Theorising Social Work Research

Researching the Social Work process Seminar topics

Researching the Social Work process 11th July 2000 Luton

Theorising from Frontline Practice Towards an Inclusive Approach for Social Work Research Professor Jan Fook Professor of Social Work Deakin University Geelong Victoria Australia

Over the last two decades I have gone through many changes in my thinking about theory, practice, and theorising from practice. Twenty years ago, I was convinced that sound formal theories only needed to be better applied, so that sound practice models could be developed. Gaps between theory and practice only existed because theory was not taught or understood well enough to apply it effectively to practice. I thought the answers lay in drawing up better frameworks which demonstrated how theory could be directly applied to practice, particularly direct practice. I therefore focused my research efforts on developing practice models from formal theories (eg. Fook, 1993).

Ten years later I began to recognise that it wasn't quite as simple as this, that practice situations are often experienced in completely different terms to the ways formal theories are expressed. The parameters of practice are often totally unlike the dictates of theory. I therefore thought the answers lay in developing better ways of understanding our practice, particularly from the point of view of practitioners. I focused my research efforts on a huge qualitative study of practitioners, seeking to trace the development of professional expertise, gleaned from practitioners' own accounts of their practice (Fook, Ryan & Hawkins, 2000).

In the year 2000 I am much less sure about any of it. I have been heavily influenced by postmodern thinking, and the need to recognise multiple and diverse perspectives. As well, my experience from undertaking the expertise study showed me that all kinds of theories are created and used by practitioners in all sorts of ways. Maybe "anything goes" when it comes to theorisation from practice. Yet on the other hand, "expert" practitioners clearly work from and with a system of thinking which has meaning for them, and there are commonalities between the ways experienced practitioners work. So it should also be with expert research practitioners - after all research is ultimately about theorising a phenomenon - all good practitioners should be able to appreciate and work with complex and changing situations, all the while being relatively comfortable with their uncertainty. They have a meaningful set of values and theories which allow them to engage confidently with new situations, but still be open to new learning.

In the year 2000 then, I ask myself what kinds of frameworks, what kinds of ways of understanding theory, practice and research, will cultivate expert practice in uncertainty? What approaches to theorising from practice can we develop which will enable the practitioner researcher and the researcher practitioner to make sense of diverse perspectives, to act confidently and responsively in changing and unpredictable situations?

A rigid, or even loose, commitment to one type of perspective, be it positivist, qualitative or deconstructive, does not seem to provide the flexibility of thinking needed to work in changing circumstances. I have long ago given up the idea that debates between different paradigms are the way to yield better practice and research. For one thing, I would argue that all perspectives and methods require critical examination, since much of our research is already practised within an embedded culture of positivism (Fook, forthcoming), whether it is explicitly so or not. The way we speak about and conceptualise research often assumes scientistic ways of thinking about knowledge building, implicit beliefs that hypothesis testing, strict sampling techniques and validated instruments are the only pathways to further understanding. For example, the accepted formats for reporting on research also assume a scientific framework, that is, relatively short articles (Fraser, 1995), which do not allow for a more discursive description of in-depth experience.

At the same time, an "anything goes" mentality, often associated with postmodern thinking, in which any type of method or approach is acceptable, also seems lacking in providing clear value-based ideals for action. It does not seem appropriate for many situations in which the moral and political dimensions might in fact be clear, and in which some methods do suggest themselves as superior to others and for good reason. Is it possible for us to have our cake and eat it too? Can we develop an approach which allows us openness, but also builds upon and uses established methods of working? Can we devise a more inclusive approach, in which different perspectives on theory and its development from practice, are used in a meaningful yet flexible way to suit the situation at hand?

In this paper I want to begin the task of mapping an inclusive approach to theorising from practice, an approach which seeks to recognise different approaches to theory-building, and to enable a more critically aware and meaningful engagement with different theories and the different ways they may be developed from practice.

I will begin by taking a step back, and reviewing broadly current postmodern thinking about theory, practice and knowledge generation. I will then look at the different elements of an inclusive approach to theorising from practice: what is theory, how it is generated, and who generates it.

