Group facilitation in a networked world

Beise, Catherine M

Abstract

Group support systems (GSS), initially developed to support problem-solving groups in face-to-face meeting settings, are extending their capabilities to support meeting participants separated geographically and temporally, as a result of advances in networking systems and application software. Facilitation is viewed as an important factor in the success of face-to-face GSS meetings. This article explores the role of the meeting facilitator in assisting distributed group meetings supported by various technologies. Interviews with 34 practicing facilitators reveal their concerns and expectations regarding benefits and limitations of distributed GSS (DGSS). The interview results offer useful insights to DGSS designers, researchers, and practitioners. The facilitator's concerns include potential loss of non-verbal signals in addressing group process issues such as participation and conflict resolution, while they perceive that DGSS can offer benefits such as focusing and structuring. The facilitator's role is likely to continue to include serving as a change agent, while evolving from individual meeting manager to that of project manager, participant trainer, and technology enabler. Traditional facilitators will likely have to increase their skill and comfort with information technology, as well as adjust and adapt to new tools and methods for accomplishing their traditional tasks.

Group Support Systems (GSS) have been used by facilitators to lead co-located teams for over a decade. A rich collection of facilitation tools and techniques have evolved to enhance the use of co-located GSS. With the development of the Internet, we are now beginning to see development of sophisticated distributed GSS software. Distributed GSS is too new for researchers to have yet collected significant data about its use. Nevertheless, it is vital that researchers begin to explore the issues and implications of facilitating distributed teams. This article is such an exploratory study. The authors have surveyed 34 experienced facilitators probing for distributed facilitation issues. The authors use the information gathered to inform both facilitation process and software development.

Daniel Mittleman

Introduction

Organizational researchers and industry practitioners widely agree that the modern enterprise is evolving towards new organizational forms. These new forms are generally characterized as team-based, global, participative, flexible, learning, and continually shifting as firms move in and out of various intra- and inter-organizational relationships (Smith and Berg, 1988; Frey, 1994; Fulk and DeSanctis, 1995; Kanter, 1995). Professional facilitators play a key role in supporting the organizational change required for firms to adapt continuously and successfully to external environmental demands. Information technology (IT) is also an important factor in catalyzing and facilitating organizational evolution, through the strategic use of data, through sophisticated analysis, and through the rapid communication and exchange of all forms of information across the globe. This article discusses the role of facilitation in supporting task accomplishment and process management of distributed teams. Teamwork and meetings distributed across place and time generally require at least some level of technology in order to occur at all, and facilitators must interact with this technology in order extend their work into this new setting.

The article first reviews the application of IT to face-to-face (FTF) meetings and the critical role of facilitation in making it work effectively. The article then extends the FTF meeting model to distributed settings and reviews some of the communication issues that arise when people must work together from different locations. A description of several types of technology that make such meetings possible sets the context for envisioning scenarios in which such technologies could be used. Next, the article discusses the results of interviews with practicing facilitators regarding their perceptions of the pros and cons of using technology to support distributed meetings. The discussion concludes with guidelines for facilitators who are working with distributed teams and a list of critical issues for DGSS researchers and designers.

Face-to-face Meetings and Information Technology

One of the ways that information technology has been applied to organizational change is in streamlining and making more productive the time that people spend in meetings. The term group support systems (GSS) is used in this article to encompass electronic support for activities and tasks required by teams and workgroups. Group support systems (GSS) have been applied to a variety of meeting tasks, including strategic planning, team development, and decision-making. Facilitation is often cited as one of the important factors in successful application of GSS to FTP meetings (McGoff, Hunt, Vogel, and Nunamaker, 1990; Bostrom, Anson, and Clawson, 1993). In this article, "facilitator" refers to professional organizational development consultants, team leaders, management development trainers, and others who have responsibility for managing meetings and developing teams. Presented in this article are the results of interviews with practicing facilitators who were questioned about their views on facilitating distributed groups.

Research evidence supports the importance of the facilitatoris role in successful group interactions using GSS (McGoff et al., 1990; Daniels, Dennis, Hayes, Nunamaker, and Valacich, 1991; Bostrom et al., 1993). Groups using GSS and a facilitator showed improved group process and cohesion for a low-conflict task, when compared to GSS groups with no formal facilitator (Anson, Bostrom, and Wynne, 1995). When either human or automated facilitation was provided to groups working on a multi-criteria modeling task, both group consensus and perceived decision quality increased (Limayem, Lee-Partridge, Dickson, and DeSanctis, 1993). An extensive field study using a GSS for strategic planning in several companies further supports the argument that the role of the facilitator is critical for successful GSS use (Ackermann, 1993).

Distributed Meetings and DGSS

The term distributed meeting refers to the interaction among a group of individuals not all located in the same room (i.e., not face-to-face/FTF), for the purpose of engaging in common activities, for example group writing or group decision making. The definition includes all technological support mechanisms for its implementation. In this study, distributed meetings are viewed as an extension of FTF electronic meeting systems.

