Discussion and Implications

---

Characteristics of a GSS Dialogue

Positive research results reported by the MIT Dialogue Project (e.g., Isaacs, 1993; Schein, 1993) support the effort of embedding the conceptual team-building framework into a GSS for supporting virtual team-building. MIT research has been conducted in a manual face-to-face (FtF) setting. When embedded with a GSS, a dialogue session may be more effective in some aspects.

First, with GSS structures of anonymity, simultaneity, electronic recording, and expandable electronic storage space, the container of dialogue can virtually contain any number of controversial ideas in a GSS dialogue. Hence, the container in a GSS dialogue can be larger than in a manual face-to-face dialogue.

Second, the suspension of dialogue may be better controlled in a GSS dialogue. Transmission of a message via a GSS is often separated from its composition and editing (McGrath & Hollinshead, 1993) whereas the message composition, editing, and transmission cannot be separated in face-to-face communication. Consequently, the communication channel of a GSS typically promotes more carefully and precisely worded communication (Daft & Lengel, 1986), which can enhance the suspension. Further, GSS can separate individual work space from team work space in communication (Bostrom & Anson, 1992), which can also make the control of suspension easier. For example, to ensure a satisfied level of suspension, a dialogue facilitator can vet the messages generated in a GSS dialogue from team members before they are sent through the electronic communication channel.

Third, with GSS structures of electronic recording and evaluation (ranking or voting), messages of preferences and values generated at Step (4) can be easier and better to be captured, edited, organized, and evaluated. As a result, shared team meanings can be more effectively and efficiently generated. In addition, GSS structures of anonymity may provide a better sense of psychological safety (Schein, 1993) and further reduce defensive behavior (Argyris, 1985) in a GSS dialogue.

In summary, a dialogue may be more effectively conducted in a GSS setting than in a face-to-face setting in terms of its three main attributes of container, suspension, and laser. Hence, a GSS embedded with the conceptual team-building framework may effectively support virtual team-building, which leads to the research propositions to be presented next. To smooth the flow of the remaining parts of this chapter, a TB-GSS is used to denote a GSS embedded with the conceptual team-building framework; and a GSS without embedding such a framework is denoted as a traditional GSS.

Research Propositions

Laboratory experiments can be conducted to test the framework proposed above. Effectiveness of team-building can be measured in terms of team trust (Larson & LaFasto, 1989; Schein, 1993), cohesion (Seashore, 1954), and conflict (Chidambaram, Bostrom, & Wynne, 1991). In general, a well-built team should perform better in task activities. Dependent variables of decision quality and satisfaction are commonly used to measure outcomes of teamwork (e.g., Benbasat & Lim, 1993; Dennis, Haley, & Vandenberg, 2001). The above five variables are used to examine whether or not a TB-GSS could enhance virtual team-building as well as the performance of virtual teamwork.

A TB-GSS can enhance team trust in two aspects. First, with the conceptual framework presented above, shared team goals could be aligned with individual goals. Shared team goals can clearly show to team members that their individual interests and needs depend on the attainment of team goals (Culbert & McDonough, 1980; Mackie & Goethals, 1987). Team members are thus motivated to understand each other, trust each other, and collaborate with each other, in order to attain shared team goals as well as individual goals.

Second, two specific guidelines for building up team trust-that a team must be fairly consistent and mature in its approach to deal with problems, and that every member must be valued and treated with respect (Larson & LaFasto, 1989)-can actually be proposed and included in the shared team ground rules at Step 4). Because shared team ground rules are generated and commonly accepted by team members themselves in a GSS dialogue, team members are more likely to follow these two guidelines. Hence, shared team ground rules in a GSS dialogue can enhance team trust and hold people together (Bohm, 1990).

Proposition #1: A TB-GSS will enhance team trust for a virtual team, relative to a traditional GSS.

Team cohesion refers to the attraction of a team and the closeness that team members feel toward each other (Seashore, 1954). Team goals consist of achievement and maintenance goals. Team maintenance goals generally involve relationships among team members. Shared team maintenance goals can help bind a team together and give the team a basic level of cohesiveness (Barker et al., 1991; Johnson & Johnson, 1987; Schultz, 1992).

Further, shared team goals generated in a GSS dialogue can enhance the sense of "we-ness" among team members (Lewin, 1951; Owen, 1985; Schultz, 1992). As a result, team members would feel closer and more attractive to each other.

Proposition #2: A TB-GSS will enhance team cohesion for a virtual team, relative to a traditional GSS.

Team conflict here refers to the personally oriented conflict that militates against team consensus and leads to a general dissatisfaction among team members (Gouran, 1992). Shared team goals can enhance the sense of "we-ness" (e.g., Schultz, 1992) and thus minimize chances of negative conflicts (Poole, Seibold, & McPhee, 1985). Further, literature review shows that team cohesion and conflict have a reciprocal causal relationship (e.g., Chidambaram, Bostrom, & Wynne, 1991; Shaw, 1981; Watson, 1987). Since a TB-GSS may enhance team cohesion (the proposition #1), it could reduce team conflict.

Proposition #3: A TB-GSS will reduce team conflict for a virtual team, relative to a traditional GSS.

