pgar.doc

Postgraduate action research: important
choices

Bob Dick

bd@uq.net.au

A draft version of a chapter prepared for Zuber-Skerritt, O. (2000) Action learning, action research and process management: theory, practice, praxis. Brisbane: Action Research Unit, Faculty of Education, Griffith University.

Action research, and its cousin action learning, often make good choices for postgraduate theses and dissertations. In this brief chapter I set out some of the advantages and disadvantages. I describe some of the choices faced by candidates who are considering the use of action research for their thesis or dissertation.
In doing so I’ll talk mostly about action research. For most candidates that will be the component most evident in the thesis or dissertation itself. Action learning may also be used as part of the thesis methodology. Its most important contribution, however, is its use to create a support network: see below.

It is not my intention to be evangelistic either about action research or about any of its many different varieties. To my mind, good research is designed to fit the research situation, the research question, and the interests and skills of those involved. A good choice will sometimes be mainstream action research. Sometimes it will be an extension or modification of action research. Sometimes it will be something else entirely.

Unlike some of my colleagues I assume that most aspects of research design are open to choice. For example I don’t personally think action research must be participative, or qualitative, or published. It often is; and I have no difficulty with that. Many writers argue for one or more those characteristics. I don’t personally believe that’s a good reason to limit my own research choices. You will make up your own mind.

If action research is not necessarily any of those, what is it? Broadly speaking, I think of it as a family of research methodologies which pursue the dual outcomes of action and research. In that sense it is true to label. This is one of the reasons that it is particularly suitable for postgraduates who wish to improve their own practice while they pursue a postgraduate qualification.

I do think action research profits from the use of a cyclic or spiral process in which action alternates with critical reflection. I regard its spiral process and its pursuit of both action and research as its defining characteristics. Do those characteristics suit what you plan to do? Then action research may be a good choice for you.

As you can see, I subscribe to the design principle that form follows function. First you decide what you want to do. Then you decide how best to do it. However, that is a choice to be made.

Action research for theses and dissertations

As I said above, I think of action research as a family of methodologies, each of which simultaneously pursues action and research. The action takes the form of change or improvement or implementation. The research consists of learning and understanding, often leading to publication.

Both sets of outcomes are likely to be important to a higher degree candidate. For theses and dissertations the research component is obligatory. For some candidates it is useful to give emphasis to the action component too. Here, a choice between theory-driven and data-driven research is relevant.

Choice 1: Theory driven or data driven?

I have come to the conclusion that the distinction between qualitative and quantitative research is less important than that between theory-driven and data-driven research. I think it is a paradigmatic choice, in many of the ways in which Kuhn (1970) and others use that term.

To make this distinction more evident, let me overstate it a little:

Do you wish to do research which takes the body of prior literature as its starting point, deriving its contribution to knowledge by assuming that literature as a given and extending or refining it?

This is theory-driven research. Those who prefer it may do better to choose some approach other than action research. I think that most current quantitative research and much qualitative research takes this form.

Or do you wish to deal with the research situation and the people as they are, as far as possible putting aside your preconceptions so that you are more open to experiencing them fully?

This is data-driven research. It is responsive to the situation, and flexible.

The form of action research that most interests me takes this second form. I argue below that it offers particular advantages to the higher degree candidate who wishes to research her own practice, improve her research situation, and add to her qualifications.

Let me be clear here. I am not saying that all action research is data-driven. Sagor (2000), for instance, clearly acknowledges that his approach is more quasi-experimental than anything else. Nor am I saying that only action research is data-driven. Glaser in particular describes a strongly data-driven form of grounded theory in his writing (for example 1992, 1998). Heuristic inquiry (Moustakas, 1990) and phenomenology (see Crotty, 1998), among other methodologies, attempt to be sensitive to the data and to minimise preconceptions.

For a higher degree candidate this is an important choice. On the one hand it does carry some risk. On the other hand it does offer advantages because of its dual goals of action and research.

The risk arises in three main forms, I think.

The candidate may well have prior training in research and this may have given her some expectation which will mislead her as she goes about her thesis work.

The way in which universities structure higher degrees often assume that good research is theory-driven. For instance, you may be expected to spend your first year perusing the literature. You may be required to write your summary of the literature before your candidature is confirmed.

Some examiners may be surprised by data-driven research because it doesn’t fit their mental image of higher degree research. This may include some examiners who regard themselves as sympathetic to and supportive of such approaches as action research.

One consequence of these risks is that the person choosing a data-driven approach has to be prepared to argue the case for her choice very clearly and cogently. Those who choose this approach because they imagine it is easier are mistaken. These are not risks to be undertaken lightly.

Despite these risks, the advantages of data-driven action research may sometimes be very attractive. I think this is particularly true when the three possible outcomes of data-driven action research are all important to you.