Various underlying hardware and software technologies allow individuals or groups in different locations to interact in a variety of communication modes. For simultaneous interaction, distributed meetings have traditionally relied on voice connection through the conference call or teleconferencing. In this format individuals at different locations exchange voice messages. Videoconferencing adds the use of cameras, display screens or monitors, and video transmission to supplement voice with pictorial information exchange. Computer data can also be transmitted between sites by itself, in conjunction with voice, or with both voice and video information. GSS software supporting distributed meetings extends beyond the simple one-to-one communication of electronic mail, for example, by providing specific functionality for group tasks, interaction, and decision-making.1

One of the biggest differences between FTF and DGSS meetings is that the processes of FTF meetings are familiar and comfortable (if not necessarily desirable). The metaphor of FTF meetings may be easier to extend into a synchronous (same time) DGSS session than into an asynchronous (different time) one. The asynchronous distributed "meeting" is a new phenomenon that may significantly change the behavioral rules, in terms of how work gets accomplished or exclude some tasks generally associated with FTF meetings such as general communication/announcements or decision making/voting. Discussions of DGSS in this article will refer mainly to synchronous distributed meetings, except where noted otherwise, and where consideration of asynchronous meeting mode increases understanding of distributed meetings in general.

Comparative studies of distributed and FTF groups have investigated whether technical limitations inhibit the group's ability to communicate in the same way that they do face-to-face, and thus potentially reduce the group's ability to accomplish a team task. Earlier studies of information richness in communication media noted that reductions in "social presence" and "media richness" would be likely to present serious constraints on interpersonal communication (Short, Williams, and Christie, 1976; Daft, Lengel, and Trevino, 1987). The limitations of electronic media in transmitting multiple types of information needed for human communication (emotional, non-verbal, facial expressions, body language) would limit the ability of people to work together from remote locations.

More recent results have challenged the range of tasks over which information richness theories might apply. Less complex, more structured tasks may be adequately served by "leaner" media (Markus, 1994b; Burke and Chidambaram, 1995; Panko and Kinney, 1995). Lack of a non-verbal channel does not significantly reduce the feedback necessary for adequate comprehension needed for team member interaction (Rao, 1995). Dispersed groups can still make effective decisions, although they may be less satisfied (Gallupe and McKeen, 1990). This may be due to having to work within the constraints of a particular tool or technology (Cass, Heintz, and Kaiser, 1992).

Distributed GSS may be adequate for certain tasks or phases in various group processes. Dispersed groups may even have advantages over FTF groups. The reduction in social cues of a distributed group may contribute to a greater task focus. A distributed environment in asynchronous mode may lead to more time to reflect (Turoff, Hiltz, Baghat, and Rana, 1993), resulting in richer ideas. Distributed groups can be more creative than FTF groups for structured tasks (Ocker, Hiltz, Turoff, and Fjermestad, 1995). DGSS in the asynchronous mode were found to make it possible for larger groups to work together, although participation may not be spread equally, depending on the task, the technology, and the perceived benefit of working on the task (Turoff et al., 1993). In the synchronous DGSS mode, even with high-quality audiovisual conferencing equipment, facilitators and group members are limited by how many screens, conversations, and threads of content they can follow. Creative application of facilitation and technology may allow expansion of group size in the synchronous DGSS mode, but it is also quite possible that effective group sizes may be comparable to or even smaller than in FTF settings.

Evidence from several case studies suggests that the social and organizational context in which electronic media are introduced and used have more influence on outcomes than do differences in media or geographic location (Lea, O'Shea, and Fung, 1995; Zack and McKenney, 1995). Groups select, adopt, and adapt media to fit their needs, their organizational context, their social structures, and their work habits (Haythornthwaite, Wellman, and Mantei, 1995; Hinds and Kiesler, 1995). Orlikowski, Yates, Okamura, and Fujirnoto (1995) describe the process of "meta-structuring," or technology-use mediation, in which local experts assist in the assimilation of new information technology through interventions in the group's processes, the context in which they use it, and even in the design of the DGSS. The consequences of such technologies will depend on how people understand and use it (Markus, 1994a). In a DGSS environment, the facilitator can play the role of the local expert assisting in the identification and assimilation of new DGSS tools and accompanying techniques.

These studies suggest that, as organizations continue to frequently re-structure themselves in response to their competitive environment, they will still be able to successfully accomplish tasks assigned to team members who are geographically and temporally dispersed. The communication and coordination processes needed to accomplish the tasks, however, may have to change, and one increasing role of the facilitator will likely be that of change agent. As GSS tools are implemented in organizations, human intervention will be needed to implement and manage their use. The facilitator's role could become more, rather than less, important.

Research Framework

Traditional (non-GSS) facilitation has been described as having three dimensions: substantive, procedural, and relational (Hirokawa and Gouran, 1989). The substantive role is concerned with facilitating acquisition, manipulation, and use of information needed for problem-solving or decision-making, thus focusing on meeting task. The procedural dimension is process-oriented, concerned with maintaining task focus via specific steps and tactics. The relational dimension involves management of interpersonal and group social interaction, in fact minimizing problems arising from such interaction in order to maintain a focus on the task (Hirokawa and Gouran, 1989).

A long history exists of analyzing meetings by considering them in two major dimensions: process and content (e.g., Bales, 1950). "Content" corresponds to the "substantive" role, while "process" encompasses both the procedural and relational roles, as defined above by Hirokawa and Gouran (1989). This concept has been extended to GSS facilitation (e.g., Fuller and Trower 1994). The complexity of the facilitation role, with or without technology, is further illuminated by sixteen critical activities identified by Clawson and Bostrom (1993). Table 1 maps these 16 activities to Hirokawa and Gouran's three dimensions. Technology is added as a fourth dimension.

