Action Learning Sets: the case for running them online

Len Bird

Work-Based Learning Unit

Coventry Business School



This paper develops an argument for using asynchronous online computer mediated conferencing (CMC) to facilitate the action learning set process. It starts with a definition of action learning and then traces the roots of action learning back to the philosophical position of social constructionism. From this analysis the paper establishes that the key feature of action learning is the promotion of reflexivity and the social construction of knowledge through language. The paper then examines the literature on online communication to answer the question, "can an action learning set community be created via a text-only CMC?" Having examined competing responses to this question the paper concludes by considering the main challenge to the argument, (i.e the loss of co-presence) and the role of the online facilitator as an ‘online discourse analyst’.


Asynchronous CMC, action learning, social constructionism, discourse analysis.


It is not difficult to see why universities are enthusiastic about developing online action learning. Online learning offers the prospect of delivering action learning courses to student groups who were traditionally excluded. If students don’t need to meet face-to-face then the geographical limitation on student recruitment is lifted. Courses can be ‘scaled-up’ without the constraints imposed by the need to assemble the learners for set meetings. However, we need to be cautious: is this demand for efficiency driving pedagogy? Is the opportunity too good to miss, such that, in the rush to increase student numbers, the essence of the complex human learning process of the set is lost? In this paper I intend to address these questions by considering a theoretical case for running the action learning set process within an asynchronous online learning community.

My argument requires an examination of the roots of the action learning set process from a social constructionist perspective and an investigation of the learning dynamic in action. I shall identify the key social ingredients that make up the rich and complex dynamic that constitutes an effective set. The argument concludes by looking at the literature on the nature and significance of the printed word – as opposed to the spoken word - to support my proposition that these key ingredients can be replicated online, in a manner that can be considered at least equal to the face-to-face version.

To begin I need to identify the concepts involved by looking at how action learning is understood and defined. This will lead me to the examination of what actually goes on in an action learning set viewed from a social constructionist viewpoint.


Raelin (1997) states that action learning is based on the straightforward pedagogical notion that people learn most effectively when working on real-time problems occurring in their own work setting. Raelin introduces the learning equation, first drafted by Revans, which helps operationalise the process.

"In Reg Revans’ original conceptualisation, learning results from the independent contributions of programmed instruction (designated P) and spontaneous questioning (designated Q). P constitutes information and skill derived from material already formulated, digested and presented, typically through coursework. Q is knowledge and skill gained by apposite questioning, investigation and experiementation."

(Raelin, 1997: 22)

The Learning Equation Learning, L = P + Q

Mike Pedler (1996: 20) cites a definition adapted from Aspinwall (Pedler & Aspinwall, 1996) in which Aspinwall states that action learning is the best way to educate managers. It is based on the premise that: "There can be no learning without action and no (sober and deliberate) action without learning"(Revans, 1983: 54). Aspinwall explains that action learning focuses on the facilitation of Q. Programmed knowledge can be helpful but it should only be sought after careful reflection on what knowledge is needed and why. Action learning sets create the arena for Q to flourish.

Koo (1999: 89) has an extensive account of the essence of action learning. Koo quotes a number of sources of Revans’ work. In particular a quote cited by Smith (1997: 366) nicely encapsulates the action learning process.

"Action learning is a means of development, intellectual, emotional or physical, that requires its subject, through responsible involvement in some real, complex and stressful problem, to achieve intended change sufficient to improve his observable behaviour henceforth in the problem field."

(Revans, 1979: 4)

Bourner, Beaty, Lawson and O’Hara (1996: 32) view action learning as a process of reflection and action, aimed at improving the effectiveness of action. The purpose is to learn from experience and to act more wisely. The process can be used wherever a group can support the learning of others in a context where experience can be reflected on. McGill and Beaty (2001) take up the importance of reflection in the process in their description:

"Action learning is a continuous process of learning and reflection, supported by colleagues, with an intention of getting things done. Through action learning individuals learn with and from each other by working on real problems and reflecting on their own experiences."

