WHEN WORLDS COLLIDE IN CYBERSPACE: HOW BOUNDARY … · Rothbard et al., 2005). Scholars have defined boundaries as “mental fences” (Zerubavel, 1991: 2) used to simplify and order

WHEN WORLDS COLLIDE IN CYBERSPACE:HOW BOUNDARY WORK IN ONLINE SOCIAL

NETWORKS IMPACTSPROFESSIONAL RELATIONSHIPS

ARIANE OLLIER-MALATERRERouen Business School and Université du Québec à Montréal

NANCY P. ROTHBARDJUSTIN M. BERG

University of Pennsylvania

As employees increasingly interact with their professional contacts in online socialnetworks that are personal in nature, such as Facebook or Twitter, they are likely toexperience a collision of their professional and personal identities unique to this newand expanding social space. In particular, online social networks present employeeswith boundary management and identity negotiation opportunities and challengesbecause they invite nontailored self-disclosure to broad audiences while offering fewof the physical and social cues that normally guide social interactions. How and whydo employees manage the boundaries between their professional and personal iden-tities in online social networks, and how do these behaviors impact the way they areregarded by professional contacts? We build a framework to theorize about howwork-nonwork boundary preferences and self-evaluation motives drive the adoptionof four archetypical sets of online boundary management behaviors (open, audience,content, and hybrid) and the consequences of these behaviors for respect and liking inprofessional relationships. Content and hybrid behaviors are more likely to increaserespect and liking than open and audience behaviors; audience and hybrid behaviorsare less risky for respect and liking than open and content behaviors but more difficultto maintain over time.

As the world becomes increasingly connectedthrough social media, employees are interactingmore with coworkers, supervisors, and otherprofessional contacts in online social networks.Some of these online networks, such as Face-book and Twitter, are social spaces where inter-actions can be personal as well as professional.Participation in such networks results in a po-tential collision of professional and personalworlds that may open up opportunities as wellas create challenges for employees as they

strive to establish and maintain respect and lik-ing in the eyes of their professional contacts(Ashforth, Kreiner, & Fugate, 2000; Kossek, Noe,& DeMarr, 1999; Phillips, Rothbard, & Dumas,2009). When interacting in online social net-works, employees move from offline interac-tions, where disclosure and feedback are tai-lored within particular conversations andguided by clear physical cues (Goffman, 1956), tointeractions characterized by open disclosure tobroad audiences, some of which are not readilyvisible (Boyd, 2007; Donath & Boyd, 2004; Ryan,2008). Thus, in the new world of work, collisionsof one’s professional and personal domains areincreasingly frequent online. This presents newopportunities and challenges for boundary man-agement and identity negotiation in cyberspacethat are not directly addressed in past theoryand research on these processes in physicalspace (e.g., Ashforth et al., 2000; Kreiner, 2006;Kreiner, Hollensbe, & Sheep, 2009; Nippert-Eng,1995; Rothbard, Phillips, & Dumas, 2005; Swann,

We thank our editor, Kevin Steensma, and the three anon-ymous reviewers for providing insightful comments; theWharton Center for Leadership and Change Management,the Wharton Center for Human Resources, and the WhartonGlobal Initiatives Research Program for funding; partici-pants at the 2011 Wharton People and Organizations Con-ference and members of the Contemporary P@thways ofCareer, Life and Learning Research Center of Rouen Busi-ness School for their helpful comments on prior versions ofthe manuscript; and Tarani Merriweather Woodson for herhelpful research assistance.

! Academy of Management Review2013, Vol. 38, No. 4, 645–669.http://dx.doi.org/10.5465/amr.2011.0235

645Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyrightholder’s express written permission. Users may print, download, or email articles for individual use only.

Johnson, & Bosson, 2009). Although scholarshave begun to investigate challenges for indi-viduals using social media more generally (Do-nath & Boyd, 2004; Ellison, Steinfield, & Lampe,2007; Forest & Wood, 2012), little is known abouthow employees manage the boundary betweentheir professional and personal identities in on-line social networks and what this means fortheir professional relationships.

To guide theory and research on onlineboundary management, we build a frameworkthat extends boundary theory into the new andexpanding world of online social networks. Wepropose that employees’ online boundary man-agement behaviors are driven by a combinationof their preferences for segmentation versus in-tegration of their professional and personalidentities (Ashforth et al., 2000; Kreiner, 2006;Rothbard et al., 2005) and their motives for self-enhancement or self-verification (Brown, 1990,1991; Kwang & Swann, 2010; Pfeffer & Fong, 2005;Sedikides & Strube, 1995; Swann, 1983, 1990;Wiesenfeld, Swann, Brockner, & Bartel, 2007). Weidentify four sets of online boundary manage-ment behaviors that employees enact in re-sponse to these drivers and theorize about howthese behaviors influence the degree to whichemployees are respected and liked by profes-sional contacts.

We focus on respect and liking as importantelements of professional relationships for tworeasons. First, respect and liking tap into thetwo core dimensions by which people formulatejudgments of others: competence and warmth(Fiske, Cuddy, & Glick, 2007; Fiske, Cuddy,Glick, & Xu, 2002; Fiske, Xu, Cuddy, & Glick,1999; Hamilton & Fallot, 1974). Moreover, judg-ments of professional contacts on these two di-mensions may diverge based on interactions inonline social networks that mix the professionaland personal. Second, respect and liking arehighly desirable in organizations and havebeen related to a number of important outcomes(Cuddy, Glick, & Beninger, 2011), including su-pervisory performance ratings (Lefkowitz, 2000;Robbins & DeNisi, 1994), group commitment (Ty-ler & Blader, 2003), leader-member exchange(Dienesch & Liden, 1986; Wayne & Ferris, 1990),and group performance (Gruenfeld, Mannix,Williams, & Neale, 1996; Jehn & Shah, 1997). Re-cent studies show that self-disclosure in onlinesocial networks can significantly affect likingand individuals’ impressions of others more

generally (Forest & Wood, 2012; Weisbuch,Ivcevic, & Ambady, 2009). In this way employees’boundary management behaviors in online so-cial networks can help or hurt their professionalrelationships and, thus, hinder their career suc-cess (Cuddy et al., 2011; Dutta, 2010).

Because research on social media is still rel-atively new, many empirical questions concern-ing employees’ experiences in online social net-works remain unanswered. However, a recentreview of Facebook-related research (Wilson,Gosling, & Graham, 2012) points to benefits suchas keeping in touch with strong and weak tiesand, in particular, crystallizing otherwiseephemeral relationships (Ellison et al., 2007). Yetthe review also points to evidence that employ-ees are struggling with issues of identity pre-sentation and relationships among groups andindividuals. For instance, studies focusing onhiring decisions show that the chances appli-cants will be offered a job increase if theirFacebook profile emphasizes family values orprofessionalism, but the chances decrease iftheir profile contains inappropriate material,such as alcohol and drugs (Bohnert & Ross,2010). While management research to datehas not directly examined the impact of employ-ees’ behavior in online social networks on theirprofessional relationships, we draw on thisemergent body of evidence to argue that theway employees interact with their professionalcontacts in online social networks impacts therespect and liking they receive from them (Stop-fer & Gosling, 2012). Thus, by building a theoret-ical framework of online boundary manage-ment, we open up new lines of research on afast-growing phenomenon that has importantimplications for employees, managers, and or-ganizations.

BOUNDARY MANAGEMENT: FROM PHYSICALSPACE TO CYBERSPACE

Boundary management of multiple identitiesis a classic organizational challenge. Followingthe Second Industrial Revolution, the boundarybetween professional and personal life wascited as one of the essential features of a bu-reaucratized society (Weber, 1968), leading em-ployees to enact different identities when theyinteracted in a professional setting versus a per-sonal setting with family and friends (Ashforthet al., 2000; Kossek et al., 1999; Nippert-Eng, 1995;

646 OctoberAcademy of Management Review

Rothbard et al., 2005). Scholars have definedboundaries as “mental fences” (Zerubavel, 1991:2) used to simplify and order the environment, or“physical, temporal, emotional, cognitive,and/or relational limits that define entities asseparate from one another” (Ashforth et al., 2000:474). Employees have come to rely on boundar-ies to delineate their professional and personaldomains and avoid the coactivation of incongru-ent facets of their identities (Rothbard & Rama-rajan, 2009). To some extent these boundariesare reified through social cues regarding normsand expectations that guide behavior in each ofthese domains (Nippert-Eng, 1995). However,boundary management is an active, ongoingprocess (Edwards & Rothbard, 2000), and em-ployees either reinforce or lessen boundariesduring social interactions. For example, theymay befriend work colleagues and invite theminto their homes.

A key reason why employees navigate theirmultiple identities in this way is to maintain orenhance their professional relationships (Du-mas, Phillips, & Rothbard, in press; Phillips etal., 2009; Roberts, 2005). On the one hand, theprofessional domain often includes strong andclear norms and expectations of what consti-tutes appropriate professional behavior (Bloor &Dawson, 1994; Pratt, Rockmann, & Kaufmann,2006). Thus, employees who enact their personalidentities in ways that are seen as inappropri-ate in the professional domain lose respect inthe eyes of their professional contacts. On theother hand, employees’ professional contactsmay appreciate seeing aspects of their personalidentities, since personal self-disclosure andfrequent interactions tend to increase liking(Collins & Miller, 1994; Saegert, Swap, & Zajonc,1973). Therefore, if employees can effectivelymanage the boundaries between their profes-sional and personal identities such that theyengage in some personal disclosure in their in-teractions, but without violating professionalnorms, they will be more respected and liked byprofessional contacts.

However, to date, little theory has shed lighton the boundary work that employees do wheninteracting in online social networks. Here wefocus on online social networks that are per-sonal in nature and pervasive enough such thatemployees receive connection requests fromprofessional contacts (e.g., Facebook or Twitter).With a variety of features, including biographi-

cal profiles, photo sharing, and public com-ments (Boyd, 2007), these online social networksare becoming a key forum for relationship de-velopment and maintenance (Ellison et al., 2007).Recent data point to a blurring of the profes-sional and personal domains in these onlinenetworks, which exacerbates the need for activeboundary management. For instance, youngeremployees are connected to an average of 16coworkers on Facebook (Millennial Branding,2012), and 41 percent of Facebook users think itis irresponsible to ignore a friend request from acoworker (Ketchum Global Research Network forLiberty Mutual, 2011). While most of the litera-ture on online social networks focuses on pri-vacy (see, for instance, Boyd, 2007; Lewis, Kauf-man, & Christakis, 2008; Martin, 2009; Solove,2007), a few studies report evidence of individu-als conducting online boundary work throughself-censoring (Lampinen, Tamminen, & Oulas-virta, 2009; Skeels & Grudin, 2009), adjusting pro-file visibility (Tufekci, 2008), customizing profilesto disclose different information to different in-dividuals (Donath & Boyd, 2004; Zhao, Gras-muck, & Martin, 2008), and creating multiple pro-files (Stutzman & Hartzog, 2012).

A great deal of personal information is oftendisclosed in these online forums. However, a keyfeature that distinguishes such online interac-tion is that the information disclosed is muchless tailored to a particular conversation. Inface-to-face and other individualized interac-tions, such as email or phone calls, employeescan manage the boundary between their profes-sional and personal identities by controlling theamount and nature of personal information theydisclose to professional contacts over time andby adapting such disclosure to particular dyadicrelationships (Collins, 1981; Goffman, 1956). Inonline social networks, personal information isdisclosed in a nontailored fashion. Comparedwith feedback and comments made in offlinework interactions, for instance, online informa-tion is shared with a potentially broader audi-ence and is also persistent in time and easilysearchable (Boyd, 2007). Because of this, whentwo professional contacts first connect in a pri-marily personal online social network like Face-book, they are granted access to a bevy of per-sonal information about each other all at once,perhaps including content the focal person wasunaware of or had forgotten. When a new rela-tionship starts offline, individuals also ex-

2013 647Ollier-Malaterre, Rothbard, and Berg

change a large amount of personal informationinitially, yet this information is tailored and di-rected in a personalized manner in the conver-sation between the individuals (Goffman, 1956;Taylor, 1968; Worthy, Gary, & Kahn, 1969). In con-trast, connecting in online social networks un-leashes a flood of self-disclosure in the form ofan archive of information that is not tailored tothe particular relationship or situation, and itsoriginal context and meaning may be skewed(Boyd, 2008). The recent “timeline” feature ofFacebook is a vivid illustration of a nontailoredchronological display of a vast amount of infor-mation. In addition, this information may beeasily searchable and retrievable using searchengines and websites that collect available in-formation on a given individual and his or herconnections (Boyd, 2007). Such a flood of nontai-lored self-disclosure—especially at the begin-ning of a professional relationship—is an expe-rience that seems to be unique to connecting inonline social networks.