Current thinking about theory and practice and the need for an inclusive approach

I am aware that postmodern thinking is not the most popular in current social work circles. However I think it is important to be familiar with some basic postmodern analysis, since it is challenging some of taken for granted assumptions in relation to theory and practice in social work. Some of these challenges can have positive ramifications for the practice of social work.

There is a challenge to our assumptions about how professional knowledge and control are created and maintained, two features which are said to define the professions (Friedson, 1970). Professions maintain their status by controlling their knowledge base and how it is disseminated. Professions are socially legitimated by their knowledge claims (Leonard, 1997, p. 97). In a postmodern world, however, specialist knowledge is challenged as the exclusive domain of a particular professional group. These challenges also arise from a broader postmodern movement to question traditional hierarchical arrangements. In this sense the traditional authority of professional knowledge is questioned, as against the legitimacy of the experience of the service-user. Similarly the privileging of the scientific knowledge of the researcher, as against the lived experience of the practitioner, is also debated. Smart (1992, p. 100) sees professional expertise as 'the citadel........which has disqualified the understanding and knowledge of "ordinary" people'.

Postmodern analysis thus points up a widening gap between theory and practice. This is old news - we have long been aware that practitioners use little formal theory in everyday practice (Sheppard, 1995) What postmodern analysis does is point out that the gap arises because of the different social positioning and ways of knowing of the different players. The disparities between knowledge and theory generated by professional researchers, and the 'on-the-ground' knowledge embodied in the daily experience of both practitioners and service users are widening. And perhaps it is not so much that the disparities are widening, as it is also that it is now more acceptable to question the taken-for-granted authority of academic, non-practitioner, researchers. A major set of questions for social work is what constitutes legitmate social work knowledge or theory, how is it best generated, and by whom?

As a follow on from this, there is also a question about what types of knowledge and theories should be the goal of professional research. Should we be striving to create generalised theories which have been validated through scientific testing? Or can universally applicable theories be generated by practitioners grounded in specific contexts? At the same time can knowledge which is developed to operate regardless of context be meaningful to specific service users and practitioners? Does the attempt to generate universal knowledge assist the social work endeavour in postmodern and changing contexts?

The many differing answers to these questions are at the heart of the differing perspectives on social research and its methodology which have arisen in the last few decades. For instance, qualitative, ethnographic and naturalistic approaches draw attention to the need for methods to research sympathetically phenonena from the perspective of the experiencer in context (Denzin & Lincoln, 1994). Grounded theory (Strauss & Corbin, 1994) methods enable the researcher to develop theory from the experience of respondents, rather than imposing preconceived theory. Reflective methods (Schon, 1983; Fook, 1996) also upend the traditional theory-practice hierarchy and encourage the identification of theory implicit in practice, by practitioners themselves. Deconstructive, semiotic and narrative research methods also encourage the development of theories implicit in practice, through the interpretation of texts representative of experience. Action research recognises that theories are generated in context, influencing, and being influenced by a context of interactions as they are in the process of being developed. Participative and collaborative forms of inquiry (Heron, 1996; Baldwin, 2000) recognise that theories are often most effectively generated from practice through an alliance and dialogue between researchers and practitioners.

How do we make sense of these many differing perspectives? Let us start by focusing on the broad themes they share, namely:

What is theory

I have often devalued much of the theorising I have done in the past by imagining that thinking could only be counted as "theory" if it provided explanation and was also generalisable. However, if we examine these ideas closely, I would argue that there are many different forms of thinking which can perform these functions, albeit in ostensibly different ways. I have now come to think that "theory" can vary from a single descriptive idea, concept or label, to more complex sets of related ideas. Often just "naming" or labelling a piece of behaviour can function to provide some explanation, or connect the behaviour with related ideas. Different theories might spell out these connections or explanations to different extents.

The literature on qualitative theory building supports this thinking. According to Strauss & Corbin (1994) theory "consists of plausible relationships proposed among concepts and sets of concepts" (p. 278) and that "grounded theory" is conceptually dense (p. 279) and grounded in the multiplicity of perspectives of the actors in a situation (p. 280). Theories also vary in terms of their degree of formality, generalisability and relevance (Strauss & Corbin, 1994, p. 281) and the degree to which they should provide explanation, or can indicate causality (Miles & Huberman, 1994, p. 432). There are also different levels of theories - Strauss & Corbin (1994, p. 281) distinguish between theories based on data gleaned from one context (substantive theory and "higher order" theories developed from numerous contexts (formal).

