Virtual Teams
Inputs of Team-Level Virtual Team Research
Effects of communication medium/technology Group/team composition (e.g., on cultural values, country of origin, gender, etc) Geographic dispersion Team size Team leadership Team training Team structure
Models of Team Virtuality (Kirkman & Mathieu, 2005)
(Kirkman & Mathieu, 2005). developed a three-dimensional model of team virtuality including: the extent to which team members utilize virtual tools to coordinate and execute team processes (versus working and meeting face-to-face); the amount of informational value provided by such tools (i.e., how important each tool is to accomplishing work, akin to communication richness for communication tools) the synchronicity of team member virtual interaction (i.e., whether or not communication or work occurs in real time between sender and receiver). Common to these conceptualizations is the degree to which teams use CMC; however, debate remains about what constitutes virtuality versus its antecedents (Kirkman & Mathieu, 2005).
Team Level Outputs Research in Virtual Teams
A noteworthy example is an examination of an in-depth case study of a European based automotive manufacturing team, May and Carter (2001) found that the introduction of video conferencing, shared whiteboard, application sharing and product data management tools increased team efficiency and flexibility including time savings, sales volume, and reduced costs. Malhotra et al. (2001) examined how an interorganizational virtual team was able to produce radical innovation. Lurey and Raisinghani (2001) found that teams' processes and team members' relations were strongly positively related to team performance and team member satisfaction; selection procedures and executive leadership styles were moderately associated; and design process, other internal group dynamics, and additional external support mechanisms were only weakly related. 5 year study Boh, Ren, Kiesler, and Bussjaeger (2007) found that dispersed projects garnered higher net earnings than local projects when there was a better match of scarce expertise to project requirements; and a curvilinear relationship was observed, such that a very high percentage of dispersed experts on a project increased coordination costs and reduced net earnings.
Various Topics in The Process Domain in Virtual Teams
Action Processes Transition Processes Interpersonal Processes All of the above are from Marks et 2001 Also cover Informational Processes and Socialization
Individual Level Inputs Virtual Teams
Anonymity Demography Individual Differences (Personality, Cognitive Style) Task Type
Effects of communication medium/technology (Additional Communication Mediums)
Baker (2002) found no decision quality differences between teams using text-only or audio-only CMC, and the addition of video to audio-only CMC resulted in significant improvement to teams strategic decision quality. Kahai and Cooper (2003) found that richer media facilitate social perceptions (i.e., total socio-emotional communication and positive socio-emotional climate) and perceived ability to evaluate others' deception and expertise; leaner media (i.e., electronic mail and electronic conferencing) facilitate communication clarity when participants have less task-relevant knowledge; richer media can have significantly positive impacts on decision quality when participants' task relevant knowledge is high; and the effects of participant deception can be mitigated by employing richer media. Lowry, Roberts, Romano, Cheney, and Hightower (2006) compared groups of three and six members each with three media options (i.e., FTF without CMC support, FTF with CMC support, and virtual with CMC support). Results indicated that smaller groups establish and maintain higher levels of communication quality, and FTF with CMC support groups have higher levels of communication quality than virtual with CMC support groups; however, no significant difference between traditional FTF groups and virtual groups with CMC support was found. Finally, a few researchers have examined predictors of media use by virtual team members. For example, in a sample of teams in sales divisions of two Fortune 100 companies in the IT industry, Watson-Manheim and Belanger (2007) found that repertoires of communication media use are influenced by institutional conditions (e.g., incentives, trust, and physical proximity) and situational conditions (e.g., urgency, task, etc.) and by routine use of the media over time.
Directions for Future Research in Virtual Teams
Below we summarize what we believe will be the most important directions for future research over the next decade, addressing both theoretical and methodological implications in each of five areas including: (a) conceptualization of virtuality- it is not a dichtonomy, need more info on what distinguishes it from its antecedents, little understanding of time and stages here and no discussion of the communication tool effects, differentiation between miles and time in terms of dispersion. (b) team development- Very few developmental models of virtual teams. Don't know if development here differs from FTF teams, don't know whether developmental cycles differ within different types of virtual teams, most of the research here is case studies (c) virtual team leadership- most of the research is case studies, move to the organizations in the real world, leadership may be more dispersed in teams like these, rely on development cycles rather than a list of leader behaviors (d) levels of analysis- hundreds of IDs have not been explored, few studies have measured individual cultural features, look at individual level components relationships with team phenomena, look at new emergent states besides trust, more research needed at organizational level (e) multi-disciplinary approaches- need more research in the field, need to read the lit of other disciplines
Research Involving Relational Demography in Virtual Teams
Bosch-Sijtsema (2007) found that when members of virtual teams are heterogeneous in organizational and cultural background, have little history of working previously together, and have different experiences working in teams, members often bring widely different expectations to the team compared to FTF teams. Expectation mismatches led to motivational problems, dissatisfaction, and lower performance in these teams, but did result in more individual learning.