In the course of doing the research, you are able to study and improve your own practice. In effect you use the higher degree as valuable personal and professional development. You and other candidates can also form a learning set for mutual support. Personal and professional development will then be further enhanced.

One of the purposes of action research is to bring about improvement in the research situation. If you research your own practice, it is very likely you will also improve your work situation. The action outcomes you achieve are likely to be of benefit to your organisation or community.

And of course, at the end of it, you gain a qualification.

I have recently co-supervised a number of practitioners who, by the end of their higher degree, had achieved all three of these outcomes. It is also worth noting that all of them worked full time in demanding jobs. Yet most completed their degree in approximately the same time as do most candidates who are enrolled full time in their degree program.

The implications of data-driven research

It may appear from what I have said above that this is an easy choice for candidates who find the three advantages appealing. Let me therefore depict in more detail two of the features of this approach which may surprise some of you (and perhaps some supervisors, committee members and examiners).

First, if you are using a flexible and responsive approach you probably can not know where the data will lead you. Choice of literature is therefore difficult. At the start of a study you may now know what literature will later come to be relevant. It makes sense to postpone reading until the relevance of literature is more easily judged. In turn, this may have consequences for the structure of the eventual thesis.

Second, it is not just the interpretation &endash; the understanding &endash; which emerges slowly from the situation. So does the methodology. You can begin action research by asking initially-fuzzy questions using initially-fuzzy methods, thereby gaining initially-fuzzy answers. You may then use those intially-fuzzy answers to refine your methods as you proceed. To say it differently, research content and research process both develop as the research proceeds.

Checkland and Holwell (1998) add a further component. They describe research as a framework F, operationalised as a methodology M, focussed on an area of concern A. In the course of the research one learns about and therefore modifies F, M and A.

I think they are talking about all research. And I think they intend their framework F to be an epistemological framework, consisting for example of assumptions about what counts as knowledge. I look forward to the day when all researchers will be expected to explain and justify their epistemology, their methodology and methods, and their conclusions.

I think it is evident that such an approach encourages researchers to be more mindful as they go about their research. (I take the adjective "mindful" from Bentz and Shapiro, 1998.) This brings us to the second choice. How is the candidate to engage with the research methodology and with the supervisor or committee.

Choice 2: Technician or performing artist?

I might as easily have taken this as the first choice, for it can apply to all forms of thesis and dissertation research. You might describe it as an existential choice.

I can phrase the choice in the following form.

Do you wish to be an apprentice who will learn thoroughly a particular approach to research from the supervisor, committee, and literature? That is, will your learning be primarily propositional? At the conclusion of such a research program you can expect to know how to do one form of research. To overstate the situation, this is research by recipe.

Or do you expect to engage in research with whatever resources and understanding you can bring to bear, learning from your experience? That is, will your learning be primarily through questioning inquiry, with supervisor and committee functioning as mentors rather than as teachers? Such an approach will engage you in examining your assumptions about the nature of knowledge, and of methodologies. This is research as performing art.

You will recognise the influence of Checkland and Holwell (1998) and also of Revans (1983) in the phrasing of this choice.

Action research lends itself to the second of these options. It alternates action with critical reflection. That critical reflection can be about the data and the interpretations emerging from it. It can also critique and improve the methodology. Beyond that, it may be used as an opportunity to examine the assumptions about knowledge which inform the research design.

There appear to be enticing parallels between action research and experimental research. (I was trained as an experimental psychologist.) It seems to me that good researchers refine their methodology and methods from study to study. This is a process which resembles the spiral nature of action research. It applies to a study at a time, whereas in action research each turn of the spiral is an opportunity for learning and change.

If I were embarking on a study with uncertain outcomes I think I would begin with action research. It would allow me to build enough understanding to decide which methodology and methods best suited my research situation and my research question. Whatever methodology I chose, I could still use action research as a "meta-methodology". I could thus enhance my learning about whatever methodology I was using.

I expect it is by now apparent that action research offers substantial flexibility and responsiveness to the situation. It is not for the faint-hearted. But it does offer a manager or a professional an opportunity to achieve some important practical outcomes. If you choose action research you can choose to embark immediately on action. This brings us to a third choice.

Choice 3: "Action research" or "action research"?

As I have said, action research is action and research. One or the other of these, however, may be emphasised.

Is your main intention to bring about change, with research outcomes as a desired but sacrificable bonus? Will this be action-oriented research?

Or is it your desire above all to do good research, with change as a wished-for but not essential outcome? Will this be research-oriented research?

I would not want you to think that action research must be one or the other. Action and research fit each other well. All else being equal, better understanding allows more effective change. For the most part, trying to change a system will give you a better understanding of it. As choices, action and research are emphases, not absolutes.