(McGill and Beaty, 2001: 11)

Peter Cusins (1996: 34) puts forward a useful model of the action learning process. He proposes an understanding of action learning as a ‘syndrome’ of four activities: experiential learning, creative problem solving, acquisition of relevant knowledge, and co-learner group support. By ‘syndrome’ Cusins means "a number of things that flow together". Each activity can be thought of as necessary but insufficient on its own. The implications are that the process is only effective when each of the elemental activities are performed effectively. When this happens then the action learning process becomes "holistically synergistic", i.e. the effectiveness of the whole activity becomes greater then the sum of each of the component activities.

The action learning set facilitator is regarded as pivotal by many authors (Casey, 1991; McGill & Beaty, 2001; Mumford, 1993; O’Neil, 1996). The function of the facilitator depends on the design of the action learning programme and the composition of the set membership. The set facilitator’s main task is to maintain the set process and model set behaviour.

As Revans (1983) has stated repeatedly, there are no universal recipes for action learning. My own synthesis of the process (drawing heavily on the definition by Aspinwall, cited by Pedler ,1996: 20) would be along the following lines.

The action learning process provides opportunities for facilitated shared reflection on individual perceptions of personally engaging, hitherto intractable problems, in order to clarify and render them more manageable, and to facilitate the creation and iterative exploration of alternative actions in the light of new insights. The learning set provides a balance of emotional support and intellectual challenge through comradeship and insightful questioning which enables each member to act and learn effectively on three levels: i.e.

about the problem being tackled;

about what is being learned about oneself; and

about the process of learning itself, i.e. ‘learning to learn’.

This definition, with its focus on public shared reflection and learning on three levels, forms the platform on which my argument begins. In my view, action learning sets form a unique type of learning community. Members come together in a voluntary and supportive way and form a ‘social contract’ to co-operate and help each other learn. The contract is often partly enshrined in the set’s ground rules. These rules – often negotiated at the set’s first meeting – form the social protocol by which the group will function. The set then has an explicit set of guidelines that controls its transactions and the terms of engagement. The ground rules represent a collective supportive framework for promoting joint learning in a fair and equitable manner.

The action learning set process is also unique by virtue of the ‘entry theory’ used by members to formulate their initial problem. By entry theory I mean the personal theoretical assumptions that underpin the structure of the problem to be addressed over the duration of the set meetings. In order to formulate the problem, learners must have a view on the nature of the task, the route to be used to address the problem and the possible shape of the solution. In essence, it can be said they have a pre-existing theoretical position that guides their initial direction and actions. As the set progresses the learner’s initial position and personal assumptions are challenged and modified through apposite questioning and public reflection.

With this starting point established I now wish to look beneath the surface of the action learning set process and uncover the social constructionist foundations of the human dynamic involved.


My definition talks of ‘shared reflection on individual perceptions of problems’. It is this ‘shared’, social dimension of the reflection process that I wish to explore further. My aim is to make a link between 'shared reflection' and a view of action learning as the 'social construction of knowledge'. Several authors have made the connection without a deep examination of the philosophical position that underpins the argument. I propose to summarise some accounts of this position and then dig deeper into the social constructionist background.

The clearest account of the link is made by Pedler (1997: 251). Whilst countering one of the criticisms of action learning - that it is too centred on the individual’s problems and the need for individuals to change - Pedler argues that by considering the action learning set process as the social construction of knowledge, it is not the individual’s problems that are important to learning but the collective relationship of the set and the wider context. By this he means that the set’s deliberations can be viewed as a collective construction of new social meanings and realities. The apposite questioning, the discussion and the public reflection lead to new thoughts, ideas and perspectives being shared collectively by the set in a communal way. Pedler suggests that the set dynamic represents, "a collective search where everyone contributes their problems and insights to achieve a shared understanding." (1997: 258) Viewed in this way the action learning set process offers a perspective in which members participate in a shared process of meaning making within the set community, so that frameworks of understanding on which to base future actions are generated. In short, knowledge for action is constructed.