Effects of Team Size on Virtual Teams
Bradner, Mark, and Hertel (2005) found that compared to members of larger teams, members of smaller teams participated more actively on their team, were more committed to their team, were more aware of the goals of the team, had greater awareness of other team members, and were in teams with higher levels of rapport; AND larger teams were more conscientious than smaller teams in preparing meeting agendas; team size was also associated with different technology choice—larger teams adopted technology to support the coordination of asynchronous work, while smaller teams adopted technology that primarily supported collaboration.
Effects of Geographic Dispersion on Virtual Teams
Cramton and Webber (2005) found that geographic dispersion significantly and negatively related to work processes (i.e., communication, coordination) and team effectiveness, and work processes partially mediated the relationship between geographic dispersion and team effectiveness Chudoba, Wynn, Lu, and Watson-Manheim (2005) found that being distributed had no impact on self-assessed team performance. Gibson and Gibbs (2006) found that geographic dispersion was negatively related to virtual team innovation, but the effects were less negative in teams with higher psychological safety Metiu (2006) found that geographical dispersion exacerbated processes related to status closure (i.e., the monopolization of opportunities of higher status groups at the expense of lower status groups), and status closure resulted in less intragroup cooperation within the overall GVT.
Research on De-identification in Virtual Teams and Its Effects
Depersonalization increases conformity to groups norms typically through its effects on group identity, more so for women than for men (Lee, 2004, 2005). Some research also found that depersonalization induced greater polarization and extreme perceptions of group norms, especially when group identity, rather than individual identity, was more salient (Lee, 2006, 2007b; Spears et al., 1990). Several studies have also found beneficial effects for the use of CMC as compared to FTF on individual level outcomes including: number of ideas, less production blocking, and less evaluation apprehension, although studies are mixed with regard to levels of satisfaction (Gallupe et al., 1992; Jessup & Tansik, 1991). Anonymous CMC brainstorming appears to have more advantages than identified CMC brainstorming for individual outputs (Jessup & Tansik; Mejias, 2007). A danger of CMC groups, however, is that self-ratings of contribution are typically more inflated and less accurate than in FTF groups; and, on the contrary, rating biases stemming from liking are evident in ratings of others in FTF groups but not in CMC groups (Weisband & Atwater, 1999).
Models of Team Virtuality Gibson and Gibbs (2006)
Gibson and Gibbs (2006) conceptualize virtuality along four dimensions including: geographic dispersion; electronic dependence; dynamic structural arrangements (i.e., frequent changes in team members, their roles, and relationships to each other) nationality diversity. Common to these conceptualizations is the degree to which teams use CMC; however, debate remains about what constitutes virtuality versus its antecedents (Kirkman & Mathieu, 2005).
Effects of Transition Processes Team Level Virtual Teams Research
Huang, Wei, Watson, and Tan (2002) found that CMC groups with an embedded goal-setting structure had higher cohesion, better team commitment, better collaboration climate, perceived decision quality, and more decision alternatives than CMC groups without a goal-setting structure. Baba, Gluesing, Ratner, and Wagner (2004) found that the process of cognitive convergence (i.e., increasing similarity of individual cognitive structures through information sharing) was related to virtual team performance only when patterns of cognitive divergence were also reversed. Facilitators of the relationship between cognitive convergence also included: similar experiences in a common context, surfacing hidden truths through knowledge brokering, and shifts in agent self-interest and negotiation of task interdependence. Munkvold and Zigurs (2007) found that swift-starting virtual teams need to structure their interaction from the onset, including introducing team members' background and competence, discussing project goals and deliverables, defining roles and responsibilities, and setting milestones; they have to pay immediate attention to familiarizing themselves with and integrating available technology, and agreeing on preferred communication media and frequency.