In any event, for thesis purposes you have a more limited choice. Almost certainly, you need research outcomes. So your choice becomes one of deciding how much change you are going to pursue. In turn, that may influence how participative you are able and willing to be.

You may have little interest in action. You may nevertheless wish to be flexible in pursuing understanding of the situation on its own terms. If so, action-research-like processes may serve you well (though you may offend those whose commitment to action research is ideological). Alternatively, you may find grounded theory a better choice. It too is a data-driven approach. I should add that this is truer of the version championed by Glaser (1992) than of Strauss and Corbin (1997, 1998).

(You may also "mix and match". There are methods used in grounded theory which may be incorporated in action research, and vice versa.)

It is important here to note that action research can be rigorous without surrendering action outcomes. The sources of rigour may be different to those which characterise other research, and especially theory-driven research. They are no less effective.

Rigour in action research

This is more of an imperative than a choice. Your thesis will be expected to make a contribution to knowledge. You will be expected to demonstrate that your claimed contribution is supported by rigorous research and evidence and argument.

The danger is that your supervisors or committee, and your examiners, may equate rigour with the criteria which are appropriate within their own paradigm. They may value quantification, precise research questions, substantial early literature review, and the like. It is important that you understand the ways in which action research achieves rigour so that you can make a good case for what you have done.

Some of the arguments you can use are to be found in the qualitative literature generally. I’m thinking particularly of the well-reasoned monograph by Kirk and Miller (1986). Also relevant are Patton (1990), and many of the papers scattered through Denzin and Lincoln’s (1994) handbook, among others.

For the most part, I have not found the action research literature particularly concerned with issues of rigour. Some of the better arguments for rigour are to be found in work which some would regard as at the margins of action research. I’m thinking here of such writers as Argyris (particularly Argyris, Putnam and McLean Smith, 1985) and Checkland (particularly the first half of his 1981 book) among others.

The tight spirals of action research, however, deserve special attention here. Their main function is to provide flexibility and responsiveness for effective change. They also confer two advantages less evident in theory-driven research.

Each turn of the spiral provides you with yet another chance to test the interpretations you have so far developed against the data you are collecting.

Within each turn you develop plans which are then tested in action. Each turn of the spiral is a miniature test of the assumptions which guided your plans.

These advantages derive from the nature of each turn of the spiral. I described this spiral earlier as an alternation between action and critical reflection. The critical reflection can also be regarded as having two components. First is a review of the results of the previous action. Second is the planning for the next action. Action research can be described as a regular cycle of planning, action and review.

Alternatively, you may describe it as Kemmis and McTaggart (1988) do: "plan, act, observe, reflect". And you may note its similarity to other cycles, such as that for experiential learning (for example see Kolb, 1984) or quality management (for example see Rothwell, Sullivan and McLean, 1995).

Not all varieties of action research emphasise the spiral nature which, to my mind, confers many of its advantages. You have different versions of action research to choose from.

Choice 4: Which action research methodology?

The choices are many. All I will attempt here is to sketch in the breadth of choice you face, and some of the varieties. They include

many varieties of participative (or "participatory") action research (PAR), for instance Whyte (1991), Greenwood and Levin (1998), Smith and Willms (1997), and Zuber-Skerritt (1996), to name just a few;

several varieties of systems methodology, including those of Checkland (Checkland and Scholes, 1999) and Flood (1997);

teacher research such as that of Mills (2000), Hubbard and Power (1999), or the "living theory" approach of Whitehead and his colleagues (Whitehead, 1993), among others;

the substantial body of work on "action science" developed over many decades by Argyris and his associates: Argyris (1999), Argyris and Schön (1996), or Argyris, Putnam and McLain Smith (1985);

approaches whose similarity to action research is more or less explicit; for instance Forrester (1999), Heron and Reason (1997), the "action inquiry" of Torbert, (Fisher and Torbert, 1995), the "action evaluation" of Rothman (1997), and some aspects of "appreciative inquiry" (Elliott, 1999);

finally, some of the participative evaluation methodologies, such as the utilisation-focussed evaluation of Patton (1997) and Guba and Lincoln’s fourth generation evaluation (1989).

This by no means exhausts the possibilities.

I had originally intended to say something about how you might choose from this rich array. But that is too large a task for me at this time. You will have to wait for a separate paper. For the moment, let me suggest the following issues which might guide your decision.

Is your chief motivation for using action research your strong commitment to participation and equity? This may indicate that the more deliberately participative approaches may suit.

Is your wish to develop your understanding of a richly interconnected and complex system? Especially if you can treat it as an information system or a decision-making system, one of the systems methodology may be indicated.

Do you work in an academic setting where action research is regarded as a fringe paradigm, and highly suspect? You may then be able to frame your research as "evaluation", which may be more respectable.

Are you drawn towards understanding how interpersonal relationships and system dynamics interact to undermine effectiveness? Action science was devised to address this issue.

