Working Together Apart

In C. D. Keen, C. Urquhart and J. Lamp (eds), Proceedings of the 7th Australasian Conference of Information Systems (ACIS96), Vol. 2, Department of Computer Science, University of Tasmania, pp. 701-712.

WORKING TOGETHER APART:

Fay Sudweeks
Key Centre of Design Computing
University of Sydney
fays@arch.su.edu.au

Marcel Allbritton
School of Information Studies
Syracuse University
mmallbri@syr.edu

In this paper, we explore the concepts of collaborative communication patterns and group developmental processes. Using a qualitative textual analysis of the content of email messages of a collaborative research group, a method for mapping types of collaborative communication and stages of group development is described. The case study describes the formation and implementation experiences of a large international group of researchers who used computer-mediated communication to study computer-mediated communication. We demonstrate that the case-study methodology provides theoretical insights and is a useful tool for allowing further investigation and deeper understanding of the role of communication patterns in group dynamics.

Collaborative groups are an important social and organisational phenomenon in communication and development. They are usually voluntary, and both the duration and focus are determined by the participants. Collaboration enables scientists to tackle problems that they are incapable of working on alone because of limitations of resources, skills, and time. Most individuals collaborate to combine intellectual and material resources to accomplish a project of mutual interest (Kraut, Egido and Galegher 1990). The study of collaborative communication may provide a way of managing several factors that otherwise inhibit communication within scientific communities: (i) extreme specialisation in topic areas among the members of a scientific community, (ii) complexity and overabundance of information, and (iii) rapid rate of change and high level of uncertainty that exists in modern scientific research environments.

Human communication is information-exchange between two or more human beings. In this context, computer-mediated communication (CMC) is one type of human communication, as is face-to-face interpersonal communication. Often, research on collaborative computer-mediated groups is more likely to focus on technology development than on communication processes. Precedence is given to the technologies linking communicating humans, rather than to how humans communicate using the technologies. Current research on computer-mediated groups relies heavily on laboratory and experimental studies; little is known about communication and development within "natural" electronic groups. In this paper, we suggest a research perspective that gives equal attention to both computer mediation and human processes. A case study of an electronic research group is used as a means to further understand the interaction of, and theoretical insights in, to communication and developmental patterns in collaborative computer-mediated groups. The findings are compared to developmental models of face-to-face groups.

Several texts in the literature examine the process of CCMC, to some degree, from different perspectives. Schrage (1990) conceptualised collaboration, and explored the dynamics of the process of collaboration. Hiltz (1984) applied sociological techniques to the study of group interaction through computer conferencing, specifically focusing on explaining how groups of knowledge workers communicate and cooperate. Sanderson (1996) distinguishes between collaborative research, in which participants jointly identify common research questions and coordinate efforts to create new knowledge, and cooperative research, in which participants simply exchange information or data.

There is a large and somewhat conflicting body of literature that explores the similarities and differences between computer-mediated and face-to-face groups. For example, compared with face-to-face groups, computer-mediated groups are more likely to: (i) make higher quality decisions (Easton, George, Nunamaker and Pendergast 1990; Gallupe, Bastianutti and Cooper 1991; Steeb and Johnston 1981), make about the same quality decisions (Hollingshead), or make lower quality decisions (McLeod); (ii) generate more ideas (Valacich, Paranka, George and Nunamaker, 1996), or generate less ideas (Easton et al. 1990); have a more equal participation rate (Easton et al. 1990, Hiltz, Johnson and Turoff 1986; Siegel, Dubrovsky, Kiesler and McGuire 1986); take longer to reach consensus (Gallupe et al. 1991; Siegel et al. 1986); and engage in more task-oriented communication (Hiltz et al. 1986), socio-emotional communication (Hollingshead, McGrath and O'Connor 1993), and inflammatory communication (Siegel et al. 1986). Most of these studies, however, are of small experimental groups and of variable duration. An interaction effect of task orientation and social behaviours observed by Walther and Burgoon (1992) could provide an explanation for the lack of consistency. Walter and Burgoon found that in extended-time asynchronous computer interactions, computer-mediated groups became less task oriented, less inflammatory, and more social than their restricted-time counterparts.