Task. In order to achieve its task (examples of tasks include strategic planning, product design, and resource allocation) a group follows a set of standard task activities. The facilitator's role is to keep the group focused and move it along in a timely manner toward its goal. The challenge in a DGSS environment is how to accomplish this without FTF contact, in either a synchronous or asynchronous environment.

Procedural and Relational (Process). The facilitator in the FTF setting has a wide array of verbal and non- verbal cues to observe and build upon, as well as a toolkit of instructions, interventions, and tools. These are all used by the facilitator to move the group in one direction or another as well as to sense, surface, and resolve a variety of interpersonal issues that may inhibit the group's progress toward its team development and goals. Team members undergo stages of team development, (forming, storming, norming/integration, performing, adjourning2) involving social processes that may require skilled facilitation. This dimension of facilitation appears to be particularly challenging in distributed interaction, given a reduced set of cues.

Technology. DGSS technology has the potential ability to replace, support, or inhibit human facilitation. Successful adoption and use of DGSS technology will include identifying the current technologies that exist, forecasting their continuing rapid evolution, acquiring/developing and implementing software that makes them useful, and adding new features as group needs evolve and change. How transferable are the skills needed in FTF facilitation when applied in distributed settings, and what are the mechanisms needed for developing those skills?

The studies referenced in the previous discussion that examine computer-mediated communication indicate a fifth dimension, the organizational context in which the group process, facilitation, task, and technology are implemented. These five dimensions provide a useful framework for examining the issues in facilitating distributed meetings supported by technology.4

GSS facilitation is generally defined as being shared among the team leader, the participants, and the technology (Bostrom et al., 1993; Fuller and Trower, 1994). This applies to the distributed situation also. However, the proportions of the facilitation task assigned to these each of these three "actors" may differ when moving from an FTF to a distributed environment, and may also differ between synchronous and asynchronous modes.

Some recent studies in DGSS facilitation assume that the bulk of the facilitation burden will have to be taken up by the technology, given the absence of FTF human interaction (Dubs and Haynes, 1992; Limayem et al., 1993). However, in a study of a hypertext system used to support a distributed group, human moderation was needed to keep hypertext discussion organized and on-going (Ocker et al., 1995). Human facilitation is both possible and likely to be helpful in both synchronous and asynchronous distributed group situations.

Research Methods

As a means of exploring these issues, a group of practicing facilitators responded to interview questions about their experiences, concerns, and expectations about facilitating distributed meetings. This study focuses on the observations of facilitators who are embedded in organizational situations involving both group meetings and different degrees of technology support. Tapping the expertise and perspectives of these experienced facilitators provides a potentially rich source of data for examining expectations and concerns about DGSS facilitation. It should be noted that these facilitators had very little experience in the actual facilitation of meetings in the DGSS context. Based on the experience of the authors, most practicing facilitators currently do not have a great deal of either GSS or DGSS experience. Developing an understanding of the expectations and concerns of pre-DGSS experienced facilitators can provide insights into issues of technology adoption and design considerations. It should not be interpreted as a substitute for examining "best practices" derived from the experiences of expert DGSS facilitators. This article aims to raise questions and anticipate emerging issues as organizations increasingly utilize distributed meetings, by addressing the following research questions:

In moving from the FTF to DGSS meeting environment,

1) what concerns do facilitators have regarding group task, group process, the technology, and their own role as a facilitator in an organization, and

2) what benefits do they as facilitators anticipate and/or desire?

A number of the more recent studies mentioned above regarding computer-mediated communication use qualitative research methods in order to focus on a few entities in depth, rather than many entities more selectively (Orlikowski et al., 1995; Zack and McKenney, 1995). This is motivated by the need to understand the complex interaction among many variables that contribute to outcomes in organizational settings. Given the number of factors and the difficulty of isolating and controlling individual variables, the use of qualitative methods are also needed in studying facilitation (Bostrom et al., 1993). This study takes a qualitative data analysis approach, where much of the data appears in words rather than numbers, is acquired through interviews and tape recordings, and is processed via transcription prior to analysis (Miles and Huberman, 1984). This allows the collection of richer, more meaningful data within a natural context.

The researchers conducted individual in-depth semi-structured interviews with practicing facilitators, in order to tap into their wealth of experience and expertise in managing groups. Data presented in this article is a subset of data collected, either in person or over the phone, with thirty-seven professional group facilitators. Three respondents did not complete questions regarding distributed meetings. Thus, analysis for this article is based on the thirty-four completed interviews. Facilitators reported varying levels of experience leading groups with and without GSS technology.

In order to analyze the results, all comments were initially transcribed from the interviews into lists under each question. Each researcher independently clustered the comments into categories. Then, the three researchers examined the results together and verified their common categorizations.5

Of the thirty-four facilitators participating in this study, fifteen (44%) stated that they have participated in one or more distributed meetings. Facilitators participated in meetings involving audio-conferencing and teleconferencing; different room, same building with facilitator running back and forth; videoconferencing; and participating in a demonstration of a distributed GSS meeting. Experience with distributed meetings was evenly distributed among facilitators with high and low levels of GSS experience.6 The sample was also relatively evenly distributed between facilitators with high and low levels of traditional meeting experience.

