Distributed Group Support Systems:
Theory Development and Experimentation

Starr Roxanne Hiltz, Donna Dufner, Jerry Fjermestad, Youngjin Kim,
Rosalie Ocker, Ajaz Rana, and Murray Turoff
New Jersey Institute of Technology
Book chapter for: Olsen, B.M., Smith, J.B. and Malone, T., eds.,
Coordination Theory and Collaboration Technology, Hillsdale NJ: Lawrence
Erlbaum Associates, 1996.
Copyright, 1996


Distributed Group Support Systems use asynchronous computer mediated communication to support anytime/anywhere group discussions and decision making. The results of five controlled experiments are described, which explored the effects of different task types, tools, and processes on the process and outcomes of group decision making in this environment. Though GDSS type tools generally appear to improve both objective and subjective outcomes, various process interventions have had little or no effect on these groups, which had one to four weeks to adapt the use of the system features to their own expectations and preferences.

One type of computer-based system to support collaborative work ("groupware;" Johnson-Lenz, 1982; Ellis et. al 1991) is most often called a Group Support System, or GSS. Other terms that have been used include "Group Decision Support Systems" ("GDSS;" DeSanctis & Gallupe, 1987) and "Electronic Meeting Systems" (Nunamaker et al, 1991).
DeSanctis and Gallupe's seminal paper, "A Foundation for the Study of Group Decision Support Systems" (1987) has been extremely influential in providing a common framework for research. They defined GDSS as combining "communication, computer, and decision technologies to support problem formulation and solution in group meetings" (p. 589). Various types of tools or structures for interaction can help a group to avoid process losses and to achieve better decisions or outcomes. For example, the use of the computer as a channel for communication can allow everyone to input simultaneously, thus encouraging greater equality of participation. Failure to quantify preference structures can be overcome by providing appropriate voting scales and tools, while failure to efficiently organize and communicate information about ideas and preferences can be overcome by the statistical analysis and display of the results of rating or voting. They also presented a "contingency" theory to help explain why GDSS is not always beneficial; it depends upon whether the nature of the technology and structuring provided is appropriate for the group size (smaller vs. larger), the type of task, and the communication mode, of which they identified two: same place (FtF, or "decision room" ) and different place, or dispersed.
The term "Group Support System" (GSS) has come to be used as more general and inclusive than "GDSS," which is often used to imply decision room, same time settings. GSS can apply to many stages and types of group work, not just "decision making," and to computer support for groups that are working asynchronously through wide area networks as well as at the same time. "Distributed Group Support Systems" embed GDSS type tools and procedures within a Computer-Mediated Communication (CMC) system to support collaborative work among dispersed groups of people. "Distributed" has several dimensions: temporal, spatial and technological. The majority of GDSS research has been conducted in "Decision Rooms," where the participants are meeting at the same time and same place. CMC based systems can be used synchronously (same or different places, but at the same time), or "asynchronously." The central focus of the program of research reported here is asynchronous groups, in which interaction is distributed in time as well as in space. The group members use the system to work together to reach a decision or complete their cooperative work over a period of time, with each person working at whatever time and place is convenient. In addition, the system used is itself "distributed;" that is, there can be more than one "server" in different places which are linked, and the user interaction may occur on a "user agent" located on the individual PC.
Asynchronous use of computer-based group support tools and processes is a unique mode of communication, different not only from Face to Face (FtF) communication, but even than synchronous use of CMC or other forms of computer support (Rice, 1984, 1993). It leads to different communication behavior (such as the tendency towards much longer entries by participants, and the discussion of many topics at once) and to unique coordination problems and opportunities (Hiltz & Turoff, 1985; Malone and Crowston, 1990; Turoff et al., 1993).
Among the key variables which were observed to influence the effectiveness of small group decision making in the FtF condition are "leadership" and "process." For example, imposing certain structures for interaction on small FtF groups, such as a strict agenda which forces "rational" decision making, or brainstorming (Osborn 1957) or Nominal Group Techniques (Van de Ven & Delbecq, 1971), can improve process and outcomes. Procedures (or "structures") for interaction which decrease process losses in the FtF condition may not be the same as those which are helpful in a computer-mediated communication condition, however, particularly in the fully distributed and asynchronous mode.
Many GDSS systems that are "decision room" based simply include a set of tools and procedures in the "package" that is always provided as part of the system; thus, the effects of medium of communication are confounded with the effects of specific tools and procedures. Our objective has been to isolate specific tools and procedures and explore their effectiveness in the asynchronous environment. Very few other GDSS experiments have looked at the asyncnronous condition. Of the 120 GSS experiments published through the end of 1995 which we have been able to identify (Fjermestad and Hiltz, 1996), besides the three NJIT experiments published to date, only five other experiments have used an asynchronous condition, plus two that compared synchronous and asynchronous conditions, and only one other investigator (Chidambaram, 1989, 1990) has conducted more than one experiment in the asynchronous mode.
New Jersey Institute of Technology's project is an integrated program of theory building, software tool development and assessment, and empirical studies (both controlled experiments, and as opportunities arise, field studies). The project investigates the effectiveness of different types of tools and procedures for various types of tasks and groups, within the distributed environment. Specific studies also contrast the distributed mode of communication with other modes. This chapter will review some of the accomplishments of the first five years of the program of research, which was partially supported by the National Science Foundation. It will first summarize the theoretical framework which was constructed to guide and integrate all of the separate research studies. Then it will briefly describe the software tools that were developed. Next, the design and results of the first five controlled laboratory experiments will be described, each of which was conducted as Ph.D. dissertation research. Several field studies, which have been published elsewhere will be briefly alluded to. Finally, it will summarize what we see as the main findings across the various studies completed thus far, and their implications for theory building and future research directions. In brief, to give a preview, the most important finding is that imposing restricted structures or procedures for interaction in the distributed GSS mode does not have the beneficial effects that have been observed in the FtF or decision room modes of communication.