Questions of generalisability and relevance relate to the varying degrees to which theories can apply more broadly, yet can also provide understanding. These are complex concepts, since a theory which is generalisable, may not necessarily apply to a specific instance. In relation to this point, Gilgun (1994, p. 122) distinguishes between two types of generalisability. The first type, "idiographic", refers to theory which is developed from specific situations and can be tested for its relevance to, its ability to provide understanding of, other situations. The second type of generalisability, called "nomothetic", is associated with the search for more "general laws, abstracted from time, place and specific person" (p. 122). According to Gilgun these different types of theory are developed differently. The advantage of idiographic theories is their openness, their ability to take into account the individual "exception to the rule", to illuminate specific experiences in different situations and contexts. The beauty of nomothetic theories is their predictive function, their ability to impose order and clear guidelines for practice. I like to think of the differences between them as the differences between "bottom up" (inductively developed) and "top down" (deductively applied) theories. Both are needed in any type of effective practice. "Bottom up" theories allow us to constantly apply our thinking in ways relevant to the situation at hand. "Top down" theories give us a starting point for engaging with a new situation, a framework which may ultimately prove inadequate but at the very least provides a beginning framework which makes new experiences initially manageable.

I think one of the difficulties for both practitioners and researchers is that an artificial distinction between idiographic and nomothetic theories has been created - it is often assumed that idiographic theories emerge from practice, and nomothetic theories from research. This has led to a devaluing of the forms of theorising associated with practice. Idiographic theorising is often thus not even viewed as research. But in fact both forms of theorising are used in, and are integral to, practice and research, indeed, to the business of living.

In order to get around the devaluing of theories which are not seen as generalisable, I like to use the term "transferability" (Fook, forthcoming) in place of generalisability, to refer to the ability of the theory to transfer meanings between different contexts. However it is important to remember that these ways of differentiating between the generalisability of theories are not mutually exclusive categories. Idiographic theories may become more nomothetic through further testing. Nomothetic theories may be constantly revised through accounting for exceptions. Transferable theories can become more generalisable. Theories can go through a succession of different types of research processes to develop ( as Strauss & Corbin term it, 1994, p. 281) from "substantive" (developed in one context) to "formal" (developed from many contexts) levels, from a description of a single concept, through the development of relationships between many concepts. At their most complex, grounded theories "connect a multiplicity of perspectives and patterns and processes of action and interaction that in turn are linked with carefully specified conditions and consequences" (Strauss & Corbin, 1994, p. 280).

Understanding the differences and relationships between generalisabiltiy and relevance or transferability, between nomothetic and idiographic theories, points up the variations between theories in terms of their content, purpose and application. For example, in social work we are familiar with the differences between practice theories, and the broader social or behavioral theories on which they are based (Fook, 1993, p. 40; Payne, 1997, pp. 38-41). In some literature, these differences might be categorised as "substantive" (knowledge about a situation or phenomenon) or "procedural" (knowledge about how to use that knowledge) (refer expertise article here). Eraut makes the very pertinent point that practising professionals are engaged in the constant process of using theories (Eraut, 1994) and that this in itself creates a theory (about how to use theory). We can therefore say that there may be as many different types of theories as there are the processes to create them (Payne, 1997, p. 49).

Miles and Huberman (1994, p. 434) note some of the many types of theory:

Given the diversity of theories, their uses and applicability, and given the diversity of social work practice experience, it makes sense to take an inclusive approach to our understanding of theory. It is this more inclusive approach I am taking in that I would argue that in simple terms what seems common in all these conceptions is the idea that "theory" at one end of the continuum, is a term we use to denote a label which gives some meaning to a phenomenon. At another end of the continuum, it is a set of labels and complex relationships between them. On another end of the continuum are theories which illuminate understanding, and on another end, those which allow prediction. These theories may be generated through a multitude of different methods, from different contexts and for different purposes.

Theorising, then, in simplest terms, is the act of developing these labels, the different processes of creating an idea or sets of ideas, from and through, different types of experiences. Thinking about theorising in this way is enabling because it allows the practitioner/researcher to locate her or his specific knowledge generation activities within a research context and discourse. It allows the practitioner/researcher to identify the specific contribution of her or his particular form of theorising, and how it fits within a broader context of professional knowledge.

