2.0 Collaboration and Groupware

For the context of this study, collaboration is broadly defined as the interaction among two or more individuals and can encompass a variety of behaviors, including communication, information sharing, coordination, cooperation, problem solving, and negotiation. Groupware refers to the set of technologies available to support collaboration through the use of computers. The field of study that examines the impact of groupware on group behavior and performance is termed computer supported cooperative work (CSCW). Computer-mediated communications is another term frequently encountered that relates to collaboration and refers to communications among individuals that are conducted through the use of computers (i.e., through the use of groupware).

Groupware consists of a wide range of technologies. One common method of categorizing groupware is the place and time of the group interactions that the tools support (Johansen, 1988). Collaboration can occur among individuals at the same time (synchronous) or through a time delay (asynchronous). Collaboration can also occur between people who are located at the same place or who are separated by physical distance. Figure 1 provides a list of common groupware tools categorized in terms of the place/time paradigm. While many of the tools can fall into more than one category, the matrix provides a useful model to think about groupware and the various collaboration activities it can support. The focus of this study is on the collaboration of groups separated by time and/or location. Therefore, technologies and issues that fall into the same place/same time cell of the matrix are not addressed.

2.1 Computer Supported Collaborative Work (CSCW)

A growing body of literature in the field of CSCW examines the impact and issues surrounding the implementation of groupware. The literature can be characterized into three categories. One area of the literature includes the results from academic research. This research generally addresses experiments conducted within a university laboratory setting with faculty and students comprising the teams under study. The teams are either formed specifically for the purpose of the evaluation and work through a contrived scenario, or existing teams of students and faculty are studied within the context of their academic responsibilities. The primary limitation of this research is the degree to which the results can be generalized to real-world teams operating within industry and government.

A second area of research emerges from academic researchers who have expanded their investigations beyond the laboratory and conducted detailed, longitudinal examinations of groupware applications in industry and government. These studies, perhaps, provide the most rigorous data available on groupware, but they are limited in number due to the extensive resources required to conduct field studies. The final category of literature stems from anecdotal reports from industry based on their experiences with groupware. Although lacking the rigor typically associated with academic research, these case studies provide significant insight to real-world accomplishments and issues associated with groupware. Similarities between observations reported across varying organizations and business problems lend credibility to these findings. This report draws on the findings from all three sources of information on groupware and its support for collaboration activities.

2.2 Virtual Teams

Virtual teams are characterized by the considerable amount of communication that is conducted online (McMahan, 1998). While geographic dispersion or temporal displacement (e.g., shift work) among team members typically drive online communications, it is the degree of online communications, not the dispersion or displacement of the team, that characterizes a team as virtual. This means that a group that is collocated but still conducts the majority of their business online may be considered a virtual team.

There is a large body of research that suggests virtual teams communicate less effectively than face-to-face groups (Chidambaraum, 1996; Hightower & Sayeed, 1996; Warkentin et al., 1997). This research proposes that the lack of paraverbal (tone, inflection, and volume) and nonverbal cues (body language such as eye contact, facial expression, and hand gestures) in computer-mediated communications significantly degrades the flow, context, and content of group interactions. A review of the research in this area found that interactions among individuals engaged in computer-mediated communications differ in several key areas from face-to-face teams (McGrath & Hollingshead, 1994). Researchers frequently observe more equal participation among members of virtual teams. This equality of participation is attributed to lower status members being less inhibited in computer-mediated environments. Virtual team interactions also tend to be more focused on task execution and less on social behaviors due to the depersonalization of the communication context and frequent failure to develop strong personal relationships. Studies have also found that individuals express more negative and uninhibited messages during computer-mediated communications. Finally, virtual teams have more difficulty reaching consensus than face-to-face teams. Researchers attribute this finding to a lack of interpersonal feedback and reduced concern with social norms.

Critics of this research argue that the findings are limited because the groups in the studies were ad hoc, and the time period was not sufficient to establish effective working relationships. Recent research on this topic suggests that the differences between virtual and face-to-face teams may not be as predominant as earlier implied. Studies have found that virtual teams may communicate as effectively as face-to-face groups provided they have sufficient time to develop strong relationships and adapt to the groupware tools (Chidambaram, 1996; Warkentin et al., 1997).

Industry experiences have shown that virtual teams required special management attention. Companies such as Hewlett-Packard and Whirlpool discovered that virtual teams required more formal communication than traditional collocated teams because there was less informal exchange of information among team members (Geber, 1995). Formal mechanisms had to be established to distribute information among team members that would be passed in more informal ways among collocated workers. More formal project management principles and tools were also deemed necessary to assign responsibilities and track the progress of virtual teams.

2.3 Potential of Collaboration

The potential of groupware tools to greatly enhance efficiencies in business operations is driving widespread interest in these tools. The time for groupware is ripe as many organizations are challenged with integrating geographically, often globally, dispersed teams. Primary motivations for deploying groupware include cost reductions, increased productivity, and leveraging professional expertise (Coleman, 1997). When groupware is successful, the benefits accrued can be dramatic.

A study of the impact of groupware on product development at a major power tools manufacturer found that the use of electronic bulletin boards and shared documents reduced the time to complete tasks by reducing the time team members spent on non-value-added activities (D'Souza et al., 1997). Examples of non-value-added activities included travel time to meetings, printing and distributing hardcopy documents, and phone calls to request information that was available online to the groupware users. Groupware reduced these non-value-added activities from 44 percent to 6 percent of the team's time.

Groupware also had an impressive impact on team operations within IBM. IBM studied 30 teams over the course of a year as they addressed production-line quality issues. Teams that used groupware tools expended an average of 50 percent fewer labor hours to work the problem than teams that used traditional methods (e.g., phone, face-to-face meetings). The teams using groupware also completed the task an average of 91 percent faster than traditional teams. To ensure that these findings were not isolated to a specific task and group of teams, IBM conducted a second study that tracked 50 teams located across six IBM sites. These teams worked a variety of business problems. Labor costs for the second set of teams using groupware were reduced by an average of 55 percent, while project completion time was reduced by an average of 90 percent (Grohowski et al., 1990).

One of the most widely publicized success stories for groupware is from the Boeing Corporation. Boeing tracked the progress of 64 teams that used groupware tools to define design requirements for the shop floor of the 777 aircraft. Team activities supported through the use of groupware included problem definition, design alternative generation and evaluation, planning, and documentation of group decisions and accomplishments. The use of groupware reduced project execution time by an average of 91 percent and labor costs by 71 percent. The estimated return on investment for the groupware tools during the first year of its pilot implementation was 170 percent (Post, 1992).

The previous examples are just a small subset of industry's success with groupware. However, successful implementation of groupware is rare. Researchers generally agree that there are more groupware failures than successes. Even when companies are able to successfully apply groupware in small pilot programs, these pilots rarely lead to full deployment in the organization. For example, despite the dramatic success at Boeing on the 777 program, groupware at Boeing is still limited to only a few teams and locations. Boeing believes that "groupware is advancing slowly because it conflicts with the existing organizational, social, computing, and network infrastructures" (Poltrock, 1996).