1. Theoretical Foundations and Integration

A specific GSS is a particular combination of communication mode, tools, and structuring of process (via a facilitator or procedural instructions/agenda). The effects of a GSS on the process and outcomes of collaborative work depend upon a number of contingencies. We started with DeSanctis and Gallupe's (1987) framework, which identified three types of contingencies: communication condition (face-to-face or dispersed-- we extend this to include synchronous vs. asynchronous); group size and task type. For task type, we currently use McGrath's (1984) "task circumplex." The graphical representation of this typology (not included here) differentiates tasks on two dimensions. The first dimension classifies tasks on the basis of outcome: intellectual (e.g., a decision) or behavioral (e.g., a "product" or action). The second dimension uses the type of behavior of group members (convergent or cooperative, vs. conflicting). We are presently focusing on the four types of "cooperative" tasks: Generating actions (Type 1: Planning and and Type 2: Creative tasks) and Choosing solutions to a specific problem (Type 3: Intellective tasks, which have a "correct" or optimum decision whose quality can be measured; and Type 4: Preference, for which the objective is to reach agreement).
We extended this initial framework to produce a more comprehensive theoretical foundation that will enable us to compare the results of different studies and to compare our results to those of other researchers. One completed paper (Fjermestad, Hiltz, & Turoff, 1993) reviews the major research models that have been used for studying GSS's and derives and presents an integrated, comprehensive model of the factors that are utilized for their investigation. This paper shows that in the short time since the publication of the DeSanctis and Gallupe (1987) foundation paper, the number of research dimensions included in various models has more than doubled (from three to seven). There has also been a shift in research emphasis from the technology to the interaction among the technology, the task, and the group to produce outcomes. The model organizes all of the variables that have been used in GSS research into four dimensions: contextual, intervening, adaptation, and outcomes. A concise overview of the theoretical framework, showing the version we began with, is shown in Chart 1. There, the contextual factors are shown at the top, then the intervening variables, and the outcomes or dependent variables are at the bottom. The model served as the theoretical framework for all individual studies carried out within the program of research.
The Contextual factors are all external or driving variables that comprise the environment or conditions for the decision making task. For any one experiment, they are (relatively) fixed or controlled. These include characteristics of the group, task, environmental and organizational context, and of the particular technology (GSS) being used.
Intervening factors are related to the emergent structuring of the group interaction, both derived from and adding to the set of conditions created by the context of the group decision sessions. For example, the methods used by the group may vary as to session length, number of sessions, and presence and role of a facilitator. These factors can change from session to session, and thus are dynamic rather than static.
Adaptation, or "modes of appropriation" is an important intervening variable in our model. According to Adaptive Structuration Theory (DeSanctis & Poole, 1991, 1994; Poole & DeSanctis, 1990, 1992; Sambamurthy and DeSanctis, 1989), group outcomes are not determined by the effects of single elements (such as technology and task characteristics), but by a complex and continuous process in which those elements are appropriated by the group. The four dimensions of the construct (level of use, attitudes toward the GSS, level of consensus, and level of control) are measured in all studies via questionnaire items designed and validated by Scott Poole. For each of these aspects of group appropriation, there can be "effective" or ineffective modes. For example, the group may use the GSS facilities little or not at all, even though they are instructed to, or they may use it in a very different manner than was intended.
Finally, the Outcomes, or dependent variables, are the result of the interplay of the intervening, adaptation, and contextual factors. They include efficiency measures (e.g., calendar time to decision), effectiveness measures (e.g., number of different ideas generated or decision quality), and subjective satisfaction measures.
It should be noted that work is completed on further developing and applying the integrated framework to a comprehensive comparison of over 100 published experiments on GSS's to date (Fjermestad and Hiltz, 1996). Different outcomes have been observed depending upon the initial set of independent variables and the group processes (influenced by intervening variables) that result in a specific adaptation or "adaptive structuration" (Poole and DeSanctis, 1990) of the technology provided. By focusing on differences in the variables controlled and studied, a foundation is provided for understanding differences in findings.

2. Facilities Development

NJIT's EIES 2 is a CMC enhanced with GDSS tools, that provides the foundation that allows continued evolution and the incorporation of additional functionality (Turoff, 1991). As a result of this capability, we were able to enhance the system and to create a "developer's kit" to allow Ph.D. students, or others, to use a version of Smalltalk to develop their own features or interface characteristics. EIES 2 is based upon an object-oriented data base and a compiler for the X.409 communication data base specification language. This base allows the evolution of new object types as they are needed.
To support group-oriented objectives a CMC system must allow other computer resources to be integrated within the CMC environment. The approach we have chosen uses the metaphor of an "activity" that can be attached to any communication item. "Doing" an activity executes a program or procedure on the host computer or the network of users' computers.
Work has been completed on many kinds of "activities." One set, "List" and "Vote," replicate the functionality provided by Minnesota's SAMM (Software Aided Meeting Management) for a group to create and revise a common list of alternatives, and then apply several types of voting procedures to this list: vote for one, vote yes or no on each alternative, and rate or weight each alternative. Another activity, "Poll," allows the construction, response to, and display of results from a poll or survey, within the CMC environment. A third, "Question/Response" Activity, supports processes such as Nominal Group Technique, dialectical inquiry and brainstorming, as well as applications to collaborative learning. Each participant must independently (and possibly anonymously) respond to a problem or question, before seeing the responses of others. Many other activities have been developed to support other kinds of group tasks, including a class "gradebook" activity for the Virtual ClassroomTM.
It should be noted that the series of studies reported here were conducted with a VT100 based full screen menu-type interface, rather than a GUI (point and click Graphical User Interface.) The VT100 type interface has the advantage of being usable on any PC or even a "dumb terminal;" whatever the subjects may have had available to use at home or at work would suffice. We are currently completing work on a Web-based GUI that can be used with browsers such as Netscape, for those who have the necessary equipment and prefer this style of interaction.

3. The First Five Experiments

Each of these experiments represents an attempt to find appropriate tools and processes to support four different types of task in the McGrath "task circumplex;" they examined:
. Voting tools and sequential procedures for a preference task;
. Conflict vs. Consensus structures plus experience for a planning task;
. The effects of FtF vs. distributed asynchronous CMC as it interacts with a
structured design procedure, for a creative task, software design
. Question-Response tool and the Polling tool for an intellective task (peer review)
. Designated leadership and sequential vs. parallel procedures for a mixed task, choosing a stock portfolio.
A concise summary of the methods and findings of the experiments is presented in Chart 2(methods) and Chart 3(findings). Each group had one or two weeks to complete each decision (depending on the experiment). This is a relatively long time period, compared to the 10 or 15 minutes that some experimental tasks used in decision room GDSS experiments have taken. Unless otherwise noted, all used as subjects undergraduate and graduate students from the Computer Science and Management degree programs at NJIT and Rutgers. Students participated as a course assignment, and were graded; alternate assignments were offered for those who chose not to participate. It should be noted that in asynchronous groups interacting over a week or more, group size cannot be truly controlled. Despite the grade incentive, some students "dropped out" of the group interaction, perhaps because of illness or computer problems, and thus decreased group size below the starting number. When "group size" is reported, it refers to the "ending" group size, not the number who were trained and began a task. In all experiments, if this "effective group size" fell below two, the group was dropped from the analysis.
Most of the studies also used expert judges to rate some aspect of the quality of the group outcomes. In all cases, at least three judges were used. They consisted of faculty members or advanced ("ABD") graduate students with expertise in the area. Coding and rating procedures were developed and refined during pilot studies, and judges were trained with pilot study transcripts before being given the experimental data to rate or code.
The brief overviews that follow are essentially extended abstracts of parts of dissertations that total 300 to 550 pages; obviously, in just a few pages, many details such as a complete list of hypotheses with justifications, and specifics of measurement of variables, must be omitted.