Research Involving Individual Differences in Virtual Teams
Individual differences have also been investigated in a number of studies, including personality characteristics such as extraversion (Straus, 1996), internal locus of control (Lee-Kelley, 2006), focus immersion (i.e., the level of attention focus and engagement; Rutkowski, Saunders, Vogel, & van Genuchten, 2007), self efficacy (Staples & Webster, 2007), cultural values/country of origin (Hardin, Fuller, & Davison, 2007; Tan, Wei, Watson, Clapper, & McLean, 1998), and gender (Bhappu, Griffith, & Northcraft, 1997; Palomares, 2004). Workman, Kahnweiler, and Bommer (2003) found that individuals' commitment to virtual teams was stronger when they had cognitive styles characterized as external (i.e., preferring group, rather than individual, problem solving), conservative (i.e., preferring structure, compliance with existing rules and procedures, familiarity, and minimal change), and global (i.e., having mental representations that have thin or fuzzy boundaries). The relationships between both external and conservative cognitive styles and commitment to virtual teams were stronger when individuals reported using richer, rather than leaner, communication media.
Trust Effects on Team Level Virtual Teams Research (Predictor)
Jarvenpaa, Shaw, and Staples (2004) found that early in a team's existence, a member's trusting beliefs have a direct positive effect on his or her trust in the team and perceptions of team cohesiveness; later on, however, a member's trust in his team operates as a moderator, indirectly affecting the relationships between team communication and perceptual outcomes. Using case studies of five virtual teams in a large petrochemical company, Baskerville and Nandhakumar (2007) found that personal trust is most effectively established or reinvigorated through geographically collocated social interaction; personal trust is an antecedent to the activation and operation of effective virtual teams; abstract trust (i.e., trust based on shared organizational norms and expertise) is an alternative to personal trust as an antecedent to the activation and operation of effective, short-term virtual teams; and personal trust gradually dissipates over time without collocated social interaction.
Effects of Team Leadership on Virtual Teams
Kayworth and Leidner (2000) identified four critical success factors for virtual team leadership including: communication, culture, technology, and project management Carte, Chidambaram, and Becker (2006) found that in a longitudinal study of virtual teams, the most successful teams displayed significantly more concentrated leadership behavior focused on performance (i.e., "producer" behavior) and shared leadership behavior focused on keeping track of group work (i.e., "monitor" behavior) than the lower performing teams; and these behaviors emerged strongly during the first half of the groups' lifespan, and stayed throughout the life of the groups, but steadily dissipated in strength over time. Malhotra, Majchrzak, Carman, and Lott's (2001) case study identified several key leadership practices for managing successful virtual teams including: establishing virtual team strategy initially, encouraging the use of knowledge management/collaborative tools, and restructuring work without changing core creative needs. Suchan and Hayzak (2001) found in a case study that to build and maintain team trust, the virtual team leader used a FTF, three-day project kickoff, a mentoring program, and a culture that promoted information sharing, team-based rewards, and employee development. Using case studies of four virtual team leaders from four international organizations, Sivunen (2006) identified four different leadership tactics employed in enhancing identification with the team: catering for the individual, giving positive feedback, bringing out common goals and working and talking up the team activities, and FTF meetings.
Effects of Action Processes Team Level Virtual Teams Research (Virtual Team Development)
Maznevski and Chudoba (2000, fs) found that virtual team dynamics were characterized by both: (a) a series of interaction incidents incorporating a set of decision processes using a particular medium that are shaped by a limited set of structural characteristics; and (b) repeating temporal patterns of regular FTF meetings in which the intensity of interaction is extremely high, followed by a period of some weeks in which interaction incidents are less intense. Drawing upon adaptive structuration theory, the authors suggest that within interaction incidents the medium and form are selected to match the function, but across incidents over time, the function is modified to match the medium and form. Majchrzak, Rice, Malhotra, King, and Ba (2000) found that to address misalignment between the organizational environment, group structure, and technology, group members initially modified their environment and group structures, whereas later in the team's life span, group members modified all three aspects including technology. Using a multimethod study of five virtual team projects in five different organizations, Ratcheva and Vyakarnam (2001) found that virtual teams follow self-energizing developmental processes that are nonlinear and not consistent with existing developmental models for FTF teams. Using surveys and interviews of seven virtual teams of five-six members engaged in distance learning in a university setting, Johnson et al. (2002) found that the developmental pattern of the teams followed Tuckman's (1965) forming-storming-norming-performing model, but the teams moved through the stages very quickly, did not experience the storming stage, and encountered a conflict resolution stage between the performing and forming stage in an interactive fashion.
