STC Nº 0050-2004 AI/TC FJ 72 47 Idem cit F J 71.

set have been introduced and issues which continue to be of interest in relation to the concept of organisational or collective learning have been identified. Given the apparently diverse nature of the empirical setting, the concept of context has been chosen as the issue with which this thesis is concerned. In the last chapter, various conceptualisations of the concept of context were considered and cultural historical activity theory was identified as a potential analytical framework for thinking about this concept in relation to collective learning in primary care teams.

The purpose of this chapter is to set out the methodological basis upon which this study of contextual collective learning in primary care has been conducted. First, there will be a brief discussion of the various paradigms and perspectives within the social sciences so that this study may be located within the overall field. This sets the scene for a discussion of the abstract nature of the phenomena of interest in this research. Some methodological challenges concerning ontological and epistemological issues, presented by the empirical setting of primary care in relation to these abstractions, will be identified prior to providing an explication of the subsequent research process. Following identification of the research questions addressed by this thesis, an account will be given of the research strategy through which the project was enacted. An explanation of the sampling framework for this instrumental study of multiple cases will be followed by the introduction of the primary care teams studied. Finally, the analytical process through which findings of this instrumental study of multiple cases were reached will be explained.

Social science research paradigms

The social sciences are characterised by a variety of paradigms and perspectives which stretch across academic disciplines and the full range of social science areas of study and topics of interest. A paradigm, or interpretive framework, has been defined as “a basic set of beliefs that guide action” which support “the researcher’s epistemological, ontological and methodological premises” (Denzin and Lincoln, 2005). Paradigms are therefore

axiological, incorporating the ethical and moral frameworks which support the researcher’s world view and the ways in which s/he thinks that world may be known. Perspectives are somewhat different from paradigms, in that they are:

…not as solidified, nor as well unified….although a perspective may share many elements with a paradigm – for example, a common set of methodological assumptions or a particular epistemology.

(Denzin and Lincoln, 2005 p.183)

This presents a series of challenges for social science researchers in terms of making methodological choices and decisions about their projects. which are not as evident in the natural sciences where debates tend to focus on the methods used, but rarely debate the ontological and epistemological aspects of research (Sechrest and Sidani, 1995). In the social sciences these issues are more complex than the somewhat simplistic, if commonly held, dichotomous quantitative/qualitative view would suggest (Guba and Lincoln, 2005). This relatively common view is neither entirely relevant to, nor accurately descriptive of, this field of academic endeavour. It is not entirely relevant because its emphasis tends to be erroneously and narrowly focused on research techniques or methods; it is not accurately descriptive, because both the quantitative or qualitative approaches may be evident within the same social science discipline, and even within a single research project (Easterby- Smith et al., 2002).

Paradigms and the perspectives they encompass may be complementary or competitive (Guba & Lincoln, 2005). They may be conceptualised as a continuum ranging from “conventional” to “alternative” (Guba and Lincoln, 1994). Those situated at either end of the continuum (positions 1 and 2 respectively) are generally held to be mutually exclusive and competitive, whilst those situated along the continuum may be complementary to a greater or lesser extent. Most social science research projects fall towards the “alternative” end of the spectrum within those paradigms regarded as interpretivist (O'Reilly, 2005) rather than positivist. However, despite claims of their demise (Wilson and Natale, 2001), positivist and post-positivist paradigms do still feature in some social science studies as identified in Chapters 3 and 4 (Hammersley, 1995). A summary of paradigms is set out in table 1.