3.1 The Effects of Voting Tools and Sequential Procedures for a Preference Task

This experiment, carried out by Dufner (Dufner, Hiltz and Turoff 1994; Dufner, 1995; Dufner, Hiltz, Johnson and Czech, 1995), is a replication (modified for implementation in asynchronous mode) and extension of the doctoral research of conducted by Watson (1987) at the University of Minnesota. The study, which was preceded by a full year of pilot studies, extends the Watson research to include an investigation of adaptive structuration, media richness, system and task expectations, and training (Dufner 1989, 1995; Hiltz et al. 1991).
The Foundation Task, developed at Minnesota and used for this experiment, can be classified as a preference allocation task based on the McGrath Circumplex model (McGrath 1984). The subjects play the role of a foundation board, and are to reach consensus on how to allocate funds among "applicants" representing very different kinds of objectives, such as cutting local taxes, helping the homeless, or improving the town library.

3.1.1. Procedures and Experimental Design

Individuals were assigned randomly to groups, as much as was possible given time constraints and schedules. These groups were then trained for approximately three hours in the use of the medium (EIES 2) and in working together to perform a group decision making task. All groups were given a suggested agenda ("define the problem," etc.), as used by Watson, and their conferences were seeded with root comments corresponding to each stage in this suggested set of activities. TOOLS groups were also given training in the use of the "List Activity" (an electronic flip chart) for group generation and management of lists, and in the use of the "Vote Activity," which provided three forms of voting on the items on the list. Groups assigned to a SEQUENCED condition were instructed as follows: "You must all work on the same agenda item together. The group decides when to move to a new agenda item. You do not have to follow the agenda order. However, you must all work on the same agenda item together. You are asked not to work ahead of or following the group." The "not sequenced" groups were not clearly instructed that they were free to work in parallel; they simply were not given these instructions.
After training, each group was given five business days to perform the experimental task. Groups were instructed to communicate only through the medium. No formal facilitation was provided to the groups, although technical assistance was given when anyone asked for help. There were a total of 31 groups with 119 subjects; group size varied from 3 to 8 subjects.

3.1.2. Hypotheses and Selected Results

From pilot studies (Dufner 1989, Hiltz et al 1991) we knew that groups in asynchronous mode encounter coordination problems (Dufner, Hiltz and Turoff 1994) that cause frustration with the medium. Therefore, we hypothesized that the TOOLS and SEQUENTIAL PROCEDURES would make significant contributions to subjectively reported perceptions of medium richness (Zmud, Lind and Young 1990) and satisfaction with the process in the asynchronous environment.
No significant difference in the SEQUENCED versus NOT SEQUENCED groups was found. Re-examining the manipulation, we decided that we could not determine whether this was because imposing a sequenced process truly makes no difference, or whether the manipulation was not strong enough. Therefore, we noted that examining the "sequenced" vs. "parallel" process should be tried again in a subsequent experiment.
The TOOLS Groups perceived more media richness, reporting that the medium was significantly more dependable, convenient, flexible, and wide- ranging than did the groups not supported with tools. The TOOLS groups also perceived the system as more personal, more rich, and as providing more feedback and more immediate feedback than did the groups not supported with tools . TOOLS groups were also found to be significantly more likely to recommend the system for future meetings; to have enjoyed their participation in the discussion; and to have a higher opinion of the overall quality of the discussion than did the groups not supported with tools.
These experimental results seem to indicate that user perceptions of media richness and of the quality of group processes can be improved by providing voting tools that support group discussions, at least for preference type tasks, where the primary goal is to reach consensus. This contrasts with the findings by Watson (1987) for the same task and the same type of tools in a synchronous (Decision Room) environment, where tools created few significant differences.
Despite the significant and consistent positive effects of providing the listing and voting tools on subjective perceptions, there were no significant results on other dependent variables measured in this experiment, including changes in level of consensus and the group's equality of influence. Because this experiment did not have decisions that could be rated on quality, the effectiveness of tools such as "List" and "Vote" should be examined for intellective tasks or other tasks for which quality measures can be obtained, in the future.

3.2. Effects of Decision Approach and Experience on Planning Tasks

The basic objective of this longitudinal experiment (Fjermestad, 1994; Fjermestad et.al, 1995) was to examine the performance and attitude changes of groups involved with strategic decision making in a computer-mediated- communications (CMC) environment. The two independent variables of interest were decision approach and experience. Decision approach consisted of dialectical inquiry (DI; Schwenk, 1990), which is a structured approach to induce conflict, and constructive consensus, which is a set of instructions telling the group to reach agreement. Experience consisted of working with a group on two related but different tasks, each taking two weeks to complete.
Previous research in the field of organizational strategic decision making has demonstrated that structured conflict can improve the quality of decisions (Mason,1969; Mitroff et al, 1979; Mitroff and Mason, 1981; Schweiger et al 1986, 1989; Schwenk, 1990; Tjosvold, 1982) and negatively affect both group perceptions and process outcomes (Schweiger et al 1986, 1989; Turoff, 1991). The two basic structured conflict methods are DI and Devils's Advocate (DA). Schwenk's meta-analysis (1990) indicates that for studies that focus on groups, DI has a slight advantage over the DA.
The tasks were unstructured decision making tasks, with no right or wrong answers; they are Type 4 (Planning) based on McGrath's (1984) task circumplex, and fit Schweiger et al's (1986) requirements for strategic decision making. The three specific tasks used in this study were developed by Chidambaram (1989) and were were modified and updated for use in an asynchronous communications mode instead of a set of discrete FtF meetings. The Threat of Takeover task was used as a training task for all groups and the Issue of Image and Product Line Expansion were the experimental tasks.