A second key feature of online interaction isthat even after the initial flood of self-disclosure,the sharing of personal information continuesbetween the contacts, without access to the vis-ible social cues—including facial expressions,vocal tones, and body language (Mehrabian,1971)—that normally help reinforce norms andscripts in physical space by providing criticalinformation about how one should behave insocial interactions (Collins, 1981, 2004; Goffman,1959). Online social networks that are personalin nature essentially put users in one large, in-visible room with “friends” or connections fromvarious social worlds, which often include boththe professional and personal. Employees inter-act with a visible audience (e.g., contacts whofrequently interact with the employee or areavailable to chat at the time the employee logsin) that makes salient particular social identi-ties (Tajfel & Turner, 1986). This prompts them touse the social scripts they associate with theseparticular ingroups (Tajfel, 1970). Meanwhile,outgroup members might be in the invisible au-dience—that is, all the other contacts, includingprofessional contacts, who may not be salient inemployees’ minds as they share information(Boyd, 2007). Employees often use the visible au-dience as a guide to recreate social cues andpost information that is appropriate for thesecontacts. However, in doing so they also unwit-tingly disclose that information to invisible au-

diences, for which the personal informationmight not be as appropriate. To some extent,employees can navigate online social networkswithout physical cues using self-categorizationprocesses (Turner, Hogg, Oakes, Reicher, &Wetherell, 1987). Yet without the physical andsocial cues that typically help communicate andenforce norms in physical space, they are morelikely to fall prey to false consensus biases(Ross, Greene, & House, 1977) and to overesti-mate how much their professional contactsshare their understanding of what constitutesappropriate disclosure of information in onlinesocial networks.

In addition to the challenges of managing theinformation employees disclose about them-selves in online social networks, employees alsoneed to manage what information their profes-sional or personal contacts disclose about them,as well as the public or semi-public feedbacktheir contacts give on information the employ-ees disclose themselves (Boyd, 2007). For in-stance, an employee may suffer reputationalconsequences from a coworker’s comment im-plying that she does not work hard, even if thatis actually not the case. Employees could alsobe embarrassed by personal comments made bytheir friends and family members that their pro-fessional contacts see as well. The asynchro-nous and public or semi-public nature of theinteractions thus constrains employees’ choicesregarding boundary management and identitynegotiation, since employees need to consider towhom it might be appropriate to connect andwhat information to share—not only in terms ofwhat they themselves disclose online but alsowhat others may disclose about them.

Taken together, these challenges suggest thatmanaging boundaries between the professionaland the personal is qualitatively different incyberspace than in physical space, and formany employees such boundary management ismore difficult online than offline (Ashforth et al.,2000; Kossek et al., 1999). If employees want aboundary between their professional and per-sonal identities in online social networks, theyhave to actively construct and maintain it them-selves. This creates ongoing challenges for em-ployees who face social pressure to enact differ-ent norms, roles, and scripts within theirprofessional and personal domains. The ways inwhich employees deal with these challengessignificantly influence the degree to which their


professional contacts like and respect them, assuggested by emerging empirical evidence (e.g.,Bohnert & Ross, 2010; Forest & Wood, 2012). How-ever, theory and research to date have not di-rectly addressed how and why employees man-age boundaries between their professional andpersonal identities in online social networks, orhow this affects the way their professional con-tacts regard them.

ONLINE BOUNDARY MANAGEMENT:A THEORETICAL FRAMEWORK

Motivational Drivers and Online BoundaryManagement Behaviors

We develop a framework to explain the moti-vational drivers of online boundary manage-ment behaviors. Drawing on social network the-ory, we build on the premise that employees,when interacting in online social networks, ad-dress two fundamental questions: how theystructure their ties in the networks and whattype of information they share with their ties(Burt, 2004; Granovetter, 1973; Podolny, 2005). Onthe one hand, employees can manage theirstructural social capital—that is, to whom theyare connected in the network (Nahapiet & Gho-shal, 1998). In particular, they can choosewhether to be connected to professional con-tacts at all or to keep professional contacts sep-arate from personal contacts. On the other hand,employees can manage their relational socialcapital—that is, the nature and richness of theinformation that is exchanged with their con-nections (Nahapiet & Ghoshal, 1998). Thus, theycan choose what and how much personal infor-mation to disclose to their professional contacts.Employees’ answers to these questions aboutwith whom they communicate and what theycommunicate in their online social networksform the basis of our framework.

Two key identity navigation processes mayguide employees’ answers to these two funda-mental questions: boundary work as informedby work-nonwork boundary preferences (Ash-forth et al., 2000; Kreiner, 2006; Rothbard et al.,2005) and identity negotiation as informed byself-evaluation motives (Sedikides, 1993; Swann,Pelham, & Krull, 1989). We use identity naviga-tion as a broad term encompassing both bound-ary work—that is, constructing mental frame-works to delineate how roles and social

identities are merged or separated (Ashforth etal., 2000)—and identity negotiation—that is, pro-cesses “whereby relationship partners reachagreements regarding ‘who is who’” (Swann etal., 2009: 81). First, we contend that employees’answers to the question of with whom they com-municate in online social networks primarilycorrespond to whether they prefer to mentallyorganize their social worlds such that facets oftheir professional and personal identities aresegmented versus integrated (Ashforth et al.,2000; Kreiner, 2006; Rothbard et al., 2005). Second,we contend that employees’ answers to thequestion of what they communicate in onlinesocial networks primarily correspond to whetherthey choose to present themselves to the worldin ways that verify or enhance their existingself-views (Sedikides, 1993; Swann et al., 1989).The interactions of these two separate cognitiveprocesses drive employees’ boundary manage-ment behaviors in important and predict-able ways.

First, based on prior work on boundary man-agement, we expect that the preference to seg-ment versus integrate professional and per-sonal identities is an important driver of withwhom employees are motivated to connect inonline social networks and how they structurethese ties (Ashforth et al., 2000; Kreiner, 2006;Rothbard et al., 2005). In particular, employeeswho have a greater desire to segment their pro-fessional and personal identities are likely moreattentive to and concerned with classifying thedomain in which their contacts belong; in re-sponse, they will engage in efforts to keep dif-ferent classifications of contacts separated intheir online social world (just as they do in thephysical world). Segmenting domains enablesemployees to avoid the psychological discom-fort and conflict caused by the coactivation ofpersonal and professional identities that are notcompatible (Rothbard & Ramarajan, 2009). Forinstance, employees who prefer to segment theiridentities may feel uncomfortable posting fam-ily pictures in online social networks if theirsupervisor, coworkers, or other professional con-tacts can see them. In contrast, employees whoprefer integrating their identities are not likelyto manage their online audience as actively.Instead, they are more likely to deliberatelyseek out network ties across domains, creating alarge pooled audience of both professional andpersonal contacts. Thus, segmentors should be


motivated to more actively construct boundariesbetween their professional and personal con-tacts in online social networks than integrators.

Second, employees’ motives to present them-selves to others in a positive and socially desir-able manner (self-enhancement) or to behave ina manner that confirms their own positive andnegative self-views (self-verification) are likelyto play an important role in shaping the contentthey disclose in online social networks. We fo-cus on self-enhancement and self-verificationbecause these two self-evaluation motives havereceived extensive attention and empirical sup-port and have been contrasted frequently in theliterature (see Anseel & Lievens, 2006; Sedikides,1993; Swann, 1990; Swann et al., 1989; see Kwang& Swann, 2010, for a meta-analysis). In addition,both involve evaluations of the self that concernhow people want to be seen by others (and thusthe content they share online), rather than moreinternally focused self-evaluation motives, suchas self-assessment and self-improvement.Whereas preferences for segmentation versusintegration lie on opposite ends of a continuum(Rothbard et al., 2005), self-enhancement andself-verification are essentially different routesto self-evaluation rather than opposite motives.Self-enhancement suggests that when peopleself-evaluate, they want to enhance their posi-tive self-concept and protect it from negativeinformation. Thus, they will selectively attend toand promote self-relevant information that hasfavorable implications for the self and avoidsuch information that has negative implications(Sedikides, 1993). In contrast, self-verificationdrives people to seek affirmation of their preex-isting self-concept such that they will attend toand promote either positive or negative self-relevant information as long as it is consistentwith their self-views (for reviews see Swann,1983, 1990). Therefore, although not on a contin-uum, self-enhancement and self-verificationrepresent very different approaches to self-evaluation and identity negotiation.1

Although online social networks create newopportunity structures to display identity cuesand select interaction partners (Swann, 1987),they also constrain the flexibility of self-presentational behaviors, because disclosureand interactions cannot be easily tailored to dy-adic interactions and roles are often enactedsimultaneously. In offline interactions wherepeople activate and enact situated identities(Alexander & Weil, 1969) and role-specific con-ceptions (Swann et al., 2009), employees canchoose to self-enhance with specific profes-sional contacts, such as supervisors, and to self-verify with others, such as peers. However, wheninteracting online, employees must make domi-nant or macro choices that apply to broadgroups of contacts. For instance, if their domi-nant choice is to self-verify, perhaps becausethey are connected with mostly peers or becausetheir personal contacts are most salient in theirminds, then their other professional contacts(supervisors, subordinates, etc.) may access thesame self-verifying information, restricting em-ployees’ ability to self-enhance toward thesecontacts. Even if they tend to first verify positiveself-views, over time they are bound to alsoshare negative information about themselves,perhaps when seeking emotional support (Elli-son et al., 2007). This self-verifying disclosure ofvulnerability may have very different implica-tions for respect and liking in the eyes of profes-sional contacts than a macro choice driven byself-enhancement.

Employees seeking self-enhancement aremore likely to try to share information that helpsmanage the impressions others form of them(Leary & Kowalski, 1990; Schlenker, 1980; Swannet al., 1989). Since employees who self-enhancehave a desire to see themselves and their ac-tions, traits, and attitudes in the most positivelight (Pfeffer & Fong, 2005), we expect that theywill seek favorable evaluations and feedback soas to achieve a high level of personal worth(Morrison & Bies, 1991; Sedikides & Strube, 1995).

1 Research examining self-enhancement and self-verifi-cation motives suggests that individual differences, such asnarcissism (John & Robins, 1994; Paulhus, 1998), self-esteem(Baumeister, Tice, Hutton, 1989), self-certainty (e.g., Pelham &Swann, 1994; Swann & Ely, 1984), and uncertainty avoidance(Roney & Sorrentino, 1995); the rushed versus calm nature ofthe environment (Hixon & Swann, 1993; Swann & Schroeder,1995); and the stage of the relationship (i.e., qualifying or

established) influence which motive will be used by anindividual at a particular time (Swann, 1990). For example,narcissists are more likely to self-enhance (John & Robins,1994; Paulhus, 1998). Also, individuals are more likely tosuspend their motive to self-verify when relationships are ina “qualifying stage,” where they are trying to prove them-selves to one another, especially in the context of asymmet-ric relationships with superiors (Swann et al., 2009).


Thus, they will actively try to make positive im-pressions (Leary & Kowalski, 1990; Roberts, 2005;Schlenker, 1980) and be more likely to regulatethe information they share in online social net-works, as well as the information disclosed byothers about them. In contrast, employees whoprefer to ensure the consistency and stability oftheir self-views (i.e., self-verification; Swann,1983) may disclose a more diverse array of pos-itive and negative self-relevant information,with the expectation that the audience theyreach, or part of this audience, will give themfeedback that confirms their positive and nega-tive self-views. In sum, self-enhancers are moti-vated to actively control the information theydisclose online as a way to positively impressprofessional contacts (Brown, 1990, 1991;Sedikides & Strube, 1995), whereas self-verifiersare more likely to share both positive and morenegative information as a way to receive confir-mation of their self-views (Kwang & Swann,2010; Swann, 1983, 1990; Swann et al., 1989;Wiesenfeld et al., 2007).2

By crossing these two identity navigation pro-cesses in a 2 ! 2 matrix, we develop a concep-tual framework of four motivational drivers ofonline boundary management and four corre-

sponding sets of archetypal online boundarymanagement behaviors (see Figure 1). While themotivational drivers capture employees’ desiresregarding with whom they connect and whatthey share in online social networks, the arche-typal behaviors capture the actions these de-sires are likely to drive employees to take inonline social networks to construct and main-tain their professional and personal identitiesover time. Table 1 provides examples of thesefour sets of behaviors, which are meant to beillustrative, not exhaustive.

