker, W. 1987. “The Social Construction of Factsand Artifacts,” in The Social Construction of TechnologicalSystems: New Directions in the Sociology and History ofTechnology, W. E. Bijker, T. P. Hughes, and T. J. Pinch (eds.),Cambridge, MA: MIT Press, pp. 17-50.

Potter, J., and Edwards, D. 2001. “Discursive Social Psychology,”in The New Handbook of Language and Social Psychology, W.P. Robinson and H. Giles (eds.), New York: John Wiley & Sons,pp. 103-118.

Powell, W., and DiMaggio, P. 1991. The New Institutionalism inOrganizational Analysis, Chicago: University of Chicago Press.

Ramiller, N. 2001. “The ‘Textual Attitude’ and New Technology,”Information and Organization (11:2), pp. 129-156.

Richards, J. 2008. “McKinsey: Ignore Second Life at Your Peril,”Times Online, April 23 (available online at http://technology.timesonline.co.uk/tol/news/tech_and_web/article3803056.ece).

Ricoeur, P. 1981. Hermeneutics and the Human Sciences: Essayson Language, Action, and Interpretation (J. B. Thompson,trans.), Cambridge, UK: Cambridge University Press.

Rogers, E. M. 1995. The Diffusion of Innovations, New York: TheFree Press.

Rose, F. 2007. “How Madison Avenue Is Wasting Millions on aDeserted Second Life,” WIRED Magazine (15.08), July 24(available online at http://www.wired.com/techbiz/media/magazine/15-08/ff_sheep).

Rosenberg, N. 1994. Exploring the Black Box: Technology, Eco-nomics, and History, Cambridge, UK: Cambridge UniversityPress.

Sairamesh, J., Lee, A., and Anania, L. 2004. “Information Cities,”Communications of the ACM (47:2), pp. 29-31.

Saldana, J. 2009. The Coding Manual for Qualitative Researchers,Thousand Oaks, CA: Sage Publications.

Simon, H. A. 1956. Administrative Behavior, New York: Macmillan.

Starbuck, W. H., and Milliken, F. J. 1988. “Executives’ PerceptualFilters: What They Notice and How They Make Sense,” in TheExecutive Effect: Concepts and Methods for Studying TopManagers, D. Hambrick (ed.), Greenwich, CT: JAI Press, pp.35-65.

Strauss, A. L., and Corbin, J. M. 1998. Basics of QualitativeResearch: Techniques and Procedures for Developing GroundedTheory, Thousand Oaks, CA: Sage Publications.

Swanson, E. B. 1994. “Information Systems Innovation amongOrganizations,” Management Science (40:9), pp. 1069-1092.

Swanson, E. B., and Ramiller, N. C. 1997. “The Organizing Visionin Information Systems Innovation,” Organization Science (8:5),pp. 458-475.

Swanson, E. B., and Ramiller, N. C. 2004. “Innovating Mindfulllywith Information Technology,” MIS Quarterly (28:4), pp.553-583.

Taylor, J. R., and Van Every, E. J. 2000. The EmergentOrganization: Communication as Its Site and Surface, Mahwah,NJ: Lawrence Erlbaum Associates.

Toulmin, S. E. 2003. The Uses of Argument, Cambridge, UK: Cambridge University Press.

Urquhart, C. 2001. “An Encounter with Grounded Theory: Tackling the Practical and Philosophical Issues,” in QualitativeResearch in IS: Issues and Trends, E. Trauth (ed.). Hershey, PA: Idea Group Publishing, pp. 104-140.

Uzzi, B. 1997. “Social Structure and Competition in InterfirmNetworks: The Paradox of Embeddedness,” AdministrativeScience Quarterly (42:1), pp. 35-67.

Weick, K. 1979. The Social Psychology of Organizing, Reading,MA: Addison-Wesley.

Weick, K. 1990. “Technology as Equivoque: Sensemaking in NewTechnologies,” in Technology and Organizations, P. S. Goodmanand L. Sproull (eds.), San Francisco, CA: Jossey Bass.

Weick, K. 1995. Sensemaking in Organizations, Thousand Oaks,CA: Sage Publications.

Weick, K., Sutcliffe, K. M., and Obstfeld, D. 2005. “Organizingand the Process of Sensemaking,” Organization Science (16:4),pp. 409-421.

Westley, F. R. 1990. “Middle Managers and Strategy: Micro-dynamics of Inclusion,” Strategic Management Journal (11:5),pp. 337-351.



Whitton, M. C. 2003. “Making Virtual Environments Compelling,”Communications of the ACM (46:7), pp. 40-46.

Yates, J. 1989. Control through Communication: The Rise ofSystem in American Management, Baltimore, MD: JohnsHopkins University Press.

Ye, R., and Johnson, P. E. 1995. “The Impact of ExplanationFacilities on User Acceptance of Expert Systems Advice,” MISQuarterly (19:2), pp. 157-172.

About the Authors

Nicholas Berente is an assistant professor of Management Infor-mation Systems at the Terry College of Business at the Universityof Georgia. He studies the intersection of organizational innovation,design, and information technologies from a social systems perspec-tive. He draws upon institutional, discursive, and complex systemstraditions in his research. Nick earned his Ph.D. and MBA fromCase Western Reserve University, and conducted his post-doctoralstudies at the University of Michigan.

Sean Hansen is an assistant professor of Management InformationSystems at the E. Philip Saunders College of Business at RochesterInstitute of Technology. He earned his Ph.D. and MBA from theWeatherhead School of Management at Case Western ReserveUniversity. His research interests include requirements engineering,

IT strategy, Web 2.0 technologies, and the application of contem-porary cognitive theory to information systems development. Hiscurrent research focuses on the emergence of new approaches torequirements engineering in response to advances in systems designand implementation.

Jacqueline C. Pike is an assistant professor of Information SystemsManagement at Duquesne University. She earned her B.B.A. fromthe Honors Tutorial College and College of Business at OhioUniversity and her Ph.D. from the Katz Graduate School of Businessat the University of Pittsburgh. Her research interests include self-presentation in social computing environments, mass collaborationsystems, green IT, and the visual display of information. She hasgiven presentations about her research and teaching at leadingconferences.

Patrick J. Bateman is an assistant professor at the WilliamsonCollege of Business Administration, at Youngstown State Univer-sity. He earned his Ph.D. in Information Systems at the KatzGraduate School of Business, University of Pittsburgh. He has abackground as a financial analyst, corporate education manager, andtechnical analyst in the banking and financial services industries.His research interests include the dynamics of online communities,social media, virtual worlds, computer-mediated communication,and individuals’ use of interactive media. He has published inInformation Systems Research and has presented his work at leadinginternational conferences.



Documents

Arguing the Value of Virtual Worlds: Patterns of ... · virtual worlds, this paper reports on a study of the 288 argu-ments used by 59 professionals in assessing the organizational