3.2.1. Procedures and Experimental Design

The research design is a 2 X 2 factorial repeated measures design. The factors are decision approach and experience. Groups in each decision approach were given two weeks (10 business days) to complete each of two tasks.
The 160 subjects used in the study were undergraduate and graduate students in computer science and management information systems at NJIT. They all had some fluency with the use of E-mail and computers and were given course credit and a grade for participation. All subjects were assigned to groups based upon availability and scheduling constraints. The ideal group size was six subjects per group, but due to the subjects' scheduling constraints, the actual group sizes ranged from four to seven (Fjermestad, 1994). Experimental conditions and task orders were randomly assigned to the groups.
The Dialectical Inquiry Approach (DI) is based upon the procedures developed by Schweiger et al (1986, 1989) and Tung and Heminger (1993), modified to support asynchronous communication and decision making (Fjermestad, 1994). The DI groups were divided into two subgroups, denoted as the Plan and Counter-plan subgroups. These groups were in separate conferences on EIES 2. All members of both groups were to initially develop an individual recommendation (including supporting facts and assumptions) within two business days and enter it in a List Activity in the CMC system.
The Plan group then had two days to develop a single recommendation. Members read the individual case recommendations and then debated and discussed them in a Question Activity which requires each participant to reply before viewing other's replies. When complete, a case leader organized and entered the subgroup's recommendation. This was then submitted to the Counter-plan sub-group, which had two days to negate the assumptions and develop a counter-plan.
The moderator then created a new conference for the full group and added the plan and counter-plan. The Full group's objective was to critically evaluate the plan and counter-plan through debate and discussion, and to develop a single final group recommendation. A Voting Activity was available if the group chose to use it, in all conditions. The time limit for this task was four business days.
The Constructive Consensus Approach (CC) follows the basic method developed by several researchers (Hall, 1971; Hiltz et al, 1991; Nemiroff et al, 1976; Schweiger et al, 1986). The CC groups functioned as one group and were in a single conference for the entire task. Their objective was to reach consensus on a single final recommendation. Each individual group member had two days to develop an individual recommendation. The group then had eight business days to examine the case situations systematically and logically, in order to develop a final recommendation through debate and discussion. A Voting Activity was available if the group chose to use it.

3.2.2. Selected Hypotheses and Results

Based on previous research in FtF conditions cited above, it was hypothesized that DI groups would be superior to consensus-structured groups in terms of effectiveness (decision quality), but would be less efficient and express less subjective satisfaction than the consensus structured groups. It was also expected that group performance would improve on the second replication of the use of the assigned decision sturcture, and that furthermore, there would be an interaction. Since DI is an unfamiliar structure, it was expected that the improvement would be more marked for the DI groups.
Contrary to these expectations, there were very few differences between "Task 1" and "Task 2", and no differences between DI and CC groups in terms of group performance. DI groups required significantly more asynchronous meeting time and communication to complete their recommendations . Depth of evaluation as rated by judges showed no difference; but perceived depth of evaluation was lower in DI than in CC groups. CC groups reported greater decision acceptance and willingness to work together again than DI groups. Relatively few experiential effects were observed. Thus, no advantages were observed for the DI approach as compared to a consensus approach that also carefully structured the interaction, but it took more work and produced less participant satisfaction.
This study of asynchronous strategic decision making and a study using decision room GSS by Tung and Heminger (1993) report that there are no differences in effectiveness between constructive consensus and dialectical inquiry groups in a GSS environment. Perhaps what is happening is that the GSS technology is significantly improving the consensus groups to the point where the outcomes are as high as the structured conflict processes in a FtF environment. Thus, GSS equalizes consensus groups' performance to that of the Dialectical Inquiry groups without affecting decision and process satisfaction and without any of the process losses.

3.3. Effects of Mode and Structure for a Creative Task: Distributed Software Design Teams

The experiment conducted by Ocker (1995; see also Ocker et. al., 1995) investigates the effects of distributed asynchronous communication on small groups performing high-level requirements analysis and design work. It is the first experiment to examine the software design process in a fully distributed environment. The experimental task is the Automated Post Office (APO); groups are required to develop and reach consensus on the initial requirements and interface design of an APO and to submit these in the form of a written report. The APO task, as used in this experiment, is a modification of the task used by the University of Michigan (Olson et al., 1991, 1992, 1993). It is primarily a creativity type task, but also contains elements of planning and decision making (McGrath, 1984). The APO task can also be categorized as occurring during the early stages of the innovation process (West, 1990).

3.3.1. Variables and Major Hypotheses

Two variables were manipulated in this experiment. The first variable, an imposed process, pertains directly to the degree of coordination required for the effective performance and satisfaction of groups working on a creative task. Groups in the imposed process conditions followed a sequence of steps adopted from research on argumentation and structured communication (IBIS; Kunz and Rittel, 1970; "Design Space Analysis," MacLean et al. 1991). The imposed process contained three main phases: (1) generation of design alternatives (2) period of critical reflection and individual evaluation of alternatives and (3) group evaluation of alternatives and consensus reaching. The second variable is mode of communication (asynchronous computer-mediated communication (CMC) vs. face-to-face). It was expected that asynchronous CMC groups would out-perform FtF groups, because of fewer process losses, and the ability of each of the participants to think and work at their own paces. Dependent variables include performance outcome (quality and creativity), and group satisfaction.
H1: Asynchronous CMC groups will produce solutions of higher quality than Face to Face (FtF) groups.
The problem solving structure was chosen due to its capability to structure communication and for its fit with the activity of design. It was felt that FtF groups would not need this added coordination, because high-level design has its own inherent structure (Olson et. al. 1992), but that it would ease the cognitive burden of distributed asynchronous groups. Therefore, an interaction was hypothesized:
H2: CMC structured groups and FtF groups will produce solutions of higher quality than CMC unstructured groups and FtF structured groups.
Based on an analysis of the task requirements (e.g. Guindon, 1990; Simon, 1973; King & Anderson, 1990; West, 1990), it was hypothesized that overall, the solutions produced by the asynchronous groups would be more creative than those produced by the face-to-face groups.
H3. CMC groups will produce more creative solutions than FtF groups
Again, based on relative coordination requirements, an interaction effect was hypothesized such that the solutions produced by the asynchronous imposed-process groups and face-to-face no-imposed-process groups would be more creative than those produced by asynchronous no-imposed-process groups and face-to-face imposed-process groups. Finally, it was hypothesized that the face-to-face groups would be more satisfied than the asynchronous groups because FtF is a "richer," more personal medium.

3.3.2. Procedures

Subjects were undergraduate students enrolled in an undergraduate upper level systems design course, or graduate students in CIS, MIS, or the MBA program. The marjority had coursework and/or job experience directly relevant to systems design.
Groups were required to reach consensus on the initial requirements of the APO and to submit these requirements in a formal report at the end of the experiment. The asynchronous design groups communicated using the EIES 2 computer conferencing system; each of these groups communicated in its own computer conference. The experiment lasted two weeks.
The asynchronous groups met together for one three hour training session, while the face-to-face groups met twice for a total of six hours, with the first and second sessions spaced exactly two weeks apart. Both the FtF groups and the asynchronous groups in the imposed process condition were trained on the process using the same script. All asynchronous groups were trained on the basic use of the EIES 2 system.
The FtF groups had a PC and word processor available for creating their final reports. (Technical difficulties led two groups to hand write their report; these were among the longer reports, so it does not seem to have negatively affected them. These were later transcribed.)
The computer conferences and FtF meetings were minimally facilitated. The facilitator played the role of a technical assistant, helping groups with equipment problems and answering questions of a technical nature.
All participants completed questionnaires, which was the source of subjective satisfaction data. All groups' final reports were printed using the same word processing package, to mask indications of mode of communication. Quality and creativity of solution were rated by an expert panel of judges, using procedures and scoring adapted from Olson & Olson.