Studies in computer-supported cooperative work (CSCW) provide a knowledge base from which to pursue the study of CMC and collaborative communication among scientific researchers. While not based on the concept of collaborative communication, CSCW provides a sizeable body of knowledge that can be applied to the study of CCMC (see, for example, Kraut, Galeher and Egido 1990; Schmidt and Bannon 1992; Suchman and Trigg 1986). Another related body of knowledge is the study of computer-supported cooperative learning from a multidisciplinary approach (Kay 1992).

The interdisciplinary flavour of collaborative computer-mediated research requires an overview of various theoretical aspects of face-to-face group dynamics. Both positive and negative effects of group processes are reported. On the positive side, face-to-face groups are typically characterised by knowledgeable decision making, diversity of opinions, social endorsement of solutions and decisions (Maier 1967; Hill 1982; Schwartz and Levin 1990), and creativeness (Osborn 1941). On the negative side, face-to-face groups display social loafing (Kravitz and Martin 1986) and risky decision making (Kogan and Wallach 1967).

Even though independence and individualism are valued, people are very much affected and influenced by norms that are developed in both large social communities and small groups. All types of social interaction are an attempt by participants to `fit in'; that is, to adopt a behaviour that the participant perceives to be appropriate for the social context (Goffman 1963). According to communication theorists, all behaviour is communication. Interactivity could be defined, for example, as a product of successful interlocutor negotiation for a communicative interaction that satisfies the needs of each. Effective communication requires not only successful negotiation but a shared culture (Hirsch 1987). To attain an understanding of social behaviour, we must study the interactions of individuals within society. The individual's purposive behaviour can best be understood in terms of interactions within smaller communities such as organisations or groups. The process of becoming acquainted with a group of strangers and forming a new social unit—a group—from a collection of individuals requires considerable interactive skills.

Groups are popularly regarded as small communities which people join voluntarily and primarily for social reasons but in fact groups are an integral structure of a society. People may join a work group but have social and emotional needs to be satisfied; similarly, groups formed specifically for social reasons often develop tasks or programs to fulfil achievement needs. A conflict therefore develops between the high performance goals and socio-emotional needs of group members. The conflict and resolution processes as these needs are satisfied appear to be general patterns throughout the life cycle of groups. Face-to-face group development research suggests that groups move through stages associated with specific turning points in group activity. These stages are described broadly as sequential, cyclical or a combination of sequential and cyclical.

Sequential stage theories assume a linear progression of stages, but the number of stages vary. Tuckman (1965) defines five stages of group development: (i) forming is the initial "getting to know you" stage; (ii) storming is a period of conflicts as group members compete for status and recognition; (iii) norming is a stable period in which conflicts are resolved and behavioural norms are established; (iv) performing is the stage during which the group focuses on the task and creates a social unit; and (v) adjournment is the period in which the task is concluded and relationships change. Hare and Naveh (1984) identify four stages in their LAIG model of problem-solving groups: (i) latent pattern maintenance and tension reduction is the initial stage, when the group tries to reach a consensus on aspects of the work task; (ii) adaptation involves identifying skills within the group and assigning roles; (iii) integration involves flexibility and compromise; and (iv) goal attainment.

Recurring phases have a cyclic progression in which phases emerge, recede and re-emerge to confront in greater depth. The `fundamental interpersonal relationships orientation' (FIRO) model (Schutz 1996) attempts to explain interpersonal behaviour in people-oriented terms. Groups develop in response to three interpersonal needs: (i) inclusion (who's `in' and who's `out'), (ii) control (who's `top' and who's `bottom'), and affection (who's `near' and who's `far'). These processes are cyclical but near the termination of the group there is a reversal of the phases with less cohesiveness and, finally, diffusion of boundaries. Bales (1950) developed a method for the continuous observation of interactions, called `interaction process analysis'. He proposed that the success of a group depends on two factors: how well it can solve the tasks facing it (task function), and how well it can keep its members satisfied with the group (integrative, or socio-emotional function). A group proceeds through a series of task-oriented phases (orientation, evaluation and control) in parallel with predictable cycles of socio-emotional behaviour. Yet another theory introduces the notion of punctuated equilibrium (Gersick 1988, 1989), that group development is underlined by periods of inertia punctuated by periods of change.