Research on Outputs of Virtual Teams at the Individual Level
Mejias (2007) found that process losses have negative effects on two dimensions of meeting satisfaction (i.e., process and outcome satisfaction) only in identified CMC environments, but not in anonymous CMC settings. Conversely, process gains generate positive effects on both outcome and process satisfaction regardless of anonymity levels. Froehle (2006) found that among survey responses from over 2000 customers of a large internet service provider, that three customer service representative characteristics typically associated with satisfaction in FTF encounters (i.e., courtesy, professionalism, and attentiveness) had no effect on customer satisfaction in either telephone or CMC contexts.
Emergent States Research in Virtual Teams
Morris, Marshall, and Rainer (2002) found that both user satisfaction and trust are positively related to job satisfaction in virtual teams. Examining antecedents of trust with 224 U.S. undergraduate students using CMC in a laboratory experiment, Ferrin and Dirks (2003) found that the effects of different types of team-based rewards (i.e., cooperative, competitive, or mixed) on individual trust in team members was mediated by perceived partner motives and performance and own information sharing
Cultural Differences in Virtual Teams
Most FTF research has demonstrated that collectivists are typically more comfortable, have less social loafing, and are more productive in team-based settings than are individualists (Earley & Gibson, 1998; Oyserman, Coon, & Kemmelmeier, 2002). However, Hardin et al. (2007) found that team members from individualistic cultures reported higher levels of self-efficacy for working in groups and virtual teams than did team members from collectivistic cultures. This finding is critical given that Staples and Webster (2007) found that self-efficacy for teamwork was important for virtual teams (even more so than FTF teams).
Effects of Team Composition on Virtual Teams
Ocker (2007) examined the emergence of dominance in virtual teams using in-depth case studies of eight teams in a university setting, finding that the dominant individual belonged to the majority sex in each team; dominance was driven by a combination of a few team member status traits; and when one or more status markers belonged to a single person—the dominant member—and were absent in other team members, dominance was most pronounced. Paul, Samarah, Seetharaman, and Mykytyn (2005) found that cultural diversity moderated the relationship between collaborative conflict and performance such that the effects are stronger for culturally heterogeneous rather than homogeneous groups. Staples and Zhao (2006) compared homogeneous and heterogeneous groups (i.e., based on individualism/collectivism values, different languages spoken, country of birth, and nationality) found that heterogeneous teams were less satisfied and cohesive and had more conflict than the homogeneous teams, and that within just the heterogeneous teams, the performance of the virtual teams was superior to that of the FTF teams. Haas (2005) examined the impact of team composition in 96 global virtual project teams in an international development agency in terms of the number of cosmopolitan members (i.e., individuals with broad experience in many countries) and local members (i.e., individuals with deep experience in the project country), finding that both cosmopolitans and locals helped to increase internal knowledge (i.e., knowledge possessed by members themselves) and while cosmopolitans (but not locals) helped to increase external knowledge (i.e., knowledge from sources outside the team) and that a mix of both locals and cosmopolitans was optimal for GVT performance.
Effects of Interpersonal Processes Team Level Virtual Teams Research (Conflict)
Poole, Holmes, and DeSanctis (1991) found that the introduction of a CMC tool had no direct effect on team outcomes (i.e., group consensus), but the effects were mediated by how the groups adapted the tools to either promote, or inhibit, productive conflict management. In a field experiment using 35 five-person university student virtual teams in the U.S. and Japan, Montoya-Weiss, Massey, and Song (2001) found that certain conflict handling strategies (i.e., competition, collaboration) were positively related to virtual team performance, while others (i.e., avoidance, compromise) were negatively related. Introducing a formal, structured plan for managing team time and activities weakened the negative effects of avoidance and compromise on virtual team performance. Using 24 product development teams in five organizations, Mortensen and Hinds (2001) found that conflict was slightly more detrimental to CMC team performance than FTF team performance. They also found that shared team identity was associated with less task and affective conflict in CMC but not in FTF teams (as a criterion conflict example). Kankanhalli, Tan, and Wei (2007) examined antecedents to conflict and found that cultural diversity is likely to contribute to both task and relationship conflict while functional diversity may result in task conflict; large volumes of electronic communication and lack of immediacy of feedback in asynchronous media can contribute to task conflict; the relationship between task conflict and team performance is likely to be contingent upon task complexity and conflict resolution approach. The influence of relationship conflict on performance may depend on task interdependence and conflict resolution approach. The conflict resolution approach may in turn be determined by the nature of conflict attribution.
