Theorizing in Qualitative Research (Part I)

By Huayi Huang

In talking with normal people, it is common to discuss ‘theory’ and see their eyes glaze over and their attention wander. In healthcare improvement practice and research, for example, the role of theory continues to be seriously under recognized . Whilst it sometimes benefits the academic agenda to mystify and obfuscate the idea of theory, the myth that some of us do not theorise is underpinned by the naïve realist idea that reality can be directly observed (without theory).

But can reality be observed directly…when you really think about it?

Reflecting on our experiences as qualitative researchers and fieldworkers, we observe usually using natural languages we know such as English. Such languages have evolved largely without premeditated planning of their lexicon and key ideas. In contrast to mathematical languages and ideas we learnt from school – from a branch of knowledge whose key lexicon and ideas largely predefined for us as impressionable school children, and often accepted as ‘true’ without serious questioning or scrutiny.

Reflecting on our common experience as learners in a changing world though – the urge to look for explanations, understandings and causes from our personal and professional experiences is almost irresistible – to the extent that American social historian Charles Tilly argues for us to be defined as ‘reason-giving animals’.

As Ravitch and Riggan (2011) point out in explaining how conceptual frameworks guide research, nearly everyone can agree on the idea of theory as ‘trying to give reason’ to things working the way they do. In another words, theorising aiming to help explain why – to the observations we rely on. The natural languages and ideas used in qualitative research continue to evolve largely without premeditated planning, so it makes sense for this evolutionary rather than premeditated development of ideas to carry through – to the explanations based on these observations.

To understand their idea of ‘theory’ more deeply, Ravitch and Riggan (ibid.) also offer the thought: of thinking about theory as moving us towards answers to our whys, by way of identifying and examining relationships among things. In one of my current academic projects on international and national primary care transformation, for example, this amounts to identifying and examining transformation ‘actors’ key to that which is being theorised, and key relationships between these key actors.

We’re all well used to identifying and examining key relationships among things and people we encounter. In our professional research lives – the general project is around ‘theorising’ on behalf of our research clients – to enable them to identify and (re)examine the key relationships, things, and people they have in mind.

At a glance, theories from the academic literature may be distinguished as Grand, Middle-range, and Small. ‘Small’ academic theories focus on identifying and examining relationships among those things and people relevant to a specific intervention project. For example, in the thinking a teacher does around preparing to affect change, in existing key relationships, things and people involved in introducing a new classroom technology into their class.

An example of middle-range theorising is in the current sociological theorising of Normalisation Process Theory (which I’m continuing to explore currently). It is a theory currently gaining traction in health care and services research and practice. This theory was intended to aid description and explanation, for thinking about how innovative practices can become routinely embedded in social contexts in healthcare. In a nutshell, Normalisation Process Theory helps us to think sociologically, in identifying and examining key relationships, things and people involved in embedding innovative practices across a broad area of work and society.

Finally, so called ‘grand’ theories are the academic equivalent of the most encompassing ideas we find ourselves returning to – across the full range of personal and professional experiences we encounter in our lives. In Merton’s words, the ‘master conceptual scheme’ by which one tries to identify key relationships, things and people involved in the local fabric of our social structures. In supplying a general lexicon and set of key ideas for thinking, talking, and sharing with each other about our worlds – ‘grand’ theorising seeks to maximise the breadth of the applicability of your most encompassing current ideas, at the expense of these ideas not necessarily being well adapted for each and every domain of personal and professional experience you try to apply them to. In context of statistical theory in quantitative research, for example – neglecting research data on how changes and outcomes at a national level occur, in favour of measureable indicators of what is happening, where it is happening, and to whom.

Will you be a naïve realist?

To further demystify and erode the naïve realist idea that reality can be directly observed (without theory), let us dip into the fact that multiple distinct theories of truth exist, and continue their life in current academic work.

According to Schwandt’s dictionary of qualitative inquiry, ‘Truth’ in qualitative research is one of the most difficult of all philosophical topics, intimately related to questions of meaning (to a knower). Schwandt describes the following theories of truth – on the premise that ‘truth’ might be a property of particular statements, propositions, beliefs, and assertions shared about our realities.

Then, how is truth established and defined under these alternative working interpretations?

1) A correspondence theory of truth proposes that statements, propositions, beliefs, and assertions become established as true if they accurately represent or match some state of affairs in the world. Statements and propositions, for example, become false if the knower sees inaccuracies in the correspondence between the proposed truth and the relevant parts of their world. Under this premise for Grand, Middle-range, or Small theorising, we move towards answers to our whys, by seeking to identify and (re)examining correspondence relationships, between particular statements and propositions, and the experiences of ourselves or our research clients. In another words, the picture painted in the knower’s mind by the statements we offer as researchers must broadly correspond to the reality experienced by the knower so far – for these statements, propositions, beliefs, and assertions to become true in the reader’s mind.

2) A consensus theory of truth brings in the idea of ‘ideal communications situations free from the distortions of everyday communication’. This ideal is popularly expressed indirectly in some academic circles, through talking about its opposite of ‘bias’ – in the everyday reality of professional quantitative health studies and research for example.