A combinational model of sequential and cyclical phases was proposed by Robbins (1993). Using the terminology of Tuckman and the theoretical underpinnings of Gersick, Robbins developed a five-stage model: (i) forming and norming; (ii) low performing, (iii) storming, (iv) high performing, and (v) adjourning. The model has been applied to a computer-mediated group by Romm and Pliskin (1995) and found to be robust.

Despite the recent development of sophisticated web browsers, powerful authoring tools, and a cohort of programming software, concomitant with a phenomenal increase in Internet use, text-based CMC (email) remains the universal network communication media (Harrison and Stephen, 1996). The utilities of text-based CMC systems are low cost, efficiency, high level of information management, high level of connectivity, platform compatibility, and transcendence of time and space. These types of systems are the primary form of network media with which researchers retrieve information, exchange information and, most importantly, engage in collaborative communication.

Collaborative communication is always communication, but communication is not always collaborative communication. The shared creation of meaning differentiates collaborative communication from communication. If this meaning is to be mutually created, it is necessary that it be created in a shared "space" or environment. This shared space can be a cocktail napkin, a blackboard, or many other possibilities. The quality and quantity of collaborative communication are enhanced by tools that facilitate the shared creation of meaning. The shared "space" or environment provided by networked computers allows for collaborative communication to take place among individuals regardless of normal communication constraints of cost, geographic location, or time. Collaborative computer-mediated communication is a process of collaborative communication in which participants use computer-mediated communication. In Table 1, we define these key terms.

The hypothetical creation of types of communication in collaborative communication allow for a mapping of categories, and provide for further examination of the process of collaborative computer-mediated communication. Communication can only hypothetically be broken down through the use of a classification of types. Given the inherent complexity of communication, the following types of communication are not intended to be mutually exclusive. A typology of communication types could be beneficial as a descriptive method of understanding collaborative communication. This type of descriptive perspective is also well suited to understanding new communication technologies which are largely unprecedented reflecting the need for descriptive rather than experimental research methodology.

Informal control of communication is the collective informal creation, management, and enforcement of communication norms. Norms are mutually acceptable definitions of communication behaviours among individuals so that interactions can be organised into an agreed upon state. For example, a norm on the Internet is that email users do not type in upper-case letters as upper-case letters are considered the equivalent of yelling. Jackson (1977) refers to norms as mechanisms of regularity that groups collectively create and impose upon themselves.

Formal control of communication is connected with the enforcement of rules. Formally controlled communication in collaboration is often needed to manage the process and product of knowledge or information generation. While informal control of communication is generally performed on a collective level, formal control of communication occurs on an individual or small-group level. In CCMC, formal control of communication is performed by the coordinators of the collaboration. Formal control addresses the use of the CMC system in order to encourage and coordinate the activities of the participants. Formal control of communication is embodied in activities such as information exchange concerning project management, management of overhead, and rule enforcement. Examples include the process of selecting committee members, enforcing time limits on discussions, drafting of compromise proposals, and implementing voting systems. In a CCMC environment, the coordinators and highly active participants are more likely to exercise formal control of communication.

Socio-emotional communication is communication content which deals with the interpersonal relationships among the communicators. Socio-emotional communication addresses the creation of relationship norms among communicators.

Conceptual communication involves the shared creation of mutual understandings and meanings among communicators. An example is the creation or agreement upon a common vocabulary by group members, and the creation or agreement of the work to be completed. Conceptual communication is the integration of socio-emotional communication and task communication. This type of communication requires a medium to high level of interactivity. Generally, it is necessary that conceptual communication be preceded by some degree of socio-emotional communication. Conceptual communication may also involve, to some degree, the creation and prescription of shared rules to follow during the collaborative process. Conceptual communication often requires that implicit communication be made explicit. Realistically, it is not always possible to have complete or full shared creation of mutual understanding of meaning, but this is what conceptual communication strives for.

Task communication is communication content which deals with the explicit work to be accomplished by participants in a communication process. Task communication focuses upon information content of communication, while conceptual communication focuses on the creation of meaning preceding the processing of information. Task communication deals with specific research activities to be completed by members, and often has to be conducted independent of other group members. Task communication can be conceptualised more accurately as information exchange rather than communication.

