Group Support Systems through the Lens of Action Research: Cases in Organisations Robert Davison Dept of Information Systems, City University of Hong Kong Tat Chee Avenue, Kowloon, Hong Kong isrobert@is.cityu.edu.hk Tel.: +852-2788-7534 Fax: +852-2788-8694 Sajda Qureshi Research Institute for Decision & Information Systems, Erasmus University, The Netherlands Email: squreshi@fac.fbk.eur.nl Gert-Jan de Vreede, School of Systems Engineering, Policy Analysis and Management Delft University of Technology, The Netherlands Email: gertjanv@sepa.tudelft.nl Doug Vogel Dept of Information Systems, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong Email: isdoug@is.cityu.edu.hk Noel Jones Capacity Building International Ireland Email: NJones@iol.ie Abstract Research into Group Support Systems (GSS) has traditionally employed laboratory, or to a lesser extent, field study methodologies. Questions have been raised as a result concerning methodological appropriateness to complex dynamic organisational situations. In this paper, we describe an approach that combines Action Research with GSS, so as to gain the benefits arising both from direct intervention with organisations, and from the use of a system that encourages greater participation in group interaction. A number of cases, from different geosocial environments on three continents, are used to illustrate how this bi-methodological combination works in practice. We find that the 1 synergism enabled creates an environment advantageous to researchers and practitioners alike. 1. INTRODUCTION The increasing occurrence of networked computers and communication technology has enabled new technologies that support groups with a multitude of organisational tasks to emerge. The term 'groupware' is often used to denote these technologies. One subset of groupware - Group Support Systems (GSS) - aims to make group meetings and group decision making more productive. We conceive of GSS as computer-based information systems that support groups of users in work that is both intellectual and collaborative, employing a variety of idea generating, evaluating and structuring tools. Since the mid-1980s, an increasing amount of research has been conducted in GSS. Much of this work was initially undertaken at Universities in North America, but since the early 1990s we have witnessed work carried out in Europe, Africa, Australasia and South East Asia. Most of the studies published have exhibited considerable inconsistencies, making it hard to draw firm conclusions about the use of GSS. Many researchers have noted these inconsistencies, Jessup et al. (1990, p.313) for example, discussing the role of anonymity in GSS research, noting "unfortunately, empirical investigations thus far provide confusing results". Various explanations have been suggested for these inconsistencies, for example: the lack of theory driven methodological research (Rao and Jarvenpaa, 1991); the prevalent use of students as subjects; the fact that groups are formed for the sole purpose of the task studied (Pinnsonneault and Kraemer, 1990); and even experimental design itself (Galliers and Land, 1987). Furthermore, Pervan (1994a) observes that there has been insufficient replication of experimental conditions to make anything more than the most tentative of generalisations about GSS performance. Another major difference that has emerged between the various studies published relates to the methodology used. Experimental work has dominated and continues to dominate the discipline (at least so far as published material is concerned), though there is an increasing awareness of the importance of field studies, as well as other methods such as case studies and action research. The design of experimental studies was often purely grounded in terms of theoretical 2 considerations, as well, perhaps, as what Galliers (1991) has alluded to as a "house style". This methodological parochialism seems illogical, given the richness and complexity of the real world, since a methodology best suited to the situation under consideration, as well as the objectives of the researcher, should be chosen (Pervan, 1994b). Indeed, much of the experimental work involved artificial problems being tackled by artificial decision makers - typically students. As a result, few studies attempted either to study issues emanating from real organisational use of GSS, or to study the actual use of GSS by organisations themselves. In fact, only a handful of published studies and PhD dissertations have investigated the in situ application of GSS in organisations, other studies being conducted with university facilities. Although the overall number of field studies conducted is relatively small, the results that have emerged do show a much higher degree of consistency, leading us to suggest that GSS do have the potential to add value to group interaction. More critically, however, it will be through research in organisations that the real issues that are important to organisations will be elicited. Thus, it is this form of research, we argue, that should be undertaken since it has the potential to produce benefits for researchers, practitioners and organisations alike. In this paper we propose that action