Delphi approaches to research can also be seen as a popular expression and implementation of a consensus theory of truth, where specific procedural checks are in place to create a communications context approximating the ideal of ‘freedom from the distortions/biases in everyday communication’. These communications are in this case usually between expert panels, convened for a research topic with considerable disagreement/uncertainties within the expert community. The consensus on key relationships, things, and people which then emerges, approximates a collective sort of truth: through enabling processes of consensus-making and construction of increased certainties in the collective statements, propositions, beliefs, and assertions eventually issued by the panel.

3) For knowers operating under a coherence theory of truth, a premium is placed on coherence in truth-judgements in a qualitative research context – where particular statements etc are regarded as ‘true’ if they cohere with the prior truths brought in by a researcher into a new project, or perhaps coheres well with the prior truths key to the research client’s understanding of their reality.

Quantitatively, this premise on coherence as proxy for ‘truth’ can be seen in research papers demonstrating ‘Construct Validity’; through demonstrating that the operationalisation of an idea/construct being measured, functions “in predictable ways in relation to other operationalizations based upon your theory of the construct”. This sort of theorising of an idea or construct is trying to get at the readers’ belief that the offered operationalisation of an idea is actually measuring what it is supposedly measuring. The belief in this statement is substantiated here through demonstrating coherent quantitative relationships between distinctly different, yet mutually coherent quantitative operationalisations of the same idea.

Stepping back to think coherently across both qualitative and quantitative research, qualitative projects tend to occur in a context where team members tend to come at things in context of substantively differing backgrounds of ‘settled’ and justified personal beliefs. Methods to enable a working collective agreement on the background of ‘settled’ and justified collective beliefs for the intellectual foundation for the project are therefore really important – to ensure that research team members, or their clients are then learning new things based on more or less the same background of ‘settled’ collective beliefs and foundations. Quantitative researchers have things much easier than qualitative researchers on this point, since the predominantly mathematically-inclined state of many current research and knowledge cultures tend to predefine for their members a broadly ‘more settled’ foundation of collective beliefs – on which to further develop statements, propositions, beliefs, and assertions founded on ideas of quantity.

4) A contextualist theory of truth is closely allied with the social constructionist focus on the (collective) constructions from our interpersonal interactions with particular knowledge communities or cultures (eg. the idea of ‘truth’ here). Here, theorising that ‘truth’ is more of a social construction than you might think (in conflict with the objectivist, impartial vision on this idea). People who try to know things this way are naturally drawn to the relative rather than universal face of ideas around truth. The need to see truth as relative to particular ways of existing understanding becomes pronounced, when encountering an area of academic literature where competition and divergence in key ideas is high – with little meaningful agreement and harmony in its conceptual foundations. In everyday contexts, the relative nature of truth and truth statements, propositions, beliefs, and assertions can be seen in the real phenomenon of ‘fake news’ we’ve recently seen, and our worries over extreme actions based on the idea of truth as a relative construct and idea.

To see the truth…in the contextualist theory or social constructionist vision of truth however, just reflect on the socio-historical (and not universal) nature of your sources of knowledge? For example in thinking about the key ideas you took away with you from early schooling in this particular school in Japan, rather than this other particular UK school say. Or in the parental competition to get your kids into Oxbridge for example. For contextual theorists (of truth), the key relationships for understanding ‘truth’ reside in observable acts of ‘truth telling’ – and questions and answers around why and how we come to believe the truths we do, through the inevitably contextually-situated key relationships, things and people involved in our ongoing personal and professional education.

5) Pragmatic theories of truth are underpinned by an idea familiar to many of us in coping with our busy lives: The truth of particular assertions we read in essence are judged through their ‘doing’ or acceptance attempt. For example, if I read that eating 5-7 fruits a day is supposedly good for my health, do I then (as a would be believer) in fact find this assertion to ‘hold’ in my attempts to live like this? In another words, I now try eating 5-7 fruits a day instead of 2-3 say. If this has no perceptible positive effects on my life as I try this ‘new fact’ out in practice (ie. if I don’t then see concretely 5-7 fruits being actually good for my health), then as a pragmatic theorist I should dismiss the truth of eating 5-7 fruits a day being good for my health (at least for me), even if national health guidance in my country continues to advocate this fact as true.

A more conceptual example would be in the various high level ideas we turn to in organising our lives. For example, does the key idea from the latest blog on personal finance I follow really enable me to more effectively manage my personal finances? If it seems to help, then I come to see the key personal finance idea I’ve just tried to apply as true!

In essence, pragmatic theorising is focussed on the key relationship between:

1) ‘incoming’ statements, propositions, beliefs, and assertions new to a knower, and

2) personal or collective judgements made on the basis of ‘trying these assertions out’, in attempting to apply them to help us understand, or accomplish meaningful work in our personal and professional realities.

The idea of truth then becomes a sort of dynamic expression and label, as a way of relating to particular incoming statements from others about our world. In another words functioning as a label for a dynamically emerging relationship of ‘being true’ and ‘being false’: between particular statements, propositions, etc and our own reality. This line of thinking harmonises with relativist visions of truth (e.g. a contextualist theory of truth), where the idea of truth as an invariant property attached to each and every ‘objective’ statement, proposition, belief, and assertion, is an idea strongly rejected by relativist knowers.

Based on what we’ve covered so far then, will you continue to choose to reproduce the myth that good researchers do not theorise, and directly observe without theory?

Articles and blog posts