Open boundary management. Employees whomake a macro choice to self-verify in online so-cial networks, perhaps because they are mostlyconnected with peers and personal contacts,and who also prefer to integrate their profes-sional and personal identities are unlikely toperceive many boundary challenges. These em-ployees may therefore engage in open behav-iors, which we define as a very simplified ap-proach in which employees do not constructboundaries separating their professional andpersonal identities but, instead, present them-selves as the person they perceive themselves tobe and fully enact their personal identity in on-line social networks to a broad audience. Openbehaviors result in no active boundary manage-ment across professional and personal identi-ties. They entail disclosure of both positive andnegative self-verifying content and acceptanceof disclosures and feedback provided by theirconnections. For instance, employees mightshow positive unconventional facets of them-selves that are not usually expressed in theirwork environment (such as an investmentbanker writing children’s comic books) or nega-tive facets of themselves (such as discussing aprofessional setback or an ongoing divorce) toreceive confirmation of their self-views (Swann,1983). In this case professional and personal do-mains are often merged as employees strive forconsistency in their self-views and integrationacross both domains (see Table 1 for examples).

Proposition 1: Employees are morelikely to engage in open boundarymanagement behaviors in online so-cial networks when they combine self-verification motives and preferencesfor integration of their professionaland personal identities.

2 We conceptualize the temporal nature of online bound-ary management behaviors as analogous to personal proj-ects, which are defined by Little as “a set of interrelated actsextending over time . . . intended to maintain or attain astate of affairs foreseen by the individual” (1983: 273). Thisconceptualization places individuals in a specific, stablecontext that shapes their motivations to engage in onlineboundary management for a meaningful but not permanentstretch of time (Little, 1983; McGregor & Little, 1998). This is inline with the notion that the two motivational drivers in ourframework have been construed as stable individual prefer-ences within a given social context but are not personalitytraits and, thus, may change over an individual’s life course.Specifically, scholars have posited that preferences for seg-mentation versus integration are stable in a given life stage(Kossek & Lautsch, 2012) and that self-enhancement andself-verification motives are associated with firmly heldself-views, which are also likely to be relatively stablewithin a given set of relationships or contexts but are notpermanent over time (Swann et al., 1989; Swann, Polzer,Seyle, & Ko, 2004). Thus, employees’ motivations may shift asa result of significant life events and changes, such asswitching to a new workgroup, job, occupation, or organiza-tion that is more or less open to personal self-disclosure thanthe previous one (Edmondson, 1999; Polzer, Milton, & Swann,2002). As such, employees are most likely to enact particularboundary management behaviors for an extended stretch oftheir career but are unlikely to enact the same behaviorsthroughout their entire career.


Audience boundary management. Not all em-ployees who make a macro choice to self-verifyin online social networks are comfortable inte-grating their professional and personal identi-ties. Some prefer to segment these identitiesand, thus, are more likely to engage in audiencebehaviors, which we define as the constructionand maintenance of a boundary between em-ployees’ professional and personal contacts byrestricting professional contacts from online so-cial networks that employees deem personal innature. An example is setting up private profilesand ignoring or denying connection requestsfrom certain professional contacts (see Table 1for more examples). Employees who prefer tokeep their professional and personal identitiesseparate (Ashforth et al., 2000; Kreiner, 2006;Rothbard et al., 2005) and seek to display self-verifying identity cues to selected interactionpartners (Swann, 1983, 1987) are likely to protecttheir online identities from unsolicited disclo-sures and feedback and to reach out to contactswho may validate their self-views. They maystrive to avoid coactivation of their professional

and personal identities that might cause thempsychological discomfort (Rothbard & Ramara-jan, 2009). In particular, these employees maywant to self-verify by posting personal and fam-ily pictures and expressing intimate feelings orviews of the world, yet at the same time not wantto share this personal information with theirprofessional contacts and have them commenton it. Indeed, self-verification does not meanthat individuals necessarily disclose self-verifying information to everyone they encoun-ter; rather, when they do disclose information,they look to confirm their self-views (Swann etal., 2004). Thus, those employees who wish toself-verify but also wish to segment their profes-sional and personal identities will see theboundary challenges in online social networksas about restricting the professional contactswith whom they are willing to connect in thosenetworks that are personal in nature so that theycan more freely share personal information andinteract with those they do accept as connec-tions. In this way, audience behaviors enablethe disclosure of positive and negative self-

FIGURE 1Online Boundary Management Drivers, Behaviors, and Consequences

Preferences for segmentation versus integration of professional and personal identities

Integration Segmentation

Self-verification

Open boundary management behaviors

Decrease average respect

Decrease average liking

Audience boundarymanagement behaviors

Protect but do not increase average respect Decrease average liking

Self

-eva

luat

ion

mot

ives

Self-enhancement

Content boundarymanagement behaviors

Increase average respect

Increase average liking

Moderated by online boundary management capabilities

Hybrid boundarymanagement behaviors

Increase average respect

Increase average liking

Moderated by online boundary management capabilities (most demanding behaviors)

••

••• •

••

•

•


verifying personal information within the per-sonal domain and prevent its disclosure to cer-tain professional contacts.

Proposition 2: Employees are morelikely to engage in audience bound-ary management behaviors in onlinesocial networks when they combineself-verification motives and prefer-ences for segmentation of their profes-sional and personal identities.

Content boundary management. In contrast toemployees who make a macro choice to self-verify in online social networks, some employ-ees present themselves online in a primarilyself-enhancing way, perhaps because of person-ality traits (e.g., John & Robins, 1994) or becauseimportant relationships are still in a qualifyingstage when they make their macro choices(Swann et al., 2009). When these employees alsoprefer integration of their professional and per-sonal identities, they are likely to engage incontent behaviors, which we define as manag-ing their identities primarily through actively

controlling what information they disclose—butnot the people to whom they disclose—in onlinesocial networks. They may, for instance, broad-cast professional achievements they are proudof or post polished family pictures that mightenhance their status or likability (see Table 1 formore examples). Their desire for integrationmeans they are motivated to connect with theirprofessional contacts online rather than excludethem from their online personal lives. However,because they strive to enhance their image inthe eyes of their professional contacts(Sedikides, 1993; Sedikides & Strube, 1995), theyare likely to be concerned with managing theinformation they share with this integrated au-dience. As such they may purposefully selectprofessional and personal information that theybelieve will enhance the impressions their pro-fessional contacts have of them and that mayelicit positive feedback (Lampinen et al., 2009).In short, employees who use content behaviorsthink of the boundary challenges as about pre-senting the most ideal image of themselves to a

TABLE 1Illustrative Examples of Boundary Management Behaviors in Online Social Networks

Archetypal BoundaryManagementBehaviors Examples of Behaviors (Not Exhaustive)

Open • Set up public searchable profile• Disclose both positive and negative information in professional and personal domain• Let others comment publicly on posts and tag one in pictures

Audience • Set up private profile and ignore or deny connection requests from certain professional contactsin online social networks deemed as personal

• Use different sites to segment audiences—for instance, a LinkedIn account for professionalcontacts and a Facebook account for personal contacts (Stutzman & Hartzog, 2012)

• Use nicknames (Tufekci, 2008) or make the profile unsearchable to avoid unsolicited requestsfrom professional contacts

Content • Disclose information that is flattering (e.g., achievements, good picture), glamorous (e.g., travelobservations and pictures), or makes one look smart (e.g., interesting news articles)

• Keep postings noncontroversial (e.g., refrain from discussing politics, religion, or sexualorientation)

• Control pictures in which one is tagged or prevent others from tagging one in pictures• Monitor comments by others on profile or prevent others from commenting on profile

Hybrid • Create and maintain lists of contacts and manage what content each of these subgroups canaccess—for example, create a “professional list” including all professional contacts and thenexclude this list from posts containing personal information (Donath & Boyd, 2004; Zhao et al.,2008; cf. Google" Circles)

• Clean up profile when transitioning from one life or career stage to another (e.g., studentstransitioning to first job; DiMicco & Millen, 2007)

• Educate connections to recognize that some conversations are personal or professional and totherefore refrain from commenting in ways that would cause embarrassment


broad set of professional and personal contacts(Roberts, 2005).

Proposition 3: Employees are morelikely to engage in content boundarymanagement behaviors in onlinesocial networks when they combineself-enhancement motives and prefer-ences for integration of their profes-sional and personal identities.

Hybrid boundary management. While audi-ence and content behaviors are relativelystraightforward, they do not work well for em-ployees who prefer to both keep their profes-sional and personal contacts separate (Ashforthet al., 2000; Kreiner, 2006; Rothbard et al., 2005)and enhance their image in the eyes of profes-sional contacts in online social networks(Sedikides, 1993; Sedikides & Strube, 1995).These employees may not want to forego thebenefit of connecting and sharing informationwith professional contacts in online social net-works. They view boundary control as entailingboth with whom they connect in online socialnetworks and what information they communi-cate to their connections. In response to thesedual challenges, such employees may craftmore sophisticated online boundary manage-ment behaviors, which we term hybrid behav-iors, defined as the concurrent management ofaudience and content boundaries such that em-ployees divide their professional and personalcontacts into separate audiences and tailor thecontent they disclose to each audience. For in-stance, managing lists in online social networks(as illustrated in Table 1) is an attempt to re-create separate spaces in which one can dis-close different information and receive feedbackin a separate and more tailored manner. In im-plicit recognition of the hybrid approach,Google", one of the newer online social net-works, has enabled users to more easily catego-rize people from different life domains into sep-arate “Circles.” Other hybrid behaviors aretemporal, such as when employees adjust thecontent in and access to their profiles whentransitioning to a new career stage (DiMicco &Millen, 2007). Employees who use hybrid behav-iors may be driven to match their content to theiraudience because they not only have a focus onpreventing inappropriate information from spill-ing over from the personal to the professionaldomain but also have an equally strong desire

to actively construct and enact a positive profes-sional image (Roberts, 2005).

Proposition 4: Employees are morelikely to engage in hybrid boundarymanagement behaviors in online so-cial networks when they combine self-enhancement motives and prefer-ences for segmentation of theirprofessional and personal identities.

Consequences of Online BoundaryManagement Behaviors for Respect and Liking

Having theorized about how and why employ-ees use different online boundary managementbehaviors, we next examine how using each ofthese behaviors influences the degree to whichemployees’ professional contacts respect andlike them. Respect and liking are separate di-mensions of positive regard that can have dif-ferent predictors and consequences (Hamilton &Fallot, 1974; Prestwich & Lalljee, 2009) and there-fore often need to be distinguished (Wojciszke,Abele, & Baryla, 2009). Drawing on Wojciszkeand colleagues’ (2009) distinction between re-spect and liking, we define respect as the degreeof deference or positive regard with which oneviews another person and liking as the degreeto which one is fond of and feels a sense ofattachment to another person. Respect judg-ments are based on agentic information, such asthe appropriateness and intellectual quality ofthe information being shared (Wojciszke et al.,2009). Respect is a key component of perceivedcompetence (Fiske et al., 2007) and contributes tomaintaining a positive social identity, which isan antecedent of self-esteem, well-being, psy-chological engagement with the group, groupinclusion and cooperation, and career success(Cuddy et al., 2011; Tyler & Blader, 2000). Likingjudgments are based on communal information,such as the warmth and valence of the informa-tion being shared (Wojciszke et al., 2009). Likingis an important component of perceived warmth(Fiske et al., 2007) and contributes to trustingrelationships and friendships at work (Brass,1984). Respect and liking, thus, contribute to pos-itive interpersonal relationships, group commit-ment, and cooperation, which, in turn, may leadto enhanced decision making and performance(Jehn & Shah, 1997).