research is particularly suitable for organisational research as it enables researchers to become deeply involved with the processes that take place in organisations. Their involvement should, in turn, enable them to glean insights into the nature of the added value that GSS can engender and so to advise organisations how they can make optimal use of GSS. The objective of this paper is to illustrate how GSS can be applied successfully in a variety of organisational settings, and how researchers and practitioners alike can develop new knowledge and understanding from this application. We propose that action research provides a unique implementation and research context in which implementation, practice and research synergistically reap benefits and inform one another. This study will make a contribution towards understanding how GSS can function in organisations in different social and cultural contexts - we will illustrate our work with cases drawn from disparate geosocial environments. We expect that this work will inform researchers as to how they can most effectively conduct GSS research in organisations in a manner that is suitable to, and respectful of, the 3 problem domain. Practitioners will also benefit through an increased awareness of the importance of research in action in their work, and how it can help to elicit knowledge that subsequently informs their practice. In the following sections, we examine the research and practice of action research, explaining how it functions and what advantages it offers to researchers and organisations. We then explain how we believe GSS and action research can work together - in research and practice - and illustrate this with several cases. We then discuss the relevant issues, drawing implications for future research, practice and application of GSS in organisations. 2. ACTION RESEARCH Action research, according to Kemmis (1980), involves the application of tools and methods from the social and behavioural sciences to practical problems with the dual intentions of improving the practice and contributing to theory and knowledge in the area being studied. Action researchers either participate directly in, or intervene in, a situation or phenomenon in order to apply a theory and evaluate the value and usefulness of that theory (Checkland, 1981, 1991; Vreede, 1995). Action research is typically carried out as part of an attempt to solve problems by allowing the researcher to become a participant in the action, the process of change itself becoming the subject of research (Checkland 1981). For action research to be successful, the organisation must be supportive of, and the processes being investigated must be conducive to, information sharing and learning. The action researcher is not alone in gathering data for the research, as the other participants are also responsible for data collection. The sharing of data feeds into a learning process in which the researcher modifies her/his theories and the other participants modify their perceptions and ways of working. Argyris et al. (1985) define action research as an inquiry into how people design and implement action in relation to each other. They state that because of this, action research is a science of practice. In action research, the researcher has a remit for action. This means that the organisation in which he or she is doing research has given him the go ahead to solve their particular problem or help manage change processes. Zuber-Skerrit (1991) describes action research as having the following four elements: plan, act, observe and 4 reflect. This way of using action research involves a cycle in which the researcher's observations and reflections feed into abstract conceptualisations and some concrete experiences which together inform the researcher's action to improve the problem situation or manage a change process. Action research is best seen as an iterative cycle in which the researcher begins with planning how to carry out the activity; this plan must also serve as the research methodology. The cycle continues to execution, also known as action or intervention, to solve the problem or manage the change process; this is where the researcher collects data. Additional data on the situation or phenomenon being studied are gathered in the next step of the cycle, observation. The following reflection entails interpretation of the data, and consequences of action that then feed into the planning stage to modify the methodology or model that then determines what action will be taken in the next cycle. Further cycles of activities continue until a desired end-state is achieved (Zuber-Skerrit, 1991). As action research often takes place within organisational research settings regarded as case studies, some researchers position action research as a subset of case study research (Galliers, 1991), but others (e.g. Vreede, 1995) observe the differences between the two approaches and appear to suggest that they should be treated as separate methods (see Table 1). Table 1 Case studies and action research (adapted from Vreede, 1995) Case Study Researcher is observer Exploratory, explanatory or descriptive Focus on "how?" and "why?" May be positivist or interpretivist