Other authors have given support to the social constructionist view of action learning and the sense of community developed in the set. Raelin (2000: 101) talks of the power of public reflection. " Through public reflection we may create a collective identity as a community of inquiry". He goes on to discuss the ‘critical consciousness' developed through public reflection, and the influence of the social context in which reflection takes place. Similarly Stephen Kemmis (1985), in discussing the politics of reflection, argues that reflection is not purely something that goes on in the head, but is a social process.

Thus a link can be found between reflection and the social constructionist view that knowledge is socially constructed within a community. From this position I can now begin to argue that the three levels of learning, included in my definition above, are brought about by the social interactions between set members. To continue the argument I need to make a brief excursion into the world of social constructionism. My purpose is to contextualise and reinforce the view of the AL set process I have just put forward, and to help explain further the concept of ‘situated learning’, which I feel is at the heart of action learning.


Social constructionism is a theoretical orientation concerning a postmodernist approach to studying human beings as social animals. Burr (1995: 3) identifies the essence of the social constructionist agenda. He describes four key canons. Firstly, it is in opposition to positivism and empiricism. Social constructionists challenge the view that observations of the world unproblematically yield its nature. Secondly, social constructionism invites us to consider the historical and cultural heritage of the ways in which we commonly understand the world. The way we see the world depends upon where and when we live. The state and form of knowledge is continually changing as new perspectives emerge and provide a new dominant orthodoxy. The third key belief is that knowledge is sustained by social processes. The social constructionists view of our knowledge of the world is that people construct knowledge between them in the goings-on of their everyday lives. Burr’s fourth cornerstone of social constructionism is that knowledge and social action go together. By this, Burr means that our socially constructed understanding of the world brings with it a corresponding action. The way we understand a phenomenon yields a corresponding social action that reflects and reinforces our viewpoint. Each different construction brings with it a different kind of action.

Kenneth Gergen (1999) in his influential book, "An Invitation to Social Constructionism", develops a convincing picture of social constructionist views and identifies the importance placed on narrative and discourse. Like Burr, Gergen (1999: 46) also describes the working assumptions underpinning social constructionism under four headings. Firstly, he suggests that for any set of circumstances there are an unlimited number of descriptions or explanations. The way we see a situation depends on our view of the world and the language we use to describe it. Secondly, Gergen argues that the meaning of the language and conversations we have about problems stem from the relationships in which the words are exchanged. From a social constructivists view, language is given meaning by relationships. Words do not exist in a vacuum: their meaning springs from their cultural use and from the actions of those with which we interact. Gergen’s third assumption concerns generative discourses. By this Gergen means that, just as the meanings of words are heavily laden with the tradition and history of their past usage, so their use in present relationships with others can generate narratives and stories that become our future. Gergen’s fourth pillar stresses the need for critical reflection in life. In trying to understand the world we must be careful about reaching the ‘right’ conclusions. Every ‘right’ or ‘moral’ position is a position seen from within a particular tradition or cultural framework: we need to show reflexivity in our considerations, which means to constantly challenge our positions and to suspend the obvious and listen to alternative constructions of reality.

In differentiating social constructionism from traditional psychology Burr (1995: 6) stresses Gergen’s view on the importance of language. He points out that our understanding of the world is not based on objective reality but is obtained from interaction with other people. We are born into a culture where people exist and hold views and perspectives on the world. As we grow up we grow within this ‘culture’ and are exposed to it through language. Language is the ‘oxygen’ of our knowledge growth. Thus, in order to create our interpretation of reality we must have language (Gergen 1999: 20). There is no thinking without language. If you don’t have the vocabulary then you cannot think about an issue. For Burr, language is more than a simple way of expressing ourselves, "When people talk to each other the world gets constructed" (1995:7). Traditional psychology views language as the passive vehicle for thoughts and emotions; social constructionists view language as a form of action, action that can be reflected on like any ‘concrete’ experience.