Table 1: Paradigm positions (adapted from Guba & Lincoln, 2005 p.196)

(NB: shading represents perspectives compatible with researcher’s view, see p84) Issue Positivism Post-

positivism Critical theories Constructivism Participatory Nature of knowledge Verified hypotheses established as facts or laws Nonfalsified hypotheses, probable facts or laws Structural/ historical insights Individual and collective reconstructions sometimes coalescing around consensus Extended epistemology: primacy of practical knowing; critical subjectivity; living knowledge Knowledge accumulation Accretion – adding to “edifice of knowledge”; generalisations & cause/effect linkages As for positivism Historical revisionism; generalisation by similarity

More informed & sophisticated reconstructions; vicarious experience In communities of inquiry embedded in communities of practice Goodness or quality criteria Conventional benchmarks of “rigor”: internal & external validity, reliability, objectivity As for positivism Historical situatedness; erosion of ignorance and misapprehensions; action stimulus Trustworthiness and authenticity including catalyst for action Congruence of experiential, presentational, propositional, practical knowing; leads to action to transform the world in the service of human flourishing Values Excluded – influence denied Excluded – influence denied Included – formative Included - formative Included - formative

Ethics Extrinsic Extrinsic Intrinsic Intrinsic Intrinsic

Inquirer posture “Disinterested scientist” as informer of decision/ policy makers and change agents As for positivism “Transformative intellectual” as advocate and activist “Passionate participant” as facilitator of multivoice reconstruction Primary voice manifest through aware self-reflexive action; secondary voices in illuminating theory, narrative, other presentational forms

Training Technical and quantitative; substantive theories Technical; quantitative & qualitative; substantive theories Resocialisation; qualitative and quantitative; history; values of altruism, empowerment & liberation As for critical theories Co-researchers are initiated into the inquiry process by facilitator/researcher and learn through active engagement in the process; facilitator/researcher requires emotional competence, democratic personality and skills

Each paradigm has its relative merits and difficulties. These may only become evident when considered in relation to the questions to which the research project is addressed.

These philosophical issues may seem somewhat esoteric initially. However, their importance lies in the nature of social science itself, in that it concerns the study of human lives (O’Reilly 2005) and related concepts which are frequently abstract rather than material (Bateson, 1941). It is this aspect of social science which requires careful consideration be given to methodological issues in all research projects.

Indeed, Mason (1996) suggests researchers ask themselves five “difficult questions” concerning the ethical/moral, ontological, epistemological and methodological issues they face, regardless of the (qualitative or/and quantitative) methods they eventually adopt, to meet this requirement:

1. What is the nature of the phenomena, or entities, or social ‘reality’, which I wish to investigate?

(Ontological issues)

2. What might represent knowledge or evidence of the entities or social ‘reality’ which I wish to investigate?

(Epistemological issues, related to1.)

3. What topic, or broad substantive area, is the research concerned with? (Methodological issues, related to 1 & 2.)

4. What is the intellectual puzzle? What do I wish to explain? What are my research questions?

(Connects what you wish to research with how you are going to research it, consistent with and related to 1, 2 & 3.)

5. What is the purpose of my research? What am I doing it for?

(Axiological issue and academic contribution, relates to paradigm positioning and encapsulates 1, 2, 3 & 4.)

These questions enable the researcher to think about the paradigm and perspective which best suits their particular research project. This facilitates the construction of a coherent research design which retains a necessary degree of flexibility (although not in the positivist or post-positivist paradigms where different criteria apply), to allow alterations to be made in light of what is learned when conducting the research whilst ensuring it remains focused and the questions are addressed. This is particularly important when considering the frequently abstract nature of the topics of interest to social scientists which will be discussed in the next section.

Researching abstract concepts

The central issue for any research project is obviously to address the research question(s) posed. However, it is important to consider the preceding process which culminates at that important point. Each step in that research process involves choices and decisions relating to the issues encapsulated by Mason’s (1996) five questions as outlined above. In the social sciences, the process is closely related to the researcher’s favoured paradigm, and within that, his or her preferred perspective on the phenomena of interest. These preferences are not whimsical but are rigourous, informed choices, which display the skills necessary to conduct research. They also relate to a key quality indicator in research projects: how appropriate is the design for studying the phenomena under scrutiny, and does the methodology (and methods used) fits the question(s) asked (Light Jr, 1979).