How theory is generated from practice

The problem of course, as we all know, is that practice, particularly frontline practice, can be unpredictable and uncontrollable, changing and contextually-based. It is often "one-off". It is hard to access (by researchers) and often practitioners themselves are not keen to undertake the research. There are also ethical considerations, of intruding on the experiences of those who often have least to gain directly from the research, service users. Frontline practice does not lend itself easily to the requirements of traditional research as we see them - the need to measure and control variables, to make predictions, to be able to generalise our findings.

However an inclusive view of theory provides for an inclusive approach to frontline practice as data, and for the analysis (theorisation) of that data. In this section I want to provide a brief overview the many possibilities for obtaining of data from frontline practice, and then to review the many different types of analytical processes available for theorising from this data.

The data for theory generation - accessing frontline practice experiences

For the purposes of our discussion here, an inclusive approach to collecting data from frontline practice means understanding frontline practice as involving many different experiences in which the practitioner is involved in delivering services directly to service users. The experiences can be many and varied, ranging from personal interactions with service users, to writing reports, making judgements in implementing policies, attending case meetings, lobbying managers, and so on. I find it most useful to talk about "accessing experiences" rather than "obtaining data", since the information we seek is the experiences themselves. Since these experiences already occur, it is more accurate to speak of accessing them in the most appropriate ways, rather than trying to collect something ("data") which does not already exist in the form we want it. We should be asking," How do we access frontline practice in ways which will best enable us to theorise from it?" rather than "What instruments allow us to collect best data on frontline practice for research purposes?" I also think this is a more inclusive way of thinking about data and its collection, because it forces us to think about practice experience at how it happens "naturalistically", so that we can approximate our methods to it, rather than necessarily creating or imposing more "artificial" methods for data collection.

However it is also important to bear in mind that because frontline practice experience is holistic and multi-perspectival, we will only ever have a partial and selective access to the whole experience. Experiences also take place in contexts, which partially influence the experience, and no one player has access to all aspects of the context at any one time. Even if it our own practice experience from which we wish to theorise, our view of it will be framed by what we have access to, and must therefore still be limited. One of the important principles therefore in researching holistic experience is to try to maximise the number of perspectives, or to access the experiences in different ways, and from different angles.

Clandinin and Connelly (1994) make some useful points about experience and our study of it. Experience at the very least is both personal and social, but it is the holistic experience we wish to study. They suggest that methods therefore need to be focussed in at least four directions - on the inwards (internal thinking, etc) and outwards (the context), the backwards (past) and forwards (present and future). They then suggest a number of different ways of accessing personal experience for study. These include: oral histories; constructed annals of particular periods; stories; historical artifacts; interviews; journals; autobiographical writing; letters; conversations; and field notes. I include the whole list because it may easily suggest similar ways of accessing frontline practice.

It is useful to divide possible ways of accessing experience into a number of main categories:

. ethnographic and observational methods can be conducted in a variety of ways depending on the slice of experience under study. The "slice" may be a defined period of a practitioner's work time (eg. one week), a series of defined activities (all assessment interviews for a period of a month), staff meetings, and so on. Participant observation may be a method most accessible for practitioners. An issue will be how observations are recorded, and there are more or less unobtrusive ways to do this (Kellehear, 1993).

. existing documents, which may include diaries, case notes, files, policy documents, minutes of meetings, position statements, reports and statistics on practice

. accounts of practice, which could include interviews, staff room conversations, debriefing sessions, supervisory sessions, or simply constructed accounts. Process or taped recordings of interviews with service users are useful.

One of my favourite tools for accessing practice is the description of critical incidents selected by practitioners (Fook, Munford & Sanders, 1999), because, if I am aiming to theorise from practice as it is experienced by practitioners, I find it best to elicit their own descriptions of their practice, rather than to study accounts which have been constructed in some other way or for some other purpose. For example, I often find that even if conducting an unstructured interview, practitioners will often tend to try to describe their practice in formal theory terms, possibly because they know I am a social work theorist, or because that is how they think they should talk about their practice. If I want to elicit practice in its "rawest" possible terms, I tend to ask for concrete as possible descriptions. This is a useful principle for accessing frontline practice experience, because it minimises the perception of evaluation of the worker's practice in formal theoretical terms.