3.3.3. Major findings

The overall quality of solution was rated by a panel of expert judges to be equally good between the asynchronous groups and the face-to-face groups. (Although the asynchronous groups were rated as consistently higher, the difference was significant only at the .07 level). Contrary to hypotheses, there was no significant interaction effect between mode of communication and the presence/absence of an imposed process in relation to quality of solution.
As for creativity, the solutions produced by the asynchronous groups were judged to be significantly more creative than those produced by the face- to-face groups. Again, there was no interaction effect between mode of communication and the presence or absence of an imposed process.
Contrary to hypotheses, there were no significant differences between CMC and FtF groups on key measures of subjective satisfaction: perceived depth of analysis, solution satisfaction, and decision scheme satisfaction.


The imposed process was hypothesized to benefit asynchronous groups by providing the added coordination which is missing in this form of communication. There are several possible explanations for why this did not occur. Concerning the design of the APO, a strong metaphor is available in the form of automatic teller machines. The problem may have been familiar enough to groups, so that the need for coordination might have been greatly reduced; upon entering the group, group members already knew how to approach the solution to this problem.
The major finding of this experiment is that groups which communicated asynchronously, whether they followed a structured problem-solving approach designed to enhance coordination or were left to their own devices to reach a decision, produced significantly more creative results than the face-to-face groups. A tentative conclusion is that asynchronous communication, in and of itself, leads to higher levels of creativity. Possible explanations for this include a greater amount of communication over an extended period of time, reduced production blocking, and the production of a collective memory.

3.4. Effects of Question and Polling Activities on an Intellective Task: Supporting the Peer Review Process

The task for this experiment consisted of review and decision on publishability of a manuscript submitted to a refereed journal or conference. Contrary to the traditional review process, where two or more experts review and rate the quality of a manuscript individually, the distributed group support system based review process as adopted in this experiment called upon reviewers to undertake the task as part of a group, or panel. This new mode for conducting a review involved performance processes that are typical of intellective, decision making, and cognitive conflict tasks (McGrath, 1984). We classify it as primarily intellective, since two criteria for rating the quality of the solution were available: the ratings of the article by the actual reviewers of the paper, and the ratings by a panel of expert judges. The desirability of a system that can support group review activities in a different-time different-place mode was evident. This study being the first to investigate the viability of a DGSS based review process, motivated a research design that would allow the study of independent and interactive effects of support tools from within a DGSS.
Two support tools, Poll and Question activities on EIES2, were made available for this experiment, and utilized in a 2 x 2 factorial design. Poll activity, especially developed for this research, allows reviewers to rate the quality of a manuscript on various scales and enables them to view summary statistics on group ratings. Several scales can be grouped into one Poll. Question activity establishes a structured form of group discussion by requesting the provision of justifications for ratings on individual scales and maintaining an independent chain of discussion on each scale. Group members' responses to Question activity are textual items with no limit on size, whereas Poll activity responses consist of numbers representing scale anchors. One important feature, common to both Poll and Question activities, is that group members cannot view others' responses before having provided their own initial responses.

3.4.1. Procedures

An EIES2 conference was established for each group. The final data set for this experiment consisted of 33 groups, with 30 groups of size three and 3 groups which began at size 3 but ended at size 2. The majority of subjects were graduate students (73%); all subjects were enrolled in courses which required them to read and critique journal articles, and to use EIES 2 as a regular part of coursework. The mean age of subjects was 30 years with an average full time work experience of slightly over five years. Since some subjects were enrolled in distance sections of courses, the manuscript and training materials instructing them how to use the tools and procedures for their condition were mailed to all subjects, rather than being explained in a face to face training session.
Subject groups reviewed a manuscript actually submitted to a refereed source with a pending editorial decision. The selected manuscript met the criteria developed as a result of three rounds of pilot studies and was judged to be commensurate with ability levels of the potential subject population.
Groups in all four conditions were to (1) individually evaluate the manuscript, (2) provide ratings and justifications for the ratings on six scales; (3) share their responses with the group, and finally (4) discuss and reach agreement on ratings. This four step process was to be completed over a period of two weeks with steps 1 through 3 completed by the end of the first week. The identities of group members were concealed through the use of "pen names".
Three measures for quality of group outcome were adopted: (1) quality of the decision (disposition recommendation); (2) quality of the review; and (3) comprehensiveness of the review. A panel of expert judges independently rated the paper on the same scales as the subjects, and their ratings were compared to those produced by each group. Disposition recommendation categories consisted of (1) Accept as is (2) Accept with minor revisions (3) Major revisions; or (4) Reject. The four expert judges were evenly divided on (3) "major revisions" or (4) "rejection." Thus, either of the latter two recommendations were considered "correct" in assessing quality of the decision in terms of the correctness of the disposition recommendation.
The quality of the review was calculated on the basis of the deviation of the group's decision from the judges' decisions on the separate aspects of the manuscript (literature review, methodology, presentation style, etc.) The comprehensiveness measure consisted of counts of the number of lines of discussion associated with each of the separate scales: e.g. "substantive emphasis" was the amount of attention paid to critiquing the literature review; methodological emphasis, stylistic emphasis, interpretive emphasis, and wisdom were lines devoted to methodological critiques, etc. (Cummings et al, 1985). Adaptive structuration was measured by a series of questionnaire items.

3.4.2. Hypotheses

Since Question activity imposes a structure for group members to engage in the group proceedings and coordinate their activities, it was expected that Question activity groups would do better than No-Question groups on all three measures of quality (Sambamurthy and DeSanctis, 1989; Easton, Vogel, and Nunamaker, 1989). (Poll activity was expected to be primarily a consensus- enhancing tool, rather than a quality-enhancing tool; these results are not included here.) There were also expected to be some interactions between Poll and Question. In all cases, a moderating variable must be pre-discussion level of agreement; if most of the individual reviewers agreed on their initial ratings, one would not expect any of the tools used to make much difference. In addition, many hypotheses were developed with the basic premise that positive forms of Adaptive Structuration (high levels of comfort, consensus, and respect regarding the tools) would be strongly related to favorable outcomes.