In summary, it can now be appreciated how social constructionism provides a paradigm of understanding that gives philosophical credibility and substance to the views put forward by various authors above, that the AL set process facilitates the social construction of knowledge and meaning through reflection and the use of language.

This philosophical view of the world has been encapsulated in the theory of ‘situated learning’ (Lave and Wenger, 1991). They argue that most accounts of learning ignore the basic social nature of the learning process. They propose that learning is a process of ‘participation’ in communities of practice that embody certain beliefs, behaviours and intersubjectively held knowledge; with the participation at first legitimately peripheral but gradually increasing in engagement and complexity. This decentering of the learning process – away from the individual and towards the community - moves the focus of the learning away from pedagogy and the skills of the teacher to the efficacy of the community’s learning resources. Thus, the problem for the learning set facilitator is not one of pedagogy – how best to teach – but one of how the learning community involved can be replicated within the confines of the set’s online social interactions. This is the issue I need to address now in order to isolate the key 'social interaction' ingredients that must be replicated if online AL sets are to work effectively. I then examine the literature to assess if the key ingredients can be replicated online.


I have explained how social interaction within communities is the key learning process within situated learning, and how social constructionism places language, dialogue and narrative at the heart of all reality construction via social interaction. Thus, for an action learning set to function online a situation needs to be created in which facilitated, shared reflection and the social construction of knowledge can proceed in a text format. The virtual medium must allow the apposite questioning, discussion, and emotional support that leads to new thoughts , ideas and wider perspectives being shared by the set in a communal way. In short, a socially constructed knowledge must be created through the key ingredient of language.

In order to convince you that this proposition is a practical one I must show that the use of text-based language can be as powerful as face-to-face language. I must show that the rich mixture of emotional and intellectual support, provided by a face-to-face set, can be replicated in a virtual environment. To do this I turn to the literature on online communities, and, in particular, those authors who comment on language and conversation as key aspects of learning. Following this, to conclude the paper, I consider the main challenge to my argument, and the role of the online tutor.


Many authors have examined the contrast between face-to-face communication and virtual text-only communication. A number of issues emerge from this literature which are relevant to my argument.

Asensio, Hodgson and Trehan (2000) challenge those who advocate the view that online communication is a second best to face-to-face. They draw on Derrida’s (1967) work concerning the relative status of speech and writing. Derrida considers writing not just as the script form of language – and thus second best – but as a rich means of conveying a considered view of the world reality as we perceive it. Thus, the written word is much more meaningful and capable of expressing emotion and feeling than is the spoken word. Asensio et al found that the text-only format of a discussion forum caused students to take much more care over their contributions in ensuring that their emotions and meanings were displayed in the right way. Clearly this is of importance in an AL set where social and emotional support is crucial to the process of constructiong reality through language alone.

Gergen (1999: 26) points out that many scholars have extended the implications of Derrida’s writings on linguistic ‘deconstruction’ in many directions. The Deconstructionist’s view is that all rationality involves a massive suppression of meaning. The binary nature of meaning, as represented in words, suggests that any rational course of action is based on the suppression of a contrasting viewpoint. Gergen explains Derrida’s theory:

"…the distinctiveness of words depends on a simple split between ‘the word’ and ‘not the word’. The meaning of ‘white’, then, depends on differentiating it from what is ‘not white’ (or black for instance). Word meaning depends, then, on differentiating a presence and an absence, that which is designated by the word against what is not designated."

(Gergen 1999: 27)

Thus, in order to deconstruct rationality, as understood by the individual, requires extensive critical reflection and reflexivity to challenge taken-for-granted assumptions. I have indicated above that AL set members come to a set with an ‘entry theory’ or ‘rational approach’ to tackling their problem based on their initial theoretical position. In order to deconstruct this entry theory and make personal progress takes time for critical reflection and pondering. An asynchronous CMC allows for this time; a point highlighted by a number of authors (Wegerif, 1998:43; Harisim, 1989:60; Doherty, 1998:online).