Effects of communication medium/technology (Meta Analytic Findings)
Rains (2005) found that CMC groups experience greater participation and influence equality, generate a larger amount of unique ideas, and experience less member dominance than do groups meeting FTF. Fjermestad (2004) found that the modal outcome for CMC compared to FTF methods is "no difference," while the overall percentage of positive effects (i.e., decision quality, depth of analysis, equality of participation, and satisfaction) for results that compare CMC to FTF is 29.2% in favor of CMC. Additionally, more detailed analysis suggests that task type, CMC type, and their interaction have a moderating effect on adaptation and outcome factors. Specifically, groups working on idea generation tasks using CMC improve to 39.6% over FTF. Conversely, asynchronous CMC groups working on decision making tasks improved to 46.4% over FTF. FTF groups show higher levels of consensus and perceived quality, communicate more, and are more efficient (i.e., requiring less time to complete the tasks). No differences, however, were observed between FTF and group support systems groups on satisfaction and usability
Research on Action Processes in Virtual Teams
Regarding action processes, several researchers have examined the effects of de-individuation or depersonalization (i.e., the process of reducing individual identity salience) on action processes in virtual teams (e.g., Lea & Spears, 1992; Spears, Lea, & Lee, 1990). For example, using a series of university student laboratory experiments, Lee (2004) found that when the group level of self-identity was salient, uniform virtual appearance of CMC partners (i.e., using computer-rendered symbols) triggered depersonalization and heightened conformity behavior; however, when virtual appearance of CMC partners was distinct, conformity actually decreased; and depersonalization enhanced adherence to group norms directly and through its effects on group identification.
Effects of Action Processes Team Level Virtual Teams Research (Knowledge Management)
Regarding action processes, the bulk of the studies can be divided into two main groups: those dealing with aspects of knowledge management such as acquisition and sharing and those dealing with various stages of virtual team development. Rafaeli and Ravid (2003) found that teams that used email to share more information performed better than teams who shared less information. From in-depth case studies of 12 virtual teams in a university setting in Norway and the U.S., Sarker, Sarker, Nicholson, and Joshi (2005) found that the volume of communication, the credibility of the communicator, and the nature of cultural values (i.e., collectivism) held by the communicator were found to significantly predict the extent of knowledge transferred; however, capability was not found to have a significant influence. Sole and Edmondson (2002) found that while virtual teams can easily access and use unique locale-specific knowledge resources to resolve problems that arise in those same locales, they encounter difficulties in uncovering and sharing situated knowledge (i.e., knowledge embedded in the work practices of a particular organizational site).
Research on Interpersonal Processes in Virtual Teams
Regarding interpersonal processes, not surprisingly, most of the existing research has examined communication (e.g., Anawati & Craig, 2006; Zhou & Zhang, 2007), led by Walther and colleagues (e.g., Walther, 1994, 1997; Walther & Bazarova, 2007; Walther & Burgoon, 1992). For example, among over 150 individuals in a university student laboratory experiment, Tidwell and Walther (2002) found that individuals interacting in CMC exhibited a greater proportion of more direct and intimate uncertainty reduction behaviors than FTF participants did, demonstrated significantly greater gains in attributional confidence over the course of the conversations, and the use of direct strategies by CMC individuals resulted in judgments of greater conversational effectiveness by partners.
Research on Transition Processes in Virtual Teams
Regarding transition processes, Forester, Thoms, and Pinto (2007), using 82 participants from 12 virtual teams in an international engineering and construction company, found that goal commitment was significantly positively related to both perceived task outcome and psychosocial outcomes, whereas quality of goal setting was related only to perceived task outcome suggesting that goal setting is related to perceptions of outcomes in virtual teams
Effects of Interpersonal Processes Team Level Virtual Teams Research
Research on virtual team interpersonal processes at the team level falls mainly into two categories: communication (e.g., Crowston, Howison, Masango, & Eseryel, 2007; Leenders, van Engelen, & Kratzer, 2003; Murthy & Kerr, 2002; Pauleen & Yoong, 2001; Rains, 2007; Reinig & Mejias, 2004; Sproull & Kiesler, 1986) and conflict management (e.g., Rutkowski et al., 2007). Within each of these categories, both communication and conflict have been treated as either a predictor or criterion variable.