3.4.3. Selected Findings and Conclusions

No differences in the quality of decision were detected due to Question or Poll activity. In fact, most groups reached a decision that the paper could not be accepted; thus, there was very little variance from a correct decision in ratings for disposition of the paper, and hence none of the independent and intervening variables were significantly associated with this measure.
With respect to the quality of the review, the results showed that mprovement depended upon the level of pre-discussion agreement. If groups started with a lower level of initial agreement, then the quality of review was enhanced by the tools. Specifically, at lower levels of initial agreement, groups with the Question activity produced significantly higher quality reviews than No-Question groups. Highly agreed upon poor quality ratings before the discussion phase left little or no opportunity for an improvement in the quality of review through discussion. Unexpectedly, the Poll activity showed a marginally significant (p = 0.0797) main effect on the quality of the review. This main effect was attributed to the fact that Poll activity groups had significantly lower levels of pre-discussion agreement than No-Poll groups. No significant effects on quality of the review were observed due to the modes of appropriation.
In terms of the effects on the measures of comprehensiveness, relatively few effects of the Question activity were supported. One of the significant finding was that groups that used the Question activity had significantly higher wisdom (concern for the paper's contribution and significance) in their reviews than No-Question groups. Mediating effects of the level of pre-discussion agreement similar to those for quality of review were observed on the amount of methodological emphasis. An unexpected, though not surprising, result was that in the absence of the Question activity, the Poll activity had a negative effect on interpretive emphasis.
Mediating effects of the modes of appropriation (Poole and DeSanctis, 1990; DeSanctis and Poole, 1994) on measures of comprehensiveness were rare. The level of challenge was observed to be one of the stronger mediating factor. Additionally, it was observed that a higher level of respect for the system did not always lead to an enhanced level of performance.
Despite the fact that the majority of the hypothesized effects were not observed, the experimental findings offer important implications for the review process. In summary, it was concluded that the strength of the DGSS based review process lies in its ability to allow for (i) disagreement among reviewers before the discussion phase, and (ii) the subsequent opportunity for resolution of the disagreements with the use of support tools. These mechanisms combined with anonymous contributions can be profitably used to avoid many commonly noted dissatisfactions with the traditional peer review process (Rana, Hiltz, & Turoff, 1995; Peters and Ceci, 1982; Mahoney, 1977, 1978; 1985; Cole, Rubin and Cole, 1977; Cole, Cole, and Simon, 1981).
Currently, a peer review system is being developed for the World Wibe Web. The system uses the object base of EIES2 and provides an easy to use and attractive http interface through popular browsers, such as NetScape. The plans are to conduct field trials of the viability and potential benefits of the collaborative review process.

3.5. Effects of Parallel vs. Sequential Procedures and of a Designated Leader for an Intellective/Mixed (Stock Selection) Task

Silver (1990) defined system restrictiveness as the degree to which and the manner in which a system limits its users' decision-making processes to a subset of all possible processes. The objective of this study is to examine how the use of coordination structures with different degrees of coordination flexibility, or system restrictiveness, affect group performance in a distributed asynchronous GSS (or DGSS for short) environment.
The investment club task, developed for this study, is classified as primarily an intellective task. A group was asked to select at least one, but no more than three stocks from a list of 15 stocks to maximize its portfolio value in six months. Six months after the experiment, all portfolio values were calculated and ranked to evaluate decision quality. The task also has aspects of a planning task, since the group had to decide what information to gather and how to evaluate this information, in order to reach a decision; and of a preference task, since the group had to reach agreement, and at the time of the decision, there was no way to actually know what decision would turn out to be best six months later.

3.5.1 Experimental Design and Procedures

The experiment (Kim, 1996) was conducted with a 2x2 factorial design. There were 212 subjects in 47 groups. The subjects were Rutgers, NJIT, and Fairleigh Dickinson university students enrolled in various degree programs. All subjects in all conditions were give the same basic agenda as a coordinating structure, consisting of root comments in their conferences which requested them to define objectives, decide on criteria, review the candidates (stocks, in the case of the experimental task), evaluate the candidates on the criteria, and reach agreement on selection. Training was given to all subjects in the form of a week long asynchronous conference, which included a practice task, the selection of a leader by following the agenda, ending with the sending of a private message to the experimenter by each member, giving a rank ordering.
Four coordination structures were created with two independent variables. The four coordination structures were different in that each structure restricted interaction in a different way. In parallel communication groups, all discussion items were presented, and discussed concurrently by all members of the group, throughout the experiment. Sequential communication groups discussed one item on the agenda at a time. Once a group moved to the next item, revisiting the previous items was not allowed.
In sequential groups, moving from one discussion item to the next item was made by a leader's decision (in groups with a leader), or by a timetable (in groups without a leader). In sequential groups without a leader, the discussion deadline for each discussion item was announced to the groups at the beginning of the experiment. In sequential groups with a leader, the leader made the summary of the discussion of the item for the group, and opened the next discussion item. In parallel groups with a leader, the only requirement for a leader was to summarize group discussion once in a while.
Groups without a designated leader heard no more about this topic after the training. For those in the Leader conditions, the experimenter used the individual rankings to arrive at the most preferred leader. This was announced in the group's conference, along with the leader's role. The leader was specifically empowered to assign a division of labor, and requested to track and summarize the group's progress. Leaders also were put in a "leadership conference" where they could ask questions about the role.

3.5.2. Major Hypotheses and Findings

It was expected that groups supported with a less restrictive structure would perform better than groups supported with a highly restrictive structure. Previous research on DGSS indicated that an imposed coordination structure can be overly restrictive due to the limited bandwidth of the interaction medium ( Hiltz, et al., 1990), and the need to synchronize individual activities. Previous research also demonstrated that a GSS with a high degree of system restrictiveness had negative impacts on group performance (Chidambaram and Jones, 1993; McLeod and Loker, 1992; Mennecke, et al., 1992). GSS's with a high degree of system restrictiveness leave no freedom for the group to adaptively structure the system to its own preferable decision strategy (DeSanctis and Poole, 1991; Poole and DeSanctis, 1990). DGSS, in which coordination of individual activities is one of the major requirements, should not be highly restrictive. Research indicates that individuals come to the group with a relatively inflexible preference for a particular decision making strategy (Putnam, 1982). Therefore, DGSS should be flexible enough to allow the individuals freedom to concentrate on aspects of the problem to which he or she can best contribute (Turoff, et al., 1993).
A previous experiment with synchronous CMC (Hiltz, Johnson & Turoff, 1991) found that a designated leader elected by the group could improve the quality of decision for an intellective task. Therefore, it was hypothesized that this would also be true in a distributed environment.
Many of the observed differences in dependent variables were not significantly related to experimental condition. However, objective decision quality, evaluated with actual portfolio values six months after the experiment, was significantly better for leader than for no-leader conditions. Parallel groups perceived that their decision quality was better than that of sequential groups. Parallel groups also had higher decision quality as objectively measured, but not significantly so (p= .14). The average length of comments in the Leader conditions was longer than in the No Leader conditions; there were no other differences in consensus and participation.
Satisfaction with a coordination process was higher in sequential groups, and higher in groups without a designated leader than groups with a leader. (As in many other studies, these subjective satisfaction results run counter to the objective quality results). Satisfaction with the group, however, was higher in parallel groups. Sequential groups reported more improved understanding of the task structure than parallel groups.