While we recognise Action Research Researcher is active participant Prescriptive, intervening Additional focus on "How to?" Interpretivist that case studies and action research are methodologically different, the reasons that Benbasat et al. (1987, p.370) put forward as making case study research viable are also true for action research, viz.: • It is necessary to study the phenomenon in its natural setting when research in it is in its formative stages; 5 • The researcher can ask "how" and "why" questions, so as to understand the nature and complexity of the processes taking place; • Research is being conducted in an area where few, if any, previous studies have been undertaken. Benbasat and his colleagues (1987) further argue that within a case study research strategy, multiple methods of data collection should be employed to gather information from people, groups and/or organisations. The boundaries of the phenomenon studied may not be clearly evident at the outset of the research and no experimental control or manipulation is used. When used with a case study, action research becomes particularly useful in generating a body of experiential knowledge which may be distilled into ideas, theories and concepts that can be used in different settings at different points in time. This may be done by identifying causal relations over a period of time, by presenting a narrative within the context of a theory or framework, by developing or modifying an existing theory or combination of theories in the light of knowledge gained in the case study, or by comparing the knowledge and explanations with other studies. In addition, action research may be used not only for theory testing, but also theory building and/or expanding (Galliers, 1991). The strength of action research is that it provides the researcher with an inside and working view of a case especially where an explicit change process becomes the subject of the research. Ideally, action research will ensure that there is synergy between research and practice where people work together to improve complex situations. However, according to Checkland (1981), problems with action research arise from the fact that it cannot be wholly planned and directed down particular paths. Where a particular situation does not present a problem situation that is perceived to require change, action research should not be seen as appropriate. At the other end, a weakness of action research, according to Benbasat et al. (1987), is that it potentially lacks objectivity stemming from the researcher's stake in effecting a successful outcome for the client organisation. They add that further generalisations to other situations where the intervention technique is applied by people less knowledgeable than the researcher may be difficult. In view of these strengths and weaknesses, a synergistic role for action research and GSS becomes apparent. This synergy is discussed in the 6 following section, using cases to illustrate the possibilities open to researchers pursuing GSS. 3. SYNERGY BETWEEN GSS AND ACTION RESEARCH The synergy between GSS and action research can work in a number of ways. GSS provide the structure and platform for systematic data collection, as well as a mechanism by which organisations can explore alternative forms of group interaction and problem solving. GSS function as a catalyst that can enable organisations to introduce effective implementation and support protocols that otherwise might go unused. The protocols can be consistently applied and scarce facilitator skills can be leveraged across a larger number of situations and groups. Overall, GSS make feasible the broad-based collection of data that action researchers can use in developing guidelines and diffusing innovation in organisations. The use of GSS may help researchers to overcome some of the weaknesses of action research. The use of anonymity in GSS, for example, appears to provide a degree of 'objectivity' allowing participants to express their genuine views. Using GSS facilitation as an intervention technique within action research may potentially reduce the bias of the researcher and allow for further generalisations of the intervention technique to other situations to be achieved. On the other hand, GSS requires support both in terms of how to support problem solving most effectively, and for theory building, testing and developing concepts relating to managing change in organisations. It appears that action research is particularly suited to strengthening GSS research by enabling more real world problem situations to be solved in situ. At the same time, action research can be used to build theory, concepts and/or insights that may directly enable GSS to be applied more effectively. The synergy that we describe is illustrated in the following section where we consider four cases where GSS has been applied as part of an action research guided solution. The four cases were conducted in three different geosocial environments, and thus give an idea of the breadth of the applicability of the "GSS + Action Research" technique. 