Each of the other methodologies mentioned above offers its own flavour and style. As I’ve said, there are reasons to avoid the literature related to your research theme if your approach is data-driven. The methodological and philosophical literature, on the other hand, is well worth your early perusal.

We will shortly address a fifth choice, to do with the level of participation. Previous experience suggests to me that this is a contentious point. Let me try to state my position clearly.

The virtuous spiral

It seems to me that there are two important dimensions to action research. The first of them relates to the action, which is more likely to be achieved if there are high levels of participation. In addition, you may choose high participation for reasons of ethics or values. The second dimension is to be found in the strategies for maintaining rigour. It is these which I wish to address now.

The strategies must be effective within a process which is very flexible and very responsive to the situation. They were developed for that reason. They are not a second best approach because the best is not available. They have virtue in their own right. The virtue is derived from different sources.

You might decide that extensive participation is not for you. You may decide it is too difficult for a thesis. It may introduce complications you wish to avoid. You might nevertheless decide that you can use action research-like processes for data collection and analysis. In other words, you might still make use of data-driven action research processes for the research, but be less concerned about the action.

This introduces our fifth choice.

Choice 5: What extent and style of participation?

As mentioned, in some quarters this has become an ideological choice. For many people it is therefore not a choice but an imperative. Even if you accept this imperative there is a spectrum of participation. It stretches from minimal participation if any (see Clark, 1972) through to "do it yourself" approaches (cf. Wadsworth, 1984).

At one extreme is much "teacher research" where an educator researches her own practice. The pupils may or may not be involved. Towards the other extreme are the approaches for which "emancipation" is a goal, and political considerations an issue. Many of these are to be found in the research which takes the ideas of Habermas (1984, 1987) as foundational. The Deakin school (Kemmis and McTaggart, 1988) has been influential in educational circles in Australia. Elsewhere there are similar approaches, for instance Toulmin and Gustavsen (1996).

Other critical approaches (for example Ghaye and Ghaye, 1998) draw on different foundations.

In any event, participation is far from being an all-or-none choice. Who will you involve? How will you go about involving them? What will you involve them in? Each of these offers further choices.

In large communities and organisations, will you involve everyone? Will you also involve those outside the community or organisation who also have an interest? If numbers are very great, will you reduce the depth of participation to make room for everyone?

How will you approach people and engage them in the process? For instance will you broadcast invitations widely? Perhaps you will slowly build relationships as you work your way through the networks. Or will you set up a steering committee, reference group, or the like, and ask them to involve others?

You will probably want to approach people to be informants. Beyond this, what will you involve them in? Will they interpret the data as well as provide it? Will they help you design the study? Of course, you will write the thesis or dissertation. But if there are other reports, who will write them?

These are important questions, I believe. Yet in much of the action research literature they are not given much attention.

Fortunately, other fields address these issues in ways which are relevant to action researchers. I include here evaluation (see Gregory, 2000) and the literatures of organisation development (for instance, Toulmin and Gustavsen, 1996, mentioned above), community development (McNair, 1998) and rural development (Chamala and Keith, 1995). And it would be remiss of me to omit the public participation literature. Sarkissian and Perlgut (1994) offer an Australian example.

Action learning also fits well. Current approaches to action learning as described by such writers as Marquardt (1999), Rothwell (1999) or Raelin (2000) bear a close resemblance to group-based participative action research. An action research group can function with benefit as an action learning group ("learning set"). Although the earlier fashion was different, current action learning mostly gathers participants from one organisation around a shared project, much as an action research study might. Although action learning has mostly been used in organisational settings it functions well in a variety of other situations too.

Two further issues ...

First, who decides the nature and extent of participation? You may be committed to involving participants as co-researchers and equals in an emancipatory relationship. That may or may not be their desire. Sometimes you may have to settle for less than you wish, at least initially (cf. Grundy, 1983).

Second, your thesis or dissertation is expected to be an independent and original contribution to knowledge. You may therefore have misgivings about high involvement for participants. You may fear it undermines the independence of your contribution.

Perry and Zuber-Skerritt (1992) have addressed this persuasively and helpfully. (Their paper is also reproduced as the following chapter of the present book.) The field study may be done as participatively as you and your participants wish. Your individual reflection upon that study then provides your own contribution.
. . .
So far in this chapter we’ve considered the strategic issues: the questions which do much to determine the overall shape of the study. A number of other choices are also worthy of note.

I begin by considering the methods to operationalise the methodology.

Choice 6: Methods for data collection and analysis?

You will find that much of the literature on action research does not explain in detail how it might be done. This need be of little concern. You can be vigilant that your chosen methods are consistent with both your action and your research aims. With this protection, many of the methods in use for other approaches can be pressed into service.