We build on the premise that the respect andliking of employees in the eyes of their profes-


sional contacts depend on employees’ ability tomanage their professional and personal identi-ties online in ways that mirror what their pro-fessional contacts would consider appropriatebehavior in offline interactions. Because workorganizations are usually strong situations (Mis-chel, 1973) where employees face institutionalpressures to behave in a rational, professionalmanner (Bloor & Dawson, 1994; Pratt et al., 2006),the norms, roles, and scripts within employees’professional domains may be especially salient,strong, and actively enforced in the minds oftheir professional contacts (Van Maanen &Schein, 1979). However, in online social net-works, professional contacts have access to per-sonal information that they otherwise would nothave seen in the context of an offline tailoredconversation. As such, respect and liking areonly protected or enhanced when employees’online behaviors conform to the norms, roles,and scripts that their professional contacts ex-pect of them in offline interactions.

Because we are interested in the conse-quences of using online boundary managementbehaviors for workplace outcomes, we focus onaverage degree of respect and liking, defined asthe aggregation of the judgments of the focalemployee by each of his or her professional con-tacts. Our propositions concern the impact on-line boundary management behaviors have onthe average degree of respect and liking as com-pared to if the focal employee did not partici-pate in online social networks at all. Thus, in ourdefinition of average degree of respect and lik-ing, we include all professional contacts, notjust those with whom an employee is connectedonline. We focus on average—as opposed to dy-adic—respect and liking because employees’online behaviors are nontailored by nature andare therefore seen by groups of multiple con-tacts (at best, separate groups of contacts in thecase of hybrid boundary management). In addi-tion, compared to dyadic respect and liking, av-erage respect and liking are more likely to im-pact collective organizational outcomes (Lincoln& Miller, 1979), such as group performance (Gru-enfeld et al., 1996), group-serving behaviors (Ty-ler & Blader, 2000), or resource mobilization andinformation sharing in groups.

There are several reasons self-disclosure inonline social networks might influence the de-gree to which a focal employee is respected andliked by his or her professional contacts. First,

professional norms may influence these rela-tionships. Norm violations are especially harm-ful to respect and liking in work contexts wherestrong norms convey clear images of what con-stitutes appropriate professional behavior(Bloor & Dawson, 1994; Chreim, Williams, & Hin-ings, 2007). Employees who are not consistent inonline social networks with how their profes-sional contacts, on average, expect them to be-have in offline interactions will lose credibilityor upset their professional contacts becausethey violate norms enforced by the group. Whatconstitutes normative acceptable self-disclo-sure will vary across professional relationships,depending on how formal each dyadic relation-ship is. However, professional norms generallyimpose limitations on personal self-disclosureoffline and online. For instance, posting a pic-ture of oneself in a bathing suit on one’s onlineprofile might be well received by a coworkerwho is also a close friend, but not by more for-mal professional contacts, who may be in theinvisible audience and, hence, not salient in theemployee’s mind. Thus, enacting one’s profes-sional and personal identities online in waysthat fail to mirror the degree and nature of per-sonal disclosure that is expected of one in off-line interactions harms the way one is regardedon average by professional contacts. Becauseindividuals tend to forget part of their audienceonline (Boyd, 2007), employees often uninten-tionally fail to mirror their offline professionalnorms by disclosing too much or inappropriatepersonal information. Conversely, highly cau-tious employees end up disclosing too littleabout themselves or being too controlling aboutthe feedback they receive (e.g., by deleting oth-ers’ comments or preventing anyone from com-menting on their profiles). Consequently, theycan be perceived as uptight or cold by onlineprofessional contacts, since warmth and likabil-ity are associated with the degree and nature ofpersonal self-disclosure in online social net-works (Forest & Wood, 2012; Rothbard, Berg, &Ollier-Malaterre, 2012; Weisbuch et al., 2009).

Even in work environments where norms areunclear, weakly enforced, or open to personaldisclosure, a second reason online social net-work behaviors impact average respect and lik-ing is that the information that employees dis-close can make them appear to be lessprototypical members of the group (Tyler &Blader, 2000) and therefore not a good fit with the


organization. For instance, coworkers in an oildrilling company might respect and like an em-ployee less who posts about active involvementin environmental protection groups. Below wefirst discuss the specific consequences of thefour sets of online boundary management be-haviors in our framework for average respectand liking and then the role of employees’ ca-pabilities, as well as social feedback and orga-nizational norms in shaping these behaviorsand consequences.

Consequences of open behaviors. Openboundary management behaviors are consis-tent with the underlying philosophy of socialmedia that privacy is obsolete and that sharingis preferable to hiding (Boyd, 2008). Yet, as dis-cussed above, this may not be the best approachin many professional contexts, where onlinepersonal disclosures may violate offline norms.Because employees who use open behaviorsmay reveal too much or inappropriate personalinformation given the norms in their profes-sional domains (Pratt et al., 2006), or let theircontacts reveal too much or inappropriate per-sonal information about them, professional con-tacts may feel that the open employee does notunderstand or is incapable of upholding thenorms and scripts required of him or her in theprofessional domain and so does not fit with thegroup (Tyler & Blader, 2002). In this way, employ-ees who use open boundary management be-haviors can inadvertently lose respect by re-vealing too much or inappropriate informationto a broad audience of professional contacts(Phillips et al., 2009) and, in turn, may be per-ceived as not worthy of trust regarding impor-tant decisions or sensitive information (Mayer,Davis, & Schoorman, 1995; Williams, 2001, 2007;Zand, 1972).

An extreme example is offered by the case ofa young professional who complained about herboss in unprofessional terms on Facebook whilehaving him as a contact and, consequently, waspublicly ridiculed by her boss on her FacebookWall and then fired (World News Australia,2009). Because of the nontailored nature of inter-actions in online social networks and the poten-tial for others to comment and provide publicfeedback, open boundary management behav-iors offer little protection for average respect inthe eyes of a professional audience.

Proposition 5a: Open boundary man-agement behaviors are likely to un-dermine average respect among one’sprofessional contacts because of thevast amount of nontailored profes-sional and personal information thatis self-disclosed or revealed by one’sconnections to a broad audience.

Open boundary management behaviors alsohave implications for the degree to which anemployee is liked by his or her professional con-tacts. Because liking is often based on homo-phily, or individuals’ attraction to similar others(Blau, 1977; Byrne, 1971), open behaviors mayincrease liking in the eyes of some professionalcolleagues—for example, when a coworkershares the focal employee’s values. However,research on homophily suggests that organiza-tions are more heterogeneous than friendshipnetworks in that organizations offer fewer op-portunities for choice homophily—the propen-sity to choose similar others—such that profes-sional contacts are likely to be much morediverse in terms of values (McPherson & Smith-Lovin, 1987). While disclosed homophilous val-ues may lead some professional contacts to likeemployees who use open behaviors, dissimilarvalues that are disclosed by employees them-selves or their contacts may lead many profes-sional contacts to like them less (Harrison, Price,& Bell, 1998; Phillips et al., 2009). Therefore, wecontend that while there are some professionalrelationships in which open behaviors do notharm and even increase liking, on average,open behaviors are more likely to undermineliking in the eyes of a broad set of professionalcontacts.

Proposition 5b: Open boundary man-agement behaviors are likely to un-dermine average liking among one’sprofessional contacts because of thevast amount of nontailored informa-tion that is self-disclosed or revealedby one’s connections to a broadaudience.

Consequences of audience behaviors. Audi-ence boundary management behaviors impactthe way one is regarded in several ways. Con-trary to open behaviors, audience behaviors pro-tect employees from losing respect by prevent-ing the overdisclosure of information by


themselves or others to a broad audience. Forinstance, young employees or new graduatesapplying for jobs maintain respect by excludingprofessional contacts from their profiles on net-works that are personal in nature, perhaps be-cause their profiles include provocative partyphotos, so that their visible online behavior mir-rors offline professional norms. In professionalsettings that value discretion, audience behav-iors help employees to be seen as worthy of trustregarding important decisions or sensitive infor-mation (Mayer et al., 1995; Williams, 2001), ratherthan as careless, indiscreet, or unaware ofboundaries. However, since audience behaviorsare based on the segmentation of professionaland personal contacts and on self-verificationmotives, they deprive employees of opportuni-ties to enhance their professional imagethrough broadcasting their competence andachievements to professional contacts. Thus,while audience behaviors help maintain respectat work, they are unlikely to help increase itrelative to the baseline of not participating inonline social networks.

Proposition 6a: Audience boundarymanagement behaviors are likely tomaintain but not increase average re-spect among one’s professional con-tacts because they do not allow one tobroadcast competencies and achieve-ments to broadly relevant workaudiences.

Audience boundary management behaviorshave costs in terms of liking because they canlead employees to exclude some professionalcontacts from the selected audience. Ignoring ordenying connection requests might be sociallyoff-putting or embarrassing because it makesthe boundary highly visible. Despite employees’own preferences for segmenting their profes-sional and personal identities, some of their pro-fessional contacts may expect to connect andshare with them in online social networks(Skeels & Grudin, 2009). When the boundary ismore visible—in particular, in triadic settingswhere professional contacts are connected to amutual acquaintance and can see that the focalemployee who uses audience behaviors has ac-cepted this mutual acquaintance’s connectionrequest and not theirs—they may feel excluded.Further, since employees who use audience be-haviors prefer to keep professional and personal

identities separate, they can miss out on inter-personal connections online that could help en-hance liking in the offline relationship. Ignoringconnection requests or accepting them only innetworks that are professional in nature, suchas LinkedIn, may signal distrust (Lewicki, McAl-lister, & Bies, 1998) and, thus, paint the employeeas cold (Fiske et al., 2007) and less likable.Therefore, we propose that audience behaviorsmay have negative consequences for averageliking.

Proposition 6b: Audience boundarymanagement behaviors are likely toundermine average liking amongone’s professional contacts becausethey make professional contacts feelexcluded from one’s online life.

Consequences of content behaviors. Employ-ees who use content behaviors make consciousefforts to present themselves in a way that islikely to generate approval from members oftheir broad audience. To do so they will typi-cally downplay attributes making them seemless attractive and emphasize attributes leadingto a positive assessment of them by others. De-ciding what content to disclose therefore en-ables employees to manage their professionalidentities, promoting an appearance of compe-tence and suitability for a professional role(Leary & Kowalski, 1990). Thus, unlike audiencebehaviors, content behaviors enable employeesto accept connection requests from professionalcontacts and to share information that might beof interest or helpful to their professional con-tacts. A key benefit to this approach is promot-ing oneself to a broad professional audience bychoosing content that is interesting, appealing,or indicative of expertise or high status—for ex-ample, flattering family pictures, impressiveachievements, or glamorous activities (Schlenker& Weingold, 1992).

Further, unlike open behaviors, content be-haviors entail a careful monitoring of the infor-mation disclosed by oneself and others, therebyenabling employees to keep potential mis-matches between their beliefs and values andthose of their professional contacts private.Thus, in professional settings that value limiteddisclosure of information, content boundarymanagement behaviors enable employees tomirror their offline behavior and to interact in asocially appropriate way (Van Maanen &


Schein, 1979). Therefore, content behaviors arelikely to enhance average respect among one’sprofessional contacts.

Proposition 7a: Content boundarymanagement behaviors are likely toincrease average respect among one’sprofessional contacts because they al-low one to broadcast self-enhancinginformation to a broad set of workaudiences.

In addition, because content behaviors enableemployees to share personal information—evenif it is carefully monitored—they can enhanceliking of employees by their professional con-tacts (Collins & Miller, 1994; Cozby, 1973). Pro-vided that the exchange of information remainsappropriate, sharing details about each other’spersonal lives, thoughts, and desires indicatestrust and allows individuals to discover com-mon perspectives (Jourard, 1959; Jourard & Lasa-kow, 1958; Worthy et al., 1969). Moreover, em-ployees may benefit from boundary crossoversthat lead to reciprocal exchanges of information,thereby enriching professional relationships(Rothbard, 2001; Rothbard & Ramarajan, 2009).For instance, an employee may build strongerbonds with long-distance coworkers by sharingselected professional and personal informationwith them and giving positive feedback—for in-stance, by regularly “Liking” their posts or com-menting on them. The fact that they interact witheach other not only on a professional basis butalso on a personal one may help build greatercloseness and liking.

Proposition 7b: Content boundarymanagement behaviors are likely toincrease average liking among one’sprofessional contacts because they al-low one to broadcast personal andprofessional information to a broadset of work audiences.