Research conducted within the positivist or post-positivist paradigm is based on deductive logic and seeks to test a priori propositions or hypotheses to either confirm or disprove them. This type of research is potentially useful when the topic relates to material, measurable objects and the questions relate to questions such as what those things are, how frequently they are encountered, or how two or more of them relate to one another (or do not). But social scientists are often interested in abstract concepts related to human interactions and their questions focus on why certain processes occur (or do not) and how they happen (Mason, 1996). This means that the field of research is not static and frequently the focus is not on material objects. Therefore, the positivist and post-positivist paradigms are used less often in the social sciences. When they are used, researchers need

to ensure they relate openly and appropriately to the phenomena of interest and enable the research questions to be addressed.

Topics of interest to social scientists may often be better served by the application of inductive logic (Mason, 1996). This allows the researcher to use data from the “real world” s/he is studying to inform the course of the research, enable new insights to be gleaned, new themes to emerge, and to form the basis for analysis and findings. Although there are methodologies which seek to be entirely inductive, for example some purist forms of grounded theory (Glaser and Strauss, 1967) where everything begins with and stems from the data generated, this is not usually the case in practice. Most people do not conduct their research knowing nothing about their topic of interest. In most instances they draw upon various informative sources prior to commencing data gathering, including their own prior experience or “pre-understanding” (Gummesson, 2000), “fore-shadowed problems” (Malinowski, 1972) gleaned from the findings of previous research about their topic, and by familiarising themselves with potentially relevant areas of theory. The process involves moving back and forth, both mentally and physically, in an iterative and inductive fashion: mentally, they move between what is already understood and what is emerging through the process of conducting the research which creates new insights; physically, they move between their field of research and their own research base. This iterative-inductive process (O’Reilly, 2005) allows the researcher to identify and address their research questions by enacting the flexible but focused and coherent research strategy devised at the outset when addressing the “five difficult questions” (Mason, 1996) from within their preferred paradigm and perspective.

However, there are circumstances in which paradigms may be inappropriately employed. In his 1941 paper, Gregory Bateson eloquently describes how this may arise, offering an illuminating account of the dangers which lie in wait for the unwary researcher who regards abstract concepts, so often of interest in social science, with a measure of “false concreteness” (Bateson, 1941 p.62). In this case, the abstract concept in question was “ethos” in relation to the culture of a group of people in New Guinea whom Bateson was observing. It is worth quoting at some length as it encapsulates the point well:

I was especially interested in studying what I called the “feel” of culture…It appeared to me that it was impossible to understand the behaviour…apart from the “feel” of their culture, and from this it seemed to follow that the “feel” of the culture was in some way causative in shaping native behaviour… But it also guided me into regarding the “feel” of the culture as much more concrete and causally active than I had any right to do. This false concreteness was reinforced later by an accident of language… I was referring to an abstraction [but the word “ethos”] helped me to go on thinking that it referred to a unit something which I could still regard as causative… as if it were a category of behaviour or a sort of factor which shaped behaviour… I multiplied my offences by creating several more concepts of about the same degree of abstraction… and all these I handled as though they were concrete entities. I pictured the relations between ethos and cultural structure as being like the relation between a river and its banks – “The river moulds the banks and the banks guide the river. Similarly, the ethos moulds the cultural structure and is guided by it.”…I was looking now for physical analogies which I could use in analysing my own concepts… But when one is seeking an elucidation of one’s own concepts, then one must look for analogies on an equally abstract level… But these similes about ethos being the river… were, as I thought, the real thing… I thought that there was one sort of phenomenon which I could call ethos and another sort which I could call “structural culture” and that these two worked together – had mutual effect one on the other. All that remained for me to do was to discriminate clearly between these various sorts of phenomena so that other people could perform the same sort of analysis that I was doing.