Results and Discussion. Respondent comments clustered naturally into the categories mentioned above: the substantive or task dimension pertaining to meeting content; process issues divided into procedural and relational, pertaining to structuring the meeting activities and interpersonal issues; technology issues; and organizational issues, that pertained to the setting within which GSS meetings are conducted. An additional category resulting from the interviews was around issues related to the role of the facilitator. Some comments clearly indicated that the respondent's perspective was influenced by their experience with a particular FTF GSS such as Group-Systems(R), and that they were probably not considering other types of tools such as shared whiteboards.

Task-related Issues. Respondents did not state many task-oriented concerns about the distributed environment. They suggested that in some cases DGSS may be the only feasible alternative to assembling a particular team made up of individuals with specific and unique skills and/or experience. This may be especially true in the non-simultaneous environment where each busy group member can contribute ideas or votes during intervals rather than wait for all to become available at exactly the same time.

Some respondents suggested that having to put messages into words would make the meetings more focused and productive. Several comments indicated a sense of increased amounts of information resulting from technology support, but there was concern with information overload or having more garbage along with more information. Some facilitators who were familiar with FTF GSS commented that future tool design should focus on consolidating rather than generating information, for example sorting, evaluating, categorizing, and summarizing.

Process Issues. Many of the process issues raised assume a synchronous DGSS perspective, reflecting the facilitator's experiences with same time meetings, with or without GSS.

Procedural. The DGSS meeting provides an opportunity to add structure to the sequencing of group activities. As one respondent noted, it may force people to take more time to think before they respond to questions. This could be true in both synchronous and asynchronous situations. However, participants may become disoriented. They may not understand the meeting structure or design, may have difficulty knowing when it is time to input ideas and when it is time to vote (or undertake other actions), and as a result may participate less than desired. The respondents expressed concern about the ability of the facilitator to intervene and control the meeting, given restricted communication channels. Again, the constraints on intervention could apply to either synchronous or asynchronous modes.

While planning is considered a key factor in the success of meetings in general (Kayser, 1990), it is even more important with GSS meetings (Clawson and Bostrom, 1993). Planning in the context of GSS includes identifying objectives, participants, tasks, and roles; developing an agenda; and mapping GSS tools to tasks. The facilitators in the present study strongly echoed the necessity for adequate time spent planning. They emphasized that synchronous distributed meetings especially will require even more planning and preparation than FTF GSS meetings, and otherwise would be more likely to be a waste of time.

Successful implementation of distributed meetings will demand higher levels of discipline, restraint, and responsibility from participants. DGSS may be designed to impose more structure on meetings and group work tasks. This may, in turn, generate better organization and more efficient "meetings." For instance, one respondent described a long-term project within her organization requiring participants, sponsors, and facilitators to develop, without computer support, detailed checklists of pre-meeting and during-meeting activities. In this case, structure was imposed through organizational mechanisms (top management edict), but once adopted as a new norm, was viewed very positively as a tool for increasing meeting effectiveness. DGSS might equally provide such a structured environment, though it is not clear how readily groups will persist with it unless it also has organizational backing (top management support).

Coordination of activities among multiple sites, particularly if there were not a facilitator at each site, was another concern. Respondents anticipated that the dynamics of the meeting would differ based on which group is co-located with the facilitator. This would be a problem mainly for groups acting in the synchronous mode. Facilitators also need to manage member subgroups and to control side conversations that can distract from the task focus. On the other hand, side conversations may be an indication of other issues that should be surfaced and addressed in a subsequent meeting, even if not relevant to the task at hand. A facilitator in a remote location will not be aware of verbal conversations among respondents at the same remote site unless they communicate the side issue electronically and the facilitator monitors that communication.

Relational. Virtually all respondents were concerned about process issues in facilitation of distributed group meetings. More than a third observed that facilitators, in much of what they do, currently use non-verbal cues7 that may be missing or changed in a distributed format. Respondents were concerned that a narrower range of information channels would restrict their ability to resolve conflicts, to gain consensus, to deal with politics, to gain commitment both to the meeting process and to post-meeting follow-through, and to build trust and rapport. They feared that their ability to manage participation (encouraging those who tend to withdraw and to tone down those who dominate) might be restricted. Several respondents suggested that they would be more receptive to using DGSS meetings for groups who had previously come together in an FTF setting where interpersonal affective relationships can be built.

Surfacing and resolving conflict among group members, particularly where tasks involve distribution of resources or setting of policy, is critical to performance. Groups that surface but do not resolve conflict can be unproductive and actually decrease the team members' desire and ability to work together. On the other hand, if genuine conflict is not surfaced, the likelihood of an agreed upon action being enthusiastically implemented is not high. Voting and similar tactics can make explicit the level of conflict that exists within the group (Poole, Holmes, and DeSanctis, 1988; Sambamurthy, 1989). The respondents in this study were concerned about the difficulty of managing this process in a distributed environment.