Most commonly used for data collection are interviewing and focus groups. These are well covered in the literature, as the following examples demonstrate.

Glesne (1999) gives a readable account of both interviewing and rapport building (the latter a much neglected topic). In his wide-ranging book, Kvale (1996) mostly treats the interview as conversation, but also addresses the philosophical and scientific context. Taylor and Bogdan (1998) give considerable attention to interviewing in their general text on qualitative research.

Focus groups have been most used in market research, though their origins were in qualitative research. The market research literature can be useful for its attention to the practicalities. For instance, see Edmunds (1999), practical despite its title. More recently, focus groups have been returned to their origins, and high quality accounts are appearing. Barbour and Kitzinger (1999) are a good example. For a practical account which also considers data analysis see Krueger and Casey (2000). There are many more.

Focus groups and interviews are not usually very participative. (There are ways of increasing participation, but space does not permit an examination here.) As you work at the participative end of the continuum you may well find other literatures of more use. I’ve already mentioned the literatures on organisation, community and rural development.

Much of your time is likely to be spent with groups of people. You will be helped here by the literature on group facilitation. Heron (1999), familiar with collaborative research methods, has much to say that is useful. I think you will find much of use in Kaner (1996). There are also structured group processes such as group feedback analysis (Heller and Brown, 1995) which are valuable.

For making sense of the data, Miles and Huberman (1994) provide some useful techniques. The descriptions are detailed and practical. The coding methods of grounded theory may also be used: see Glaser (1998) and Strauss and Corbin (1997). In my view, the latter overcode their data; but probably that is a matter of judgment.

As you make these choices about methods, I assume you will continue to press for appropriate rigour in what you do. Further choices then present themselves about issues of the criteria for "good" research, and about generalisability.

Choice 7: Rigour or relevance?

Fuelled by the growth of new paradigms, there has been much debate about the criteria by which research can be judged. There are those who believe that qualitative approaches and constructivist philosophies require their own criteria, distinct from those which characterise quantitative research. Lincoln (for instance 1990) argues this vigorously.

These views, though, can not safely by accepted without question. I have argued that the most important research choice is not qualitative versus quantitative. Theory-driven or data-driven: that is a more influential choice.

Further, action research is intended to act on the world. This presumes that the world which is to be acted on is something more than a mental construction. What we know of it may be a mental construction &endash; and I assume it is. I assume also that the mental construction stands in some relationship to the world "out there". And it is "out there" that we wish to influence. Cherry (1999) examines such issues as they apply to action research.

On the grounds explored above, it seems to me that reliability and validity are still useful concepts. When I am interviewing, for example, it is still my intention that as far as possible the information I collect is not idiosyncratic. That is, it is to some extent reliable. I attempt also to discover as accurately as I can the experience that my informant describes. That is, it is to some extent valid. The greater its validity and reliability, the greater the assurance I can have when I draw on my understanding to take action.

In addition to this, generalisability is a key issue. I agree with Baskerville and Lee (1999) that qualitative researchers have too easily given up on this. Their point is that there are several forms of generalisability. I would add that there are ways of extending the generalisability of case studies. (Most action research studies can be regarded as case studies, I think.)

The experimentalists have some grounds for claiming that their findings at best have the status of general laws. Suppose they find that variable A consistently influences variable B in a certain way. The claim can then be made that all else being equal, variable A always influences variable B in that way. As far as it goes, this is fine.

You and I might reply that this is less useful than might be imagined. Generalisability is not imbued with as much virtue as claimed. In the situations we deal with, it is seldom that only variable A and variable B are operative. The status of variable B is then indeterminate.

My own approach is to acknowledge that there is some trade-off involved between discovering general truths and being relevant to the local situation. My choice of "action research" or "action research" may tempt me to favour one or the other of these.

However, I can aim for a "best of both worlds" combination of them. There are grounds I can use to argue for the generalisability of my results. These grounds can be developed through logical analysis, multiple case studies, or highly diverse samples. I can also compare my interpretations to those in the literature.

So far, we have talked about choices which influence the shape of the research as it is carried out. Except indirectly, that isn’t the basis on which examiners will assess your research. No matter how well you do the research, you will be judged on what you capture on paper in your thesis or dissertation.

Choice 8: How will I write it up?

Here, the choice you face is between convention and a more tailored approach. Based on their training, many supervisors, committee members and examiners will have a mental image of thesis structure. It probably has chapters titled something like Introduction, Methods, Results, Discussion, and Conclusion. This is the convention, and may be the safer way to do it.

However, if you treat action research as data-driven and responsive, your study may not fit this structure without some trimming. An alternative is to devise a structure which is intended to reflect the conduct and style of action research.