7 4. THE CASES 4.1. IBM (USA) GSS was introduced at IBM in 1987 with the assistance of university researchers. A series of field studies ensued, demonstrating that GSS technology could be effectively introduced in organisational environments for a variety of problem solving tasks (Grohowski, et al., 1990; Vogel, et al., 1990). Based on success at the first facility, IBM installed the technology at six more sites over the following year and similarly expanded their internal facilitation support capabilities (Martz, et al., 1992). IBM continued expanding internally to 24 sites and beyond with the same format of use i.e., pre-planned session agendas with facilitation support throughout the meeting process. The facilitation role has been institutionalised with several generations of facilitators emerging from a wide variety of backgrounds and levels of experience with group and organisational dynamics. The series of GSS field and case studies at IBM could, from a different perspective, be seen as action research. As such, University of Arizona researchers worked with IBM to get the GSS technology operational and trained IBM facilitators (Grohowski, et al., 1990). Advice was continuously given to IBM over a period of years during which time a number of empirical studies took place (e.g., Martz, et al., 1992). University of Arizona researchers additionally conducted advanced facilitation training sessions as organisational use spread internally. Although the focus of the researchers was primarily informed by a positivist perspective, it could also be argued that much of what was done, especially in the early years, was driven by an interpretivist perspective in that few preconceptions and no experience existed with respect to the operational use of GSS in a corporate setting. The researchers were truly in an exploratory mode as they examined human actors within their social settings through field and case studies, noted by Orlikowski and Baroudi (1991) as appropriate to generating valid interpretative knowledge. 4.2 Hong Kong Police During 1996-1997, the Junior Command Course of the then Royal Hong Kong Police Force used a GSS as part of its management training programme for Junior Inspectors. A total of nine groups of officers undertook management case problems 8 that examined real issues confronting the police in Hong Kong, viz.: CD ROM piracy and the repatriation of illegal immigrants to Vietnam. The facilitator (researcher) of the GSS meetings consciously adopted an action research framework to enable the sharing of data among the various parties involved, and to improve the practice of providing a valuable learning experience for the officers. Several action research cycles were conducted, with both the facilitator of the meetings and the police force as an organisation becoming progressively more aware of the importance of both the GSS and the need to regulate the manner in which the officers undertook their cases. An easily identifiable end-state did not exist in the case, and so potentially the meetings could run ad infinitum, so long as the officers were perceived to be benefiting from the activity. However, major improvements were made to the facilitation style over the duration of the case, with the result that the officers were able to work more effectively and efficiently in the later groups. It was not always possible to stick to a rigid action research path as full cooperation was not forthcoming from the senior officer who "owned" the training course. For example, he provided very little in the way of feedback on the activities undertaken by the officers, nor did he attend any meetings in the last two months of the case, relying on the facilitator to run all aspects of the meetings. In initial meetings, the senior officer insisted that the meeting management style be entirely ad hoc, with facilitation adjusted to meet the ongoing needs of the meeting. The ad hoc nature was enhanced by his propensity to provide crisis management information on the fly to the officers, e.g. informing them that the death penalty would be imposed for CD ROM pirates in future, or that the Vietnamese government proposed to employ its soldiers to escort returning migrants. In later meetings, the ad hoc style was moderated, with a more formal agenda adopted by the facilitator in response to junior officer suggestions that the meeting was too unstructured. Although the amount of time used by the later groups for the different activities was generally shorter, more work was accomplished with the improved meeting efficiency. 9 4.3 The World Bank in Africa At World Bank headquarters in Washington, DC, a dedicated GSS facility was opened in May, 1993 as a joint venture between two support units, the information technologies department and the department charged with organisational design, planning and related services. The technology has been used successfully in the intervening years for a variety of tasks for intact management groups, country teams drawn from across organisational boundaries with ongoing accountability in a particular country, and special project teams. The action research reported here focuses on World Bank use of GSS in Malawi and Zimbabwe as a part of the bank's efforts to take the GSS out into the field. GSS use was targeted at country assistance strategies and related stakeholder