Consequences of hybrid behaviors. Hybridboundary management behaviors, driven byboth self-enhancement motives and segmenta-tion preferences, may help individuals span do-main boundaries when role identities are com-patible and yet keep information separate whenprofessional and personal identities are incom-patible (Rothbard & Ramarajan, 2009). As a re-sult, employees may be able to customize thematch between content and audience, which

should maximize respect and liking (Phillips etal., 2009). Because employees are aware of whatcontent they share with which audience, hybridbehaviors can help alleviate online social net-works’ challenges by encouraging more tailoreddisclosures. In particular, hybrid behaviorsmake invisible audiences more salient and limitthe potential for one’s contacts to provide inap-propriate information across domains, since adisclosure would be seen only by contacts forwhom the initial information was intended. Hy-brid behaviors thus enable employees to bestmirror the tailored nature of offline relation-ships, where some professional relationships—for instance, with friendly coworkers—are lessformal and allow for more personal disclosurethan others—for instance, with supervisors, sub-ordinates, or customers. Hybrid behaviors thuscombine the benefits of positive professionalimpression management (Roberts, 2005) withcareful and personalized segmentation of pro-fessional and personal identities (Ashforth etal., 2000) and, as such, are likely to increase bothaverage respect and liking among professionalcontacts.

Proposition 8a: Hybrid boundary man-agement behaviors are likely to in-crease average respect among one’sprofessional contacts because of thesharing of self-enhancing information.

Proposition 8b: Hybrid boundary man-agement behaviors are likely to in-crease average liking among one’sprofessional contacts because of thetailored disclosure of personal andprofessional information.

Trade-offs Between Online BoundaryManagement Behaviors

So far we have proposed that content and hy-brid behaviors potentially have more positiveconsequences for average respect and liking byprofessional contacts than open and audiencebehaviors. However, employees using contentand hybrid behaviors need to keep potentialmismatches between their beliefs and valuesand those of their professional contacts private(Sedikides & Strube, 1995). This requires ongoingeffort because the underlying philosophy andtechnical structure of online social networks en-courages unrestricted and transparent sharing


of information as the norm and default (Boyd,2008). Open behaviors are very easy to use, andaudience behaviors do not require much skill inthat they only demand a series of one-time de-cisions about whether to include a contact ornot. In contrast, content and hybrid behaviorsrequire ongoing effort and skillful decision mak-ing about what content to share. And hybridbehaviors also entail decisions about withwhom to share what content, requiring that in-dividuals invest effort and skill in setting upand maintaining lists and privacy settings todelineate their separate audiences. Moreover,content and hybrid behaviors can be under-mined by one’s own mistakes as well as un-wanted disclosures by one’s connections, whichcan decrease respect and liking in the eyes ofprofessional contacts if the information dis-closed is incongruent with the self-enhancedimage one has been promoting such that oneappears inauthentic (Brumbaugh, 1971).

To cope with the complexity of executing con-tent and hybrid behaviors effectively and to cap-italize on the potential benefits offered by themfor respect and liking, we contend that employ-ees need online boundary management capa-bilities, which consist of the time, effort, andtechnical skill required to avoid the accidentaldisclosure of too much or inappropriate contentto professional contacts. Consistent with thestrategic management literature (e.g., Nelson &Winter, 1982; Teece, Pisano, & Shuen, 1997), weuse the term capabilities to capture not onlytechnical ability but also required resources(i.e., time and energy). We argue that the effec-tiveness of content and hybrid behaviors de-pends on these capabilities such that comparedwith open and audience behaviors, content andhybrid behaviors have greater potential for in-creasing average respect and liking yet mayactually decrease respect and liking if employ-ees are low in online boundary managementcapabilities.

Content behaviors may expose employees toaccidental overdisclosure of personal informa-tion. Overdisclosure may occur when employeesinadvertently reveal information to the wrongprofessional contacts because they have forgot-ten about the invisible audience (Boyd, 2007). Inoffline interactions individuals tend to discloseintimate information only once they haveformed a dyadic boundary with someone (Der-lega & Chaikin, 1977), ensuring that information

disclosed to the person will not be leaked tomutual acquaintances. In online social net-works, however, employees may mistakenlythink they are disclosing information to only aspecific person (while answering a question onsomeone’s Facebook Wall, for instance). Numer-ous incidents illustrate poor online boundarymanagement capabilities. A particularly color-ful example was when a fraud fugitive was ar-rested because he boasted on Facebook aboutenjoying himself in a luxury resort, the locationof which he disclosed, forgetting that policeagents were part of the audience (BBC News,2009). In addition, personal contacts can acci-dentally or purposefully violate the boundarybetween the employee’s professional and per-sonal identities. Skeels and Grudin (2009: 101)mention an informant whose personal friends,in an effort at humor, “poked” her manager onFacebook, which she felt was embarrassing andmay have lessened her manager’s respect forher. Thus, online boundary management capa-bilities may well extend to managing one’s con-nections by, for instance, restricting the abilityof others to post information on one’s profile orby educating one’s personal contacts not to vio-late the boundary.

Compared with content behaviors, hybrid be-haviors are costlier in terms of the time, effort,and skill required. In essence, hybrid behaviorsare about pairing the appropriate content withthe corresponding audience (either professionalor personal). To enact hybrid behaviors effec-tively, employees need to be familiar with pri-vacy settings in the given online social networkand be able to set them up and manage themover time (Zhao et al., 2008). They must make theeffort to devise and implement the chosen be-haviors (e.g., assign connections to sublists) andthen make myriad small decisions on an ongo-ing basis to pair information with ties, and theyalso need to adjust their behaviors to frequentwebsite changes (Strater & Lipford, 2008). Fur-ther, as employees’ career or life stages changeand they gather more online contacts, changejobs, move organizations, or perhaps switch in-dustries, the way they have segmented theironline audiences may no longer be ideal (e.g., aformer peer could become a subordinateor boss).

In sum, monitoring and managing the contentthat is available to different audiences may re-quire considerable time, effort, and technical


skill, and, thus, the positive impact of hybridbehaviors on respect and liking depends on em-ployees’ online boundary management capabil-ities. Mistakes easily happen, such as employ-ees accidentally posting negative commentsabout a coworker in another department (thatthe coworker can see or hear about from some-one else) or accidentally posting family picturesto the wrong sublist of contacts, thus disclosingmore than intended and prompting their profes-sional contacts to realize that the employeesprobably share more online than they thought,yet not with them. In addition to mistakes, fauxpas also occur when employees’ contacts sharepictures or comments that are dissonant withwhat the employees themselves share. Such in-cidents decrease professional contacts’ respectfor employees if they reveal behavior thatdoes not mirror the contacts’ expectations (e.g.,inappropriate disclosure) or makes the employ-ees appear inauthentic (Brumbaugh, 1971; Hal-pin & Croft, 1966). These mistakes can also de-crease liking by signaling distrust and makingprofessional contacts feel excluded from thepersonal online world of the employee, much aswith the audience behaviors (Lewicki etal., 1998).

Proposition 9a: Employees’ onlineboundary management capabilitiesmoderate the consequences of contentand hybrid behaviors such that con-tent and hybrid behaviors will in-crease average respect and liking ifemployees are high in these capabil-ities but decrease average respect andliking if they are low in thesecapabilities.

In addition, because the goal of online socialnetworks is to connect people, and their primarymodus operandi is therefore to integrate indi-viduals’ social worlds (Donath & Boyd, 2004),identity segmentation in online social networksoften requires more time, effort, and skill thanidentity integration. As a result, employees whowant to segment their personal and professionalidentities in online social networks must devoteenergy and develop knowledge to make the on-line tools work for them.

Proposition 9b: Hybrid behaviors re-quire greater online boundary man-

agement capabilities than contentbehaviors.

Extending the Framework: The DynamicNature of Online Boundary Management

We have argued thus far that employees’ on-line boundary management behaviors are rela-tively stable in given social contexts, such asspecific phases of the career and life cycles (Lit-tle, 1983). An important aspect of this is that onceemployees begin to enact a set of online bound-ary management behaviors, path dependenciesand, eventually, inertia (e.g., Sydow, Schreyögg,& Koch, 2009) are established as a result of themacro choices they make about whom their con-tacts should be and what information theyshould disclose to whom. For example, havingaccepted connections with professional contactsmakes it difficult to then exclude them if oneshifts to audience behaviors; removing a con-nection may hurt liking in the relationship, andclosing one’s account in order to remove connec-tions before starting again may raise questionsand suspicions. Also, having disclosed or hav-ing let others disclose inappropriate informa-tion makes it difficult to shift to content behav-iors unless one goes back to clean up one’sprofile and asks connections to clean up theircomments too.

However, we also suggest that the macrochoices employees make based on their self-evaluation motives and on their preferences forsegmentation or integration of their profes-sional and personal identities are not set instone but, rather, evolve and shift over time. Thetemporal nature of online boundary manage-ment is likely to follow a pattern of punctuatedequilibrium, in which employees engage in thesame basic set of behaviors for multiple weeks,months, and perhaps even years, but then shifttheir behavior in response to significant eventsor changes in their lives. Thus, we suggest thatit takes concerted time, thought, and effort toshift one’s online boundary management behav-iors and that this is not likely to happen on aday-to-day basis. What, then, prompts employ-ees to reexamine their macro choices and corre-sponding online boundary managementbehaviors?

First, there are important turning points thatpeople encounter, stemming from significantchanges to their roles, careers, or life stages or


from critical events that are likely to compelthem to reexamine their online boundary man-agement. A clear example of this is when stu-dents apply for jobs. They often shift from asituation where the majority of their interactionsare characterized by self-verification motives toone where self-enhancement is the dominantmotive. It may not be until they have passed the“qualifying” stage of obtaining a new job, set-tling into the company, and feeling comfortablewith others that they shift back to a state wheretheir self-verification motives are dominantagain. In this scenario we would suggest thatindividuals are likely to change their online be-haviors from either open or audience behaviorsto either content or hybrid behaviors, dependingon their preferences for integration or segmen-tation. Another example of how going through aparticular career or life stage alters one’s onlinebehaviors is employees who have opted formore time-consuming behaviors, such as hybridbehaviors, making conscious decisions tochange to content behaviors, which are simplerand easier to use, when they enter a career orlife stage in which they lack time. Such a shiftmay be reversed when they regain enough timeto act in line with their full preferences. In ad-dition, critical events can make employeesaware of the need to manage boundaries in on-line social networks. For instance, an employeemay realize that she has been disclosing moreinformation than intended, such as the songsshe was listening to on the Internet or photosfrom long ago. Likewise, witnessing a colleagueor friend’s online faux pas might prompt em-ployees to revisit their macro choices so thatthey disclose less information (content behav-iors) or disclose to a more targeted audience(audience or hybrid behaviors).

Second, the feedback employees receive fromtheir professional and personal contacts alsotriggers shifts in macro choices. Such feedbackis provided through processes of (1) iterativeidentity negotiation (Swann et al., 2009) and (2)enforcement of organizational and occupationalnorms (Bloor & Dawson, 1994; Pratt et al., 2006).While online social networks offer new opportu-nities for employees to make identity claims,these claims are made in a public or semi-public context where others assess their reli-ability and provide feedback accordingly (Boyd,2007). Thus, online boundary management istypically not a one-way imposition of the self’s

desires on the audience but, rather, an iterativeprocess that is negotiated between self and au-dience over time. Since clarity, continuity, andcompatibility are important premises of suc-cessful identity negotiation (Swann et al., 2009),employees may, for instance, tone down exag-gerated self-enhancement behaviors if their pro-fessional contacts make comments that under-mine or fail to support their impressionmanagement efforts. In contrast, employeesmay disclose more because of interaction rituals(Collins, 1981, 2004) in which sharing personalinformation and receiving positive feedbackabout it, such as “Likes” and encouraging com-ments from professional contacts, generate emo-tional energy.

In addition, feedback stemming from organi-zational and occupational norms (Bloor & Daw-son, 1994; Pratt et al., 2006) interacts with em-ployees’ individual preferences as they chooseonline boundary management behaviors. In par-ticular, employees may tone down their self-verification strivings if they receive feedback inthe workplace that information they had dis-closed violated organizational norms. Similarfeedback processes can lead employees to shiftfrom integration behaviors to more segmentedones, or vice versa. Employees might reviseopen boundary management behaviors thathave backfired and transition to audience be-haviors where they can self-verify but with asegmented audience. Or, conversely, they mighttransition from audience to content or hybridbehaviors if they become aware of strong normsto be connected online after receiving profes-sional connection requests. Skeels and Grudin(2009: 101) cite such a situation where an infor-mant felt he had to accept customers’ connectionrequests on Facebook and subsequently startedmonitoring his information disclosure. Last,norms around technology, particularly aroundwhat is appropriate to disclose to whom in agiven industry or workplace, may evolve andeven be codified into formal rules, triggeringreexaminations of employees’ online boundarymanagement behaviors.