There are two papers on a web site at Southern Cross University which reflect these two strategies. One is by Perry (nd). It offers suggestions for structuring the paper so that it appears conventional. I have written the other paper (Dick, 1993), with the intention of capitalising on the particular virtues of action research. I describe a thesis structure which organises the thesis around the contribution to knowledge, at the cost of being less conventional in form. I have reason to believe that candidates have found these valuable, some choosing one approach and some choosing the other.
. . .
I have now considered most of the design choices that are important in approaching a thesis or dissertation. In the remainder of this chapter I turn my attention to the actual conduct of the study. Immediately below you will find a discussion of some of the traps which await the unwary candidate. Finally, I briefly describe an action learning set &endash; a supportive group of colleagues &endash; as an effective way of helping to overcome these traps.

Avoiding the traps of procrastination

I hope it is now clear that action research has distinct qualities which set it apart from most other research. If you take into your action research study the assumptions that were developed within other approaches, those assumptions may inhibit you from making the most of your chosen paradigm. In particular, I think there are four traps which might be described as forms of procrastination:

postponing data collection until you have read all the relevant literature;

postponing analysis until you have collected all the data;

postponing action until you have interpreted all the data;

postponing writing until everything else is complete.

Action research requires neither extensive pre-reading, nor extensive early data collection, nor complete analysis. It lends itself to early action. It does not even require a research question or "thematic concern" (Kemmis and McTaggart, 1988), though one may well be useful. It is enough to have a research situation. The thematic concern will identify itself soon enough.

Data-driven research provides an approach which suits activists and managers and other people who wish to act on the world and understand how they do so. Each turn of the action research spiral can contain data collection, data analysis, action planning, action, and evaluation:

data are interpreted as they are collected;

interpretations are turned into action plans, which are then immediately acted on;

the results of the action can be reviewed immediately;

though you will wish to revise it later, you can capture on paper your conclusions as you develop them.

It is useful to read enough of the methodological literature at the beginning. You don’t wish to find, later, that your methodology wasn’t adequate. The content literature can be left until later. The early imperative is to find a suitable research situation, and achieve early ethical approval. (This latter topic deserves a paper of its own.)

As I have said, each turn of the spiral tests your assumptions in action. Each turn gives you another opportunity to test your emerging interpretations by vigorously seeking disconfirming evidence.

There is also a final trap. If you succeed in doing all of this, you may then suffer the opposite procrastination: allowing yourself to be distracted from the theory by the action. As I said, action research can be appealing to those who wish to act on the world. It is also tempting for them, when an action has been successful, to press ahead to the next action.

Regular, systematic and critical reflection allows an escape from this final trap. This provides the understanding which eventually accrues to become your contribution to knowledge. It is also where you develop the understanding to improve your practice and your work situation.

Careful but economical documentation, done continuously, also helps. In addition, you can give more attention to the planning phase of each turn of the spiral. During planning you can become conscious of your assumptions (about epistemology, about methodology, about the situation, about action). Having made them explicit, you are more likely to notice the evidence which allows you to refine them.

The thesis or dissertation is an individual contribution. Nevertheless, you don’t have to do the study or the report in isolation.

Creating a support network

As already mentioned, an action research study may draw on action learning methods to engage participants in the research and action. Action research and action learning have grown together despite their different origins and early history.

In addition, you may use the processes of action learning to organise a support group. You can meet regularly with other higher degree candidates in an action learning set. There you can provide each other with mutual challenge and mutual support. You can thus overcome the loneliness and isolation which afflicts many postgraduates. You and your colleagues can help each other to be more effective and more satisfied in your thesis work. You will learn much from each other.

This can be particularly valuable if you are doing your higher degree research by distant education. Davies and his colleagues (2000) have described such an approach where regular but infrequent meetings with supervisors are supplemented with regular and frequent learning set meetings.

Summary

Are you planning to do a higher degree by research? If so, action research may be a choice open to you. I have then suggested a number of choices which you can use to guide you through the design:

Will your study be theory-driven or data driven: will you take the literature or the research situation as your starting point?

Will you engage with the research primarily as technician or as performing artist: will you closely follow a recipe, or design and conduct your study to fit the situation and the theme?

Will this be "action research" or "action research": how action oriented will your study be?

Which of the many available methodologies will you use?

Who will you involve as participants in the study, and how, and to what extent?

What methods for data collection and analysis will you use to operationalise your methodology?

Facing a trade-off between rigour and relevance, what will you do?

Will you structure your thesis or dissertation conventionally, or will you design a structure which makes the most of your chosen methodology?

Finally, I identified some traps which face most higher degree candidates. I offered some ways of avoiding those traps, including the formation of a group of colleagues to act as an action learning set.

References

Argyris, C. (1999). On organizational learning, second edition. Oxford, UK: Blackwell.

Argyris, C., and Schön, D.A. (1996). Organisational learning II: theory, method and practice. Reading: Addison-Wesley.