consultation as well as preparation of environmental strategy. Jones and Miller (1997) report on results from over 120 stakeholders including government ministry representatives, donor sector groups, private sector participants, religious groups, academicians, journalists, parlimentarians, regional officials, and project directors. Action researchers had to overcome considerable scepticism in their attempts to use GSS to get the depth of coverage and range of stakeholder insights that they felt was appropriate. Department management teams at the bank headquarters were particularly sceptical about taking computer based technology to Africa and using it with a wide range of stakeholders with widely varying keyboarding skills and a historical preference for face to face dialogue with World Bank staff. As it turned out, the stakeholder groups were fascinated by the technology and, without exception, requested that the World Bank leave it behind when staff returned to headquarters. Almost all participants welcomed the opportunity to share their ideas with the World Bank via GSS. Quality and quantity went hand in hand as participants generated a wealth of information that went far beyond World Bank needs. The World Bank was better able to efficiently and effectively learn what stakeholders felt was important. 4.4 Ventana Corp., USA Ventana Corporation was created by the University of Arizona to transfer GSS technology to the private and public sectors. Needless to say, the technology is used widely within the organisation of approximately 50 employees for a host of 10 problem solving and ongoing project activities. In the project reported here, a PhD candidate from the university was working with Ventana Corp. to examine the application of GSS coupled with process modelling and animation tools for generating insights for business process re-engineering. A task environment was chosen around the cycle of order handling and product shipping. The goal was to arrive at a more cost effective approach to deal with expanding company sales. The approach taken was to use the GSS to develop a validated activity model reflecting broad-based stakeholder input and a shared understanding of the "as-is" situation. The process model was then interfaced with a commercially available simulation package to gain insights into aspects of the broader process that particularly warranted re-engineering. Additional details can be found in Walsh and Vreede (1997). During model development, it became clear that numerous opportunities existed to improve business processes. Processes that had evolved over a number of years were clearly inefficient and in need of re-engineering. Stakeholders, using the GSS as part of the model development, began to document improvement ideas. The simulation of the finished model also amply demonstrated opportunities for improvement that became immediately visible. A re-engineering team was formed to design a radically improved process. Disagreements began to arise with respect to incremental improvements through automation versus a more "big bang" approach based on communicated vision and major departmental reorganisation focused on a new concept of customer support. At that time, the then president of the organisation intervened, shelving continuation of the re-engineering process. The researchers wrote up their results without having seen a radically new system implemented. 5. DISCUSSION In this section, we attempt to analyse what might have occurred had action research not been an integral component of the studies reported in this paper. In the case of IBM, it is doubtful that the researchers would have been granted access to groups had they not been seen as providing immediate value. The promise of a research report at some future date would have not provided sufficient motivation to gain entry. IBM was looking for assistance in fitting group 11 support technology to the needs of the organisation. In fact, researchers were not allowed to use the name IBM in publications until a sense of trust had been established based on researchers having helped the organisation implement the technology. Thus, there would have been no data had the action research style of interaction not been applied. It can also be suggested, however, that the data that was collected was not tainted by researcher presence. In fact, some of the data reported in one of the studies (Vogel, et al., 1990) was collected by IBM personnel without researcher knowledge. A particular focus of the research was to enable IBM to attain self-sufficiency in facilitating sessions and sustaining GSS innovation diffusion such that the organisation would be increasingly less dependent on the university researchers. In the Hong Kong Police case, the researcher became a value-added component of the team filling a void that otherwise would not have been covered. The project would otherwise have terminated after a few sessions, given the senior officer's non-involvement. The teams simply did not have sufficient skills to facilitate the meeting by themselves, while the senior officer was unable to allocate a sufficiently high priority to the project. Furthermore, as a result of the facilitator's close involvement in