Another important issue is that of reflexivity. If we recognise that our own perspectives (as researchers and practitioners) are integral to how we access practice experience, then this provides guidelines as to the sorts of methods we might use in "collecting the data" itself. Indeed it might also influence what counts as "data", that is, whether the perspective of the practitioner/researcher might count as an integral part of the experience being studied. These questions of course open up a complex number of issues for consideration. However for our purposes here it is useful to point up that the practitioner/ researcher, in focusing on practice experience, might usefully include situations which allow access to the perspectives of the practiitoner/researchers themselves as part of the phenomenon. A reflexive stance might in fact open up a number of different ways of accessing practice experience (Fook, 1999). An example of this is referred to by Steier (1991, pp. 166-7), in which family therapist's constructions are studied by researchers co-viewing tapes of interviews with therapists, and therapists stopping the tapes at crucial points to discuss reasons for their actions.

By way of summary, in an inclusive approach to theorising from frontline practice, the important principles in relation to ways of accessing frontline practice, are: to minimise the influence of pre-existing formal theory; to maximise the number of perspectives available; and to maximise the fit between the method for accessing the experience, with the practice experience itself; and where appropriate, to include the perspectives of the practitioners/researchers.

Types of analysis - theorising from frontline practice

In an inclusionary framework, where theorising can be seen as anything from the assignation of a single label or idea, to the development of complex sets of inter-related ideas, the analysis of practice experience can therefore take many forms. In conventional ways of viewing research, it is common to talk about two main ways of analysing data, deductive and inductive. Deductive methods involve the application of pre-existing frameworks to the data, whereas inductive approaches involve a development of theory from the data itself. The latter form is most commonly associated with qualitative approaches, but in an inclusionary framework, I would suggest that both types of analysis are necessary (and not mutually exclusive). In an inclusionary approach therefore, frontline experience can be understood or theorised in both broad ways, each of which can involve a number of different methods. Again, I find it useful to view these as being on a continuum ranging from more deductive to more inductive methods.

At the most deductive end might be placed more statistical forms of analysis, where the theory being developed is a hypothesis or single idea being tested through, for example, statistical correlations or tests of significance. At the most inductive end might be those forms of analysis involving the creation of meaningful frameworks from ostensibly unrelated sets of ideas. Among non-statistical forms of analysis, it is helpful to identify three broad categories of analysis. All involve a type of pattern recognition, an attempt to impose and evolve some kind of meaningful "coding" or theorising of the data from the experiences under study. Kellehear (1993) summarises them roughly as follows, and I have added a few of my own as well:

Types of pattern recognition:

Content analysis involves a more deductive thinking, in which patterns in the data are compared to a pre-exisiting framework, perhaps developed by someone else, perhaps developed by the researcher from other means. This type of analysis may be useful, for example, if trying to demonstrate how current practices fit current stated expectations of competencies. Content analyses can involve the identification of patterns through simple statistical measures (counting percentages and mapping trends), or a type of counting of trends which emerge from non-numerical responses to surveys. For instance, one of the most common forms of published research in Australian Social Work in the last ten years has been a content analysis of mailed or telephone surveys involving short answers to structured questions (Fook & Briskman, forthcoming). In the expertise study I undertook with my colleagues Martin Ryan and Linette Hawkins, we undertook a type of deductive analysis of our material, coding the Dreyfus & Dreyfus (1986) pre-existing theory of expertise which included elements like the use of "context-free rules", and examining out material to see whether and how often similar features were demonstrated by our participants.

Thematic analysis is probably the most commonly practised form of analysis which broadly involves analysing the material for recurrent patterns which emerge, and which broadly fit the experiences being analysed. The analysis often involves developing preliminary categories, which are tested and changed repeatedly as new themes and patterns emerge. Miles & Huberman (1994, p. 432) have constructed a 13 point list of the procedures to follow in constructing theory by this process, developing roughly from concrete descriptive to more abstract explanatory theory. The list begins with noting patterns and themes, clustering and making metaphors, counting, making contrasts and comparisons, partitioning variables and noting relationships, and which ends with a comparison of the emergent theory with pre-existing ideas. This theoretical sensitivity (Strauss & Corbin, 1994, p. 280) is a necessary part of the theory-making and interpretative process of thematic analysis. In this way, grounded theories are never developed in a vacuum, but ultimately in relation to current thinking and discourse around a phenomenon. This form of theorising thus always involves a knowledge and development of existing theory as well.