Argyris, C., Putnam, R., and McLain Smith, S. (1985). Action science: concepts, methods and skills for research and intervention. San Francisco, Ca.: Jossey-Bass.

Barbour, R.S., and Kitzinger, J., eds. (1999). Developing focus group research: politics, theory and practice. London: Sage.

Baskerville, R., and Lee, A.S. (1999) Distinctions among different types of generalizing in information systems research. In Introna, Ngwenyama L., Myers, M., and DeGross, J.I., New information technologies in organizational processes: field studies and theoretical reflections on the future of work. New York: Kluwer. [pp 49-65]

Bentz, V.M., and Shapiro, J.J. (1998). Mindful inquiry in social research. Thousand Oaks: Sage.

Chamala, S., and Keith, K., eds. (1995). Participative approaches for landcare: perspectives, policies, programs. Brisbane: Australian Academic Press.

Checkland, P. (1981). Systems thinking, systems practice. Chichester: Wiley.

Checkland, P., and Holwell, S. (1998). Information, systems, and information systems: making sense of the field. Chichester, UK: Wiley.

Cherry, N. (1999). Action research: a pathway to action, knowledge and learning. Melbourne: RMIT University Press.

Clark, P. (1972). Action research and organisational change. London: Harper and Row.

Crotty, M. (1998) The foundations of social research: meaning and perspective in the research process. St. Leonards: Allen and Unwin.

Davies, A., Dick, B., Hase, S., Sankaran, S., Gloster, M., and Kwok, R. (2000) Problem based learning with academic accreditation: a flexible postgraduate program for managers and practitioners using action research at the workplace. Paper presented at the HERDSA/ASET Flexible Learning Conference, University of Southern Queensland, Monday 3 June 2000.

Denzin, N.K., and Lincoln, Y.S. (1994). Handbook of qualitative research. Thousand Oaks, Ca.: Sage.

Dick, B. (1993). You want to do an action research thesis? -- How to conduct and report action research. Chapel Hill, Qld.: Interchange (mimeo). Available on line at http://www.scu.edu.au/schools/gcm/ar/arthesis.html

Edmunds, H. (1999) The focus group: research handbook. Chicago, Ill.: NTC Business Books.

Elliott, C. (1999). Locating the energy for change: an introduction to appreciative inquiry. Winnipeg, Canada: International Institute for Sustainable Development.

Fisher, D., and Torbert, W.R. (1995). Personal and organizational transformations: the true challenge of continual quality improvement. London: McGraw-Hill.

Flood, R.L. (1997). Action research and the management and systems sciences. Systemic Prac tice and Action Research, 2(1). Also available on line at http://www.hull.ac.uk/cfss/spar/ARMSS.html

Forester, J. (1999). The deliberative practitioner: encouraging participatory planning processes. Cambridge, Mass.: MIT Press.

Ghaye, A. and K. (1998). Teaching and learning through critical reflective practice. London: David Fulton.

Glaser, B.G. (1992). Basics of grounded theory analysis: emergence vs forcing. Mill Valley, Ca.: Sociology Press.

Glaser, B.G. (1998). Doing grounded theory: issues and discussions. Mill Valley, Ca.: Sociology Press.

Glesne, C. (1999). Becoming qualitative researchers: an introduction, second edition. New York: Longman.

Greenwood, D., and Levin, M. (1998). Introduction to action research: social research for social change. Thousand Oaks, Ca.: Sage.

Gregory, A. (2000). Problematizing participation: a critical review of approaches to participation in evaluation theory. Evaluation, 6(2), 179-199.

Grundy, S. (1983). Three modes of action research. Curriculum Perspectives, 2(3), 23-24.

Guba, E.G., and Lincoln, Y.S. (1989). Fourth generation evaluation. Newbury Park, Ca.: Sage.

Habermas, J. (1984). The theory of communicative action, volume 1: reason and the rationalisation of society. Boston: Beacon Press.

Habermas, J. (1987). The theory of communicative action, volume 2: the critique of functionalist reason. Oxford: Polity Press.

Heller, F.A., and Brown, A. (1995) Group Feedback Analysis applied to longitudinal monitoring of the decision making process. Human Relations, 48(7), 815-836.

Heron, J. (1999). The complete facilitator’s handbook. London: Kogan Page.

Heron, J., and Reason, P. (1997). A participatory inquiry paradigm. Qualitative Inquiry, 3(3), 274-294.

Hubbard, R.S., and Power, B.M. (1999) Living the questions: a guide for teacher-researchers. York, Maine: Stenhouse Publishers.

Kaner, S., with Lind, J., Toldi, C., Fish, S. and Berger, D. (1996) Facilitator’s guide to participatory decision-making. Gabriola Island, BC: New Society Publishers.