The differences between online communication and face-to-face communication have been identified by Graddol (1989; 239) and McConnell (1994: 73) who both suggest that CMC supports a more ethical mode of communication. On this point the German theorist Jurgen Habamas (1990) describes a prescription for ‘discourse ethics’ that provides an ethical foundation for productive dialogues to resolve conflicts. He advocates the elimination of all barriers so that all can participate equally without power differentials. Habamas also suggests that by allowing any participant to bring any viewpoint to the discussion a consensus can be agreed in which everyone’s interests are satisfied. What Habamas’ ideal speech situation portrays is a situation similar to an action learning set governed by the set’s ground rules. These rules I have previously described as, ‘the social protocol by which the set functions’, and as such resonate with Habamas’ recommended ethics-governed exchange process. It could well be that a CMC action learning set is the only forum in which this type of protocol could exist. In a face-to-face encounter the power differentials evident in such things as turn taking, rules and gender stereotypes etc impinge on the egalitarian nature of the discussion (Harisim, 1989).

Other authors have commented on the power of online text communication in various circumstances. Cherney (1999) provides an in-depth treatise on the nature of the research questions raised by online communities. In explaining that online chat can be dysfunctional as well as constructive she makes the comment that, "conversation online can also provide all of the good things we expect from conversations face-to-face," (1999: 2). Winiecki (1999) also argues that online asynchronous discussions can be as effective as face-to-face. In examining the linguistic processes in action during conversations, and how to replicate them online, Winiecki proposes a rationale for adapting conversational practice that compensates for any online shortcomings. McLellan (1997: 186) describes Schrage's (1991) model which stresses the need to generate mutual respect, tolerance and trust in online groups. Gunawardena, Lowe and Anderson (1997: 424) conclude their comprehensive analysis of an online debate by demonstrating that CMC did facilitate the co-construction of knowledge between international contributors.

‘Engagement Theory’ has a great deal to offer in the understanding of good online CMC design. Hodgson and Asensio (2001) examined an example of the use of videoconferencing for management learning that shed a great deal of light on the ‘need for co-presence’ debate. The student reaction to the course was that the approach was very much second best to face-to-face teaching. The lack of physical presence of the tutor was the main issue for students. However, upon examination of the courses Hodgson and Asensio found that the nature of the ‘engagement’ (Kearsley and Shneiderman, 1999) that was set up in the course was far more important than the loss of physical presence. Engagement Theory, according to Kearsley and Shneiderman, states that for a learning experience to be successful students must be meaningfully engaged in learning activities through interaction with others and worthwhile tasks, (very similar constructs to those used in action learning). The case studies that Hodgson and Asensio examined paid little attention to the student’s engagement in the design of the experience, and it was this that gave rise to the student’s criticism of the course rather than the simple explanation of the loss of co-presence. Engagement Theory, with its roots in situated learning and social constructionism offers a useful framework for designing how technology can facilitate learning communities through the appropriate use of language.


The debate, just mentioned, concerning the loss of co-presence is at the heart of the opposition to my argument and thus deserves further comment.

To the social constructionist it is the nature of the whole narrative that a person uses to describe their world that is important, not the minor nuances that body language or social cues contribute to the story. As Derrida maintains, text is more than a record of speech. It is not an equivalent to co-present conversation; it has a far richer and deeper significance which reflects our perceptions and comprehension of the world. It conveys our considered and internalised view of world reality.