In sum, despite their stability and inertia inthe relative short term, online boundary man-agement behaviors may be dynamic when ex-amined over a longer term. Employees may shifttheir behaviors because of pivotal turningpoints, feedback, or norms. In turn, these shiftsare likely to come with important implications


for the average respect and liking employeesreceive from their professional contacts. Whileaverage respect and liking may decreaseabruptly because of an accidental egregiousdisclosure by employees who lack the capabili-ties to effectively execute content or hybrid be-haviors, average respect and liking may alsochange more gradually over time as employees’connections respond to their behaviors online.

DISCUSSION

Online social networks pose new, complex,and often challenging opportunity structures forenacting professional and personal identities.How employees manage boundaries withinthese structures may help or harm their profes-sional relationships in important ways. We haveproposed a conceptual framework of onlineboundary management, theorizing about thekey underlying drivers of four archetypal onlineboundary management behaviors and the con-sequences of these behaviors for average re-spect and liking among employees’ professionalcontacts. In so doing, we contribute to existingtheory and research on boundary work (Ashforthet al., 2000; Nippert-Eng, 1995; Rothbard et al.,2005) and self-evaluation motives that guideidentity negotiation (Brown, 1990; Sedikides &Strube, 1995; Swann, 1983; Wiesenfeld et al.,2007) by extending both to the virtual world ofcyberspace, where these processes are shapedby dynamics different from those in physicalspace. In particular, this framework guides fu-ture research on boundary work and identitynegotiation in a social space characterized byopen disclosure to broad audiences and publicor semi-public feedback (Boyd, 2007; Donath,2007; Donath & Boyd, 2004), rather than the tai-lored disclosure and private feedback that char-acterize interactions in physical space. In addi-tion, by examining employees’ boundarypreferences and self-evaluation motives to-gether, this framework highlights the double-edged sword of online boundary management—although all four combinations in the 2 ! 2framework offer potential benefits, none of thefour is free of significant risk, challenge, ordifficulty.

Further, this framework opens up new lines ofresearch on the dynamics between liking andrespect or warmth and competence more gener-ally (Fiske et al., 2007; Hamilton & Fallot, 1974;

Prestwich & Lalljee, 2009; Wojciszke et al., 2009).Our framework is consistent with past workshowing how it is often difficult for individualsto be seen as both warm and competent at thesame time (Cuddy et al., 2011; Fiske et al., 2002,2007), but it also sheds new light on how thisnotion may be further complicated when itcomes to online boundary management.Whereas past work on social judgments in phys-ical space has suggested that the trade-off be-tween warmth and competence can be over-come by sending nonverbal cues to signal bothwarmth and competence (Cuddy et al., 2011), ourframework suggests that online social networksthat are personal in nature present opportuni-ties for sending both warmth and competencecues to professional contacts through the infor-mation one discloses. However, employees mayonly be able to do so through content and hybridbehaviors, which are effective only when theyare implemented with few mistakes. This sug-gests that employees’ online boundary manage-ment capabilities are critical for these behav-iors to successfully enhance both liking andrespect. Compared to hybrid and content behav-iors, audience and open behaviors are less ben-eficial overall, in that audience behaviors likelyhave neutral outcomes for average respect andnegative outcomes for average liking, whereasopen behaviors likely have negative outcomesfor both average respect and liking because ofthe risks they entail. A key insight derived fromour framework is that hybrid behaviors, whenwell executed, are the most likely to maximizeboth liking and respect, since hybrid behaviorsenable employees to more closely approximatethe customized disclosure prevalent in offlineinteractions by sending both warmth and com-petence cues tailored to their professionalcontacts.

Self-verification and self-enhancement mo-tives are key in shaping whether employeeswho participate in online social networks standto increase respect and liking among their pro-fessional contacts, in that behaviors driven byself-verification (open and audience) may beless likely to increase respect and liking thanbehaviors driven by self-enhancement (contentand hybrid). However, preferences for segmen-tation versus integration are key in determiningthe ratio between risk and difficulty involved inonline boundary management, in that the twobehaviors stemming from a preference for seg-


mentation (audience and hybrid) may be lessrisky but more difficult to maintain over timecompared to the two behaviors stemming from apreference for integration (open and content).This is because the technical structure of onlinesocial networks is typically geared toward inte-gration behaviors such that segmentation re-quires an ongoing set of macro choices regard-ing the contacts with whom one is connected.This is also because initial macro choices createinertia and path dependencies, which may bedifficult to revise as employees experiencechanges in their professional and personal livesor try to revisit their online boundary manage-ment behaviors to address social feedback andworkplace norms.

Future Research

The framework we have proposed opens upimportant new issues for researchers consider-ing boundary work, identity negotiation, and re-spect and liking in cyberspace. First, it may befruitful to apply our framework at the dyadiclevel. For instance, research is needed on therole that social exchange and reciprocity norms(Gouldner, 1960) may play in shaping respectand liking in a dyad, as well as the impact onrespect and liking in a dyad when an employeeusing one set of behaviors, such as open behav-iors, interacts with an employee using anotherset, such as content behaviors. This may be par-ticularly interesting for key dyads, such as thesupervisor-subordinate dyad, which is impor-tant for supervisory performance ratings(Lefkowitz, 2000; Robbins & DeNisi, 1994) andleader-member exchange (Dienesch & Liden,1986; Wayne & Ferris, 1990). In addition, futureresearch should pay attention to triadic andgroup dynamics in online social networks,which can impact respect and liking, as dis-cussed in the example of an employee’s bossbeing “poked” by her personal friends. For in-stance, it is likely that in online social networksthat are personal in nature, the pattern of con-nection requests being sent and accepted in ateam provides information on cohesion and in-group and outgroup dynamics. Furthermore, inaddition to average respect and liking, futureresearch should investigate the potential vari-ance in respect and liking that online boundarymanagement behaviors may produce. For in-stance, some behaviors, such as audience or

open behaviors, may polarize professional con-tacts’ respect and liking for an employee.

Second, our premise that employees need tomirror their offline professional relationshipswhen online prompts further examination ofboundary work across the different socialspaces in which individuals interact. In partic-ular, occupational and organizational normsmay interact with individual preferences (Stutz-man & Hartzog, 2012). Some occupations, such asjournalism, entail a greater need to collect infor-mation from a broad variety of sources and tomonitor the information one discloses, whereasothers, such as real estate agents, novelists, orself-employed freelance consultants, involve agreater need to disseminate information and ad-vertise one’s accomplishments. Since content orhybrid boundary management behaviors mightbe better suited to this latter set of occupationsthan audience or open behaviors, employeeswho recognize these norms may either adapttheir online boundary management behaviorsor change occupations if the conflict with theirpreferences is too strong. In contrast, some oc-cupational norms require strict professionalism,where audience behaviors may be optimal. Forinstance, physicians are advised to not breakthe physician-patient boundary in online socialnetworks (McLean Hospital, 2010), and employ-ees of the U.S. Department of Justice are warnedagainst connecting with judges, defense coun-sels, jurors, and witnesses, as well as discuss-ing a case in online social networks (U.S. De-partment of Justice, 2010). Future research couldextend the focus of our framework on individualpreferences and investigate how organizationaland occupational norms might interact with in-dividual preferences, perhaps depending on thestrength of employees’ professional and organi-zational identification (Ashforth & Mael, 1989;Kreiner, Hollensbe, & Sheep, 2006; Van Maanen& Schein, 1979) and on the relative salience oftheir professional and personal identities (Ash-forth et al., 2000; Rothbard & Edwards, 2003).Also, while our framework focuses on onlinesocial networks that are personal in nature,boundary management may also be needed onprofessional collaboration sites, such as Yam-mer, where employees may choose to keep theirinteractions strictly professional (i.e., segmenta-tion) or to engage in some level of personaldisclosure (i.e., integration).


Third, future research may be able to providefurther nuance to our framework by consideringthe intensity with which employees engage inonline social networks. Indeed, employees mayexperience the need to self-enhance or self-verify with greater or lesser intensity, and theymay have stronger or weaker preferences forsegmentation or integration of their profes-sional and personal identities. Thus, employeeswho experience these motives with low inten-sity may enact passive and possibly less consis-tent online boundary management behaviors,whereas employees who experience these mo-tives with high intensity might be torn betweenself-enhancement and self-verification or be-tween segmentation and integration and be un-able to satisfy all their motives simultaneously.These employees may experience ambivalenceand devise even more integratively complex(Tetlock, 1983) ways to manage their boundariesin the effort to self-enhance and self-verify or tosegment and integrate their identities.

Fourth, future research should examine otherboundaries, such as hierarchical or age bound-aries, that are likely to require careful manage-ment, since they, too, collide in online socialnetworks. Hierarchical boundaries are typicallyinstitutionalized—for instance, by physical bar-riers at the workplace, such as distinct floors orareas for different types of employees, or bynorms that restrict socializing to members ofone’s own rank (Ashforth et al., 2000; Goffman,1956). This institutionalization does not extend toonline social networks such that professionalcontacts from other ranks (supervisors or subor-dinates) may create unique challenges for on-line boundary management (Boyd, 2007; Ryan,2008; Tufekci, 2008)—for example, perhaps theymight use online social networks to crossboundaries that are otherwise well established(Lewis et al., 2008). Also, social norms regardingacceptable disclosure of information in onlinesocial networks might differ according to one’sgeneration and/or life stage (Pfeil, Arjan, &Zaphiris, 2009).

Last, while technology is likely to change, thepsychological motives and challenges that peo-ple face given the realities of this new onlineworld are fairly fundamental. Since the purposeof social media is to connect people and shareinformation, the default technical structure andphilosophy of social media platforms hasbeen—and appears will continue to be—disclo-

sure to as broad an audience as possible (Boyd,2007, 2008; Donath, 2007; Donath & Boyd, 2004).Thus, participating in online social networkswill involve tension between disclosure and dis-cretion (Petronio, 2002) and between the profes-sional and personal, regardless of the specificsof the technological platforms. The technologyitself may evolve, changing the specific tacticsemployees use, but the underlying motives andarchetypical behaviors we have outlined willlikely endure. As social media technology devel-ops, these questions should be revisited, andnew opportunities for research on online bound-ary management may be created.

Practical Implications and Conclusion

Our framework of online boundary manage-ment offers practical implications for employeesand organizations. Being able to create andmaintain appropriate boundaries and to negoti-ate one’s identities online are quickly becomingcritical skills that most employees now need tomaster. Our framework highlights that employ-ees are faced with a menu of different boundarymanagement options, which may have positiveor negative consequences for the way they areregarded by professional contacts and whichentail varying degrees of risk, opportunity, anddifficulty. Online boundary management mayeven lead employees to revisit offline interac-tions at the workplace. For instance, they mayreconsider who they view as trustworthy or notdiscuss online social networks with coworkersfrom whom they would not want to receive anonline connection request. Another implicationof our framework is that rather than assume thattheir own choices are shared by others (Ross etal., 1977), employees and employers should beaware that others’ online behaviors might bedriven by motives and circumstances differentfrom their own. To help employees with onlineboundary management, organizations may con-sider implementing policies regarding onlineconnections and interactions with professionalcontacts, as well as provide training to supportemployees as they develop online boundarymanagement capabilities. As the vast amount ofsocial interaction that already occurs in cyber-space continues to grow, employees will inevi-tably encounter collisions of their professionaland personal identities online. We hope that ourframework serves as a spark and guide for fu-


ture research on how employees manage theseinevitable—yet highly consequential— colli-sions of their social worlds within the relativelynew world of social media.

REFERENCES

Alexander, C. N., Jr., & Weil, H. G. 1969. Players, persons, andpurposes: Situational meaning and the prisoner’s di-lemma game. Sociometry, 32: 121–144.

Anseel, F., & Lievens, F. 2006. Certainty as a moderator offeedback reactions? A test of the strength of the self-verification motive. Journal of Occupational and Orga-nizational Psychology, 79: 533–551.

Ashforth, B. E., Kreiner, G. E., & Fugate, M. 2000. All in a day’swork: Boundaries and micro role transitions. Academyof Management Review, 25: 472–491.

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

Baumeister, R. F., Tice, D. M., & Hutton, D. G. 1989. Self-presentational motivations and personality differencesin self-esteem. Journal of Personality, 57: 547–578.