Kemmis, S., and McTaggart, R. (1988). The action research planner, third edition. Geelong, Victoria: Deakin University.

Kirk, J., and Miller, M.L. (1986). Reliability and validity in qualitative research. Beverly Hills, Ca.: Sage.

Kolb, D.A. (1984). Experiential learning: experience as the source of learning and development, Englewood Cliffs, NJ.: Prentice-Hall.

Krueger, R.A., and Casey, M.A. (2000) Focus groups: a practical guide for applied research, third edition. Newbury Park: Sage.

Kuhn, T.S. (1970). The structure of scientific revolutions, 2nd edition. Chicago: University of Chicago Press.

Kvale, S. (1996). InterViews: an introduction to qualitative research interviewing. Thousand Oaks: Sage.

Lincoln, Y.S. (1990). The making of a constructivist: a remembrance of transformations past. In Egon S. Guba, ed., The paradigm dialog. Newbury Park, Ca.: Sage. [pp. 67-87]

Marquardt, M.J. (1999). Action learning in action: transforming problems and people for world-class organizsational learning. Palo Alto, Ca.: Davis-Black.

McNair, R.H., ed. (1998). Research strategies for community practice. New York: The Haworth Press.

Miles, M.B., and Huberman, A.M. (1994) Qualitative data analysis: an expanded sourcebook, second edition. Thousand Oaks, Ca.: Sage.

Mills, G.E. (2000). Action research: a guide for the teacher researcher. Upper Saddle River, NJ: Merrill.

Moustakas, C. (1990). Heuristic research: design, methodology, and applications. Newbury Park: Sage.

Patton, M.Q. (1997). Utilization-focussed evaluation, third edition. Thousand Oaks, Ca.: Sage.

Patton, M.Q. (1990). Qualitative evaluation and research methods, second edition. Newbury Park, Ca.: Sage.

Perry, C. (nd). Structure of a five chapter PhD thesis: candidates’ notes. Brisbane: Queensland University of Technology, unpublished paper. Available on line at http://www.scu.edu.au/schools/gcm/ar/art/cperry.html

Perry, C., and Zuber-Skerritt, O. (1992). Action research in graduate management research programs. Higher Education, 23(2), 195-208.

Raelin, J.A. (2000) Work-based learning: the new frontier of management development. Upper Saddle, NJ: Prentice-Hall.

Revans, R.W. (1983). The ABC of action learning: empowering managers to act and to learn from action, third edition. London: Lemos & Crane.

Rothman, J. (1997). Resolving identity-based conflict in nations, organisations, and communities. San Francisco, Ca.: Jossey-Bass.

Rothwell, W.J. (1999). The action learning guidebook: a real-time strategy for problem solving training design and employee development. San Francisco, Ca.: Jossey-Bass/Pfeiffer.

Rothwell, W.J., Sullivan, R., and McLean, G.N. (1995). Chapter 2: models for change and steps in action research. In Rothwell, William J., Sullivan, Roland, and McLean, Gary N., eds., Practicing organization development: a guide for consultants. Jossey-Bass/Pfeiffer. [pp47-72]

Sagor, R. (2000). Guiding school improvement with action research. Alexandria, Va.: Association for Supervision and Curriculum Development.

Sarkissian, W., and Perlgut, D., eds. (1994). The community participation handbook: resources for public involvement in the planning process, second edition. Murdoch: Institute for Science and Technology Policy, Murdoch University.

Smith, S.E. and Willms, D.G., with Johnson, N.A., eds. (1997). Nurtured by knowledge: learning to do participatory action-research. New York: Apex.

Strauss, A., and Corbin, J., eds. (1997). Grounded theory in practice. Thousand Oaks, Ca.: Sage.

Strauss, A., and Corbin, J. (1998). Basics of qualitative research: techniques and procedures for developing grounded theory, second edition. Newbury Park: Sage.

Taylor, S.J., and Bogdan, R. (1998) Introduction to qualitative research methods: a guidebook and resource, third edition. New York: Wiley.

Toulmin, S., and Gustavsen, B. (1996). Beyond theory: changing organisations through participation. Amsterdam: Benjamins.

Wadsworth, Y. (1984). Do it yourself social research. Melbourne: Victorian Council of Social Service and Melbourne Family Care Association.

Whitehead, J. (1993). The growth of educational knowledge: creating your own living educational theories. Bournemouth: Hyde.

Whyte, W.F. (1991). Participatory action research. Newbury Park: Sage.

Zuber-Skerritt, O. (1996). New directions in action research. London: Falmer.

Copyright Bob Dick 1993-2000. This document may be copied if it is not included in material
sold at a profit, and this and the following notice are included. It may be cited as follows:

Bob Dick (2000) Postgraduate action research: important choices. Available on line at
http://www.uq.net.au/action_research/art/pgar.html