GSS research literature indicates that although GSS can generally increase team decision quality, it fails to increase or sometimes even decreases decision satisfaction and decision process satisfaction (e.g., Benbasat & Lim, 1993; Dennis, Haley, & Vandenberg, 2001). This issue has become one of major concern in GSS research (e.g., McGrath & Hollingshead, 1994). A TB-GSS might be able to rectify this issue by enhancing both decision satisfaction and decision process satisfaction.

In a team's decision making process, A TB-GSS may increase team cohesion (the proposition #2) and decrease team conflict (the proposition #3). Hence, team members in such a cohesive environment with fewer negative conflicts should feel more comfortable to work together on decision-makings. As a result, team members may be more satisfied with such a cohesive decision-making process.

Proposition #4: A TB-GSS will increase decision process satisfaction for a virtual team, relative to a traditional GSS.

From a goal perspective, final decision of a team is generally for the purpose of attaining team goals. In a GSS dialogue, shared team goals are generated based on individual goals (see Step (2) of the proposed framework), and consequently, the isomorphism between individual and team goals can be fostered. Thus, the final decision for attaining shared team goals may be perceived by team members as simultaneously attaining their own individual goals (Mackie & Goethals, 1987). Therefore, team members are likely to be more satisfied with the final team decision.

Proposition #5: A TB-GSS will increase decision satisfaction for a virtual team, relative to a traditional GSS.

Review of research literature indicates that team cohesion and team performance are often co-related (Littlepage, Cowart, & Kerr, 1989; Shaw, 1981), which is further confirmed by a meta-analysis (Evans & Dion, 1991). A highly cohesive team can be productive while team value is placed on accomplishing team task (e.g., Goodman, Ravlin, & Schminke, 1987; Schultz, 1992; Shaw, 1983). The team value of accomplishing team task can be reflected in the form of team achievement goals in a GSS dialogue. As a result, a more cohesive team supported by a TB-GSS (Proposition #2) may lead to a better team performance such as team decision quality.

Proposition #6: A TB-GSS will increase decision quality for a virtual team, relative to a traditional GSS.

A controlled laboratory experiment may be used to empirically test the above-mentioned research propositions in future studies. A Web-based GSS, TCBWorks (Dennis, Pootheri, & Natarajan, 1997), can be adopted to embed the team-building framework to support virtual team-building. Teams with and without the embedded team-building framework can be used to test whether the team-building framework could help a team develop more effectively and efficiently.

Further Discussion

The proposed framework can be used for different purposes in different ways. For example, it can be used to address one general team issue such as enhancing virtual team-building; and it can also be used to address one specific team issue such as role clarification, interpersonal relationship, or team socio-emotional interaction issues.

For a new virtual team, the framework could be even more important. Initial meetings of a new team are so unique and important that they set the tone for the team's future actions (Chidambaram & Bostrom, 1996; Gersick, 1989). A new team at this early stage can be most vulnerable. Team members are eager to know and familiarize with each other, but due to lack of shared meanings and values, they may not know how to start and how to co-orientate further from the start. Team members may have various high expectations towards others and the team, but due to the least available structures as a necessary guidance for interactions at this stage, they may not know how to adjust themselves to fit practical situations. These, if not resolved properly, can result in low member participation, high anxieties, and much dependence on team leaders (Bales & Strodbeck, 1951; Chidambaram & Bostrom, 1996; Mann, Gibbard, & Hartman, 1967). Consequently, team-building may be hindered.

However, a TB-GSS can help generate the necessary shared meanings, values, and provide team structures (the framework itself can be such a structure) to guide processes of initial team co-orientations and interactions. Further, with the container and suspension of dialogue, team members can interact with each other with less fear of social offenses or negative feedback from others. As a result, a new team may be better built up and more likely to function as a cohesive and effective team in the future (Gersick, 1989).

Further, team needs/goals (as well as individual goals) often change overtime, and thus, communication protocols, rules, and structures already existing in a team may need to be changed to match these changed goals (Johnson-Lenz & Johnson-Lenz, 1981; Turoff et al., 1993). As a result, a TB-GSS could be periodically used to support dynamically changing needs/goals of teams.

The conceptual framework also suggests some research questions for future studies. First, empirical studies can be conducted to directly test the conceptual team-building framework. For example, laboratory research can be used to study impacts of each component of the framework on virtual team-building.

Second, review of GSS literature indicates that most prior GSS research focuses on supporting task-oriented teamwork and largely neglects socio-emotional activities of a team. However, research reports that even in task-oriented computer system design meetings, only 40% of meeting time are spent in task-focused discussion (Olson et al., 1992). Task and socio-emotional activities coexist in and are equally important to a team (e.g., Bales & Strodbeck, 1951; Blake & Mouton, 1978; Cartwright & Zander, 1968; Huang & Wei, 2000). More research is therefore needed to study how to support a team's socio-emotional activities using a GSS.

A TB-GSS provides a virtual gathering place as well as structures for team members to interact socially, where team members can disclose and exchange personal experiences, preferences, values, likes/dislikes, and internal needs/goals. Further, a TB-GSS with dialogue structures of container and suspension can result in team members having a good sense of psychological safety (Schein, 1993). In this way, team members might feel more comfortable to share and be open to personal and controversial information (Olive & Langford, 1987). This kind of socialization of sharing personal information is important to the trust building of a team, since it can bring team spirit with it (Conlin, 1989; Hare, 1992). Therefore, future research can specifically study how a TB-GSS can be used to support socio-emotional activities of a team, a largely neglected aspect in GSS research.

---