Information technology has been a catalyst for change in many social, academic and commercial organisations. Computers have altered the way in which we form relationships and the way in which we work. It has been argued that computers have become the communication tool for collaboration (Rogers 1988). It has also been pointed out, however, that collaboration usually involves initial face-to-face meetings: "clearly, email ... [plays] an important role in facilitating ongoing collaborative efforts, but it ... [does] not engender completely new collaborative efforts among strangers" (Carley and Wendt 1991).

We now describe a research study, ProjectH, in which group email did, in fact, engender a completely new collaborative effort among strangers. Although a few of the participants met during the course of the study, the collaborating group never met face-to-face.

ProjectH is a naturally distributed collaborative group of more than one hundred researchers whose goal was an examination of selected qualitative and quantitative characteristics of CMC (Rafaeli, Sudweeks, Konstan and Mabry 1994; Sudweeks and Rafaeli 1995). The group was grounded in common membership of a computer-networked discussion group, a subset of whom proposed a research study to satisfy a shared desire to understand more clearly the nature of communication, culture and community on the network. Although other projects of similar magnitude have been conducted (see, for example, Sanderson 1996), ProjectH was perhaps the largest effort in a totally computer-mediated environment. The group used text-based asynchronous CMC as a tool for project management and for the distribution, collection and verification of data. Participant recruitment, distribution of information, coordination, formulation of policies, decision making, encouragement and data exchange took place through public and private email.

The scope of the twenty-three-month study can be traced from the diversity of its participants who represent:

There are various methodological tools available to scientists to gain knowledge and understanding of societal and communicative phenomena. One may choose quantitative or qualitative methods; one may apply interpretive or positivist theoretical paradigms; one may study societies, organisations, groups, individuals, or single messages; one may study cross-sectionally or across time. In general, the research questions of interest guide the choice of methodological tools. If the unit of analysis is the individual, applying a statistical analysis of data obtained from a sample of subjects within a population is a rigorous method of testing predefined hypotheses and determining generalisability of results. Research questions of social and organisational theories, though, seem best suited to inductive analyses. In this paper, we adopt an inductive qualitative approach to reveal communication patterns and group development in collaborative work.

During the course of the study, 1016 email messages were distributed to the group (see Figure 1 for a frequency distribution of activity), which formed the corpora for analysis. The corpora was archived by one of the authors in addition to being included in the publicly accessible archives at Comserve (comserve@vm.its.rpi.edu).

In a preliminary review of the corpora, we identified five salient dimensions: (i) themes—the discussion topics among participants; (ii) leadership—the inclination to conform or reject leadership and authority; (iii) debate—argumentativeness, criticism or aggression among participants, (iv) relationships—expressions or avoidance of friendship or intimacy among participants; and (v) action—goal-directed or task-directed activity. Following a technique developed by Romm and Pliskin (1995), each occurrence of a dimension was highlighted and labelled. The dimensions provided the means for observing the emergence of "turning points". Turning points are defined as a point in the discussions at which changes occur in the presence of a combination of dimensions. A turning point delineates the beginning and end of a phase in the group development. Using Romm and Pliskin's methodology, we identified six major phases. We adopt Robbins' (1993) terminology for the purpose of comparing his model with the emerging phases in this study: forming, storming, reflecting, low performing, high performing, and adjourning.

In a second examination of the corpora, we identified types of collaborative computer-mediated communication. The high frequency of one or more types over others indicated that such communication type(s) is characteristic of a particular developmental phase. Cell 1A in Table 2, for example, indicates that in Phase 1, the communication type A (informal control of communication) was one of the primary modes of communication in this developmental phase.

*Table 2:* Types of communication featuring in group developmental stages.
communication types		1	2	3	4	5	6
		stages
		forming	storming	reflecting	low performing	high performing	adjourning
A	informal		blank		blank	blank
B	formal	blank	blank	blank			blank
C	socio-emotional	blank			blank	blank
D	conceptual			blank	blank	blank	blank
E	task	blank	blank	blank			blank

Model of a Collaborative Computer-Mediated Group

In the following sections, we describe each of the developmental phases and include excerpts from ProjectH discussions to illustrate the communication types typical of each phase.

The project commenced when an email message about group dynamics was posted to a Comserve discussion list (CMC-L). The post caught the imagination and enthusiasm of subscribers to the list and within a few days, a group of some forty CMC-L members agreed to collaborate on a research study to capture the nature of "community" formation and examine characteristics of CMC. In the initial phase (25 May–10 June 1992), there was a process of getting to know other group members and defining the research goal.