A more semiotic process comes at analysis slightly differently. Whilst also searching for dominant themes, a deconstruction of material also involves a search for missing themes, enabling a "deconstruction" of taken for granted constructions. Elsewhere (Fook, 1996 & 2000) I have likened the reflective approach to such a process, in which the theory implicit in practice can be elicited by deconstructing accounts of it. Through a series of reflective (deconstructive) questions which are designed to uncover hidden assumptions, practitioners are able to reconstruct a desired theory of practice. The process thus involves a theorising directly from practice itself, rather than through espoused notions of it. The beauty of this reflective approach of course is that we are able to get closer to the practice as practised, rather than the formal theory, as practised, so to speak.

It is interesting to note that in the expertise study we found that the deductive analysis was useful in linking our work with pre-exiting expertise theories which gave it some academic legitimacy. We did find, however, that there were many themes which did not fit our coded categories of the Dreyfus and Dreyfus (1986) framework, in which case we had to devise new stages to add to the theory (we added a stage of "pre-student" and "expert"). However, in looking more closely at the characteristics exhibited by practitioners, we also felt that exisiting formulations of social work practice did not necessarily do justice to many of the humbling experiences we were told about. In the end we also developed a theory of social work expertise which labelled the many more complex skills and practices demonstrated by workers.

The form and content of this theory building however went through many stages. In preliminary stages the descriptions of expertise tended to be more mundane, in that we coined descriptions from exisiting social work terminology like "generic" and "holistic" (Fook, Ryan & Hawkins, 1997). However in later stages, as we had more accounts to compare and devise patterns from, and as we became familiar with a greater range of theories (our theoretical sensitivity increased) we became bold enough to develop newer terminologies for features of expertise like "connectedness" and "contextuality" (Fook, Ryan & Hawkins, 2000).

It is helpful to see the different types of pattern recognition as being in a continuum, with possibilties for "mixing and matching" different types of analysis as appropriate. I suspect that, rightly or wrongly, much of what we termed "grounded theory" in the 1980's might now be called "discourse analysis", since the recognition of commonly used language is involved in both types of analysis. For example, in the expertise study, we began with a more deductive type of content analysis, which allowed us to make some early sense of the material which was extremely daunting in its size and complexity. We felt most comfortable with applying and using practice terminologies with which we were already familiar, like "generic" and "holistic". However as our familiarity with the material increased, we became bolder in seeing patterns emerge, which we had not anticipated, like a seeming reluctance to work with men (Ryan et al, 1994). However as we connected patterns in our material with our own changing thinking around practice, we coined new terms like "contextuality" or used non-traditional terms like "reflexivity" in relation to practice (Fook, Ryan & Hawkins, 2000).

Elsewhere (Fook, forthcoming) I have termed this process of interweaving levels of theoretical analysis as a type of theoretical reflexivity, in which the researcher themselves becomes part of the research, because the lens through which they frame their material becomes integral to the way material is constructed. Brid Featherstone (2000) relates a very nice account of how shifts in her theory enriched her interpretation of interviews. For me this is a major issue at the heart of theorising from practice as a research activity. I have now become convinced, as a result of my experience with the expertise study, that the implicit aspects of practice and the theorising from it, must be included in the focus and process of research.

In this inclusive approach which I am mapping, therefore, any type of theorising involves some type of interrogation of what is already known, since knowledge building involves at least some construction of theory using frameworks which are already known. And in order to communicate our theories, to translate the findings of our research into terms which can be understood by others, we need to use some kind of shared, pre-existing discourse. Theorising from frontline practice as a research activity, therefore, needs to incorporate this dimension.