Effects of Interpersonal Processes Team Level Virtual Teams Research (Communication)
Saphiere (1996) found that the most successful teams communicated more often in informal, social ways; utilized more task and affect behaviors; frequently disagreed with one another, critically analyzing issues in meetings and focusing on task in a positive manner in writing; acted as cultural interpreters and mediators; and unanimously desired to work together again in the future. With longitudinal data from 38 teams of university students, Yoo and Kanawattanachai (2001) found that the influence of a team's early communication volume on team performance decreases as teams develop transactive memory systems and a collective mind. Using the electronic communication records of 13 teams of university graduate students, Cramton (2001) identified five key GVT interpersonal process problems—failure to communicate and retain contextual information, unevenly distributed information, differences in the salience of information, relative differences in speed of access to information, and interpreting the meaning of silence—that contribute to a lack of mutual knowledge among team members; structural and task-related factors (e.g., feedback lags, task interdependence, tight time frames) exacerbate the relationship between the five key interpersonal problems and GVT performance. In an example of research in which communication was treated as a criterion variable, Rasters, Vissers, and Dankbaar (2002) found in an in-depth case study of a single virtual team based in Europe, that virtual team communication was affected far more by the stage of group development (using Gersick's, 1988, punctuated equilibrium model) than by the particular communication media used.
Research Involving Anonymity in Virtual Teams
Sassenberg, and Boos (2003) found that high anonymity (compared to low anonymity) as well as CMC (compared to FTF communication) cause more conformity to individual needs or goals when personal identity was salient; and higher conformity to a socially shared superordinate category norm was the result when social identity was salient. Regarding the use electronic communication more generally, Aiello and Kolb (1995) compared the effects of individual- and work group-level electronic performance monitoring on participants working on a data entry task alone, in noninteracting groups, and cohesive groups, finding that highly skilled, monitored participants keyed more entries than highly skilled, nonmonitored participants, with the opposite pattern for low-skilled participants; no signs of social loafing were detected among group-monitored participants; and nonmonitored workers and members of cohesive groups felt the least stressed.
Research on Informational Processes in Virtual Teams
Several studies examined information processing in virtual teams (e.g., Griffith & Sawyer, 2006; Heninger, Dennis, & Hilmer, 2006). Belanger and Watson-Manheim (2006) examined how virtual team members structure their use of multiple media to attain strategic goals in complex work environments based on interviews with 40 information technology workers in two organizations. Results identified two primary structures individuals employ when using multiple media (i.e., sequential and concurrent) and that individuals strategically use multiple media to accomplish specific communication goals beyond simply transmitting the message, such as message acknowledgement, enhancement of mutual understanding, and participation in multiple communication interactions.
Trust Effects on Team Level Virtual Teams Research (Antecedent)
Sharifi and Pawar (2002) summarized the many challenges associated with building and maintaining virtual teams and, specifically, trust is essential to virtual team success with co-location as a necessary but not sufficient ingredient for building trust. Using surveys of student teams participating in on-line courses, Coppola, Hiltz, and Rotter (2004) found that leaders of virtual groups build swift trust by establishing early communication, developing a positive social atmosphere, reinforcing predictable patterns, and involving team members in tasks Stewart and Gosain (2006) found that among 67 established open source software project teams, adhering to the ideological tenets of the open source software community (i.e., values, norms, and beliefs) positively affected both trust and communication quality which, in turn, affected team performance. Using 45 university student virtual teams, Polzer, Crisp, Jarvenpaa and Kim (2006) examined the effects of a combination of different types of physical co-location and different configurations of nationality composition on team trust. The lowest levels of trust occurred in virtual teams consisting of two same-nationality co-located subgroups providing evidence for faultline effects along nationality.