Technology Issues. The most frequently mentioned issue was system reliability. Some facilitators have experienced technical difficulties when using GSS during FTF meetings. In situations where group members are hesitant about using computers or are dealing with critical organizational issues, technical disruptions in the meeting can be disastrous. The respondents emphasized the importance of thoroughly testing all aspects of the technology that can be invoked during the meeting. In an FTF GSS setting, ideally, someone will be present to ensure that the technology is supplemented with instructions (both what to use and how to use it) and that minor glitches can be quickly remedied. To the extent that DGSS are used across heterogeneous platforms, a facilitator or network technician at each site would need to check on the appearance of the application display. In the asynchronous DGSS meeting, general network management software should be helpful in noting that messages are stored before transmission, checked for reception, and resent when necessary.

Facilitators expressed concern that the technology would become the central focus of the group; that it would be distracting; and that group members would not use it appropriately. In an FTF setting, facilitators may observe group members drifting away from the common task toward exploration of the system itself, but can easily address this by requesting a return of the participant's attention to the group. On the other hand, drifting from the common task may be an indicator of other issues, such as a feeling of disenfranchisement from the process, or a sense that the issue is not relevant (that the individual should not be in that group). In a DGSS setting, particularly in a dedicated setting such as an electronic conference room, this same issue could be difficult to resolve without a facilitator at each site. In the asynchronous DGSS setting, it may be even harder to track involvement. Postponing completion of a task (on which other group members rely) can become a strategic political maneuver aimed at influencing the outcome.

Another interpretation of the potential for technology to overshadow task is the potential for technology to determine or limit the method(s) of addressing the problem. Not all problems are amenable to a single sequence of activities. Yet a highly structured DGSS might force group members into following a prescribed path. A more flexible system would allow the facilitator and/or group members to rearrange activities for a given problem. However, in an FTF setting, a creative facilitator might generate a new method or drop out of the agenda to discuss the meta-issue of how to proceed. This may be difficult or impossible in the DGSS meeting.

Facilitators expressed concerns but also made suggestions for technology design that could help them. Rather than being daunted by "less rich" (Daft and Lengel, 1986) communication channels, they suggested screen windows for viewing graphical and biographical information about other individual participants. They also suggested the development of communication protocols for different types of messages, such as separate channels for technical versus task-related issues. They looked forward to video as well as audio as enriching supplements to the electronic channels of current DGSS. The rapidly improving speed and capacity of network technologies is likely to reduce the importance of this issue.

Organizational Issues. In general, the respondents viewed DGSS as potentially providing a lower cost alternative to meeting at the same location, and in reducing the need for FTF meetings. Only one facilitator specifically mentioned cost as a potential obstacle to installing the hardware, software, personnel, and procedures to support distributed meetings. Currently, particularly with the installation of dedicated facilities, start-up costs can be quite high. However, the degree to which distributed meeting procedures can utilize the existing and growing computer and communication networks already within firms is probably increasing.

Respondents were concerned that DGSS might inhibit the team and community building within an organization needed for the organizational change process. On the other hand, DGSS could provide individuals with team membership opportunities that otherwise would not occur at all, due to time, geographic dispersion, and/or prohibitive travel costs (e.g., multinational corporations and global partnerships). As a by-product, the ability to include more stakeholders and to extend a discussion over a longer period of time could result in higher quality outcomes and increased individual commitment to organizational goals.

The Facilitator Role. Some respondents expected the facilitator role to diminish, perhaps even disappear. Others thought the role would expand and become even more difficult. Perhaps in the narrow sense of the facilitator standing before the group and conducting the meeting, the role will shrink. However, in the larger sense of applying expertise and direction to group process during meetings, the role will evolve into one with new responsibilities and different tools for handling them.

Guidelines and Caveats for Facilitators

The transition for the facilitator from FTF meetings to DGSS meetings offers both risks and opportunities. This section reflects on the respondents' views on how the role of the facilitator might change. Table 8 summarizes and compares the key issues as groups move from FTF to distributed contexts.

Task. With respect to task, facilitator expertise will be needed for

* selecting organizational problems that lend themselves to the DGSS approach considering the abilities of the group at a given time;

* decomposing projects so that segments can be performed using DGSS then integrated into the overall task;

* initiating meetings and matching tools to activities;

* assuring that group process is addressed so that the group's attention is on task performance.

DGSS technology could in fact streamline the decision process such that decisions are made more quickly and painlessly, but without the richness of consideration that would lead to a better solution.

Process/Procedural. With DGSS, the facilitator must become more of an information coordinator and gatekeeper, monitoring and integrating activities at multiple sites, moving group members from one activity to another, detecting and intervening in counterproductive behavior, and running complex software. If the DGSS meeting grows to sufficient size, it may become necessary to use a hierarchy of facilitators operating local nodes combining into larger accumulations. Facilitators may have to develop new control strategies and/or accommodate a less controllable environment.

Facilitators sometimes form subgroups to divide up appropriate meeting subtasks. Knoll and Jarvenpaa (1995) found participants of asynchronous meetings naturally forming into "subgroups" generally consisting of the more highly participating members. Defining separate mailing lists can generate the effect of subgroups; saved output from these subgroups can later be compared and contrasted (if replicated) or merged (if pertaining to separate segments of a larger project). Excellent DGSS facilitators will need to learn, based on the text stream of the meeting, when to break the group into subgroups, on what basis to create the subgroups, and how to reconnect their outcomes.

Process/Relational. Several respondents suggested that DGSS would be more appropriate for established groups whose members have at least met FTF and preferably have experienced some team development. Since this is not always feasible, participants can at least be contacted individually by phone prior to distributed meetings and provided with necessary meeting information, with what to expect (and not expect) from the technology, and with training in both the technology and in group process.