3.5.3. Discussion

It is interesting to notice that sequential groups showed higher satisfaction with the coordination process and more improved task understanding than parallel groups. This is contrary to what was expected, but consistent with some previous research, which indicates that GSS should be designed with some degree of restrictiveness (Dickson, Partridge & Robinson, 1993). Too much freedom in group interaction decreases group cohesiveness. This, in turn, increases the decision cost either by generating a lower quality decision or taking more time to make a decision. Therefore, a coordination structure in Distributed GDSS (DGSS) should impose some restrictions on interaction to maintain a certain level of group cohesiveness.
In future research, the degree of system restrictiveness of a coordination structure needs to be defined more precisely. In this study, sequential coordination was assumed to be more restrictive than parallel simply because it has more procedural order. However, the findings of this study can not be generalized, or compared to the findings of other research, unless the degree of restrictiveness can be objectively determined. Very little is known about what determines the perceived degree of system restrictiveness.
Though leadership is the process of coordinating the activities of group members (Jago, 1982), there were not many significant findings related to the leader variable. One explanation may be that, though research on leadership explains the variety of leadership styles (House, 1971; Stogdill, 1959), the leader function in this study was too narrowly defined. All leaders were expected to behave exactly the same as they were instructed, regardless of their natural leadership style. Implementing leader's functions in DGSS as process structuring tools is one of the requirements in designing DGSS. The successful implementation of leadership functions in DGSS, however, is dependent on the further understanding of coordination effectiveness of different leadership styles within different contingent factors of DGSS (Turoff, et al., 1993). Little research has been done in this area.

4. Summary and Conclusions

On the basis of prior studies, several measures were taken to help to assure the effectiveness of the CMC groups in these experiments, regardless of the experimental condition or manipulation which they represented. All received substantial training and practice before being left on their own for a week or two to do their experimental task (with the exception of the Peer Review experiment, in which all subjects were already system users). All were at least technically facilitated, with the facilitator checking in daily to see if there were any problems requiring assistance (some also had designated group leaders). All had a clearly stated task, objective, and deadline, and all subjects considered the task at least minimally important, since it was a graded assignment. If any of these conditions were omitted, we suspect that the results would be negative (Hiltz and Turoff, 1991).
There are three basically different conceptualizations about the nature of CMC, and of asynchronous CMC in particular. One point of view, becoming less prevalent now that millions of people are spending hundreds of millions of hours "surfing the net" for "fun," is that it is a "poverty stricken" and "cold" medium. This point of view focuses on what it is not: it does not have some of the channels of communication of the face-to-face medium.
Most of those scholars who have spent time developing and studying CMC as a support for group interaction share the assumption that it can be an effective and sociable form of communication, but they differ on how this can best come about. One group views such systems essentially as a technological mechanism, feeling that effective CMC must be built into a feature-rich and highly structured and restricted environment. Groups need to have the technology essentially "force" them to behave in what are seen as effective ways to use the medium, in order to minimize process losses and maximize process gains (Johnson-Lenz, 1991). An example of this approach is the Coordinator (Flores, e.g., 1988), or software to force a completely sequential mode of coordination of interaction.
The second approach to building CMC systems conceives them as a context for interaction, "containers" so to speak, just as rooms are. This conception is based on a social theory that human systems are self-organizing and arise out of the unrestricted interaction of autonomous individuals. From this perspective, the role of the computer system is to provide a place for people to meet and self organize (Johnson-Lenz, 1991).
CMC is a very different form of communication than face to face meetings, and it takes some time for individuals to learn to use both the mechanics and the social dynamics of such systems effectively. All of the experiments presented here have included at least one condition in which groups used asynchronous CMC, but without time pressure. They had adequate training and at least a week to complete their discussions and produce their group product or decision. Under these conditions, it appears that groups do not need a restrictive, "mechanistic" approach to coordinating their interaction. They are capable of organizing themselves and will tend to feel frustrated by overly restrictive structures or procedures, and/or to become more inefficient.
Almost all of our attempts at a mechanistic process intervention had no significant positive effects on outcomes. For example, there were no significant differences in the major dependent variables measuring outcomes, between the "imposed sequential" process and the no-process or parallel process groups for the preference task, or for the investment task. Likewise, for a creativity task, there was no difference between groups that followed an imposed procedure, and those that did not; and for a planning task, there was no difference between groups that used Dialectical Inquiry and those that used a consensus approach.
On the other hand, the presence of "tools" that a group can use when it is ready to, does seem to improve the perceived richness of CMC, and can improve process and outcomes. The tools that we have provided in various experiments include the ability to build a common list, a set of voting options, the "question-response activity" that structures the exchange of ideas and opinions similar to Nominal Group Process, the possibility of anonymity, and a "polling" tool which can allow a group to construct any sort of questionnaire type item, and display results of the polling. One must choose the tools made available to the group very carefully, to match them to the nature of the task and the size of the group, we suspect, though we have not experimented with the option of just "throwing" all the available tools at a group and letting it decide on its own what might be appropriate and how to use it. We suspect that even two weeks is too short a time to expect a group to deal effectively with this much complexity, but that very long term groups which interact for months to years, would do perfectly well with such a tool chest at their disposal.
The results of these experiments support the assertion that asynchronous CMC is not like any other form of group communication; not only is it not like Face to Face unsupported meetings, but it also has very different dynamics than a computer-supported meeting in a "decision room." Coordination mechanisms and tools that "work" or "don't work" in other media tend to have very different effects in the distributed environment.
Things that "work" in an FtF environment may not help coordination and thus improve outcomes in the distributed environment. For example, DI has generally been found to be very beneficial to FtF groups. On the other hand, things that "do not work" in the FtF or Decision Room mode of communication, may be beneficial in the distributed mode. For example, though Watson (1987) did not find any significant benefits for Listing and Voting tools in a Decision Room, Dufner (1995; results summarized above) did observe many aspects of significant enhancement of results associated with the use of these tools.
The results of the experiments also confirm that measures of adaptive structuration are very important in the study of distributed, asynchronous GSS. Particularly since the group has no facilitator physically present to enforce suggested procedures, and because they have so long a period to evolve, they may not behave at all like what was intended and expected in regard to the use of suggested tools and procedures.