Socialization in Virtual Teams
Socialization has also been examined. For example, using a case study of one virtual team in the field, Ahuja and Galvin (2003) examined socialization processes by content analyzing members' email over a three-month period and found that, as expected, newcomers exhibited an information seeking mode while established members exhibited an information providing mode. While newcomers also actively engaged in discussions regarding cognitive information (i.e., construction of task-based information based on the social reality of the group) via email, the medium did not facilitate the exchange of normative (i.e., understanding what the group values or expects) or regulative (i.e., structure, procedures, or processes that currently exist within a group) information. In a second example, Tanis and Postmes (2003) conducted a series of laboratory experiments with university students in the Netherlands and found that the limited availability of social cues (independent of the communication medium) negatively affected individuals' ability to reduce ambiguity and make positive impressions on collaboration partners; and that the effects of the limited capacity to convey social cues on selecting collaboration partners depended on the social identity of the parties involved.
Effects of communication medium/technology (Multicultural Research)
Taiwan, Li (2007) found that FTF groups, compared to CMC groups, performed better on critical functions such as problem analysis and criteria establishment and were more efficient in communication effectiveness; however, there were no significant differences on most objective and perceived group outcomes. Daily and Teich (2000) found that among 27 four-five person multicultural groups in a university student laboratory experiment involving perceptions of contributions involving ratings of self, ratings of others, and ratings by others, ratings were higher in a nongroup decision support system (GDSS) environment compared to a GDSS-supported environment; ethnic minorities' ratings of self were significantly lower than ethnic minorities' ratings of others in the nonGDSS environment but not in the GDSS environment; and the average number of ideas generated were greater in the GDSS environment.
Team Training Effects on Virtual Teams
Tan, Wei, Huang, and Ng (2000) found that virtual teams trained on using a dialogue technique (i.e., engaging in small talk, discussing good communication practices, and building a team mental model of good communication practices) before interacting exhibited stronger cohesion, collaboration, perceived decision quality, and decision satisfaction. Cornelius and Boos (2003) examined the impact of training on CMC use and found the best performance scores in the FTF condition, performance scores in CMC with training approximating those of the FTF condition, and the lowest performance scores in CMC without training. Kirkman, Rosen, Tesluk, and Gibson (2006) found that among 40 virtual teams from a high technology service organization, virtual team CD-ROM based training proficiency was more strongly related to customer satisfaction when teams had higher levels of trust and technology support, and when team leaders had longer, rather than shorter, tenure with their teams.
Team Structure Effects on Virtual Teams
Timmerman and Scott (2006) examined a comprehensive model of virtual team performance. The authors investigated the impact of both communication (e.g., communication styles, channel selection) and structural (e.g., size, number of member locations and time zones, organization type) factors on actual technology use and virtual team performance finding that structural features primarily related to technology use, whereas communication considerations were associated more so with team outcomes. Workman (2005) gathered data from 436 virtual team projects over a 27-month period in a global information technology firm headquartered in France, and found that team boundary permeability strengthens the positive effects of various aspects of virtual team culture on team performance Strijbos, Martens, Jochems, and Broers (2004) found that , CMC groups with formally designated roles, versus those in a non-role condition, have higher team coordination, a larger amount of task-content focused statements, and increased member awareness of team efficiency. Barkhi (2005) Examined the interactive effects of communication medium (i.e., FTF GDSS vs. distributed GDSS) and incentive structure (i.e., individual- vs. group-based) on group performance, finding that groups with group-based incentives engage in free-riding behavior by selecting low effort levels less often in FTF-GDSS groups than in distributed-GDSS groups, however, the performance difference between FTF-GDSS groups and distributed-GDSS groups is not statistically significant when the incentive is individual based.
Emergent States in Team Level Virtual Teams Research
Trust Team potency/collective efficacy Cohesion Group identity Team empowerment Communication climate
Group Identity, Team Empowerment, and Communication Climate Effects on Team Level Virtual Teams Research
Using 31 stable and 29 reconfigured university student groups meeting and working over a seven-week period, Bouas and Arrow (1996) found that group identity was significantly lower for CMC versus FTF groups, the effects were stronger in the reconfigured groups, but the differences between groups in identity narrowed over the course of the study. Using 35 virtual teams in a field study of a high technology travel company, Kirkman, Rosen, Tesluk, and Gibson (2004) found that team empowerment was related to both process improvement and customer satisfaction, but that team empowerment was more important for virtual teams that met FTF less, rather than more. In a comprehensive examination of 56 global virtual engineering teams in the aerospace defense industry, Gibson and Gibbs (2006) found that a psychologically safe communication climate mitigated the negative effects of virtuality on innovation.