Experienced facilitators report two challenges regarding participation: getting quiet participants to "speak up" and restraining overly dominant participants. In an FTF GSS meeting, the technology itself helps the facilitator to encourage even participation, to reduce domination, to draw out all relevant ideas, and to encourage those with the most expertise or stake to participate fully enough to feel ownership for the results. In a DGSS meeting, the technology can continue to support these efforts. The exchange of messages in the distributed setting can be monitored electronically. Those with lower participation rates can be directly queried and encouraged by the facilitator, also electronically. Those with higher participation rates can be prompted to summarize their comments. The facilitator's challenge will be to incorporate into their toolkit these technology-supported approaches to monitoring and intervention.

Techniques for conflict resolution such as focusing on principles or goals rather than positions can be reinforced in the DGSS setting through structured prompting. Facilitator expertise will still be required to draw out disagreements, to get participants to continue to interact, and to highlight areas of mutual interest and emerging consensus.

Technology. Facilitators will need to monitor usage of connected stations allowing specification of particular tools for all (perhaps each) group member(s) to use. They will use tools to monitor whether individuals have gone down an alternate branch and will need to develop the expertise to know when to encourage or limit such exploration. For tasks requiring a high level of focus on the central activities, facilitators may benefit from the ability to lock participants into or out of particular features (and the ability to turn off those locks). Facilitators should also have a response delay to force a group to slow down and think through their answers.

The distributed group meeting obviously needs to provide the best possible transmission of data, voice, and/or video, with minimal delays and downtime. Underlying computer and communications technologies that support distributed meetings are complex and difficult, while at the same time are also rapidly changing and improving. Distributed meetings are likely to become a matter of choice rather than availability. Public networks such as the Internet are already providing resources for distributed meetings (Fellers, Clifton, and Handley, 1995). Intranets, initially used for broad dissemination of up-to-date information within organizations, are increasingly likely to provide interactive capabilities, some of which may benefit from facilitation. Facilitators can and should influence the evolving design, implementation, and technical support of DGSS.

Studies of asynchronous distributed teams have noted that in many cases, groups generally benefit (or would have benefited) from adding structure to the group's efforts and activities (Turoff et al., 1993; Knoll and Jarvenpaa, 1995). Synchronous meetings, given much more severe time constraints, would probably benefit even more from added structure. A window that lists the agenda items and highlights the current meeting status, if part of the system, could diminish potential confusion. A hierarchy of activity (such as a form of parliamentary procedure) could be built into the software to separate and structure activities aimed at treating the task and agenda. Too much structure, however, could potentially inhibit the generation of innovative ideas and creativity. This would undermine the value of DGSS in stimulating a broad diversity of viewpoints. At the same time, the group must be able to tolerate that diversity (Turoff et al., 1993). The DGSS facilitator can serve as a potential source of intervention to maintain a productive balance.

Organizational. One approach to viewing meetings is that they are necessary only in certain stages and activities of a project team life cycle (Johansen, 1989). Distributed meetings, either same time or different times, may be adequate or even preferable for other stages of the team's progress. The facilitator's focus is likely to move more to the project and less to individual meetings. As one of the interviewed respondents argued, the facilitator will become more of a project manager. The facilitator will assist and guide the group through these stages, by structuring their activities, monitoring and feeding back their progress, mapping appropriate tools and technology to tasks, and managing technology use.

Some of the responsibilities currently undertaken by human facilitators may shift toward meeting participants and to guidance features built into the technology. This implies the need for participant training in self-facilitation and for distributed GSS design to incorporate automated facilitation help for both the facilitator(s) and the participants.

Organizations should leverage the facilitator's positive perceptions of DGSS by providing facilitators with training in the technology. Facilitators offer a rich potential source of expertise and experience for "technology-use mediation" (Orlikowski et al., 1995), and should be provided with encouragement, support, and resources for their role as organizational change agents.

Quantitative cost benefit analyses reported about the use of GSS for FTF meetings in organizations (Nunamaker et al., 1989; Daniels et al., 1991; Post, 1991; Di Pietro, 1992) may also be applicable to distributed meetings. Over time, as use of technology becomes a new norm, the organization as a whole may benefit from more standardized, data-intensive, broadly inclusive, and well documented meeting activities. Also, organizations may have existing resources such as individual workstations or PCs that can support DGSS without additional cash expenditures. Similarly, with changes in personnel departments and training staffs, human resources may be shifted from other areas to perform facilitation tasks.

Conclusions and Suggestions for Research

Research that contributes to a greater understanding of the facilitation task should assist DGSS designers in eventually overcoming some technology-based restrictions, as well as in designing automated tools to assist or even replace facilitator roles. Questions that arise for future research include:

* What is the role of facilitation in distributed meetings ?

* How is facilitation of a distributed meeting different from that of a face-to-face group, and how is facilitation of a synchronous DGSS meeting different from an asynchronous DGSS meeting?

* How much can and should facilitation of distributed meetings be automated?

* What do limitations of interpersonal interaction across electronic channels of communication mean for facilitation?