Who generates the theory

This leads on nicely to the third issue, the question of who are the legitimate researchers in the process of theorising from practice. There has been a lot of argument about whether practitioners can legitimately research their own practice, or whether their role and interests are too divergent to do proper justice to the research act (Padgett, 1998; White, forthcoming). On the other hand, the ability of non-practitioner researchers to sympathetically represent the world of practice from their relatively privileged position is also questionable. Yet again, collaborative researchers argue that processes and structures which use the combined efforts of practitioners and researchers, particularly in group settings, create the most effective climate for the generation of theory from practice. This is particularly the case I think, as Mark Baldwin found (2000), when it is important that practitioners own and use the theory which is to be generated by such a process.

However I wonder whether the question of who (whether practitioner or researcher) conducts the research, is really missing the point? As I argued early in the piece, the type of theorising produced and undertaken will surely depend on the position and perspective of the researcher. Neither practitioner or researcher can be said to have the "purest" view. I also think there is also a question of ability - there are many different ways of theorising practice, involving many different cognitive skills, value positions, personality traits and ways of seeing. Not every individual can simultaneously and equitably undertake all possible types of theorising. Some people have a head for numbers and fine detail. Others revel in drawing the big picture. Some people can almost intuitively connect ideas, others might arrive there through a series of painstaking steps. Some people should not do some types of research, and there are other people who are pefectly suited to particular methodologies.

Surely therefore there are many perspectives and capabilities needed to understand the complexity of the experience which is frontline practice. Again, I take an inclusive view of who the legitimate theorisers from practice should be. I would argue that the domain of theorising from practice does not exclusively belong to either practitioners or researchers. What is at stake are other questions: whether the theory which is developed from practice is both relevant and helpful in illuminating our path, in opening up new courses of action, in minimising ineffective or harmful practice; and in making us a more accountable professional group.

These goals depend on other sorts of questions like:

. does the type of data available,

and, the methods for accessing the practice experience suit

.the processes of theorising,

. the purpose and context of the research and

. the characteristics of the researcher ?

I would argue that there are some types of research that only practitioners can perform on their own practice. There are some theoretical perspectives that only certain researchers will bring to bear in theorising practice experience. There are some theories that will only be developed and used through a collaborative effort between practitioners and researchers. What makes theorising from frontline practice a legitimate research activity is ultimately the social contribution it makes: the accountability and transparency of the theorising method; the communicability of the theory to others; and its ability to transfer meanigs and transform practice.

Bye way of conclusion, I am going to have my cake and eat it too. Whilst I do think, from an inclusive viewpoint, that practice can be theorised in a diversity of ways, I do think there is one type of theorising which must take place (along with others) if the practice of social work is to be accountable, and social work practice knowledge improved. I refer here to the theory or knowledge which is implicit in action, the hidden assumptions enacted in practice. Lincoln and Guba (1985, pp. 195-198) refer to this as "tacit" knowledge, the knowledge which cannot necessarily be expressed in language form, but which must be experienced to be understood. We are all familiar with this type of knowledge, whether we call it "practice wisdom", "life experience" or whatever. Often we do not talk about it because we do not have the frameworks with which to discuss it, or it does not fit with accepted fashionable discourse. Evidence however suggests that this forms a large component of the type of theory which practitioners use (Harrison, 1991; Fook, Ryan & Hawkins, 2000), that which is built up in their own private store, devised, developed and adapted from a variety of sources, most implicit. What also is developed in this cache of knowledge are theories about how to use and apply more explicit knowledge, in changing and different contexts. Our only access to this knowledge may be through the experience of the experiencers' themselves. If we are going to advance social work knowledge in order to make it more accountable and responsive in changing and uncertain contexts, there is an onus on practitioners to theorise their practice in ways which are accessible to themselves and the broader profession.

I think this is the gift of postmodernism to social work - that we value and include the voice of practitioners and their own contribution in theorising from their own practice experience. It is our responsibility to the profession that we enable and create cultures and environments in which this can happen.


Baldwin, M. (2000) Care Management and Community Care, Ashgate, Aldershot.

Clandinin, D.J. & Connelly, F.M. (1994) "Personal Experience Methods" in Denzin, N. & Lincoln, Y. (eds) Handbook of Qualitative Research, Sage, Thousand Oaks, pp. 413-427.

Denzin, N. & Lincoln, Y.S. (1994) Handbook of Qualitative Research, Sage, Newbury Park.