BBC News. 2009. Fraud fugitive in Facebook trap. http://news.bbc.co.uk/2/hi/americas/8306032.stm; accessed March 18,2012.

Blau, P. M. 1977. Inequality and heterogeneity. New York:Free Press.

Bloor, G., & Dawson, P. 1994. Understanding professionalculture in organizational context. Organization Studies,15: 275–295.

Bohnert, D., & Ross, W. H. 2010. The influence of social net-working web sites on the evaluation of job candidates.Cyberpsychology, Behavior, and Social Networking, 13:341–347.

Boyd, D. 2007. Social network sites: Public, private, or what?Knowledge Tree, 13: http://kt.flexiblelearning.net.au/tkt2007/?page_id#28.

Boyd, D. 2008. Taken out of context: American teen socialityin networked publics. Doctoral dissertation, School ofInformation, University of California, Berkeley. http://www.danah.org/papers/TakenOutOfContext.pdf; ac-cessed May 4, 2010.

Brass, D. 1984. Being in the right place: A structural analysisof individual influence in organizations. AdministrativeScience Quarterly, 29: 518–539.

Brown, J. D. 1990. Evaluating one’s abilities: Shortcuts andstumbling blocks on the road to self-knowledge. Journalof Experimental Social Psychology, 26: 149–167.

Brown, J. D. 1991. Accuracy and bias in self-knowledge. InC. R. Snyder & D. F. Forsyth (Eds.), Handbook of socialand clinical psychology: The health perspective: 158–178. New York: Pergamon Press.

Brumbaugh, R. B. 1971. Authenticity and theories of admin-istrative behavior. Administrative Science Quarterly, 16:107–112.

Burt, R. S. 2004. Structural holes and good ideas. AmericanJournal of Sociology, 110: 349–399.

Byrne, D. 1971. The attraction paradigm. New York: AcademicPress.

Chreim, S., Williams, B. E., & Hinings, C. R. 2007. Interlevelinfluences on the reconstruction of professional roleidentity. Academy of Management Journal, 50: 1515–1539.

Collins, N. L., & Miller, L. C. 1994. Self-disclosure and liking:A meta-analytic review. Psychological Bulletin, 116: 457–475.

Collins, R. 1981. On the microfoundations of macrosociology.American Journal of Sociology, 86: 984–1014.

Collins, R. 2004. Interaction ritual chains. Princeton, NJ:Princeton University Press.

Cozby, P. C. 1973. Self-disclosure: A literature review. Psy-chological Bulletin, 79: 73–91.

Cuddy, A. J. C., Glick, P., & Beninger, A. 2011. The dynamicsof warmth and competence judgments, and their out-comes in organizations. Research in Organizational Be-havior, 31: 73–98.

Derlega, V. L., & Chaikin, A. L. 1977. Privacy and self-disclosure in social relationships. Journal of Social Is-sues, 33: 102–115.

Dienesch, R. M., & Liden, R. C. 1986. Leader-member ex-change framework of leadership: A critique and furtherdevelopment. Academy of Management Review, 10: 527–539.

DiMicco, J. M., & Millen, D. R. 2007. Identity management:Multiple presentations of self in Facebook. In GROUP‘07: Proceedings of the 2007 International ACM Confer-ence on Supporting Group Work: 383–386. New York:Association of Computing Machinery.

Donath, J. 2007. Signals in social supernets. Journal of Com-puter-Mediated Communication, 13: 231–251.

Donath, J., & Boyd, D. 2004. Public displays of connection. BTTechnology Journal, 22(4): 71–82.

Dumas, T. L., Phillips, K. W., & Rothbard, N. P. In press.Getting closer at the company party: Integration expe-riences, racial dissimilarity and workplace relation-ships. Organization Science.

Dutta, S. 2010. What’s your personal social media strategy?Harvard Business Review, 88(11): 127–130.

Edmondson, A. 1999. Psychological safety and learning be-havior in work teams. Administrative Science Quarterly,44: 350–383.

Edwards, J. R., & Rothbard, N. P. 2000. Mechanisms linkingwork and family: Clarifying the relationship betweenwork and family constructs. Academy of ManagementReview, 77: 85–129.

Ellison, N. B., Steinfield, C., & Lampe, C. 2007. The benefits ofFacebook “friends”: Social capital and college students’use of online social network sites. Journal of Computer-Mediated Communication, 12: 1143–1168.


Fiske, S. T., Cuddy, A. J., & Glick, P. 2007. Universal dimen-sions of social cognition: Warmth and competence.Trends in Cognitive Sciences, 11(2): 77–83.

Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. 2002. A model of(often mixed) stereotype content: Competence andwarmth respectively follow from status and competition.Journal of Personality and Social Psychology, 82: 878–902.

Fiske, S. T., Xu, J., Cuddy, A. J. C., & Glick, P. 1999. (Dis)re-specting versus (dis)liking: Status and interdependencepredict ambivalent stereotypes of competence andwarmth. Journal of Social Issues, 55: 473–491.

Forest, A. L., & Wood, J. V. 2012. When social networking isnot working: Individuals with low self-esteem recognizebut do not reap the benefits of self-disclosure on Face-book. Psychological Science, 23: 295–302.

Goffman, E. 1956. Embarrassment and social organization.American Journal of Sociology, 62: 264–271.

Goffman, E. 1959. The presentation of self in everyday life.Garden City, NY: Doubleday Anchor Books.

Gouldner, A. W. 1960. The norm of reciprocity: A preliminarystatement. American Sociological Review, 25: 161–178.

Granovetter, M. S. 1973. The strength of weak ties. AmericanJournal of Sociology, 78: 1360–1380.

Gruenfeld, D. H., Mannix, E. A., Williams, K. Y., & Neale, M. A.1996. Group composition and decision making: Howmember familiarity and information distribution affectprocess and performance. Organizational Behavior andHuman Decision Processes, 67: 1–15.

Halpin, A. W., & Croft, D. B. 1966. The organizational climateof schools. In A. W. Halpin (Ed.), Theory and research inadministration: 131–249. New York: Macmillan.

Hamilton, D. L., & Fallot, R. D. 1974. Information salience asa weighting factor in impression formation. Journal ofPersonality and Social Psychology, 30: 444–448.

Harrison, D. A., Price, K. H., & Bell, M. P. 1998. Beyond rela-tional demography: Time and the effect of surface-versus deep-level diversity on group cohesiveness.Academy of Management Journal, 41: 96–107.

Hixon, J. G., & Swann, W. B., Jr. 1993. When does introspectionbear fruit? Self-reflection, self-insight, and interper-sonal choices. Journal of Personality and Social Psychol-ogy, 64: 35–43.

Jehn, K. A., & Shah, P. P. 1997. Interpersonal relationshipsand task performance: An examination of mediatingprocesses in friendship and acquaintance groups. Jour-nal of Personality and Social Psychology, 72: 775–790.

John, O. P., & Robins, R. W. 1994. Accuracy and bias inself-perception: Individual differences in self-enhance-ment and the role of narcissism. Journal of Personalityand Social Psychology, 66: 206–219.

Jourard, S. M. 1959. Self-disclosure and other-cathexis. Jour-nal of Abnormal and Social Psychology, 59: 428–431.

Jourard, S. M., & Lasakow, P. 1958. Some factors in self-disclosure. Journal of Abnormal and Social Psychology,56: 91–98.

Ketchum Global Research Network for Liberty Mutual. 2011.The Responsibility Project. http://responsibility-project.libertymutual.com/infographics/rp-survey-online-etiquette-the-workplace#fbid#UNJZtmX1Lts; accessedNovember 10, 2012.

Kossek, E. E., & Lautsch, B. 2012. Work-family boundary man-agement styles in organizations: A cross-level frame-work. Organizational Psychology Review, 2: 152–171.

Kossek, E. E., Noe, R. A., & DeMarr, B. J. 1999. Work-familyrole-synthesis: Individual and organizational determi-nants. International Journal of Conflict Management, 10:102–129.

Kreiner, G. E. 2006. Consequences of work-home segmenta-tion or integration: A person-environment fit perspec-tive. Journal of Organizational Behavior, 27: 485–507.

Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. 2006. On theedge of identity: Boundary dynamics at the interface ofindividual and organizational identities. Human Rela-tions, 59: 1315–1341.

Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. 2009. Balancingborders and bridges: Negotiating the work-home inter-face via boundary work tactics. Academy of Manage-ment Journal, 52: 704–730.

Kwang, T., & Swann, W. B., Jr. 2010. Do people embracepraise even when they feel unworthy? A review of crit-ical tests of self-enhancement versus self-verification.Personality and Social Psychology Review, 14: 263–280.

Lampinen, A., Tamminen, S., & Oulasvirta, A. 2009. All mypeople right here, right now: Management of group co-presences on a social networking site. In GROUP ’09:Proceedings of the 2009 International ACM Conferenceon Supporting Group Work: 281–290. New York: Associ-ation of Computing Machinery.

Leary, M. R., & Kowalski, R. M. 1990. Impression manage-ment: A literature review and two-component frame-work. Psychological Bulletin, 107: 34–47.

Lefkowitz, J. 2000. The role of interpersonal affective regardin supervisory performance ratings: A literature reviewand proposed causal framework. Journal of Occupa-tional and Organizational Psychology, 73: 67–85.

Lewicki, R. J., McAllister, D. J., & Bies, R. J. 1998. Trust anddistrust: New relationships and realities. Academy ofManagement Review, 23: 38–458.

Lewis, K., Kaufman, J., & Christakis, N. 2008. The taste forprivacy: An analysis of college student privacy settingsin an online social network. Journal of Computer-Mediated Communication, 14: 79–100.

Lincoln, J. R., & Miller, J. 1979. Work and friendship ties inorganization: A comparative analysis of relational net-works. Administrative Science Quarterly, 24: 181–199.

Little, B. R. 1983. Personal projects. Environment and Behav-ior, 15: 273–309.

Martin, K. E. 2009. Privacy revisited: From Lady Godiva’sPeeping Tom to Facebook’s Beacon Program. Paper pre-sented at the annual meeting of the Academy of Man-agement, Chicago.


Mayer, R. C., Davis, J. H., & Schoorman, F. D. 1995. An inte-grative framework of organizational trust. Academy ofManagement Review, 20: 709–734.

McGregor, I., & Little, B. R. 1998. Personal projects, happi-ness, and meaning: On doing well and being yourself.Journal of Personality and Social Psychology, 74: 494–512.

McLean Hospital. 2010. Breaking the physician patientboundaries on Facebook. http://www.mclean.harvard.edu/news/press/current.php?kw#breaking-the-physician-patient-boundaries-on-facebook&id#149;accessed March 18, 2012.

McPherson, J., & Smith-Lovin, L. 1987. Homophily in voluntaryorganizations: Status distance and the composition offace-to-face groups. American Sociological Review, 52:370–379.

Mehrabian, A. 1971. Silent messages. Belmont, CA: Wads-worth.

Millenial Branding. 2012. Gen Y and Facebook white paper.http://millennialbranding.com/white-paper-purchase/.

Mischel, W. 1973. Toward a cognitive social learning recon-ceptualization of personality. Psychological Review, 80:252–283.

Morrison, E. W., & Bies, R. J. 1991. Impression management inthe feedback-seeking process: A literature review andresearch agenda. Academy of Management Review, 16:522–541.

Nahapiet, J., & Ghoshal, S. 1998. Social capital, intellectualcapital, and the organizational advantage. Academy ofManagement Review, 23: 242–266.

Nelson, R. R., & Winter, S. G. 1982. Organizational capabili-ties and behavior. In R. R. Nelson & S. G. Winter (Eds.),An evolutionary theory of economic change: 96 –136.Cambridge, MA: Harvard University Press.

Nippert-Eng, C. E. 1995. Home and work: Negotiating bound-aries through everyday life. Chicago: University of Chi-cago Press.

Paulhus, D. L. 1998. Interpersonal and intrapsychic adaptive-ness of trait self-enhancement: A mixed blessing? Jour-nal of Personality and Social Psychology, 74: 1197–1208.

Pelham, B. W., & Swann, W. B., Jr. 1994. The juncture ofintrapersonal and interpersonal knowledge: Self-certainty and interpersonal congruence. Personality andSocial Psychology Bulletin, 20: 349–357.

Petronio, S. 2002. Boundaries of privacy: Dialectics of disclo-sure. Albany: State University of New York Press.