Results of this study should be generalized to other facilitators and groups only with great caution. Findings are based on the views of a relatively small convenience sample of facilitators with little DGSS experience. In spite of these concerns, the facilitators interviewed collectively represent many years of experience with group work and provide a broad range of insight into what promises to be an important emerging technology. Systematic observation of human facilitators using DGSS will reveal whether highly experienced FTF meeting facilitators will be able to translate their knowledge into successful facilitation of distributed meetings, or whether newcomers unbound by learning from another context will be more successful.

Designers of DGSS need to address both task and process issues. Windowing technology can provide distributed groups with process information including instructions and individual messages from a facilitator or coordinator in one window and task information including text messages and voting or other quantitative models in another window. Hypertext techniques can be used both to create paths through a large information set and to maintain a trail for returning to issues of central focus. Expert systems can be designed to aid in selection and execution of complex mathematical models where appropriate for the group's task. They might also be designed to aid the facilitator with mechanical tasks such as agenda planning, estimation of when enough divergent thinking has been applied to an issue, and warning that the group does not seem to be nearing consensus. As noted in Turoff et al. (1993), tool support should be general and adaptable to a wide variety of problems, rather than focusing on specific decision support methods. Otherwise, the facilitator and participants may not have the flexibility they need to adapt to unanticipated meeting requirements, by changing the agenda, direction, methods, or tools.

As information technology expands its presence in organizations and in the facilitator's job, the facilitator will become more of a systems analyst, in gaining understanding of team and organizational needs, and in matching, designing, and helping to implement DGSS technology to satisfy those needs. At the same time, facilitators can serve as process re-engineers in helping teams learn how to work in new ways in distributed settings using DGSS and increasing an organization's ability to adapt quickly to change, to perform coordinated work, and to compete successfully.

Acknowledgements: We would like to thank the professional facilitators who contributed their time, experience, and knowledge to this, discussion. Appreciation is also extended to Hossein Arsham, Rao Vemuganti, Roger Volkema, and Lynne Markus for their comments on earlier versions of this text, and to Daniel Mittleman and several anonymous reviewers for their helpful suggestions.

1 The issue of selecting from among the various combinations of media for DGSS will not be addressed. The authors expect that due to steadily declining costs of hardware and networking, eventually all of these media will be available to employees within most organizations. Therefore, while some individuals' usage patterns may favor one or another media, designers of DGSS will simply provide rather than select among the available media.

2 See, for example, Schermerhorn, Hunt, Osborn (1997). Modern Organizational Behavior. New York: John Wiley & Sons, for a more extended description of this common model.

3 Item numbers indicate Clawson's original ordering of activities.

4 These dimensions are consistent with GSS research frameworks which divide GSS variables into dimensions of task, technology, group, and organizational (Pinsonneault and Kraemer 1989; Nunamaker, Dennis et al. 1991).

5 More details of the larger study can be found in Niederman, F., Beise, C. M. & Beranek, M. (1996). Issues and concerns in GSS: The facilitator's perspective. MIS Quarterly, 20(1), 1-22.

6 Less than 30 sessions was considered low for either GSS supported or non-GSS supported meetings.

7 Non-verbal cues are used by facilitators to detect a variety of messages from the group, such as boredom, frustration, confusion, understanding (or lack thereof), lack of involvement or participation. They also send non-verbal cues to the group to help create an amenable meeting environment, to manage individual participation, and to help keep participants interested and motivated.

References

Ackermann, F. (1993). The role of facilitators in GDS systems. University of Strathclyde, Management Science, Theory, Method, and Practice Series. Working Paper 93/1.

Anson, R., Bostrom, R. P., & Wynne, B. (1995). An experiment assessing group support system and facilitator effects on meeting outcomes. Management Science, 50(5), 11-55.

Bales, R.F. (1950). A set of categories for the analysis of small group interactions. American Sociological Review, 15, 257-263.

Bostrom, R. P., Anson, R., & Clawson, V. (1993). Group facilitation and group support systems. In R. P. Bostrom, R. T. Watson & S. T. Kinney, Computer Augmented Teamwork: A guided tour. New York: Van Nostrand Reinhold.

Burke, K. & Chidambaram, L. (1995). Developmental differences between distributed and FTP groups in electronically supported meeting environments: An exploratory investigation. Group Decision & Negotiation, 4, 213-233.

Cass, K., Heintz, T. J., & Kaiser, K. (1992). An investigation of satisfaction when using a voice-synchronous GDSS in dispersed meetings. Information & Management, 173-182.

Clawson, V. K. & Bostrom, R. P. (1993). The facilitation role in group support system environments. ACM SIGCPR Conference, St. Louis, MO.

Daft, R. L. & Lengel, R. H. (1986). Organizational information requirements: media richness and structure design. Management Science, 32(5).

Daft, R. L., Lengel, R. H., & Trevino, L. (1987). Message equivocality, media selection and manager performance: Implications for information systems. MIS Quarterly, 11(3), 355-366.

Daniels, R. M., Dennis, A. R., Hayes, G., Nunamaker, J., & Valacich, J. (1991). Enterprise analyzer: Electronic support for group requirements elicitation. 25th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

DeSanctis, G. (1993). Shifting foundations of group support system research. In L. Jessup & J. Valacich, Group support systems: New perspectives. New York: MacMillan Publishing Company.

Di Pietro, C. (1992). Groupware meetings that work. GroupWare '92 Conference, San Jose, CA.