(Bateson, 1941 pp.62-63, italics in original)

This account demonstrates how easy it could be to fall accidentally into the trap of choosing an inappropriate paradigm and research methodology in social science research because abstract concepts have been inappropriately categorised as things which exist independently in the world. However, sometimes this happens less by accident and more because of what might be termed ‘cultural expectations’: at a time in which the term “evidence-based” applies not only to scientific issues, but also to other areas of life concerning (and through) political and social policies and policy-making, there remains a

continuing trend within academia and society more generally to tacitly endorse some research paradigms - notably the “scientific”, empiricist approach exemplified by positivism and post-positivism - above others (Dickson, 1988; Denzin and Lincoln, 2005). At least part of the explanation for this may be the perception that such research paradigms have the capacity to discover “the truth”, defined as:

…the accurate representation of an independently existing reality.

(Smith and Hodkinson, 2005 p.917)

Although the veracity of such a claim for science itself may well be problematic, it is not raised here to provoke such a discussion. Rather, it is mentioned for the purposes of this discussion because it offers a possible insight into the allure such paradigms and perspectives hold for some social scientists who may, inappropriately, adopt them when researching abstract concepts relating to human interaction and behaviours. In a culture dominated by the Cartesian split (Tolman, 1999; Oxford Dictionary of Philosophy, 2005) between the subjective and objective realms (exemplified by views of dichotomous relations between mind/body, subject/object), where knowledge relates to objective entities, and where the social science researcher must still persuade the reader about the validity of his or her findings, the appeal of such truth claims is perhaps understandable. But the temptation to indulge in verisimilitude (O’Reilly, 2005) by making such claims based on “a fallacious bit of thinking” (Bateson, 1941 p.62), through which abstractions are treated as objective entities, should be resisted. Apart from misleading readers who might take the findings literally, this may not only cause confusion, but ultimately discredit the research. It is interesting to note that Descartes’ notion of separating the subjective from the objective, originally intended to free thinking from the metaphysical problems of the day (Oxford Dictionary of Philosophy, 2005), has recreated the same problem in a different guise by trying to “explain mental processes by principles analogous to those which govern the physical interaction of material objects” (Bakhurst, 1997 p.157), thereby encouraging methodological inconsistency and rendering its process opaque.

A key methodological issue in the social sciences concerns how abstractions may be intellectually apprehended based on such conceptualisations, and how these may then be studied in the complexity of social interactions and dynamic processes in the social and

material world. Therefore, given the possible paradigm variation in social science transparency is required about the conceptualisation and interpretation of abstract notions in research projects, in relation to the research questions asked. This forms an important aspect of quality criteria in the social sciences where rigour involves demonstrating the reliability and accuracy of the methods used, and the validity of data acquired by those means, as part of a coherent methodology. It is upon these criteria that the quality of the conclusions from research may be assessed (Mason, 1996).

The remainder of this chapter addresses these issues in relation to this project. Based upon the discussion in the preceding chapters, the thinking behind the methodological choices made will be explicated and their philosophical “underlabourer” aspects explained (O’Reilly, 2005). The research questions will then be introduced and the research strategy through which they have been addressed, based upon these methodological choices, will be described.

Thesis methodology

This research project concerns the phenomena of organisational and collective learning and context in relation to Scottish primary healthcare teams. The preceding three chapters have shown that these phenomena are not unproblematic and that each may be viewed as an abstract concept. As discussed, there are diverse views about the nature of these contested phenomena and therefore about how their existence in the social world may be characterised and identified. There are examples of most, if not all, of the research paradigms summarised in table 1 within the existing literature, which reflects diversity of axiological, ontological and epistemological views about these phenomena. These philosophical issues, performing in their informative “underlabourer” role, rather than the foundational and deterministic role sometimes unquestioningly claimed for them within the positivist/post-positivist paradigms (Hammersley, 1995; O'Reilly, 2005), are therefore crucially important when making methodological choices to address research questions concerning such concepts. In the next section the choices made in relation to the research under discussion here will be explicated, informed by the sources of prior knowledge outlined in the Foreword to this thesis.

Ontological issues - The nature of the phenomena of interest

In setting out to establish “organisational learning” and “context” as abstractions, I have given an indication of my ontological position about the nature of things in the social world.

I believe that the social world is a product of interactions between people: me and others,