The major dimensions during the initial phase were themes (what to study, how to study) and leadership (who will coordinate). Group members endeavoured to agree on the research to be undertaken, proposing two projects: (i) a quantitative content analysis of a representative sample of discussion lists, and (ii) a qualitative textual analysis of one list as a case study. During this phase, most members participated in the discussions to create a structure and develop norms within group. Leadership emerged, not from dominating the discussions or parading qualification, but by taking initiative and demonstrating a deep understanding of the research field. Communication primarily concerned the collective management and enforcement of communication norms (informal control), and group members reaching an agreement on the task to be undertaken (conceptual).

The second phase (10 June–26 June 1992) was a period of intense and heated discussions, mostly concerned with the ethics of using archived group discussions for research purposes. The first post raising the ethics issue marked the conclusion of the forming phase and the beginning of a period of low performing and high storming. The dimensions during this phase were themes (ethics issue), leadership (supporting and challenging the coordinators), and debate (opinionated and aggressive messages). The coordinators posted infrequently to avoid becoming entangled in the conflict and to have greater impact when emotions subsided. The communication modes during this period were a continuation of conceptual communication and an increase in socio-emotional communication, reflecting the high incidence of flaming. This phase included the day on which the highest level of messaging was reached—a total of 87 postings on 15 June.

After the previous intense phase, there was a period of equilibrium (26 June–6 August 1992)—a calm after the storm, both in tone and frequency of postings. Group members engaged in introspection and reflection, and the desire to satisfy socio-emotional needs surfaced. Dimensions during this phase were leadership (the coordinators facilitating the group to define its research focus) and relationships (people of similar interests communicating and supporting). Predictably, the typical communication modes were informal control and socio-emotional.

After months of brainstorming and conceptualising, the coordinators began to send more formal summaries of the status of the project which were intended to keep the project alive and encourage the group to implement the research ideas. Progress was made, albeit slow at first, and punctuated with periods of silence. The dimensions in Phase 4 (7 August 1992–20 January 1993) were themes (pretest, coding, copyright, sampling), leadership (coordinators more salient), and action (pilot coding completed and analysed). Communication in this low performing phase was formal and task-oriented.

When the project had a clear-cut agenda and well-defined methodology, it became obvious that more people were needed to reduce the workload since all members were volunteers and had to give priority to paid commitments. During the fifth phase (21 January–19 August 1993), discussion lists were canvassed which resulted in ProjectH almost tripling its membership. New members added vigour and heightened activity. The coordinators adopted new management strategies designed to maximise efficiency which included implementing a voting system on contentious topics and nominating committee members to discuss and draft proposals for specific issues. The effect of these strategies was to decentralise coordination and diffuse responsibilities.

During this work period, the dimensions were themes (coding, ethics, reliability, sampling, technology, oracles), leadership (diffusion), debate (the ethics issue revisited), and action (coding). Communication was primarily task-oriented and formally controlled.

In January 1994, the coding was completed and the database was available to coders soon after. During the final phase (20 August 1993–29 October 1994), people reflected about the long collaboration and voiced their experience of being part of a cohesive group. This belonging period was followed by a protracted period of adjourning. The group "hotline" remains open and occasional posts keep the group informed of other projects spawned by the collaboration, such as joint papers, graduate theses, a conference panel, a book, an electronic journal and so forth. The main dimension during this final phase was relationships—people had formed close associations both online and offline and were reluctant to part. The communication was socio-emotional and informally controlled.

The development of ProjectH as a collaborative group could be described as a six-phase process, each phase having different combinations of communication types. Figure 2 shows the similarities between ProjectH, a CMC group, and face-to-face groups, as described by Robbins (1993).

Figure 2. Mapping ProjectH developmental phases (top) and Robbins' (1993) stages (bottom)

There are also quite distinct differences that need to be explored further. ProjectH passed through three stages before settling down to a low level of performance in the fourth, whereas Robbins' model predicts some performance results much earlier. It appears that CMC groups take longer to develop norms and social relationships, and plan and assign tasks. After the initial stages of group development, however, researchers find that CMC groups come to resemble face-to-face groups in their development. One could pose a number of explanations for the delayed development of CMC groups, such as inexperience with using CMC as a group tool, slower information processing ability, absence of appearance screening, and a disorienting feeling that one is never "there" with group members because there is no "there" to be at. Even though researchers such as Walter, McGrath and Hollingshead base their predictions on artificially-created groups in a laboratory environment, the research presented in this paper tentatively supports the results obtained in a comparison of CMC and face-to-face groups.