Dubs, S. & Haynes, S. C. (1992). Distributed facilitation: A concept whose time has come? Proceedings of Computer Supported Cooperative Work Conference '92, Toronto, Canada, ACM.

Fellers, J. W., Clifton, A., & Handley, H. (1995). Using the Internet to provide support for distributed interactions. 29th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Frey, L. R. E. (1994). Group communication in context: Studies of natural groups. Hillsdale, NJ: Lawrence Erlbaum Associates.

Fulk, J. & DeSanctis, G. (1995). Electronic communication and changing organizational forms. Organization Science 6(4), 337-349.

Fuller, M. A. & Trower, J. (1994). Facilitation, systems, and users: The complete socio-technical system. 28th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Gallupe, R. B. & McKeen, J. D. (1990). Enhancing computer-mediated communication: An experimental investigation into the use of a GDSS for face-to-face vs. remote meetings. Information & Management, 18(1), 1-13.

Haythornthwaite, C., Wellman, B., & Mantei, M. (1995). Work relationships and media use: A social network analysis. Group Decision & Negotiation, 4, 193-211.

Hinds, P. & Kiesler, S. (1995). Communication across boundaries: Work, structure, and use of communication technologies in a large organization. Organization Science, 6(4), 373-393.

Hirokawa, R. Y. & Gouran, D. S. (1989). Facilitation of group communication. Management Communication Quarterly, 3(1), 71-92.

Johansen, R. (1989). Groupware: Future directions and wild cards. Conference on Organizational Computing, Coordination, and Collaboration, Austin, TX.

Kanter, R. M. (1995). World class. New York: Simon and Schuster.

Kayser, T. A. (1990). Mining group gold. El Segundo, CA: Serif Publishing.

Knoll, K. & Jarvenpaa, S. (1995). Learning to work in distributed global teams. 29th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Lea, M., O'Shea, T., et al. (1995). Constructing the networked organization: Content and context in the development of electronic communications. Organization Science, 6(4), 462-478.

Limayem, M., Lee-Partridge, J.E., Dickson, G.W., & DeSanctis, G. (1993). Enhancing GDSS effectiveness: Automated vs. human facilitation. Proceedings of the 24th Hawaii International Conference on Systems Sciences, January 1993, 95-101.

Markus, M. L. (1994a). Electronic mail as the medium of managerial choice. Organization Science, 5(4), 502-527.

Markus, M. L. (1994b). Finding a happy medium: Explaining the negative effects of electronic communication on social life at work. ACM Transactions on Information Systems, 12(2), 119-149.

McGoff, C., Hunt, A., Nunamaker, J., & Vogel, D. (1990). The role of the facilitator in the IBM decision support center process. Unpublished manuscript.

Miles, M. B. & Huberman, A. M. (1984). Qualitative Data Analysis. Newbury Park, CA: Sage.

Nunamaker, J., Vogel, D., Heminger, H., & Martz, B. (1989). Experiences at IBM with group support systems: A field study. Decision Support Systems, 5(2), 183-196.

Ocker, R., Hiltz, S. R., Turoff, M., & Fjermestad, M. (1995). Computer support for distributed asynchronous software design teams: Experimental results on creativity and quality. 29th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Orlikowski, W. J., Yates, J., Okamura, K., & Fujimoto, M. (1995). Shaping electronic communication: The metastructuring of technology in the context of use. Organization Science, 6(4), 423-444.

Panko, R. R. & Kinney, S. T. (1995). Working relationships: Assessing the need for media to support close personal relationships at work. 29th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Poole, M. S., Holmes, M., & DeSanctis, G. (1988). Conflict management and group decision support systems. Proceedings of the Second Conference on Computer Supported Cooperative Work, Portland, OR: ACM.

Post, B. Q. (1991). Building the business case for group support technology. 25th Hawaii International Conference on Systems Sciences, Hawaii, IEEE.

Rao, V. S. (1995). Effects of teleconferencing technologies: An exploration of comprehension, feedback, satisfaction, & role-related differences. Group Decision & Negotiation, 4, 251-272.

Sambamurthy, V. (1989). Supporting group performance during stakeholder analysis: The effects of alternative computer-based designs. Unpublished doctoral dissertation, Minneapolis, MN: University of Minnesota.

Short, J., Williams, E., & Christie (1976). The social psychology of telecommunications. London: John Wiley & Sons.

Smith, K. K. & Berg, D. N. (1988). Paradoxes of group life. San Francisco: Jossey-Bass.

Turoff, M., Hiltz, S. R., Baghat, A., & Rana, A. (1993). Distributed group support systems. MIS Quarterly, 399-417.

Zack, M. H. & McKenney, J. L. (1995). Social context and interaction in ongoing computer-supported management groups. Organization Science, 6(4), 394-422.

Catherine M. Beise

Kennesaw State University, 1000 Chastain Rd, Kennesaw, GA 30144, Cbeise@ksumail.kennesaw.edu

Fred Niederman

University of Baltimore, 1420 North Charles, Baltimore, MD 21201, Fniederman@ubmail.ubalt.edu

Peggy M. Beranek

Bentley College, Waltham, MA 02154, pberanek@bentley.edu

Copyright International Association of Facilitators Winter 1999
Provided by ProQuest Information and Learning Company. All rights Reserved