Trust, warmth, openness and emotional ‘feeling’ towards another individual are integral parts of the action learning set process. A smile, a frown, a hesitation or a nervous scratch can all convey meaning in a face-to-face scenario. They add to the communication process and perhaps lead the conversation in a new direction when detected by the skilled listener. But do they help generate trust, warmth and openness? What really matters in developing an interpersonal relationship between two human beings? The answer, I would argue, lies in personal disclosure. In order to develop social intimacy with someone you need to ‘give of yourself’. By this I mean you need to share confidences and private hopes, dreams and desires, in a way that demonstrates that you trust the receiver. If one never gets beyond small talk then a meaningful relationship will not develop. I do not mean that online communicators need to pour out their innermost problems and private issues in order to create a trusting community, but they do need to share their personal theoretical positions with their co-learners in an open and honest fashion. Disclosure like this is just as possible online as it is face-to-face but may take more time (Walther, 1996). It could be argued that an asynchronous CMC provides for this type of disclosure much more easily than a co-present exchange. Indeed, Asensio, Hodgson and Trehan (op cit) found that students took more care over the emotional level of their disclosures in an online format.

A further counter argument to the primacy of co-presence concerns the role of the online facilitator. An online facilitator cannot detect the smiles, frowns or hesitations of the face-to-face set members; however, she can deconstruct and analyse the words used by online participants. This role is one of an ‘online discourse analyst’, who, using a social constructionist perspective, can unpack the associated metaphors and history of the word usage displayed. As Gergen (op cit) points out, the meanings of words are heavily laden with the tradition and history of their past usage and are given rich interpretation by the nature of the relationship and context in which they are used. Narratives and stories can be revealed by a critical ‘reading’ of what people write. This hermeneutical interpretation is only possible by a considered reflection and contemplation of the words used. An asynchronous CMC allows the online discourse analyst the time to uncover these meanings and respond accordingly. The role changes from that of a skilled, face-to-face, listener, to one of a more reflective online discourse analyser. The requirements are different but the end result is just as rich in meaning and substance.


Is the case conclusive? I think not. The literature I have examined shows that language, dialogue and narrative play a major part in the learning process and that the online medium can replicate the conditions that allows these issues to converge in the social construction of knowledge. I have tried to show that face-to face communication is not the only meaningful way to create knowledge and meaning, and many authors feel that the alternative – text based communication – can in fact be superior in creating the type of learning communities which are at the heart of action learning. The deconstructionist process outlined by Derrida requires the extensive reflexivity that asynchronous CMC provides, and this goes some way in compensating for the loss of co-presence. For learners to shed their ‘entry theory’ and move on requires time; time for reflection and time for the online discourse analyst to uncover the narratives and stories underpinning student’s views. Habamas’ ideal speech conditions may only be capable of replication in an online format and the principles of Engagement Theory can help guide the online set facilitator in encouraging a meaningful use of language. All-in-all, online action learning sets appear well worth striving for and the concept is well supported in the literature.


Asensio, M., Hodgson, V., & Trehan, K., (2000) ‘Is there a difference?: Contrasting experiences of face to face and online learning.’ Proceedings of the 2nd International Conference, Network Learning, Edited by M. Asensio, V. Hodgson, J. Foster and D. McConnell, April 2000, 12-19.

Bourner, T., Beaty, L., Lawson, J., & O'Hara, S., (1996) ‘Action learning comes of age: questioning action learning’, Education + Training, 38(8), 32-35.

Burr, V., (1995) An Introduction to Social Constructionism, London: Routeledge.

Casey, D., (1991) ‘The role of the set advisor’, in Pedler, M., (ed) Action Learning in Practice, Brookfield, VT: Gower.

Cherney, L., (1999) Conversation and Community - Chat in a virtual world. Stanford: CSLI Publications.

Cusins, P., (1996) ‘Action learning revisited’, The Journal of Workplace Learning,. 8 (6), 19-26.

Derrida, J., (1967) De la Grammatologie, Paris: Minuit. Translated by Chakravorty Spivak, G (1967) Of Grammatology: The John Hopkins University Press.

Doherty, P., (1998) ‘Learner control in asynchronous learning environments’, ALN Magazine, 2 (2), online at

Gergen, K., (1999) An Invitation to Social Constructionism, London: Sage.