Groups are often characterised as democratic with well-established lines of communication. However, tensions are created by conflicting task-oriented and socio-emotional needs. When action is high and the group is focused on tasks, there tends to be security in the structure of working relationships; when the organisational work-oriented structure is less clearly defined and deadlines are not imposed or not urgent, the desire for intimacy and social interaction surfaces. "It is the rare group that can effectively combine social-emotional interests with those necessary for getting the job done" (Napier and Gershenfeld 1993: 479). In a model of group development, the interplay of task and socio-emotional processes need to be understood and accounted for. Knowing how to manipulate communication types at different developmental stages will enable scientific researchers to avoid excessive stress and conflict, and to improve both satisfaction and productivity in collaborative groups.

This paper raises more questions than are answered. What key variables facilitate the process of CCMC among scientific researchers? What are the most effective controls that facilitate CCMC? What exactly is the relationship between communication patterns and developmental processes? Do different types of communication drive group dynamics or do group developmental phases affect the types of communication used?

We have endeavoured to understand the phenomena of collaboration and CMC with the use of qualitative research methodologies. We presented a framework for categorising the types of communication in the process of CCMC. Phases of group development were examined and integrated with communication modes. The premise is that more can be understood about the process by breaking down the types of communication used, and the phases that a group progress through when its members work collaboratively.

Acknowledgements: The authors wish to thank Celia Romm and Everett Rogers for their inspiration and motivation. Most of all, we are indebted to the ProjectH group who provided the stimulus and material for understanding communication and developmental patterns.

Bales, R. F. (1950). An Interactive Process Analysis: A method for the study of small groups, Addison-Wesley, Reading, Mass.

Carley, K. And Wendt, K. (1991). Electronic mail and scientific communication: A study of the Soar Extended Research Group, Knowledge: Creation, Diffusion, Utilization, 12(4), 406-440.

Easton, G., George, J., Nunamaker, J. and Pendergast, M. (1990) Using two different electronic meeting system tools for the same task: An experimental comparison, Journal of Management Information Systems, 7(1), 85-100.

Galegher J., Kraut R. E. , and Egido C. (Eds) (1990). Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Lawrence Erlbaum, Hillsdale, NJ.

Gallupe, R. B., Bastianutti, L. M. and Cooper, W. H. (1991) Unblocking brainstorms, Journal of Applied Psychology, 76(1), 137-142.

Gersick, C. J. K. (1988). Time and transition in work teams: Towards a new model of group development, Academy of Management Journal, 31(1), 9-41.

Gersick, C. J. K. (1989). Marking time: Predictable transitions in task groups, Academy of Management Journal, 32(2), 274-309.

Hare, P. and Naveh, D. (1984). Group development at the Camp David Summit, Small Group Behavior, 15(3), 299-318.

Harrison T. M. and Stephen T. D. (eds) (1996). Computer Networking and Scholarly Communication in the 21st Century University, SUNY Press, New York.

Hill, G. W. (1982). Group versus individual performance: Are N=1 heads better than one? Psychological Bulletin, 91(3), 517-539.

Hiltz, S. R., Johnson, K. and Turoff, M. (1986). Experiments in group decision making, 1: Communications process and outcome in face-to-face vs. computerized converneces, Human Communication Research, 13(2), 225-252.

Hollingshead, A. B., McGrath, J. E. and O'Connor, K. M. (1993) Group task performance and communication technology: A longitudinal study of computer-mediated vs face-to-face work groups, Small Group Research, 24(3), 307-333.

Jackson, D. D. (1977). The study of the family, in P. Watzlawick and J. H. Weakland (eds), The Interactional View, W. W. Norton, New York, pp 2-21.

Kay, A. R. (1992). Collaborative Learning through Computer Conferencing: The Najaden Papers, Springer-Verlag, New York.