the case, useful lessons can be drawn that inform researchers and practitioners alike. The value of GSS and action research for educational environments involving the study of cases, notably those that involve real life problems with participants who have vested interests in the problem solution, is demonstrated. As researchers often work in educational environments, ample opportunities for research of this type should be readily available. Organisations outside the education sector still make considerable internal use of learning techniques, the police being but one example. In the case of the World Bank, initial scepticism by management turned out to be unfounded, with suspicions of computer illiteracy proving to be untrue. Without action research, however, the project might not have been approved to proceed, since the technology would not have been effectively adapted to fit the circumstances of the African environment. The use of GSS in World Bank headquarters had not prepared researchers to deal with situations such as vagaries in power supply and language translation reflecting local terminology and conditions. Without an action research approach that enabled researchers to adapt 12 quickly to local circumstances, the project, in all likelihood, would not have succeeded. Group dynamics were also noted to be different as a function of local culture which resulted in the need to modify session design attributes. The in situ sessions provided not only a wealth of content feedback but also process knowledge relative to the effective application of GSS in African contexts. It is doubtful that more traditionally controlled research approaches would have resulted in the amount of researcher learning or participant satisfaction attained. In the case of Ventana Corp., the researchers were considerably more focused on applying and evaluating tool use than in actually implementing broadbased organisational change. Thus, the extensive challenges that occur when new automated systems are implemented were not the primary focus of the researchers, although they certainly were interested. Unfortunately, the company was also not prepared, nor was there the broad-based support necessary, to implement the dramatic changes suggested. Lack of direct involvement of the then company president throughout the re-engineering activity and failure to actively endorse and promote suggested changes were key contributors to the project's demise. Nevertheless, the project demonstrated the research potential of exploring two relatively new technologies - animated simulation and GSS - that are not normally used simultaneously to support organisational redesign. It is interesting to query whether the inclusion of action researchers with a focus on systems implementation and related behavioural issues would have brought about a different outcome. It may have been the case that the more radical actions suggested were inappropriate for the organisation at that time in its history. Nonetheless, the organisation has subsequently implemented many of the suggestions, albeit at a leisurely pace. These cases offer a mix of cultures, conditions, geosocial circumstances and varied organisational forms that are indicative of the broad range of situations appropriate to action research. A consistent characteristic of all of these cases is the establishment of trust and confidence between researcher and organisation features that action research tends to engender. As an organisation sees a researcher work actively and productively, it then becomes more willing to cooperate and share information that might otherwise not be obtained. At the very least, researchers would have been unable to carry out the more data intensive field research reported in the literature without intervening. Furthermore, action research 13 was employed to assist the organisation in adapting the technology to fit existing needs. The action research acted as a catalyst in sustaining a broader program of field and case studies, as well as focused laboratory experiments, that provided feedback to extended development of the GSS product. Certainly, action research is not appropriate in every situation nor should it be applied without careful consideration. In the words of the ancient Chinese military strategist Sun Tzu, "Do not repeat the tactics that won you a victory, but vary them according to the circumstances" (Wee et al., 1991). As previously noted, if no change is in order, action research is inappropriate. Furthermore, researchers need to ensure that, in their enthusiasm, they do not lose their objectivity and become swallowed by the circumstances that the action research was initiated to address in the first place. In the presence of 'groupthink', the researcher has the opportunity and responsibility to bring in alternative perspectives that might not otherwise be noted. The researcher also needs to work diligently to maintain sufficient distance from the situation to study it effectively whilst maintaining organisational trust and confidence and the perception of being value-added. It indeed becomes a balancing act that, on occasion, can be very difficult and challenge the researcher in ways not encountered in traditional positivist research. Action researchers need to reflect periodically on their degree of involvement and its impact on data integrity and seek to be vigilant and cognisant of the potential for bias that would adversely affect data analysis and interpretation of results. 