Graddol, D., (1989) ‘Some CMC discourse properties and their educational significance,’ in Mason, R., & Kaye, A., (1989) Mindweave: Computers, Communication and Distance Education, Oxford: Pergamon, 236-241.

Gunawardena, C., Lowe, C., & Anderson, T., (1997) ‘Analysis of a global online debate and the development of an interactional analysis model for examining social construction of knowledge in computer conferencing’, Educational Computing Research, 17(4), 397-431.

Habamas, J., (1990) Moral Consciousness and Communicative Action, Cambridge: Polity Press.

Harasim, L., (1989) ‘On-line Education: a new domain’, From Mason, R., & Kaye, A., (eds) (1989) Mindweave: Communication, Computers and Distance Education, Oxford: Pergamon, 50-62.

Hodgson, V., & Asensio, M., (2001) ‘ Exploring the implications of videoconferencing for management learning’, paper presented to the EDINEB Conference, June 20-22, 2001, Nice: France.

Kearsley, G., & Shneiderman, B., (1999) ‘Engagement Theory: A framework for technology-based teaching and learning’, online at Viewed September 2001

Kemmis, S., (1985) ‘Action research and the politics of reflection’ in Boud, D., Keogh, R., & Walker, D., (eds) Reflection: Turning Experience into Learning, London: Kogan Page, 139-163.

Koo, L., (1999), ‘Learning action learning’, Journal of Workplace Learning, 11(3), 89-94.

Lave, J., & Wenger, E., (1991) Situated Learning: Legitimate peripheral participation. Cambridge: Cambridge University Press.

McConnell, D., (1994) Implementing Computer Supported Co-operative Learning, London: Kogan Page.

McGill, I., & Beaty, L. (2001) Action Learning: A guide for professional , management and educational development, revised second edition, London: Kogan Page.

McLellan, H., (1997) 'Creating virtual learning communities via the web', in Khan, B., (ed) Web-based Instruction, NJ: Educational Technology Publications, 185-190.

Mumford, A., (1993) How Managers Can Develop Managers, Brookfield, VT: Gower.

O’Neil, J., (1996) ‘A study of the role of learning advisor in action learning’, Employee Couselling Today, 8(6), 39-44.

Pedler, M., (1996) Action Learning For Managers, London: Lemos & Crane.

Pedler, M., & Aspinwall, K., (1996) Perfect Plc: The purpose and practice of organisational learning, Maidenhead: McGraw-Hill.

Pedler, M., (1997) ‘Interpreting Action Learning’, in Management Learning: Integrating perspectives in theory and practice, J, Burgoyne & M, Reynolds (eds), London: Sage, 248-264.

Raelin, J., (1997) ‘Action learning and action science: are they different?’ Organizational Dynamics, 26(1), 21-34.

Raelin, J., (2000) Work-Based Learning: The New Frontier of Management Development. NJ: Prentice Hall

Revans, R., (1979) ‘The nature of action learning’, Management Education and Development, 10, 3-23.

Revans, R., (1983) ABC of Action Learning, Middlesex: Chartwell-Bratt Ltd.

Schrage, M., (1991) Shared minds: The new technologies of collaboration, New York: Random House.

Smith, P,(1997) ‘ "Q"ing action learning: more on minding our Ps and Qs’, Management Decision, 35(5), 365-372.

Walther, J., (1996) ‘Computer-mediated communication: impersonal, interpersonal and hyperpersonal interaction’, Communication Research, 23(1), 3-43

Wegerif, R., (1998) ‘ The social dimension of asynchronous learning networks’ Journal of Asynchronous Learning Networks, 2(1), 34-49.

Winiecki, D., (1999) ‘Keeping the thread: adapting conversational practice to help distance students and instructors manage discussions in an asynchronous learning network’. DEOSNEWS, 9(2). (Online at , 12.2 1999)