Kraut, R. E., Egido, C. and Galegher J. (1990). Patterns of contact and communication in scientific research collaboration, in J. Galegher, R. E. Kraut and C. Egido (eds), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Lawrence Erlbaum, Hillsdale, NJ, pp. 149-171.

Kogan, N. and Wallach, M. A. (1967). Risk taking as a function of the situation, the person, and the group, New Directions in Psychology, Vol. 3, Rinehart and Winston, New York.

Kravitz, D. A. and Marting, B. (1986). Ringelmann rediscovered: The original article, Journal of Personality and Social Psychology, May, 936-941.

Maier, N. R. F. (1967). Assets and liabilities in group problem solving: The need for an integrative function, Psychological Review, 239-249.

McLeod, P. L. and Liker, J. K. (1992) Electronic meeting systems: Evidence from a low structure environment, Information Systems Research, 3(3), 195-223.

Napier, R. W. and Kershenfeld, M. K. (1993). Groups: Theory and Experience, Houghton Mifflin, Boston.

Osborn, A. F. (1941). Applied Imagination: Principles and Procedures of Creative Thinking, Scribner, New York.

Rafaeli, S., Sudweeks, F., Konstan, J. and Mabry, E. (1994). ProjectH overview: A quantitative study of computer-mediated communication, Technical Report, Key Centre of Design Computing, University of Sydney, available by ftp from ftp.arch.su.edu.au/pub/projectH/papers/techreport.txt.

Rogers, E. M. (1986). Communication Technology: The New Media in Society, Free Press, New York.

Rogers, E., M. and Kincaid, L. D. (1981). Communication Networks: Toward a New Paradigm for Research, Free Press, New York.

Romm, C. T. and Pliskin, N. (1995). Group development of a computer-mediated community, Working Paper, Department of Management, University of Wollongong, Australia.

Sanderson, D. (1996). Cooperative and collaborative mediated research, in T. M. Harrison and T. D. Stephen (eds), Computer Networking and Scholarly Communication in the 21st Century University, SUNY Press, New York, pp. 95-114.

Schmidt, K. and Bannon, L. (1992). Taking CSCW seriously: Supporting articulation work, Computer-Supported Cooperative Work, 1(1-2), 7-40.

Schrage, M. (1990). Shared Minds: The New Technologies of Collaboration, Random House, New York.

Schutz, W. (1966). The Interpersonal Underworld, Science and Behavior Books, Palo Alto CA.

Schwartz, A. E. and Levin, J. (1990). Better group decision making, Supervisory Management, June, 4.

Siegel, J., Dubrovsky, V., Kiesler, S. and McGuire, T. W. (1986) Group processes in computer-mediated communication, Organizational Behavior and Human Decision Processes, 37, 157-187.

Steeb, R. and Johnston, S. C. (1981) A computer-based interactive system for group decision-making, IEEE Transactions on Systems, Man, and Cybernetics, SMC-11(8), 544-552.

Suchman, L. and Trigg, R. (1986) A framework for studying research collaboration, Proceedings of CSCW'86, Austin, TX, pp. 221-228.

Sudweeks, F. and Rafaeli, S. (1996). How do you get a hundred strangers to agree: Computer mediated communication and collaboration, in T. M. Harrison and T. D. Stephen (eds), Computer Networking and Scholarship in the 21st Century, SUNY Press, New York, pp. 113-136.

Tuckman, B. W. (1965). Developmental sequence in small groups, Psychological Bulletin, 384-399.

Valacich, J. S., Paranka, D., George, J. F. and Nunamaker, J. R. (1996) Communication concurrency and the new media: A new dimension for media richness, Communication Research (in press).

Walther, J. B. and Burgoon, J. K. (1992). Relational communication in computer-mediated interaction, Human Communication Research, 19(1), 50-58.

Communication	Communication is a process in which participants create and share information with one another to reach mutual understanding (Rogers and Kincaid 1981).
Computer-mediated communication	Computer-mediated communication is human communication between two or more individuals through the use of central computers that store and process message content, and are connected to users in a communication network.
Collaborative communication	Collaborative communication is a process of communication in which participants share in the process of creating meaning and mutual understanding of meaning, in a shared space for a specific purpose (Schrage 1990).
Collaborative computer-mediated communication	Collaborative computer-mediated communication is a process of collaborative communication in which participants use computer-mediated communication.