Team potency/Collective efficacy and Cohesion Effects on Team Level Virtual Teams Research
Van den Bossche, Gijselaers, Segers, and Kirschner (2006) found that: interdependence, task cohesion, group potency, and psychological safety were all related to team learning behaviors; task cohesion was positively related to mutually shared cognition; and task cohesion and group potency were positively related to team effectiveness. Fuller, Hardin, and Davison (2007) found that group potency and computer collective efficacy act as antecedents to virtual team efficacy. Virtual team efficacy in turn predicted both perceptual and objective measures of performance; and effort fully, and communication partially, mediated the effects of virtual team efficacy on performance Dennis and Garfield (2003) compared scores on satisfaction, cohesion, and perceived effectiveness over a seven-week period finding that attitudes in the CMC groups, which were initially lower than FTF groups, increased positively over time to reach FTF group levels.
What Constitutes a Virtual Team?
Virtual teams are variously defined as groups of interdependent coworkers who are geographically dispersed (i.e., consisting of members spread across more than one location), electronically dependent (i.e., communicating using electronic tools such as e-mail or knowledge repositories), structurally dynamic (i.e., in which change occurs frequently among members, their roles, and relationships to each other), or nationally diverse (i.e., consisting of members with more than one national or cultural background; Gibson & Gibbs, 2006; Griffith, Sawyer, & Neale, 2003; Kirkman & Mathieu, 2005; Martins et al., 2004).
Research Involving Task Type in Virtual Teams
Walther and Bunz (2005) found that the degree of self-reported adherence to a set of team work rules (i.e., start immediately, communicate frequently, acknowledge others, explicitness, multitasking, and observe deadlines) was related to a host of positive individual outcomes including trust, task attraction, social attraction, and self-rated task success. Using 263 individuals working in 54 global virtual teams, Majchrzak, Malhotra, and John (2005) found that when individuals perceive their task as non-routine, there is a positive relationship between an individual's perceived degree of IT support for communicating context information and his/her collaboration know-how development; however, when individuals perceive their task as routine, partial IT support for contextualization is associated with lower levels of collaboration know-how development
Research on Other Facets of Interpersonal Processes in Virtual Teams
Walther, Slovacek, and Tidwell (2001) found that in new, unacquainted teams, seeing one's partner (in a photograph) promotes affection and social attraction, but in long-term virtual teams, the same type of photograph dampens affinity. Panteli and Davison (2005) found that subgroups exert different degrees of impact on the team as a whole; where the impact was high, boundaries were created between team members in different subgroups while the development of team cohesiveness was restricted; and all teams were able to produce high quality outcomes, suggesting that the emergence of subgroups may not always have a negative influence on team performance.
Organizational Level Virtual Teams Research
Zack and McKenny (1995) found that the inputs of social context (i.e., the culture, distribution of power, and the social norms, habits, practices, expectations and preferences held by a group regarding its present and past interaction) influenced each group's use of email and FTF communication such that there was greater similarity of interaction within groups across communication mode than within communication mode across groups. Andersen (2005) found that the use of CMC moderated the relationship between decentralized strategic decision making and firm performance such that the relationship was stronger with greater, rather than less, use of CMC; and that CMC was also positively related to firm performance. Finally, McDonough, Kahn, and Barczak (2001) found that among 100 firms in new product development, global teams presented companies with many more challenges than either collocated or virtual teams, and that performance is also lower for these types of teams.
Research on Organization Level Virtual Teams Research
Zack and McKenny (1995) found that the inputs of social context (i.e., the culture, distribution of power, and the social norms, habits, practices, expectations and preferences held by a group regarding its present and past interaction) influenced each group's use of email and FTF communication such that there was greater similarity of interaction within groups across communication mode than within communication mode across groups. Andersen (2005) found that the use of CMC moderated the relationship between decentralized strategic decision making and firm performance such that the relationship was stronger with greater, rather than less, use of CMC; and that CMC was also positively related to firm performance. McDonough, Kahn, and Barczak (2001) found that among 100 firms in new product development, global teams presented companies with many more challenges than either collocated or virtual teams, and that performance is also lower for these types of teams
Team Emergent States
defined as "constructs that characterize properties of the team that are typically dynamic in nature and vary as a function of team context, inputs, processes, and outcomes" (Marks et al., 2001, p. 357) Emergent states represent important mediational influences with explanatory power accounting for variability in team performance, which scholars have suggested are critical in addressing the limitations of IPO models