Dreyfus, H. and Dreyfus, S. (1986) Mind Over Machine: The Power of Human Intuition and Expertise in the Era of the Computer, Basil Blackwell, Oxford.

Eraut, M. (1994) Developing Professional Knowledge and Competence, Falmer, London.

Everrit, A., Hardiker, P., Littlewood, J. & Mullender, A. (1992) Applied Research for Better Practice, Macmillan, London.

Featherstone, B. (2000) "researching into Womens' Violence: Some Thoughts on the Process" in Fawcett, B., Featherstone, B., Fook, J. & Rossier, A. (eds) Practice and Research in Social Work: Postmodern Feminist Perspectives, Routledge, London, pp.120-135.

Fook, J. (1993) Radical Casework; A Theory of Practice, Allen & Unwin, Sydney.

Fook, J. (1996) (ed) The Reflective Researcher: Social Workers' Theories of Practice Research, Allen & Unwin, Sydney.

Fook, J. (1999) "Reflexivity as Method" in Daly, J., Kellehear, A. & Willis, E. (eds) Annual review of Health Social Sciences, vol. 9, pp. 11-20.

Fook, J. (2000) "Critical Perspectives in Social Work Practice" in O'Connor, I., Smyth, P. & Warburton, J. (eds) Contemporary Perspectives on Social Work and the Human Services: Challenges and Changes, Addison Wesley Longman, Melbourne, pp. 128-137.

Fook, J. (forthcoming) "Qualitative Practitioner Research" in Shaw, I & Gould, N. (eds) Qualitative Research in Social Work, Sage, London.

Fook, J. & Briskman, L. (forthcoming) "Australian Social Work at the Crossroads" in Shardlow, S. & Payne, M. (eds) Contemporary Social Work: A Global View, Jessica Kinsgley, London.

Fook, Munford & Sanders (1999) "Interviewing and Evaluating" in Shaw, I. & Lishman, J. (eds) Evaluation and Social Work Practice, Sage, London, pp. 164-181.

Fook, J., Ryan, M. & Hawkins, L. (1997) "Towards a Theory of Social Work Expertise", British Journal of Social Work Vol. 27, pp. 399-417.

Fook, J., Ryan, M. & Hawkins, L. (2000) Professional Expertise: Practice, theory and Education for Working in Uncertainty, Whiting & Birch, London.

Fraser, M.W. (1995) "Rich, Relevant and Rigorous: Do Qualitative methods Measure Up?" Social Work Research, Vol. 19, No. 1, pp. 25-27.

Friedson, E. (1970) The Profession of Medicine, New York, Dodd Mead.

Gilgun, J. (1994) "Hand in Glove: The Grounded Theory Approach and Social Work Practice Research" in Sherman. E. & Reid, W.J. (eds) Qualitative Research in Social Work, Columbia University Press, New York, pp. 115-125.

Harrison, D. (1991) Seeking Common Ground; A Theory of Social Work in Social Care, Avebury, Aldershot.

Heron, J. (1996) Co-operative Inquiry: Research into the Human Condition, Sage, London.

Kellehear, A. (1993) The Unobtrusive Researcher: A Guide to Methods, Allen & Unwin, Sydney.

Leonard, P. (1997) Postmodern Welfare, Sage, London.

Lincoln, Y.S. & Guba, E.G. (1985) Naturalistic Inquiry, Sage, Newbury Park.

Padgett, D. (1998) Qualitative Methods in Social Work Research, Sage, Thousand Oaks.

Payne, M. (1997) Modern Social Work Theory, Macmillan, London.

Schon, D. (1983) The Reflective Practitioner, Basic Books, New York.

Sheppard, M. (1995) (Social Work, Social Science and Practice Wisdom", British Journal of Social Work, vol. 25, pp. 265-293.

Steier, F. (1991) Reflexivity and Research, Sage, London.

Strauss, A. & Corbin, J. (1995) "Grounded Theory methodology: An Overview", in Denzin, N. & Lincoln, Y. (eds) Handbook of Qualitative Research, Sage, Thousand Oaks, pp. 273-285.

White, S. (forthcoming)"(Auto)Ethnography as Reflexive Inquiry: The Research Act as Self-Suveillance" in Shaw, I. & Gould, N. (eds) Qualitative Research in Social Work, Sage, London.