6 - CONCLUSIONS Academic research has been accused of having a focus on 'sifting through the remains in the stable after the horses have left on the crusade'. Action research puts academics on the front line with the organisation, exploring new concepts and removing uncertainties. If properly consummated, action research can add to knowledge while proactively assisting the organisation under fire. Reasonable rigor can be sustained in the presence of high degrees of relevance. It is argued here that this form of research fits especially well with rapidly changing situations in complex environments, typical of GSS use in organisational contexts. The studies discussed above illustrate the potential of the action research approach to 14 harmonise the technology with national and organisational cultures. It is also argued that this form of research is particularly conducive to gaining the respect and confidence of organisations without sacrificing academic objectivity. Overall, action research provides a synergistic solution to link organisational needs and academic research desires. We conclude that action research is a viable and desirable alternative to more traditional research approaches for a multitude of contemporary organisational situations and urge extended use and reporting in the literature. 7 - REFERENCES Argyris, C., Putnam, R. and Smith, D.M. (1985). Action Science - Concepts, Methods, and Skills for Research and Intervention, San Francisco: JosseyBass. Benbasat, I.G., Goldstein, D.K. and Mead, M. (1987). The Case Research Strategy in Studies of Information Systems, Management Information Systems Quarterly, 11, 3, 369-386. Checkland, P. (1981). Systems Thinking, Systems Practice, New York: John Wiley & Sons. Checkland, P. (1991). From Framework through Experience to Learning: The Essential Nature of Action Research, in: Nissen, H.-E., Klein, H. and Hirschheim, R. (eds.) Information Systems Research: Contemporary Approaches and Emergent Traditions, North Holland: Elsevier Publishers, 397-403. Galliers, R.D. (1991). Choosing Information Systems Research Approaches, in: Galliers, R.D. (Ed) (1992) Information Systems Research: Issues, Methods and Practical Guidelines, Henley-on-Thames: Alfred Waller, 144-162. Galliers, R.D. and Land, F.F. (1987). Choosing Appropriate Information Systems Research Methodologies, Communications of the ACM, 30, 11, 900-902. Grohowski, R., McGoff, C. Vogel, D.R., Martz, B. and Nunamaker, J.F. (1990). Implementing Electronic Meeting Systems at IBM: Lessons Learned and Success Factors, Management Information Systems Quarterly, 14, 4, 369383. 15 Jessup, L.M., Connolly, T. and Galegher, J. (1990). The Effects of Anonymity on GDSS Group Process With an Idea-Generating Task, Management Information Systems Quarterly, 14, 3, 313-321. Jones, N. and Miller, E. (1997). Using Technology for Stakeholder Consultation: The World Bank's Use of GroupSystems V in Developing Countries, Proceedings of the 30th Hawaii International Conference on System Sciences, II, 464-468. Kemmis, S. (1980). Action Research in Retrospect and Prospect, paper presented to the annual meeting of the Australian Association for Research in Education, Sydney. Martz, W.B., Vogel, D.R. and Nunamaker, J.F., (1992). Electronic Meeting Systems: Results from the Field, Decision Support Systems, 8, 2, 141-158. Orlikowski, W.J. and Baroudi, J.J. (1991). Studying Information Technology in Organisations: Research Approaches and Assumptions, Information Systems Research, 2, 1, 1-28. Pervan, G.P. (1994a). The Measurement of GSS Effectiveness: A Meta-Analysis of the Literature and Recommendations for Future GSS Research, Proceedings of the 27th Hawaii International Conference on Systems Science, IV, 562-571. Pervan, G.P. (1994b). A Case for More Case Study Research in Group Support Systems, TC8 AUS IFIP Conference, Bond University, Gold Coast, Qld, May 8-11, 485-496. Pinsonneault, A. and Kraemer, K.L. (1990). The Effects of Electronic Meetings on Group Processes and Outcomes: An Assessment of the Empirical Research, European Journal of Operational Research, 46, 2, 143-161. Rao, V.S. and Jarvenpaa, S.L. (1991). Computer Support of Groups: Theory Based Models for GDSS Research, Management Science, 37, 10, 1347-1362. Vogel, D.R., Martz, W.N., Nunamaker, J.F., Grohowski, R.B. and McGoff, C. (1990). Electronic Meeting System Experience at IBM, Journal of Management Information Systems, 6, 3, 25-43. Vreede, G.J. de (1995). Facilitating Organisational Change: The Participative Application of Dynamic Modelling, published PhD Dissertation, Delft University of Technology. 16 Walsh, K. and de Vreede, G.J. (1997). Generating Insights for Reengineering, 30th Hawaii International Conference on System Sciences, II, 210-218. Wee, C.H., Lee, K.S. and Hidajat, B.W. (1991). Sun Tzu: War and Management, Singapore: Addison-Wesley. Zuber-Skerrit, O. (1991). Action Research for Change and Development, Aldershot: Gower Publishing. 17