5. The Present and the Future

We remain optimistic about the potential benefits of asynchronous, distributed Computer Mediated Communication for supporting groups. The one experiment in this series so far that directly compared CMC and FtF groups, found that CMC groups produced significantly more creative results, for a creativity task (Ocker, 1995, reported above). We hope to do many more cross-media comparisons in the future, particularly if we are ever successful in our quest to obtain the necessary equipment for a state of the art "Decision Room" environment. As a start in this direction, the Ocker research has been extended thus far to examing two additional communication conditions, synchronous CMC groups, and "mixed mode" groups which have two hours of FtF meeting, two weeks of aysnchronous CMC, and a final face to face meeting.
We have concluded that the use of an asynchronous CMC system for GSS allows for a much wider range of possible coordination modes and tool support than is effective for synchronous meetings. All of the experiments to date have confirmed that even the most extreme asynchronous structures do not reduce the quality of the solutions when compared to more classical coordination and group approaches. The reasons for this cannot as yet be confirmed by any of the experiments, but they are hinted at from some of the results:
* All individuals are free to participate as they see fit and as much as they desire to.
* The freedom of participation as an individual seems to encourage:
* a greater expression of ideas
* more reflection
* less inhibition of ideas
* consideration of more options
To truly understand what is taking place there is a need to have groups deal with more complex and involved problems and to augment the typical statistical analysis with detailed content analysis of the discussions. Though extremely tedious, content analysis could resolve the alternative explanations for our results. In addition, the nature of benefits listed above are such that they may make a significant difference in quality only when dealing with fairly complex tasks that also instill a high degree of motivation for the group members.
While controlled experiments are informative for very specific issues, much of the insight that is needed for the support of asynchronous group communications has and will come from field trials and quasi experiments (Turoff et al 1993; Hiltz & Turoff, 1993). One key example of this has been our work in the area of collaborative learning (the Virtual Classroom (TM), Hiltz, 1994; see also Worrell et al, 1995). Its purpose is to use computer support to increase both access to and the effectiveness of education, at all levels. Rather than being built of steel and concrete, the Virtual Classroom consists of a set of group communication and work "spaces" and facilities that are constructed in software. Thus it is a "virtual" facility for interaction among the members of a class, rather than a physical space. It is "asynchronous," meaning that students and teachers may connect through the networks and participate any time, day or night, seven days a week. The software activities developed for this application stress collaborative learning approaches.
Field trials of various types with collaborative learning have been taking place at NJIT since 1980. Currently NJIT offers complete undergraduate degree programs in Information Systems and in Computer Science and many additional graduate courses through a remote learning program utilizing asynchronous group communications.
What this example and others have taught us, when combined with the experimental work in GSS, is that the key to successful systems is to discard many of the biases that come from making comparisons to face-to-face approaches and trying to adapt an approach of automating the face-to-face environment. Rather, the factors that seem to be crucial to enhancing our understanding of this area and in the future design of the functionality for such systems include:
* Providing a “non linear agenda” that allows the individual members of the group to focus on the contributions that each can best make, independent of the work of other members of the group at that moment in time (Turoff, 1991).
* Allowing a group to tailor the relationships structure of comments to fit the application domain as they perceive it. This can only be done by freeing fixed comment relationship structures to provide a full collaborative Hypertext capability (Turoff, Rao, & Hiltz, 1991; Rao & Turoff, 1990).
* Providing “reciprocal” coordination structures (Turoff & Hiltz, 1993) which will be able to check on consistency and agreement at the group level and inform participants when they need to reconsider their inputs based upon more recent contributions of others.
The reason why these factors have not played a significant role in most current GSS work has been the typical lack of complexity of the problem being examined.
The other area that our research is focusing on is the software development process and tools to support that task. Initially, the primary objective of this ongoing project is to increase knowledge about how to create more productive systems to support distributed, collaborative groups, particularly for complex software design and planning type tasks. The sub- tasks in the software development area span a wide range of critical problem areas:
* The need for enhanced creativity in the design process.
* Greater understanding of requirements between users and designers (e.g. experts who sometimes speak different languages)
* The planning of projects and efforts.
* Complex project management, which includes the tracking and monitoring of what has been accomplished, the detection of potential problems and the handoff and coordination of work between different individuals and sub-groups.
Within systems development, it is recognized that the stages of requirements definition and high-level design are important, and even crucial to the development of effective software. Collaborative designers work to achieve some consensus on the general characteristics of the new system in question (Olson, 1991). Ineffective communication during the requirements definition process is consistently associated with user dissatisfaction and lower quality systems, while effective communication is associated with improved productivity and higher quality systems (Curtis, 1988). Additionally, it has been increasingly suggested that the development of information systems and the definition of high-level requirements and design, could benefit from the infusion of creative and innovative solutions (e.g., Couger, 1993; Telem, 1988).
Particularly critical to this area will be the adding of additional tools and processes (such as group hypertext/ hypermedia authoring capabilities). Daft and Lengel were certainly right when they pointed out that the objective of most work meetings is to reduce both uncertainty and equivocality (Daft & Lengel, 1986) in unstructured problem solving.
While most work in the Hypertext area appreciates the utility of non- linear relationships in the content of the material to reduce uncertainty; however, it also seems self-evident that the problem of equivocality can only be handled by allowing people to perceive one another’s reactions to the information. This has always been clear in the context of asynchronous communications, where it is critical that each participant needs to know the status of the other members and the group as a whole. Within the context of a collaborative Hypertext environment, it becomes necessary for the individuals to be able to perceive how others traverse the network and how they modify it in a thought process type of temporal sequence.
The concept of utilizing Hypertext to support individuals to integrate the different domains supporting software engineering analysis, design and development is not new (Isakowitz, 1993). However, the equally important concept of supporting group processes and collaboration (Turoff, 1991) has received only a limited amount of attention. The specific goal of our research will be to focus on all the processes associated with software development that may be aided by Collaborative Hypertext Systems (Balasubramanian & Turoff, 1995). There have only been a few specific systems prototyped in this area (Marshall, 1992).
The current emergence of a whole new generation of implementation tools means that in the future it will be much easier to develop specific Decision Support functionality and Hypertext capabilities. It also means that there will be a major shift back to more internal development of tailored user software within organizations, rather than the current emphasis on purchased software. An objective of future research in the Group Support Systems area should be to provide a kind of "checklist" of what kinds of tools and procedures are likely to be helpful for different types of tasks, so that organizations can be guided in their self-tailoring of software to fit their needs.


This research was supported by grants from the National Science Foundation program on Coordination Theory and Collaboration Technology (NSF IRI 9015236 and NSF-IRI-9408805). The opinions expressed do not necessarily represent those of the National Science Foundation. Among the many people who have contributed to the program of research, in addition to the co-authors, are Raquel Benbunan, Robert Czech, Kenneth Johnson, Cesar Perez, Ronald Rice, Scott Poole, James Whitescarver, and William Worrell.