Pfeffer, J., & Fong, C. T. 2005. The self-enhancement motiveand understanding power and influence. OrganizationScience, 16: 372–388.

Pfeil, U., Arjan, R., & Zaphiris, P. 2009. Age differences inonline social networking—A study of user profiles andthe social capital divide among teenagers and olderusers in Myspace. Computers in Human Behavior, 25:643–654.

Phillips, K. W., Rothbard, N. P., & Dumas, T. L. 2009. Todisclose or not to disclose? Status distance and self-

disclosure in diverse environments. Academy of Man-agement Review, 34: 710–732.

Podolny, J. M. 2005. Status signals: A sociological study ofmarket competition. Princeton, NJ: Princeton UniversityPress.

Polzer, J. T., Milton, L. P., & Swann, W. B., Jr. 2002. Capitalizingon diversity: Interpersonal congruence in small workgroups. Administrative Science Quarterly, 47: 296–324.

Pratt, M. G., Rockmann, K. W., & Kaufmann, J. B. 2006. Con-structing professional identity: The role of work andidentity learning cycles in the customization of identityamong medical residents. Academy of ManagementJournal, 49: 235–262.

Prestwich, A., & Lalljee, M. 2009. Determinants and conse-quences of intragroup respect: An examination within asporting context. Journal of Applied Social Psychology,39: 1229–1253.

Robbins, T. L., & DeNisi, A. S. 1994. A closer look at interper-sonal affect as a distinct influence on cognitive process-ing in performance evaluations. Journal of Applied Psy-chology, 79: 341–353.

Roberts, L. M. 2005. Changing faces: Professional image con-struction in diverse organizational settings. Academy ofManagement Review, 30: 685–711.

Roney, C. J., & Sorrentino, R. M. 1995. Self-evaluation motivesand uncertainty orientation: Asking the “who” question.Personality and Social Psychology Bulletin, 21: 1319–1329.

Ross, L., Greene, D., & House, P. 1977. The false consensuseffect: An egocentric bias in social perception and attri-bution processes. Journal of Experimental Social Psy-chology, 13: 279–301.

Rothbard, N. P. 2001. Enriching or depleting: The dynamics ofengagement in work and family roles. AdministrativeScience Quarterly, 46: 655–684.

Rothbard, N. P., Berg, J. M., & Ollier-Malaterre, A. 2012. OMG,my boss just friended me: Hierarchical rank, self-disclosure, and gender in online social networking. Pa-per presented at the 2011 New Directions in LeadershipConference, Duke University, Durham, NC.

Rothbard, N. P., & Edwards, J. R. 2003. Investment in work andfamily roles: A test of identity and utilitarian motives.Personnel Psychology, 56: 699–730.

Rothbard, N. P., Phillips, K. W., & Dumas, T. L. 2005. Manag-ing multiple roles: Work-family policies and individu-als’ desires for segmentation. Organization Science, 16:243–258.

Rothbard, N. P., & Ramarajan, L. 2009. Checking your identi-ties at the door: Positive relationships between nonworkand work identities. In L. M. Roberts & J. E. Dutton (Eds.),Exploring positive identities and organizations: Buildinga theoretical and research foundation: 125–148. NewYork: Routledge.

Ryan, J. A. 2008. The virtual campfire: An ethnography ofonline social networking, Master’s thesis, WesleyanUniversity, Middletown, CT; available at http://www.thevirtualcampfire.org/thevirtualcampfiresm.pdf.


Saegert, S., Swap, W., & Zajonc, R. B. 1973. Exposure, context,and interpersonal attraction. Journal of Personality andSocial Psychology, 25: 234.

Schlenker, B. 1980. Impression management: The self-concept, social identity, and interpersonal relations.Monterey, CA: Brooks-Cole.

Schlenker, B. R., & Weingold, M. F. 1992. Interpersonal pro-cesses involving impression regulation and manage-ment. Annual Review of Psychology, 43: 133–168.

Sedikides, C. 1993. Assessment, enhancement, and verifica-tion determinants of the self-evaluation process. Journalof Personality and Social Psychology, 65: 317–338.

Sedikides, C., & Strube, M. J. 1995. The multiply motivatedself. Personality and Social Psychology Bulletin, 21:1330–1335.

Skeels, M., & Grudin, J. 2009. When social networks crossboundaries: A case study of workplace use of Facebookand LinkedIn. In GROUP ‘09: Proceedings of the 2009International ACM Conference on Supporting GroupWork: 95–104. New York: Association of Computing Ma-chinery.

Solove, D. J. 2007. The future of reputation: Gossip, rumor andprivacy on the internet. New Haven, CT: Yale UniversityPress.

Stopfer, J. M., & Gosling, S. D. 2012. Online social networks inthe work context. In D. Derks & A. Bakker (Eds.), Thepsychology of digital media at work: 39–59. London: Psy-chology Press–Routledge.

Strater, K., & Lipford, H. R. 2008. Strategies and struggles withprivacy in an online social networking community. Pa-per presented at the British HCI Group Annual Confer-ence on People and Computers, Swinton, UK.

Stutzman, F., & Hartzog, W. 2012. Boundary regulation insocial media. Proceedings of ACM Conference on Com-puter Supported Cooperative Work: 769–778.

Swann, W. B., Jr. 1983. Self-verification: Bringing social real-ity into harmony. In J. Suls & A. G. Greenwald (Eds.),Social psychological perspectives on the self, vol. 2: 33–66. Hillsdale, NJ: Lawrence Erlbaum Associates.

Swann, W. B., Jr. 1987. Identity negotiation: Where two roadsmeet. Journal of Personality and Social Psychology, 53:1038–1051.

Swann, W. B., Jr. 1990. To be adored or to be known: Theinterplay of self-enhancement and self-verification. InR. M. Sorrentino & E. T. Higgins (Eds.), Foundations ofsocial behavior, vol. 2: 408 – 448. New York: GuilfordPress.

Swann, W. B., Jr., & Ely, R. J. 1984. A battle of wills: Self-verification versus behavioral confirmation. Journal ofPersonality and Social Psychology, 46: 1287–1302.

Swann, W. B., Jr., Johnson, R. E., & Bosson, J. K. 2009. Identitynegotiation at work. Research in Organizational Behav-ior, 29: 81–109.

Swann, W. B., Jr., Pelham, B. W., & Krull, D. S. 1989. Agreeablefancy or disagreeable truth? How people reconcile theirself-enhancement and self-verification needs. Journal ofPersonality and Social Psychology, 57: 782–791.

Swann, W. B., Jr., Polzer, J. T., Seyle, D. C., & Ko, S. J. 2004.Finding value in diversity: Verification of personal andsocial self-views in diverse groups. Academy of Man-agement Review, 29: 9–27.

Swann, W. B., Jr., & Schroeder, D. G. 1995. The search forbeauty and truth: A framework for understanding reac-tions to evaluations. Personality and Social PsychologyBulletin, 21: 1307–1318.

Sydow, J., Schreyögg, G., & Koch, J. 2009. Organizational pathdependence: Opening the black box. Academy of Man-agement Review, 34: 689–709.

Tajfel, H. 1970. Experiments in intergroup discrimination.Scientific American, 223: 96–102.

Tajfel, H., & Turner, J. C. 1986. The social identity theory ofintergroup behavior. In S. Worchel & W. G. Austin (Eds.),Psychology of intergroup relations: 7–24. Chicago: Nel-son-Hall.

Taylor, D. A. 1968. The development of interpersonal rela-tionships: Social penetration processes. Journal of So-cial Psychology, 75: 79–90.

Teece, D. J., Pisano, G., & Shuen, A. 1997. Dynamic capabili-ties and strategic management. Strategic ManagementJournal, 18: 509–533.

Tetlock, P. 1983. Accountability and complexity of thought.Journal of Personality and Social Psychology, 45: 74–83.

Tufekci, Z. 2008. Can you see me now? Audience and disclo-sure management in online social network sites. Bulle-tin of Science and Technology Studies, 28: 20–36.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., &Wetherell, M. S. 1987. Rediscovering the social group: Aself-categorization theory. Oxford: Blackwell.

Tyler, T. R., & Blader, S. L. 2000. Cooperation in groups:Procedural justice, social identity, and behavioral en-gagement. Philadelphia: Psychology Press.

Tyler, T. R., & Blader, S. L. 2002. Autonomous vs. comparativestatus: Must we be better than others to feel good aboutourselves? Organizational Behavior and Human Deci-sion Processes, 89: 813–838.

Tyler, T. R., & Blader, S. L. 2003. The group engagementmodel: Procedural justice, social identity, and coopera-tive behavior. Personality and Social Psychology Re-view, 7: 349–361.

U.S. Department of Justice. 2010. Obtaining and using evi-dence from social networking sites. http://www.eff.org/f i l e s / f i l e n o d e / s o c i a l _ n e t w o r k / 2 0 1 0 0 3 0 3 _ c r i m _socialnetworking.pdf; accessed March 18, 2012.

Van Maanen, J., & Schein, E. G. 1979. Toward a theory oforganizational socialization. Research in Organiza-tional Behavior, 1: 209–264.

Wayne, S. J., & Ferris, G. R. 1990. Influence tactics, affect, andexchange quality in supervisor-subordinate interac-tions: A laboratory experiment and field study. Journalof Applied Psychology, 75: 487–499.

Weber, M. 1968. Bureaucracy In G. Roth & C. Wittich (Eds.),Economy and society: An outline of interpretive sociol-ogy: 956–1005. New York: Bedminister Press.


Weisbuch, M., Ivcevic, Z., & Ambady, N. 2009. On being likedon the web and in the “real world”: Consistency in firstimpressions across personal webpages and spontane-ous behavior. Journal of Experimental Social Psychol-ogy, 45: 573–576.

Wiesenfeld, B. M., Swann, W. B., Jr., Brockner, J., & Bartel, C.2007. Is more fairness always preferred? Self-esteemmoderates reactions to procedural justice. Academy ofManagement Journal, 50: 1235–1253.

Williams, M. 2001. In whom we trust: Group membership asan affective context for trust development. Academy ofManagement Review, 26: 377–396.

Williams, M. 2007. Building genuine trust through interper-sonal emotion management: A threat regulation frame-work of trust and collaboration across boundaries.Academy of Management Review, 32: 595–621.

Wilson, R. E., Gosling, S. D., & Graham, L. T. 2012. A review ofFacebook research in the social sciences. Perspectiveson Psychological Science, 7: 203–220.

Wojciszke, B., Abele, A. E., & Baryla, W. 2009. Two dimensionsof interpersonal attitudes: Liking depends on commu-nion, respect depends on agency. European Journal ofSocial Psychology, 39: 973–990.

World News Australia. 2009. Woman fired via Facebook afterrant. http://www.sbs.com.au/news/article/1070187/Woman-fired-via-Facebook-after-rant; accessed March 18, 2012.

Worthy, M., Gary, A. L., & Kahn, G. M. 1969. Self disclosure asan exchange process. Journal of Personality and SocialPsychology, 13: 59–83.

Zand, D. E. 1972. Trust and managerial problem solving.Administrative Science Quarterly, 17: 229–239.

Zerubavel, E. 1991. The fine line: Making distinctions in ev-eryday life. New York: Free Press.

Zhao, S., Grasmuck, S., & Martin, J. 2008. Identity constructionon Facebook: Digital empowerment in anchored rela-tionships. Computers in Human Behavior, 24: 1816–1836.

Ariane Ollier-Malaterre ([email protected]; [email protected]) is associate profes-sor of management at Rouen Business School and professor of management at Uni-versité du Québec à Montréal. She earned her Ph.D. at ESSEC Business School andCNAM. Her research focuses on the work-nonwork interface at the individual, theorganizational, and the country levels.

Nancy P. Rothbard ([email protected]) is the David Pottruck AssociateProfessor of Management at the Wharton School, University of Pennsylvania. Herresearch examines how factors outside the workplace influence people’s engagementwith their work. She has also examined how people cope with these potential spill-overs by segmenting work and nonwork roles.

Justin M. Berg ([email protected]) is a doctoral candidate in management atthe Wharton School, University of Pennsylvania. His research focuses on creativity,proactivity, and the meaning of work in organizations.


Documents

WHEN WORLDS COLLIDE IN CYBERSPACE: HOW BOUNDARY … · Rothbard et al., 2005). Scholars have defined boundaries as “mental fences” (Zerubavel, 1991: 2) used to simplify and order